Datasets:

michaelsyao

/

acrac

like 0

acrac_3102253_4

Cerebrovascular Disease Child PCAs

CT Head CT is less sensitive than MRI for demonstrating acute infarctions but can be useful in evaluating rare instances of intracranial hemorrhage in moyamoya [43]. CTA Head CTA, including multiphase CTA, is an alternative to MRA and catheter angiography in the initial diagnosis and follow-up of children with moyamoya. MRI Head Proposed screening guidelines for moyamoya include diagnostic imaging using MRI/MRA in at-risk populations, particularly if there are symptoms of cerebral ischemia. Screening of first-degree relatives of patients with moyamoya is probably not warranted in the absence of symptoms or unless multiple family members are already known to be affected. T2-weighted fluid-attenuated inversion-recovery imaging may show high signal in the sulci (ivy sign), indicative of slow flow in affected vascular territories. A similar appearance may be evident on T1- weighted contrast-enhanced studies. SWI may be used to demonstrate microhemorrhage that has been reported in up to 52% of patients [44]. Arteriopathy can also be further evaluated with the technique of vessel-wall imaging [15,45]. MRA Head MRA is an alternative to CTA in the initial and follow-up imaging evaluation of children with moyamoya and is the preferred noninvasive vascular imaging modality. MRA is typically performed together with brain MRI to evaluate both the cerebral vasculature and the brain parenchyma. CT Head Perfusion The use of perfusion CT, including xenon-enhanced CT, in children with moyamoya is feasible but requires repetitive imaging of the brain [28]. MR Head Perfusion MR cerebral blood flow evaluation includes perfusion-weighted MR, either using arterial spin-labeling techniques without IV contrast or dynamic susceptibility contrast techniques with IV contrast. Perfusion imaging is also used to assess functional improvement after treatment [39,46-49].

Cerebrovascular Disease Child PCAs. CT Head CT is less sensitive than MRI for demonstrating acute infarctions but can be useful in evaluating rare instances of intracranial hemorrhage in moyamoya [43]. CTA Head CTA, including multiphase CTA, is an alternative to MRA and catheter angiography in the initial diagnosis and follow-up of children with moyamoya. MRI Head Proposed screening guidelines for moyamoya include diagnostic imaging using MRI/MRA in at-risk populations, particularly if there are symptoms of cerebral ischemia. Screening of first-degree relatives of patients with moyamoya is probably not warranted in the absence of symptoms or unless multiple family members are already known to be affected. T2-weighted fluid-attenuated inversion-recovery imaging may show high signal in the sulci (ivy sign), indicative of slow flow in affected vascular territories. A similar appearance may be evident on T1- weighted contrast-enhanced studies. SWI may be used to demonstrate microhemorrhage that has been reported in up to 52% of patients [44]. Arteriopathy can also be further evaluated with the technique of vessel-wall imaging [15,45]. MRA Head MRA is an alternative to CTA in the initial and follow-up imaging evaluation of children with moyamoya and is the preferred noninvasive vascular imaging modality. MRA is typically performed together with brain MRI to evaluate both the cerebral vasculature and the brain parenchyma. CT Head Perfusion The use of perfusion CT, including xenon-enhanced CT, in children with moyamoya is feasible but requires repetitive imaging of the brain [28]. MR Head Perfusion MR cerebral blood flow evaluation includes perfusion-weighted MR, either using arterial spin-labeling techniques without IV contrast or dynamic susceptibility contrast techniques with IV contrast. Perfusion imaging is also used to assess functional improvement after treatment [39,46-49].

3102253

acrac_3102253_5

Cerebrovascular Disease Child PCAs

HMPAO SPECT or SPECT/CT Brain Single-photon-emission computed tomography (SPECT) studies can be used to assess perfusion in patients with moyamoya and can be augmented with acetazolamide challenge to assess perfusion instability and vascular reserve. Perfusion imaging is also used to assess functional improvement after treatment [39]. Arteriography Cervicocerebral Catheter angiography is required for definitive diagnosis of moyamoya and is typically obtained as part of a preoperative assessment prior to surgical revascularization or following revascularization to assess the development of surgically created collaterals. US Duplex Doppler Transcranial Head US does not typically play a role in the management of children with moyamoya. Variant 4: Child. Known or suspected cervicocranial arterial dissection based on clinical or imaging findings. Next imaging study. Cervicocranial artery dissection occurs in 2.5/100,000 children per year and is up to 4 times more common in children than in adults [50-52]. Purely intracranial dissections are also more common in children than adults [50]. Patients with cervical artery dissection can present with headache, neck pain, or ischemic infarction due to emboli, whereas patients with intracranial artery dissection can present with ischemic infarction, subarachnoid hemorrhage (SAH), or symptoms due to local mass effect [50,52]. A history of antecedent trauma may or may not be present. CT Head CT can show areas of infarction or SAH although the sensitivity of CT is less than that of MRI with DWI for demonstrating acute infarction [43]. A crescent-shaped hyperattenuating area in a cranial or cervical artery is consistent with an intramural hematoma and is highly suggestive of dissection [52]. CTA Head and Neck CTA can demonstrate vessel narrowing, pseudoaneurysm, or an intimal flap in dissection [50,52]. MRI Head and Neck MRI is the preferred study for the demonstration of infarction due to cervicocranial arterial dissection [50,52,53].

Cerebrovascular Disease Child PCAs. HMPAO SPECT or SPECT/CT Brain Single-photon-emission computed tomography (SPECT) studies can be used to assess perfusion in patients with moyamoya and can be augmented with acetazolamide challenge to assess perfusion instability and vascular reserve. Perfusion imaging is also used to assess functional improvement after treatment [39]. Arteriography Cervicocerebral Catheter angiography is required for definitive diagnosis of moyamoya and is typically obtained as part of a preoperative assessment prior to surgical revascularization or following revascularization to assess the development of surgically created collaterals. US Duplex Doppler Transcranial Head US does not typically play a role in the management of children with moyamoya. Variant 4: Child. Known or suspected cervicocranial arterial dissection based on clinical or imaging findings. Next imaging study. Cervicocranial artery dissection occurs in 2.5/100,000 children per year and is up to 4 times more common in children than in adults [50-52]. Purely intracranial dissections are also more common in children than adults [50]. Patients with cervical artery dissection can present with headache, neck pain, or ischemic infarction due to emboli, whereas patients with intracranial artery dissection can present with ischemic infarction, subarachnoid hemorrhage (SAH), or symptoms due to local mass effect [50,52]. A history of antecedent trauma may or may not be present. CT Head CT can show areas of infarction or SAH although the sensitivity of CT is less than that of MRI with DWI for demonstrating acute infarction [43]. A crescent-shaped hyperattenuating area in a cranial or cervical artery is consistent with an intramural hematoma and is highly suggestive of dissection [52]. CTA Head and Neck CTA can demonstrate vessel narrowing, pseudoaneurysm, or an intimal flap in dissection [50,52]. MRI Head and Neck MRI is the preferred study for the demonstration of infarction due to cervicocranial arterial dissection [50,52,53].

3102253

acrac_3102253_6

Cerebrovascular Disease Child PCAs

DWI and perfusion-weighted MRI provide information on regions of infarction. SWI can demonstrate emboli, and increased oxygen extraction leads to prominence of the veins in infarcted regions. T1-weighted imaging with fat suppression or T2-weighted imaging may demonstrate an intramural hematoma in 76% to 91% of patients with dissection [50,52]. MRA Head and Neck MRA without or with IV contrast may be used to assess for cervicocranial artery dissection to demonstrate an intimal flap, vessel narrowing, or pseudoaneurysm formation [52,53]. Arteriography Cervicocerebral Catheter angiography remains the gold standard for the radiologic evaluation of dissection but is typically reserved for patients in whom dissection is suspected but not confirmed on noninvasive imaging [52,53]. Angiography may reveal vessel narrowing, pseudoaneurysm formation, or an intimal flap in regions of dissection and distal vessel occlusions due to emboli. US Duplex Doppler Transcranial and Carotid Artery Doppler US may be used to evaluate the cervical internal carotid arteries for dissection. The technique is of limited utility for carotid dissections at or above the skull base and in the evaluation of the vertebral arteries because of a lack of a good acoustic window [52]. Variant 5: Child. Clinical presentation suggestive of acute stroke, known or suspected central nervous system vasculitis. Initial imaging. Childhood vasculitis is as an inflammatory process only affecting the intracranial vessels and can result in stroke. It can present as childhood primary angiitis of the central nervous system or as a secondary phenomenon in systemic rheumatologic infections and neoplastic conditions [54]. Childhood primary angiitis of the central nervous system is subcategorized into large-medium vessel vasculitis (angiography positive) or small vessel (vessel abnormality too small to be demonstrated on MRA, CTA, or conventional angiography) best diagnosed by brain biopsy [55].

Cerebrovascular Disease Child PCAs. DWI and perfusion-weighted MRI provide information on regions of infarction. SWI can demonstrate emboli, and increased oxygen extraction leads to prominence of the veins in infarcted regions. T1-weighted imaging with fat suppression or T2-weighted imaging may demonstrate an intramural hematoma in 76% to 91% of patients with dissection [50,52]. MRA Head and Neck MRA without or with IV contrast may be used to assess for cervicocranial artery dissection to demonstrate an intimal flap, vessel narrowing, or pseudoaneurysm formation [52,53]. Arteriography Cervicocerebral Catheter angiography remains the gold standard for the radiologic evaluation of dissection but is typically reserved for patients in whom dissection is suspected but not confirmed on noninvasive imaging [52,53]. Angiography may reveal vessel narrowing, pseudoaneurysm formation, or an intimal flap in regions of dissection and distal vessel occlusions due to emboli. US Duplex Doppler Transcranial and Carotid Artery Doppler US may be used to evaluate the cervical internal carotid arteries for dissection. The technique is of limited utility for carotid dissections at or above the skull base and in the evaluation of the vertebral arteries because of a lack of a good acoustic window [52]. Variant 5: Child. Clinical presentation suggestive of acute stroke, known or suspected central nervous system vasculitis. Initial imaging. Childhood vasculitis is as an inflammatory process only affecting the intracranial vessels and can result in stroke. It can present as childhood primary angiitis of the central nervous system or as a secondary phenomenon in systemic rheumatologic infections and neoplastic conditions [54]. Childhood primary angiitis of the central nervous system is subcategorized into large-medium vessel vasculitis (angiography positive) or small vessel (vessel abnormality too small to be demonstrated on MRA, CTA, or conventional angiography) best diagnosed by brain biopsy [55].

3102253

acrac_3102253_7

Cerebrovascular Disease Child PCAs

The diagnosis of childhood primary angiitis of the central nervous system requires an acquired neurologic deficit, angiographic or histologic features of central nervous system vasculitis, and no evidence of a systemic condition associated with the central nervous system findings. CT Head Evaluation of central nervous system vasculitis by CT is typically negative [56]. CTA Head CTA can provide a noninvasive assessment of intracranial vessels in arteriopathies as an alternative to catheter angiography in some patients but is typically negative in small-vessel vasculitis (eg, childhood primary angiitis of the central nervous system) [17]. MRI Head MRI typically shows progressive multifocal parenchymal lesions on T2-weighted imaging. Gadolinium enhancement of lesions is inconsistent. MRA Head MRA can provide a noninvasive assessment of intracranial vessels in arteriopathies as an alternative to catheter angiography in some patients but is typically negative in small vessel-vasculitis (eg, childhood primary angiitis of the central nervous system) [17]. Arteriography Cervicocerebral Catheter angiography is the most sensitive imaging method in the assessment of cerebral vasculitis but is typically negative in small-vessel vasculitis (eg, childhood primary angiitis of the central nervous system) [17]. US Duplex Doppler Transcranial Head US in not usually indicated in the evaluation of cerebral vasculitis. Variant 6: Child. Nontraumatic intraparenchymal hemorrhage (hematoma) found on CT or MRI. Unknown etiology. Next imaging study. Approximately half of all strokes beyond the perinatal period are hemorrhagic, whereas only 6.5% to 13% of strokes in adults are hemorrhagic [57,58]. Hemorrhagic strokes are associated with a variety of etiologies, including arteriovenous fistula or arteriovenous malformation (48%), brain tumors (15%), genetic or acquired coagulopathy (9%), thrombocytopenia (6%), cavernous malformation (5%), and coagulopathy and aneurysm (2%) [58].

Cerebrovascular Disease Child PCAs. The diagnosis of childhood primary angiitis of the central nervous system requires an acquired neurologic deficit, angiographic or histologic features of central nervous system vasculitis, and no evidence of a systemic condition associated with the central nervous system findings. CT Head Evaluation of central nervous system vasculitis by CT is typically negative [56]. CTA Head CTA can provide a noninvasive assessment of intracranial vessels in arteriopathies as an alternative to catheter angiography in some patients but is typically negative in small-vessel vasculitis (eg, childhood primary angiitis of the central nervous system) [17]. MRI Head MRI typically shows progressive multifocal parenchymal lesions on T2-weighted imaging. Gadolinium enhancement of lesions is inconsistent. MRA Head MRA can provide a noninvasive assessment of intracranial vessels in arteriopathies as an alternative to catheter angiography in some patients but is typically negative in small vessel-vasculitis (eg, childhood primary angiitis of the central nervous system) [17]. Arteriography Cervicocerebral Catheter angiography is the most sensitive imaging method in the assessment of cerebral vasculitis but is typically negative in small-vessel vasculitis (eg, childhood primary angiitis of the central nervous system) [17]. US Duplex Doppler Transcranial Head US in not usually indicated in the evaluation of cerebral vasculitis. Variant 6: Child. Nontraumatic intraparenchymal hemorrhage (hematoma) found on CT or MRI. Unknown etiology. Next imaging study. Approximately half of all strokes beyond the perinatal period are hemorrhagic, whereas only 6.5% to 13% of strokes in adults are hemorrhagic [57,58]. Hemorrhagic strokes are associated with a variety of etiologies, including arteriovenous fistula or arteriovenous malformation (48%), brain tumors (15%), genetic or acquired coagulopathy (9%), thrombocytopenia (6%), cavernous malformation (5%), and coagulopathy and aneurysm (2%) [58].

3102253

acrac_3102253_8

Cerebrovascular Disease Child PCAs

Presenting symptoms of hemorrhagic stroke differ with respect to the age of the child but may include mental status changes, seizures, or focal neurologic deficit. CT Head Unenhanced head CT is useful to assess the location and size of hemorrhage, evidence of mass effect, and hydrocephalus [59]. IV contrast may be helpful in demonstrating an underlying cause for the hemorrhage, such as vascular malformation, aneurysm, or tumor. CTA Head CTA may be used to delineate the vascular anatomy and may be used to demonstrate a vascular malformation or aneurysm but lacks the temporal information available from catheter angiography. MRI Head Unenhanced MRI can delineate the location and size of hemorrhage, evidence of mass effect, and hydrocephalus and may show evidence of an underlying structural lesion, such as vascular malformation, aneurysm, or tumor. SWI and arterial spin-labeling imaging may improve the delineation of the draining veins and arteriovenous shunting, respectively, compared with conventional MR sequences [24]. Contrast may be required to demonstrate small malformations, such as those that may occur in hereditary hemorrhagic telangiectasia, or other causes of parenchymal hematoma, such as tumor. MRA Head MRA may be used to delineate the vascular anatomy and may be used to demonstrate a vascular malformation or aneurysm but lacks the temporal information available from catheter angiography. Arteriography Cervicocerebral Catheter angiography is the gold standard for imaging of high-flow vascular malformations and aneurysms [60]. Although developmental venous anomalies (DVAs) can be demonstrated on catheter angiography, cavernoma and capillary telangiectasia are angiographically occult [61]. Variant 7: Child. Nontraumatic subarachnoid hemorrhage (SAH) detected by noncontrast CT. Next imaging study. Fifty-seven percent of SAH in children is due to aneurysm; however, only 2% to 15% of hemorrhagic strokes in children are due to aneurysm [58,62].

Cerebrovascular Disease Child PCAs. Presenting symptoms of hemorrhagic stroke differ with respect to the age of the child but may include mental status changes, seizures, or focal neurologic deficit. CT Head Unenhanced head CT is useful to assess the location and size of hemorrhage, evidence of mass effect, and hydrocephalus [59]. IV contrast may be helpful in demonstrating an underlying cause for the hemorrhage, such as vascular malformation, aneurysm, or tumor. CTA Head CTA may be used to delineate the vascular anatomy and may be used to demonstrate a vascular malformation or aneurysm but lacks the temporal information available from catheter angiography. MRI Head Unenhanced MRI can delineate the location and size of hemorrhage, evidence of mass effect, and hydrocephalus and may show evidence of an underlying structural lesion, such as vascular malformation, aneurysm, or tumor. SWI and arterial spin-labeling imaging may improve the delineation of the draining veins and arteriovenous shunting, respectively, compared with conventional MR sequences [24]. Contrast may be required to demonstrate small malformations, such as those that may occur in hereditary hemorrhagic telangiectasia, or other causes of parenchymal hematoma, such as tumor. MRA Head MRA may be used to delineate the vascular anatomy and may be used to demonstrate a vascular malformation or aneurysm but lacks the temporal information available from catheter angiography. Arteriography Cervicocerebral Catheter angiography is the gold standard for imaging of high-flow vascular malformations and aneurysms [60]. Although developmental venous anomalies (DVAs) can be demonstrated on catheter angiography, cavernoma and capillary telangiectasia are angiographically occult [61]. Variant 7: Child. Nontraumatic subarachnoid hemorrhage (SAH) detected by noncontrast CT. Next imaging study. Fifty-seven percent of SAH in children is due to aneurysm; however, only 2% to 15% of hemorrhagic strokes in children are due to aneurysm [58,62].

3102253

acrac_3102253_9

Cerebrovascular Disease Child PCAs

Aneurysms in children account for <1% of SAH due to aneurysms in all age groups [62]. The mortality in pediatric aneurysm has been reported to be 1.3%, with morbidity including infarction (8%) and seizures (4%) [63]. Only 0.6% of ruptured aneurysms occur in patients <19 years of age [51,60]. Aneurysms in children are idiopathic (45%), post-traumatic (20%), or due to a variety of conditions causing abnormal vessel-wall hemodynamic stress [62]. In contrast to adults, aneurysms in children are more likely to be giant (>25 mm) or fusiform [62]. Children with a positive family history of aneurysm account for <5% of pediatric aneurysms, and fewer than 2% of patients with a positive family history of aneurysm develop an aneurysm in the first two decades of life [62]. Fusiform aneurysms are more likely to increase in size over time than are saccular aneurysms [64]. CT Head The initial imaging study in suspected SAH is a noncontrast head CT. If the CT is negative for SAH, lumbar puncture may be performed. Unenhanced head CT is useful in ruptured aneurysms to assess the location and size of hemorrhage, evidence of mass effect, and hydrocephalus [59]. Small aneurysms are difficult to evaluate with unenhanced CT whereas a giant aneurysm, if present, appears as a hyperdense mass. CTA Head CTA is a noninvasive alternative to catheter angiography to show an underlying vascular cause of SAH with reported sensitivity (96.5%) and specificity (88%) for aneurysms of all sizes with sensitivity (98.4%) and specificity (100%) for aneurysms >3 mm [65]. CTA is used to confirm the presence of aneurysm and may be used as an adjunct to catheter angiography in the pretreatment assessment of aneurysm. CTA provides an alternative to catheter angiography for assessing residual or recurrent aneurysm but may be limited by streak artifact from the treatment device.

Cerebrovascular Disease Child PCAs. Aneurysms in children account for <1% of SAH due to aneurysms in all age groups [62]. The mortality in pediatric aneurysm has been reported to be 1.3%, with morbidity including infarction (8%) and seizures (4%) [63]. Only 0.6% of ruptured aneurysms occur in patients <19 years of age [51,60]. Aneurysms in children are idiopathic (45%), post-traumatic (20%), or due to a variety of conditions causing abnormal vessel-wall hemodynamic stress [62]. In contrast to adults, aneurysms in children are more likely to be giant (>25 mm) or fusiform [62]. Children with a positive family history of aneurysm account for <5% of pediatric aneurysms, and fewer than 2% of patients with a positive family history of aneurysm develop an aneurysm in the first two decades of life [62]. Fusiform aneurysms are more likely to increase in size over time than are saccular aneurysms [64]. CT Head The initial imaging study in suspected SAH is a noncontrast head CT. If the CT is negative for SAH, lumbar puncture may be performed. Unenhanced head CT is useful in ruptured aneurysms to assess the location and size of hemorrhage, evidence of mass effect, and hydrocephalus [59]. Small aneurysms are difficult to evaluate with unenhanced CT whereas a giant aneurysm, if present, appears as a hyperdense mass. CTA Head CTA is a noninvasive alternative to catheter angiography to show an underlying vascular cause of SAH with reported sensitivity (96.5%) and specificity (88%) for aneurysms of all sizes with sensitivity (98.4%) and specificity (100%) for aneurysms >3 mm [65]. CTA is used to confirm the presence of aneurysm and may be used as an adjunct to catheter angiography in the pretreatment assessment of aneurysm. CTA provides an alternative to catheter angiography for assessing residual or recurrent aneurysm but may be limited by streak artifact from the treatment device.

3102253

acrac_3102253_10

Cerebrovascular Disease Child PCAs

Using digital subtraction angiography as the gold standard, CTA has a reported sensitivity of 74% and specificity of 96% for the demonstration of residual aneurysm post-treatment [66]. MRI Head MRI is not typically an initial imaging study in patients with suspected SAH. However, SWI is sensitive to the presence of subarachnoid blood. MRI may be useful in classifying the aneurysms into saccular, dissecting, giant, and infectious [60]. MRI may be used to evaluate potential complications of treatment, including cerebral infarction or hemorrhage [63]. Only patients with MRI-safe aneurysm clips should be imaged with MR. MRA Head MRA is a noninvasive alternative to catheter angiography to show an underlying vascular cause of SAH. MRA of the brain is the examination of choice for children with unruptured aneurysms being followed prior to treatment [51]. A meta-analysis of the literature shows the sensitivity and specificity of MRA for the detection of intracranial aneurysm to be 95% and 89%, respectively [67]. Given the limited number of children who will develop an aneurysm under the age of 20, routine screening with MRA of children with a positive family history of aneurysm is not supported in the literature [68]. Arteriography Cervicocerebral Catheter angiography remains the gold standard for the demonstration of high-flow vascular anomalies and aneurysms as a cause of SAH. Catheter angiography can be performed for pretreatment planning and for post- treatment assessment. Catheter angiography is associated with a low incidence of procedural complications in children [51,60]. Catheter angiography is the gold standard for the diagnosis and pretreatment evaluation of intracranial aneurysm and may also be required for post-treatment follow-up in some patients [60]. US Duplex Doppler Transcranial Head US is not sensitive for the presence of subarachnoid blood or vascular causes of intracranial hemorrhage.

Cerebrovascular Disease Child PCAs. Using digital subtraction angiography as the gold standard, CTA has a reported sensitivity of 74% and specificity of 96% for the demonstration of residual aneurysm post-treatment [66]. MRI Head MRI is not typically an initial imaging study in patients with suspected SAH. However, SWI is sensitive to the presence of subarachnoid blood. MRI may be useful in classifying the aneurysms into saccular, dissecting, giant, and infectious [60]. MRI may be used to evaluate potential complications of treatment, including cerebral infarction or hemorrhage [63]. Only patients with MRI-safe aneurysm clips should be imaged with MR. MRA Head MRA is a noninvasive alternative to catheter angiography to show an underlying vascular cause of SAH. MRA of the brain is the examination of choice for children with unruptured aneurysms being followed prior to treatment [51]. A meta-analysis of the literature shows the sensitivity and specificity of MRA for the detection of intracranial aneurysm to be 95% and 89%, respectively [67]. Given the limited number of children who will develop an aneurysm under the age of 20, routine screening with MRA of children with a positive family history of aneurysm is not supported in the literature [68]. Arteriography Cervicocerebral Catheter angiography remains the gold standard for the demonstration of high-flow vascular anomalies and aneurysms as a cause of SAH. Catheter angiography can be performed for pretreatment planning and for post- treatment assessment. Catheter angiography is associated with a low incidence of procedural complications in children [51,60]. Catheter angiography is the gold standard for the diagnosis and pretreatment evaluation of intracranial aneurysm and may also be required for post-treatment follow-up in some patients [60]. US Duplex Doppler Transcranial Head US is not sensitive for the presence of subarachnoid blood or vascular causes of intracranial hemorrhage.

3102253

acrac_3102253_11

Cerebrovascular Disease Child PCAs

Head US provides limited information on intracranial aneurysm, but Doppler US may be used to confirm the vascular nature of a mass seen on head US performed in neonates and young infants with open fontanels. CT Head Unenhanced head CT is useful in hemorrhagic high-flow vascular anomalies to assess the location and size of hemorrhage, evidence of mass effect, and hydrocephalus [59]. High-flow vascular malformations that have not bled are slightly hyperdense relative to brain on unenhanced head CT. The detection of a high-flow vascular anomaly that has not bled may be improved with the use of IV contrast. CTA Head CTA is often used to delineate the anatomy of a high-flow vascular anomaly but lacks the temporal information available from catheter angiography. CTA may be used to monitor the lesion prior to treatment or to assess for growth or recurrence after treatment. MRI Head MRI is used to determine the location and size of high-flow vascular anomalies and is often preferred to CT for use in children. SWI and arterial spin-labeling imaging may improve the delineation of the draining veins and arteriovenous shunting, respectively, compared with conventional MRI sequences [24]. Contrast may be required to demonstrate small malformations. MRA Head MRA, like CTA, can be used to delineate the anatomy of a high-flow vascular anomaly but lacks the temporal information available from catheter angiography. MRA is often preferable to CTA in children for following vascular anomalies serially. The use of IV contrast may be required to assess the venous outflow of the anomaly. MRA may be used to monitor the anomaly prior to treatment or to assess for growth or recurrence after treatment. Arteriography Cervicocerebral With the increased accuracy of the noninvasive imaging studies CTA and MRA, catheter angiography is not typically used as the initial diagnostic imaging for suspected high-flow vascular anomaly.

Cerebrovascular Disease Child PCAs. Head US provides limited information on intracranial aneurysm, but Doppler US may be used to confirm the vascular nature of a mass seen on head US performed in neonates and young infants with open fontanels. CT Head Unenhanced head CT is useful in hemorrhagic high-flow vascular anomalies to assess the location and size of hemorrhage, evidence of mass effect, and hydrocephalus [59]. High-flow vascular malformations that have not bled are slightly hyperdense relative to brain on unenhanced head CT. The detection of a high-flow vascular anomaly that has not bled may be improved with the use of IV contrast. CTA Head CTA is often used to delineate the anatomy of a high-flow vascular anomaly but lacks the temporal information available from catheter angiography. CTA may be used to monitor the lesion prior to treatment or to assess for growth or recurrence after treatment. MRI Head MRI is used to determine the location and size of high-flow vascular anomalies and is often preferred to CT for use in children. SWI and arterial spin-labeling imaging may improve the delineation of the draining veins and arteriovenous shunting, respectively, compared with conventional MRI sequences [24]. Contrast may be required to demonstrate small malformations. MRA Head MRA, like CTA, can be used to delineate the anatomy of a high-flow vascular anomaly but lacks the temporal information available from catheter angiography. MRA is often preferable to CTA in children for following vascular anomalies serially. The use of IV contrast may be required to assess the venous outflow of the anomaly. MRA may be used to monitor the anomaly prior to treatment or to assess for growth or recurrence after treatment. Arteriography Cervicocerebral With the increased accuracy of the noninvasive imaging studies CTA and MRA, catheter angiography is not typically used as the initial diagnostic imaging for suspected high-flow vascular anomaly.

3102253

acrac_3102253_12

Cerebrovascular Disease Child PCAs

Catheter angiography can be used for pretreatment evaluation and post-treatment assessment of any residual anomaly [59,71]. US Duplex Doppler Transcranial Head US with Doppler may be used to show the presence of large high-flow vascular malformations, such as vein of Galen malformation and dural arteriovenous fistula or malformation in neonates and young infants. US is useful in very young children because it can be performed without sedation or anesthesia, which may be required for MRI examinations. Variant 9: Child. Clinical presentation suggestive of acute stroke, known or suspected low-flow vascular anomaly. Initial imaging. Low-flow vascular anomalies include cavernous malformation, DVA, and capillary telangiectasia. Cavernomas are comprised of endothelial spaces containing venous blood and account for 17% of hemorrhagic stroke in children [18]. In a large series, 62% of children with cavernomas presented with hemorrhage, 35% with seizures with or without hemorrhage, and 26% had cavernomas discovered incidentally [72]. Patients with cavernoma (15%) have multiple lesions and 10% have a family history of cavernoma and may have mutations in CCM1, CCM2, and CCM3 genes [72-74]. Although 9% are associated with prior cranial irradiation [72], 86% are supratentorial and 14% infratentorial in location. Cavernomas have an annual hemorrhage rate of 3.3% to 4.5% [72,75]. DVA is an anomalous vein draining normal brain parenchyma, and 20% of DVA are associated with cavernoma [72]. Most hemorrhage in patients with DVA is believed to be due to bleeding from an associated cavernoma [61,72]. CT Head Noncontrast CT can show acute hemorrhage with cavernoma. Nonhemorrhagic cavernomas and DVA may be faintly hyperdense on noncontrast imaging [61]. IV contrast increases the conspicuity of DVA. CTA Head CTA plays a limited role in the assessment of cavernoma but may be used to demonstrate a DVA. MRI Head MRI is the imaging study of choice for cavernoma. Various stages of hemorrhage may be evident.

Cerebrovascular Disease Child PCAs. Catheter angiography can be used for pretreatment evaluation and post-treatment assessment of any residual anomaly [59,71]. US Duplex Doppler Transcranial Head US with Doppler may be used to show the presence of large high-flow vascular malformations, such as vein of Galen malformation and dural arteriovenous fistula or malformation in neonates and young infants. US is useful in very young children because it can be performed without sedation or anesthesia, which may be required for MRI examinations. Variant 9: Child. Clinical presentation suggestive of acute stroke, known or suspected low-flow vascular anomaly. Initial imaging. Low-flow vascular anomalies include cavernous malformation, DVA, and capillary telangiectasia. Cavernomas are comprised of endothelial spaces containing venous blood and account for 17% of hemorrhagic stroke in children [18]. In a large series, 62% of children with cavernomas presented with hemorrhage, 35% with seizures with or without hemorrhage, and 26% had cavernomas discovered incidentally [72]. Patients with cavernoma (15%) have multiple lesions and 10% have a family history of cavernoma and may have mutations in CCM1, CCM2, and CCM3 genes [72-74]. Although 9% are associated with prior cranial irradiation [72], 86% are supratentorial and 14% infratentorial in location. Cavernomas have an annual hemorrhage rate of 3.3% to 4.5% [72,75]. DVA is an anomalous vein draining normal brain parenchyma, and 20% of DVA are associated with cavernoma [72]. Most hemorrhage in patients with DVA is believed to be due to bleeding from an associated cavernoma [61,72]. CT Head Noncontrast CT can show acute hemorrhage with cavernoma. Nonhemorrhagic cavernomas and DVA may be faintly hyperdense on noncontrast imaging [61]. IV contrast increases the conspicuity of DVA. CTA Head CTA plays a limited role in the assessment of cavernoma but may be used to demonstrate a DVA. MRI Head MRI is the imaging study of choice for cavernoma. Various stages of hemorrhage may be evident.

3102253

acrac_3102253_13

Cerebrovascular Disease Child PCAs

T2-weighted gradient-echo imaging or SWI may show additional cavernomas not seen on spin-echo imaging. The presence of acute and subacute blood products increases the likelihood of future hemorrhage [75]. DVAs are visible on both spin-echo and gradient-echo imaging and are occasionally associated with gliosis or cortical malformation. Contrast may increase the conspicuity of DVA. MRA Head MRA is not usually helpful in the assessment of cavernoma, capillary telangiectasia, and DVA. Arteriography Cervicocerebral Catheter angiography is rarely required for the evaluation of incidentally discovered DVA but may be used to exclude the possibility of an associated high-flow vascular malformation. An isolated DVA appears as an abnormal cluster of veins draining into a single collector and appears only on the venous phase of the arteriogram [61]. Cavernomas and capillary telangiectasia are usually angiographically occult. US Duplex Doppler Transcranial Head US is not usually helpful in the assessment of cavernoma or capillary telangiectasia. Head US may occasionally be used to demonstrate large DVA in neonates and young infants with open fontanels. Variant 10: Child. Clinical presentation suggestive of acute stroke, known or suspected cortical vein or dural venous sinus thrombosis. Initial imaging. Cerebral sinovenous thrombosis has an incidence of 1/200,000 children per year. It is less common in children than is arterial ischemic or hemorrhagic stroke and is most often associated with infection, trauma, dehydration, cancer, oral contraceptives, and prothrombotic disorders [76,77]. The risk is highest in the first year of life, with neonates accounting for 61% of cerebral sinovenous thrombosis [76-78]. Prothrombotic disorders are present in more than half of children, and multiple risk factors are commonly present. An acute illness with sepsis and dehydration is present in up to one-third of patients. Trauma is more common in older children.

Cerebrovascular Disease Child PCAs. T2-weighted gradient-echo imaging or SWI may show additional cavernomas not seen on spin-echo imaging. The presence of acute and subacute blood products increases the likelihood of future hemorrhage [75]. DVAs are visible on both spin-echo and gradient-echo imaging and are occasionally associated with gliosis or cortical malformation. Contrast may increase the conspicuity of DVA. MRA Head MRA is not usually helpful in the assessment of cavernoma, capillary telangiectasia, and DVA. Arteriography Cervicocerebral Catheter angiography is rarely required for the evaluation of incidentally discovered DVA but may be used to exclude the possibility of an associated high-flow vascular malformation. An isolated DVA appears as an abnormal cluster of veins draining into a single collector and appears only on the venous phase of the arteriogram [61]. Cavernomas and capillary telangiectasia are usually angiographically occult. US Duplex Doppler Transcranial Head US is not usually helpful in the assessment of cavernoma or capillary telangiectasia. Head US may occasionally be used to demonstrate large DVA in neonates and young infants with open fontanels. Variant 10: Child. Clinical presentation suggestive of acute stroke, known or suspected cortical vein or dural venous sinus thrombosis. Initial imaging. Cerebral sinovenous thrombosis has an incidence of 1/200,000 children per year. It is less common in children than is arterial ischemic or hemorrhagic stroke and is most often associated with infection, trauma, dehydration, cancer, oral contraceptives, and prothrombotic disorders [76,77]. The risk is highest in the first year of life, with neonates accounting for 61% of cerebral sinovenous thrombosis [76-78]. Prothrombotic disorders are present in more than half of children, and multiple risk factors are commonly present. An acute illness with sepsis and dehydration is present in up to one-third of patients. Trauma is more common in older children.

3102253

acrac_3102253_14

Cerebrovascular Disease Child PCAs

Cerebral sinovenous thrombosis can cause elevation in venous pressure, increased intracranial pressure, and venous infarction. Hemorrhagic infarction (40%) and hydrocephalus (10%) may complicate cerebral sinovenous thrombosis [77]. Infarctions are more frequently hemorrhagic in neonates (72%) than in older children (48%) [79]. The lateral dural venous sinuses (73%) and superior sagittal sinus (35%) are most commonly affected [76- 78]. CT Head CT Head Unenhanced CT shows increased density, often accompanied by distension, of the thrombosed sinus or vein [77,80]. CT is less sensitive than MRI for the demonstration of early ischemic infarction but readily shows areas of brain hemorrhage [77,79]. MRV Head MRV is used to diagnose and follow cortical vein and dural venous sinus thrombosis and is typically performed in conjunction with anatomic MRI without and with IV contrast for a comprehensive assessment of the intracranial venous drainage. MRV without IV contrast is commonly used to confirm absence of flow in a thrombosed dural venous sinus. MRV with IV contrast is less susceptible to the flow artifacts that may occur because of turbulent flow in the dural venous sinuses with noncontrast MRV [80]. Arteriography Cervicocerebral Catheter angiography is not routinely used for the diagnosis of cortical venous or dural venous sinus thrombosis but is used when endovascular treatment is required. US Duplex Doppler Transcranial Head US with Doppler may be used to assess patency of the dural venous sinuses in the neonate. Following closure of the fontanels, CT and MRI are more commonly used. Variant 11: Child. Clinical presentation suggestive of acute stroke, sickle cell disease. New focal fixed or worsening neurologic defect. Initial imaging. SCD constitutes one of the main etiologies of pediatric stroke. Eleven percent of children with SCD not receiving primary stroke prevention therapy will have a stroke by 20 years of age [36,82,83].

Cerebrovascular Disease Child PCAs. Cerebral sinovenous thrombosis can cause elevation in venous pressure, increased intracranial pressure, and venous infarction. Hemorrhagic infarction (40%) and hydrocephalus (10%) may complicate cerebral sinovenous thrombosis [77]. Infarctions are more frequently hemorrhagic in neonates (72%) than in older children (48%) [79]. The lateral dural venous sinuses (73%) and superior sagittal sinus (35%) are most commonly affected [76- 78]. CT Head CT Head Unenhanced CT shows increased density, often accompanied by distension, of the thrombosed sinus or vein [77,80]. CT is less sensitive than MRI for the demonstration of early ischemic infarction but readily shows areas of brain hemorrhage [77,79]. MRV Head MRV is used to diagnose and follow cortical vein and dural venous sinus thrombosis and is typically performed in conjunction with anatomic MRI without and with IV contrast for a comprehensive assessment of the intracranial venous drainage. MRV without IV contrast is commonly used to confirm absence of flow in a thrombosed dural venous sinus. MRV with IV contrast is less susceptible to the flow artifacts that may occur because of turbulent flow in the dural venous sinuses with noncontrast MRV [80]. Arteriography Cervicocerebral Catheter angiography is not routinely used for the diagnosis of cortical venous or dural venous sinus thrombosis but is used when endovascular treatment is required. US Duplex Doppler Transcranial Head US with Doppler may be used to assess patency of the dural venous sinuses in the neonate. Following closure of the fontanels, CT and MRI are more commonly used. Variant 11: Child. Clinical presentation suggestive of acute stroke, sickle cell disease. New focal fixed or worsening neurologic defect. Initial imaging. SCD constitutes one of the main etiologies of pediatric stroke. Eleven percent of children with SCD not receiving primary stroke prevention therapy will have a stroke by 20 years of age [36,82,83].

3102253

acrac_3102253_15

Cerebrovascular Disease Child PCAs

Most infarctions are due to vasculopathy involving the supraclinoid portion of the internal carotid artery or branches of the circle of Willis and occurring in a watershed distribution. Clinically silent infarctions are known to occur in 17% of children with SCD [84]. The National Heart Lung and Blood institute recommends that children with SCD between 2 and 16 years of age undergo transcranial Doppler US screening every 6 months. The treatment of acute stroke symptoms in children with SCD is transfusion. CT Head CT is relatively insensitive compared with MRI for the demonstration of infarction and is typically only used in the acute setting if MRI is unavailable or to document hemorrhage or evaluate for mass effect [26]. CTA Head CTA is not used in routine neurovascular screening in children with SCD. CTA can be used in children with contraindication to MRA. MRI Head MRI can show T2 prolongation in watershed vascular territories or cortex in patients with SCD. MRA Head MRA is the study of choice to screen for vessel stenosis in children with SCD with elevated transcranial Doppler US velocities in the intracranial vessels [39,44]. Although MRA can accurately show vessel stenosis, it is susceptible to false-positive results when long echo times are used or when severe anemia is present, resulting in turbulent blood flow and localized artefactual signal loss resembling focal stenosis [26,84,85]. Although gadolinium-enhanced MRA can be used, it is generally not required and is often avoided because of concerns that it may potentiate hemolysis in patients with SCD [26]. MRI of the brain without IV contrast is typically performed during the same examination as the MRA to assess the brain parenchyma for ischemic injury. Arteriography Cervicocerebral Catheter angiography is generally not indicated except in uncommon circumstances, including preoperative assessment for revascularization for moyamoya syndrome and suspected aneurysm complicating SCD [26].

Cerebrovascular Disease Child PCAs. Most infarctions are due to vasculopathy involving the supraclinoid portion of the internal carotid artery or branches of the circle of Willis and occurring in a watershed distribution. Clinically silent infarctions are known to occur in 17% of children with SCD [84]. The National Heart Lung and Blood institute recommends that children with SCD between 2 and 16 years of age undergo transcranial Doppler US screening every 6 months. The treatment of acute stroke symptoms in children with SCD is transfusion. CT Head CT is relatively insensitive compared with MRI for the demonstration of infarction and is typically only used in the acute setting if MRI is unavailable or to document hemorrhage or evaluate for mass effect [26]. CTA Head CTA is not used in routine neurovascular screening in children with SCD. CTA can be used in children with contraindication to MRA. MRI Head MRI can show T2 prolongation in watershed vascular territories or cortex in patients with SCD. MRA Head MRA is the study of choice to screen for vessel stenosis in children with SCD with elevated transcranial Doppler US velocities in the intracranial vessels [39,44]. Although MRA can accurately show vessel stenosis, it is susceptible to false-positive results when long echo times are used or when severe anemia is present, resulting in turbulent blood flow and localized artefactual signal loss resembling focal stenosis [26,84,85]. Although gadolinium-enhanced MRA can be used, it is generally not required and is often avoided because of concerns that it may potentiate hemolysis in patients with SCD [26]. MRI of the brain without IV contrast is typically performed during the same examination as the MRA to assess the brain parenchyma for ischemic injury. Arteriography Cervicocerebral Catheter angiography is generally not indicated except in uncommon circumstances, including preoperative assessment for revascularization for moyamoya syndrome and suspected aneurysm complicating SCD [26].

3102253

acrac_3127236_0

Pancreatic Cyst

Introduction/Background Incidental pancreatic cysts are now increasingly detected on imaging studies performed for unrelated indications. [1-3]. Both increased imaging utilization and improved cross-sectional technique are responsible for the more frequent detection of progressively smaller cysts [4-6]. The most commonly encountered pancreatic cysts include intraductal papillary mucinous neoplasms (IPMNs), serous cystadenoma, mucinous cystic neoplasm (MCN), and pseudocysts [6,7]. There is a very small risk that an incidental pancreatic cyst may be malignant [8]. For instance, an incidental pancreatic cyst on MRI has a 10 in 100,000 chance of being a mucinous invasive malignancy and a 17 in 100,000 chance of being a ductal carcinoma [8]. The risk of malignant transformation in pancreatic cysts is estimated to be 0.24% per year [9], varying according to histologic subtype [7,10]. Yet there is considerable overlap in the imaging appearance of histologically distinct pancreatic cysts, particularly those <3 cm in size, with over 60% of cysts lacking a specific radiologic appearance on CT or MRI [6]. Another important feature in the natural history of pancreatic cysts is the small risk of pancreatic adenocarcinoma developing at a separate site within the pancreas [4,7,11-13]. Although the risk of cyst-related or concomitant pancreatic malignancy is small, there is a need to characterize incidental pancreatic cysts effectively at initial imaging in order to guide management. Appropriate imaging evaluation of incidental pancreatic cysts is critical because morphology determines management. As an example, surveillance is generally recommended for cysts <3 cm in size without worrisome features or high-risk stigmata [7,13]. Cysts with worrisome features undergo sampling with endoscopic ultrasound fine-needle aspiration (EUS-FNA) [8-10] and those with high-risk stigmata are typically resected [8-10].

Pancreatic Cyst. Introduction/Background Incidental pancreatic cysts are now increasingly detected on imaging studies performed for unrelated indications. [1-3]. Both increased imaging utilization and improved cross-sectional technique are responsible for the more frequent detection of progressively smaller cysts [4-6]. The most commonly encountered pancreatic cysts include intraductal papillary mucinous neoplasms (IPMNs), serous cystadenoma, mucinous cystic neoplasm (MCN), and pseudocysts [6,7]. There is a very small risk that an incidental pancreatic cyst may be malignant [8]. For instance, an incidental pancreatic cyst on MRI has a 10 in 100,000 chance of being a mucinous invasive malignancy and a 17 in 100,000 chance of being a ductal carcinoma [8]. The risk of malignant transformation in pancreatic cysts is estimated to be 0.24% per year [9], varying according to histologic subtype [7,10]. Yet there is considerable overlap in the imaging appearance of histologically distinct pancreatic cysts, particularly those <3 cm in size, with over 60% of cysts lacking a specific radiologic appearance on CT or MRI [6]. Another important feature in the natural history of pancreatic cysts is the small risk of pancreatic adenocarcinoma developing at a separate site within the pancreas [4,7,11-13]. Although the risk of cyst-related or concomitant pancreatic malignancy is small, there is a need to characterize incidental pancreatic cysts effectively at initial imaging in order to guide management. Appropriate imaging evaluation of incidental pancreatic cysts is critical because morphology determines management. As an example, surveillance is generally recommended for cysts <3 cm in size without worrisome features or high-risk stigmata [7,13]. Cysts with worrisome features undergo sampling with endoscopic ultrasound fine-needle aspiration (EUS-FNA) [8-10] and those with high-risk stigmata are typically resected [8-10].

3127236

acrac_3127236_1

Pancreatic Cyst

For management recommendations please refer to the ACR White Paper on Management of Incidental Pancreatic Cysts [7]. The following recommendations refer to the initial imaging evaluation of pancreatic cysts incidentally detected and incompletely evaluated on imaging studies performed for unrelated indications. The recommendations below apply Reprint requests to: [email protected] Pancreatic Cyst irrespective of the imaging modality in which the cyst was initially detected. CT abdomen, MRI abdomen with MRCP, and US abdomen endoscopic are included in the discussion. These are the three conventional imaging modalities used in the workup of pancreatic cysts. Although we acknowledge the added value of US (especially contrast-enhanced US) in select cases in which workup with conventional imaging is inconclusive or incomplete, US has been omitted from the discussion because it is not routinely used in this setting. Discussion of Procedures by Variant Variant 1: Incidentally detected pancreatic cyst less than or equal to 2.5 cm in size. Initial evaluation. CT Abdomen MRI is preferred over contrast-enhanced CT in this setting. The sensitivity and specificity of CT for distinguishing IPMN from other cystic pancreatic lesions is 80.6% and 86.4% compared with 96.8% and 90.8% for MRI [3,18]. Advantages of CT include its ease of implementation and excellent spatial resolution. Multidetector CT provides critical diagnostic information pertaining to the presence or absence of calcifications (both in the background parenchyma and in the cyst proper), ductal dilation, intralesional septations, mural nodules, and pancreatic duct communication [6,15]. MRI Abdomen with MRCP Contrast-enhanced MRI with MR cholangiopancreatography (MRCP) is considered the procedure of choice in this setting because of its superior soft-tissue contrast and superior ability to demonstrate ductal communication [6-10].

Pancreatic Cyst. For management recommendations please refer to the ACR White Paper on Management of Incidental Pancreatic Cysts [7]. The following recommendations refer to the initial imaging evaluation of pancreatic cysts incidentally detected and incompletely evaluated on imaging studies performed for unrelated indications. The recommendations below apply Reprint requests to: [email protected] Pancreatic Cyst irrespective of the imaging modality in which the cyst was initially detected. CT abdomen, MRI abdomen with MRCP, and US abdomen endoscopic are included in the discussion. These are the three conventional imaging modalities used in the workup of pancreatic cysts. Although we acknowledge the added value of US (especially contrast-enhanced US) in select cases in which workup with conventional imaging is inconclusive or incomplete, US has been omitted from the discussion because it is not routinely used in this setting. Discussion of Procedures by Variant Variant 1: Incidentally detected pancreatic cyst less than or equal to 2.5 cm in size. Initial evaluation. CT Abdomen MRI is preferred over contrast-enhanced CT in this setting. The sensitivity and specificity of CT for distinguishing IPMN from other cystic pancreatic lesions is 80.6% and 86.4% compared with 96.8% and 90.8% for MRI [3,18]. Advantages of CT include its ease of implementation and excellent spatial resolution. Multidetector CT provides critical diagnostic information pertaining to the presence or absence of calcifications (both in the background parenchyma and in the cyst proper), ductal dilation, intralesional septations, mural nodules, and pancreatic duct communication [6,15]. MRI Abdomen with MRCP Contrast-enhanced MRI with MR cholangiopancreatography (MRCP) is considered the procedure of choice in this setting because of its superior soft-tissue contrast and superior ability to demonstrate ductal communication [6-10].

3127236

acrac_3127236_2

Pancreatic Cyst

The reported sensitivity of thin-slice 3-D MRCP acquisitions for demonstrating communication of a cyst with the pancreatic duct is as high as 100% [6,22]. Communication with the main pancreatic duct is suggestive of IPMN, although this may also be seen in the setting of pseudocysts [6]. The sensitivity of MRI for detection of internal septations is 91% [6,22], and its diagnostic accuracy for distinguishing between malignant and nonmalignant lesions ranges between 73.2% and 91% [6,23,24]. In distinguishing IPMN from other cystic lesions, studies have reported a sensitivity of 96.8% and a specificity of 90.8% [3,18]. Pancreatic Cyst MRI Abdomen with MRCP Because of its superior soft-tissue resolution and noninvasive approach, contrast-enhanced MRI with MRCP is generally favored over CT or EUS-FNA in this setting [6-10]. The diagnostic accuracy of MRI for distinguishing between malignant and nonmalignant lesions ranges from 73.2% to 91% [6,23,24]. Its accuracy at diagnosing the specific type of cyst is slightly lower at 50% [23]. However, an exception may be the distinction of IPMN from other cystic lesions, in which studies have reported a sensitivity of 96.8% and specificity of 90.8% [3,18]. In a study of over 300 patients with pancreatic cysts, the addition of EUS-FNA to the diagnostic workup significantly altered the management strategy in nearly 72% of patients [28]. Management algorithms integrating clinical data, imaging, and fluid analysis have reported cyst classification sensitivities of 90% to 100% and specificities of 92% to 98% [7,25]. The addition of EUS-FNA to management algorithms combining clinical history and imaging may also reduce unnecessary surgeries by 91% [7]. Variant 3: Incidentally detected pancreatic cyst greater than 2.5 cm in size. High-risk stigmata or worrisome features. Initial evaluation.

Pancreatic Cyst. The reported sensitivity of thin-slice 3-D MRCP acquisitions for demonstrating communication of a cyst with the pancreatic duct is as high as 100% [6,22]. Communication with the main pancreatic duct is suggestive of IPMN, although this may also be seen in the setting of pseudocysts [6]. The sensitivity of MRI for detection of internal septations is 91% [6,22], and its diagnostic accuracy for distinguishing between malignant and nonmalignant lesions ranges between 73.2% and 91% [6,23,24]. In distinguishing IPMN from other cystic lesions, studies have reported a sensitivity of 96.8% and a specificity of 90.8% [3,18]. Pancreatic Cyst MRI Abdomen with MRCP Because of its superior soft-tissue resolution and noninvasive approach, contrast-enhanced MRI with MRCP is generally favored over CT or EUS-FNA in this setting [6-10]. The diagnostic accuracy of MRI for distinguishing between malignant and nonmalignant lesions ranges from 73.2% to 91% [6,23,24]. Its accuracy at diagnosing the specific type of cyst is slightly lower at 50% [23]. However, an exception may be the distinction of IPMN from other cystic lesions, in which studies have reported a sensitivity of 96.8% and specificity of 90.8% [3,18]. In a study of over 300 patients with pancreatic cysts, the addition of EUS-FNA to the diagnostic workup significantly altered the management strategy in nearly 72% of patients [28]. Management algorithms integrating clinical data, imaging, and fluid analysis have reported cyst classification sensitivities of 90% to 100% and specificities of 92% to 98% [7,25]. The addition of EUS-FNA to management algorithms combining clinical history and imaging may also reduce unnecessary surgeries by 91% [7]. Variant 3: Incidentally detected pancreatic cyst greater than 2.5 cm in size. High-risk stigmata or worrisome features. Initial evaluation.

3127236

acrac_3127236_3

Pancreatic Cyst

CT Abdomen The presence of high-risk stigmata or worrisome features significantly increases the risk of malignancy, and therefore EUS-FNA is favored over CT in this setting [8,9,14]. In cases in which EUS-FNA cannot be performed and the patient is not a candidate for MRI with MRCP, a dual-phase contrast-enhanced pancreatic protocol CT may still be of value for cyst characterization or presurgical planning [1,7-10,13,19-21]. The use of IV contrast improves detection of worrisome features and high-risk stigmata and the assessment of surrounding anatomy [1,7-10,13,19- 21]. Pancreatic Cyst US Abdomen Endoscopic When high-risk stigmata or worrisome features are present, the appropriate initial imaging study is EUS-FNA [7- 10]. The unique advantage of EUS-FNA is its ability to distinguish mucinous from nonmucinous lesions by means of biochemical markers assayed from cyst fluid samples. This facilitates a specific diagnosis in many cases [25,28- 30]. For instance, the presence of nongut mucin supports a diagnosis of mucinous neoplasm. Carcinoembryonic antigen levels <5 ng/mL suggest pseudocyst or serous cystadenoma. A carcinoembryonic antigen threshold level in the range of 192 to 200 ng/mL is 80% accurate for diagnosis of a mucinous cyst [13,29]. Amylase levels of >250 IU/L suggest a pseudocyst. Molecular assays for markers such as K-ras, GNAS, PTEN, VHL, TP53, and PIK3CA may also assist in differentiating neoplastic cystic lesions and predicting cyst behavior. When performed in centers with expertise in EUS-FNA, cytological evaluation can identify atypia, dysplasia, or neoplasia [7,13,25]. Studies have demonstrated that the presence of high-grade epithelial atypia in IPMN detects approximately 30% more cancers than the presence of worrisome imaging features alone [13]. Pancreatic Cyst Variant 5: Follow-up imaging of pancreatic cyst. CT Abdomen The risk of malignant transformation of a pancreatic cyst is approximately 0.24% per year [9].

Pancreatic Cyst. CT Abdomen The presence of high-risk stigmata or worrisome features significantly increases the risk of malignancy, and therefore EUS-FNA is favored over CT in this setting [8,9,14]. In cases in which EUS-FNA cannot be performed and the patient is not a candidate for MRI with MRCP, a dual-phase contrast-enhanced pancreatic protocol CT may still be of value for cyst characterization or presurgical planning [1,7-10,13,19-21]. The use of IV contrast improves detection of worrisome features and high-risk stigmata and the assessment of surrounding anatomy [1,7-10,13,19- 21]. Pancreatic Cyst US Abdomen Endoscopic When high-risk stigmata or worrisome features are present, the appropriate initial imaging study is EUS-FNA [7- 10]. The unique advantage of EUS-FNA is its ability to distinguish mucinous from nonmucinous lesions by means of biochemical markers assayed from cyst fluid samples. This facilitates a specific diagnosis in many cases [25,28- 30]. For instance, the presence of nongut mucin supports a diagnosis of mucinous neoplasm. Carcinoembryonic antigen levels <5 ng/mL suggest pseudocyst or serous cystadenoma. A carcinoembryonic antigen threshold level in the range of 192 to 200 ng/mL is 80% accurate for diagnosis of a mucinous cyst [13,29]. Amylase levels of >250 IU/L suggest a pseudocyst. Molecular assays for markers such as K-ras, GNAS, PTEN, VHL, TP53, and PIK3CA may also assist in differentiating neoplastic cystic lesions and predicting cyst behavior. When performed in centers with expertise in EUS-FNA, cytological evaluation can identify atypia, dysplasia, or neoplasia [7,13,25]. Studies have demonstrated that the presence of high-grade epithelial atypia in IPMN detects approximately 30% more cancers than the presence of worrisome imaging features alone [13]. Pancreatic Cyst Variant 5: Follow-up imaging of pancreatic cyst. CT Abdomen The risk of malignant transformation of a pancreatic cyst is approximately 0.24% per year [9].

3127236

acrac_3127236_4

Pancreatic Cyst

Once a pancreatic cyst has been characterized on a dedicated baseline examination, subsequent follow-up may be performed with either CT or MRI. There is no evidence to suggest that MRI is superior to CT for detection of new or developing worrisome features or pancreatic ductal adenocarcinoma, and cysts that change at follow up typically do so by increasing size, which is well assessed by either modality [7], although modality concordance between baseline and follow-up examinations may facilitate comparison. For CT follow-up, a dual-phase contrast-enhanced pancreatic protocol CT, including late arterial and portal venous phases, should be performed [7,8]. The frequency and duration of follow-up is controversial and depends on multiple factors, including patient age, family history of pancreatic ductal adenocarcinoma, cyst size, and whether or not there has been prior surgical resection of a pancreatic cyst. For patients with a nonspecific pancreatic cyst without a history of prior surgery, the surveillance plan will depend upon patient age and the cyst size. Follow-up intervals are generally in the range of 6 months to every 2 years for a minimum of 5 to 10 years [7-9]. Development of high-risk stigmata or worrisome features during the surveillance period should prompt EUS-FNA or surgical evaluation. For patients with a previous history of surgery for IPMN or invasive MCN without residual disease, continued surveillance is recommended, in view of the small yearly risk of pancreatic ductal adenocarcinoma of 0.7% to 0.9% [8]. For patients with known IPMN in the remnant pancreas, residual IPMN at the surgical margins, or new postoperative recurrence of IPMN, surveillance recommendations are less well defined [8]. To date, there is no evidence basis for the recommended size threshold to follow-up cysts. Based on limited clinical and published experience, a cyst <5 mm may require one follow-up CT or MRI at 2 years. Demonstrating stability at 2 years is sufficient to stop surveillance.

Pancreatic Cyst. Once a pancreatic cyst has been characterized on a dedicated baseline examination, subsequent follow-up may be performed with either CT or MRI. There is no evidence to suggest that MRI is superior to CT for detection of new or developing worrisome features or pancreatic ductal adenocarcinoma, and cysts that change at follow up typically do so by increasing size, which is well assessed by either modality [7], although modality concordance between baseline and follow-up examinations may facilitate comparison. For CT follow-up, a dual-phase contrast-enhanced pancreatic protocol CT, including late arterial and portal venous phases, should be performed [7,8]. The frequency and duration of follow-up is controversial and depends on multiple factors, including patient age, family history of pancreatic ductal adenocarcinoma, cyst size, and whether or not there has been prior surgical resection of a pancreatic cyst. For patients with a nonspecific pancreatic cyst without a history of prior surgery, the surveillance plan will depend upon patient age and the cyst size. Follow-up intervals are generally in the range of 6 months to every 2 years for a minimum of 5 to 10 years [7-9]. Development of high-risk stigmata or worrisome features during the surveillance period should prompt EUS-FNA or surgical evaluation. For patients with a previous history of surgery for IPMN or invasive MCN without residual disease, continued surveillance is recommended, in view of the small yearly risk of pancreatic ductal adenocarcinoma of 0.7% to 0.9% [8]. For patients with known IPMN in the remnant pancreas, residual IPMN at the surgical margins, or new postoperative recurrence of IPMN, surveillance recommendations are less well defined [8]. To date, there is no evidence basis for the recommended size threshold to follow-up cysts. Based on limited clinical and published experience, a cyst <5 mm may require one follow-up CT or MRI at 2 years. Demonstrating stability at 2 years is sufficient to stop surveillance.

3127236

acrac_3127236_5

Pancreatic Cyst

MRI Abdomen with MRCP The risk of malignant transformation of a pancreatic cyst is approximately 0.24% per year [9]. Once a pancreatic cyst has been characterized on a dedicated baseline examination, subsequent follow-up may be performed with either CT or MRI. There is no evidence to suggest that MRI is superior to CT for detection of new or developing worrisome features or pancreatic ductal adenocarcinoma [7], although modality concordance between baseline and follow-up examinations may facilitate comparison. The issue of whether IV contrast is necessary for MRI follow-up of pancreatic cysts remains controversial. Noncontrast MRI is associated with shorter scan times, with some sources citing little difference in the ability to detect evolving dysplastic changes compared with a contrast-enhanced study [7,33,34]. However, the use of IV contrast may permit detection of high-risk stigmata such as enhancing mural nodules. An abbreviated protocol MRI, including T2-weighted sequences and dual-phase (late arterial and portal venous phase) contrast-enhanced acquisitions, has been shown to be equivalent to standard pancreatic protocol MRI for detection of evolving dysplasia [7,34]. The frequency and duration of follow-up is controversial and depends on multiple factors, including patient age, family history of pancreatic ductal adenocarcinoma, cyst size, and whether or not there has been prior surgical resection of a pancreatic cyst. For patients with a nonspecific pancreatic cyst without a history of prior surgery, the surveillance plan will depend upon patient age and the cyst size. Follow-up intervals are generally in the range of 6 months to every 2 years for a minimum of 5 to 10 years [7-9]. Development of high-risk stigmata or worrisome features during the surveillance period should prompt EUS-FNA or surgical evaluation [8,9,14,26].

Pancreatic Cyst. MRI Abdomen with MRCP The risk of malignant transformation of a pancreatic cyst is approximately 0.24% per year [9]. Once a pancreatic cyst has been characterized on a dedicated baseline examination, subsequent follow-up may be performed with either CT or MRI. There is no evidence to suggest that MRI is superior to CT for detection of new or developing worrisome features or pancreatic ductal adenocarcinoma [7], although modality concordance between baseline and follow-up examinations may facilitate comparison. The issue of whether IV contrast is necessary for MRI follow-up of pancreatic cysts remains controversial. Noncontrast MRI is associated with shorter scan times, with some sources citing little difference in the ability to detect evolving dysplastic changes compared with a contrast-enhanced study [7,33,34]. However, the use of IV contrast may permit detection of high-risk stigmata such as enhancing mural nodules. An abbreviated protocol MRI, including T2-weighted sequences and dual-phase (late arterial and portal venous phase) contrast-enhanced acquisitions, has been shown to be equivalent to standard pancreatic protocol MRI for detection of evolving dysplasia [7,34]. The frequency and duration of follow-up is controversial and depends on multiple factors, including patient age, family history of pancreatic ductal adenocarcinoma, cyst size, and whether or not there has been prior surgical resection of a pancreatic cyst. For patients with a nonspecific pancreatic cyst without a history of prior surgery, the surveillance plan will depend upon patient age and the cyst size. Follow-up intervals are generally in the range of 6 months to every 2 years for a minimum of 5 to 10 years [7-9]. Development of high-risk stigmata or worrisome features during the surveillance period should prompt EUS-FNA or surgical evaluation [8,9,14,26].

3127236

acrac_3158166_0

Supplemental Breast Cancer Screening Based on Breast Density

Although overall sensitivity of mammography is in the range of 70% to 85%, the sensitivity can vary significantly with breast density [5-7]. The mammographic sensitivity is higher in women with fatty breast parenchyma; however, the sensitivity may decrease to as low as 30% in women with dense breast tissue [5-9]. Although the overall performance of digital mammography is similar to film screen mammography, digital mammography has better performance in specific subgroups, such as in women with dense breasts [8,9]. In addition, even with regular mammographic screening, the interval cancer rate may be as high as 30% [10-13]. Given the limitations of mammography, supplemental screening has been advocated for women with dense breast tissue. In order to bring uniformity to the language, a federal law passed in February 2019, enables the FDA to develop a statement that the effect of breast density on mammographic sensitivity be included in all mammography report. aPerelman School of Medicine of the University of Pennsylvania, Philadelphia, PennsylvaniabPanel Chair, Boston University School of Medicine, Boston, Massachusetts. cPanel Vice-Chair, New York University School of Medicine, New York, New York. dBoston University Schools of Medicine and Public Health, Boston, Massachusetts. eYale School of Medicine, New Haven, Connecticut; Society of Surgical Oncology. fThomas Jefferson University Hospital, Robbinsville, New Jersey; American College of Obstetricians and Gynecologists. gAlpert Medical School of Brown University, Providence, Rhode Island. hFeinberg School of Medicine, Northwestern University, Chicago, Illinois; American College of Physicians. iUniversity of California San Francisco, San Francisco, California. jSutter Medical Group and Sutter Cancer Center, Roseville, California. kThe University of Texas MD Anderson Cancer Center, Houston, Texas. lEmory University Hospital, Atlanta, Georgia. mSanford Health of Northern Minnesota, Bemidji, Minnesota.

Supplemental Breast Cancer Screening Based on Breast Density. Although overall sensitivity of mammography is in the range of 70% to 85%, the sensitivity can vary significantly with breast density [5-7]. The mammographic sensitivity is higher in women with fatty breast parenchyma; however, the sensitivity may decrease to as low as 30% in women with dense breast tissue [5-9]. Although the overall performance of digital mammography is similar to film screen mammography, digital mammography has better performance in specific subgroups, such as in women with dense breasts [8,9]. In addition, even with regular mammographic screening, the interval cancer rate may be as high as 30% [10-13]. Given the limitations of mammography, supplemental screening has been advocated for women with dense breast tissue. In order to bring uniformity to the language, a federal law passed in February 2019, enables the FDA to develop a statement that the effect of breast density on mammographic sensitivity be included in all mammography report. aPerelman School of Medicine of the University of Pennsylvania, Philadelphia, PennsylvaniabPanel Chair, Boston University School of Medicine, Boston, Massachusetts. cPanel Vice-Chair, New York University School of Medicine, New York, New York. dBoston University Schools of Medicine and Public Health, Boston, Massachusetts. eYale School of Medicine, New Haven, Connecticut; Society of Surgical Oncology. fThomas Jefferson University Hospital, Robbinsville, New Jersey; American College of Obstetricians and Gynecologists. gAlpert Medical School of Brown University, Providence, Rhode Island. hFeinberg School of Medicine, Northwestern University, Chicago, Illinois; American College of Physicians. iUniversity of California San Francisco, San Francisco, California. jSutter Medical Group and Sutter Cancer Center, Roseville, California. kThe University of Texas MD Anderson Cancer Center, Houston, Texas. lEmory University Hospital, Atlanta, Georgia. mSanford Health of Northern Minnesota, Bemidji, Minnesota.

3158166

acrac_3158166_1

Supplemental Breast Cancer Screening Based on Breast Density

nUniversity of Washington, Seattle, Washington. oDonald and Barbara Zucker School of Medicine at Hofstra/Northwell, Manhasset, New York. pNorthShore University HealthSystem, Evanston, Illinois; American College of Surgeons. qSpecialty Chair, NYU Clinical Cancer Center, New York, New York. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] Supplemental Screening Based on Breast Density Discussion of Procedures by Variant Variant 1: Supplemental breast cancer screening. Average-risk females with nondense breasts. Mammography remains the only validated screening tool for breast cancer screening. Despite widespread mammographic screening, breast cancer represents the leading cause of cancer mortality in women. Although multiple studies have demonstrated improved survival and reduction in breast cancer mortality by up to 30% with regular mammographic screening, there continues to be approximately 40,000 breast cancer deaths annually [10- 13]. Women who have <15% lifetime risk are considered to be at average risk [15]. In women with nondense breast tissue, the sensitivity of mammography is high [5]. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18].

Supplemental Breast Cancer Screening Based on Breast Density. nUniversity of Washington, Seattle, Washington. oDonald and Barbara Zucker School of Medicine at Hofstra/Northwell, Manhasset, New York. pNorthShore University HealthSystem, Evanston, Illinois; American College of Surgeons. qSpecialty Chair, NYU Clinical Cancer Center, New York, New York. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] Supplemental Screening Based on Breast Density Discussion of Procedures by Variant Variant 1: Supplemental breast cancer screening. Average-risk females with nondense breasts. Mammography remains the only validated screening tool for breast cancer screening. Despite widespread mammographic screening, breast cancer represents the leading cause of cancer mortality in women. Although multiple studies have demonstrated improved survival and reduction in breast cancer mortality by up to 30% with regular mammographic screening, there continues to be approximately 40,000 breast cancer deaths annually [10- 13]. Women who have <15% lifetime risk are considered to be at average risk [15]. In women with nondense breast tissue, the sensitivity of mammography is high [5]. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18].

3158166

acrac_3158166_2

Supplemental Breast Cancer Screening Based on Breast Density

Mammography With IV Contrast CEDM combines the techniques of conventional mammography with administration of IV contrast, thus leveraging functional imaging by assessing for lesion vascularity. A dual-energy technique is utilized to acquire the images in the conventional craniocaudal and mediolateral oblique projections. The acquired data are processed and produce a low-energy image and a diagnostic recombined image. There is limited but emerging literature regarding the use of CEDM in the screening setting [26,27]. However, in the diagnostic setting, CEDM has been shown to demonstrate improved sensitivity and specificity over 2-D mammography [28-31]. The greatest improvement in the sensitivity and specificity is seen in women with dense breast tissue [29,30]. However, at this time there is no relevant literature regarding the use of mammography with IV contrast for supplemental screening in average-risk women with nondense breasts. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in average-risk women with nondense breasts. MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in average-risk women with nondense breasts. MRI Breast Without and With IV Contrast There are limited data regarding screening average-risk women with breast MRI with and without IV contrast. In a prospective observational trial, after negative mammographic screening, Kuhl et al [33] reported an additional CDR of 15.5/1,000 with MRI screening in average-risk women across all densities. However, the authors did not analyze the added CDR by breast density. FDG-PET Breast Dedicated There is limited relevant literature regarding the use of FDG-PET breast dedicated for supplemental screening in average-risk women with nondense breasts. This is not currently widely used in clinical practice.

Supplemental Breast Cancer Screening Based on Breast Density. Mammography With IV Contrast CEDM combines the techniques of conventional mammography with administration of IV contrast, thus leveraging functional imaging by assessing for lesion vascularity. A dual-energy technique is utilized to acquire the images in the conventional craniocaudal and mediolateral oblique projections. The acquired data are processed and produce a low-energy image and a diagnostic recombined image. There is limited but emerging literature regarding the use of CEDM in the screening setting [26,27]. However, in the diagnostic setting, CEDM has been shown to demonstrate improved sensitivity and specificity over 2-D mammography [28-31]. The greatest improvement in the sensitivity and specificity is seen in women with dense breast tissue [29,30]. However, at this time there is no relevant literature regarding the use of mammography with IV contrast for supplemental screening in average-risk women with nondense breasts. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in average-risk women with nondense breasts. MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in average-risk women with nondense breasts. MRI Breast Without and With IV Contrast There are limited data regarding screening average-risk women with breast MRI with and without IV contrast. In a prospective observational trial, after negative mammographic screening, Kuhl et al [33] reported an additional CDR of 15.5/1,000 with MRI screening in average-risk women across all densities. However, the authors did not analyze the added CDR by breast density. FDG-PET Breast Dedicated There is limited relevant literature regarding the use of FDG-PET breast dedicated for supplemental screening in average-risk women with nondense breasts. This is not currently widely used in clinical practice.

3158166

acrac_3158166_3

Supplemental Breast Cancer Screening Based on Breast Density

Sestamibi MBI There is limited relevant literature regarding the use of Tc-99m sestamibi MBI for supplemental screening in average-risk women with nondense breasts. This is not currently widely used in clinical practice. Variant 2: Supplemental breast cancer screening. Intermediate-risk females with nondense breasts. Women at intermediate risk for breast cancer are defined as having a 15% to 20% lifetime risk [15]. Although there are clear screening guidelines for women with >20% lifetime risk, the screening guidelines have not clearly been defined for women who are at intermediate risk. Women in this category may include patients who have been diagnosed with lobular neoplasia, atypical ductal hyperplasia, previous history of breast cancer, or have a family history of breast cancer without known genetic mutations such as breast cancer gene (BRCA)1/2. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18]. Supplemental Screening Based on Breast Density In women with a personal history of breast cancer, the supplemental CDR of screening ultrasound (US) has been reported to be 2.88/1,000 [34]. There was no difference in the CDR based on breast density or age. However, the authors reported an interval cancer rate of 1.5/1,000, which was higher in women who were <50 years of age and in those with dense breast tissue, indicating the failure of screening US in these 2 subgroups.

Supplemental Breast Cancer Screening Based on Breast Density. Sestamibi MBI There is limited relevant literature regarding the use of Tc-99m sestamibi MBI for supplemental screening in average-risk women with nondense breasts. This is not currently widely used in clinical practice. Variant 2: Supplemental breast cancer screening. Intermediate-risk females with nondense breasts. Women at intermediate risk for breast cancer are defined as having a 15% to 20% lifetime risk [15]. Although there are clear screening guidelines for women with >20% lifetime risk, the screening guidelines have not clearly been defined for women who are at intermediate risk. Women in this category may include patients who have been diagnosed with lobular neoplasia, atypical ductal hyperplasia, previous history of breast cancer, or have a family history of breast cancer without known genetic mutations such as breast cancer gene (BRCA)1/2. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18]. Supplemental Screening Based on Breast Density In women with a personal history of breast cancer, the supplemental CDR of screening ultrasound (US) has been reported to be 2.88/1,000 [34]. There was no difference in the CDR based on breast density or age. However, the authors reported an interval cancer rate of 1.5/1,000, which was higher in women who were <50 years of age and in those with dense breast tissue, indicating the failure of screening US in these 2 subgroups.

3158166

acrac_3158166_4

Supplemental Breast Cancer Screening Based on Breast Density

Cortesi et al [35] evaluated the efficacy of biennial screening US examination in women who were BRCA mutation carriers, high-risk (non-BRCA1/2), and intermediate-risk patients. Overall, MRI had sensitivity of 93.7%, followed by mammography with sensitivity of 55.0% and US with 29.4% sensitivity. In the nondense breast, the sensitivity of mammography was 82.5% versus 10% for US. In the dense breast, the sensitivity of mammography was 50% versus 42.6% for US. Sensitivity analysis by risk level was also performed. The US sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 22.7%, 24.5%, and 33.6%, respectively. The mammographic sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 25.0%, 66.4%, and 56.6%, respectively. Only the BRCA1/2 mutation carriers underwent MRI screening, which demonstrated a sensitivity of 93.7%. The authors did not analyze the efficacy of US screening based on both density and risk. Mammography With IV Contrast CEDM combines the techniques of conventional mammography with administration of IV contrast, thus leveraging functional imaging by assessing for lesion vascularity. A dual-energy technique is used to acquire the images in the conventional craniocaudal and mediolateral oblique projections. The acquired data are processed, producing a low- energy image and a diagnostic recombined image. There is limited but emerging literature regarding the use of CEDM in the screening setting. However, in the diagnostic setting, CEDM has been shown to demonstrate improved sensitivity and specificity over 2-D mammography [28-31]. The greatest improvement in the sensitivity and specificity is seen in women with dense breast tissue [29,30]. However, at this time, there is no relevant literature regarding the use of mammography with IV contrast for supplemental screening in intermediate-risk women with nondense breasts.

Supplemental Breast Cancer Screening Based on Breast Density. Cortesi et al [35] evaluated the efficacy of biennial screening US examination in women who were BRCA mutation carriers, high-risk (non-BRCA1/2), and intermediate-risk patients. Overall, MRI had sensitivity of 93.7%, followed by mammography with sensitivity of 55.0% and US with 29.4% sensitivity. In the nondense breast, the sensitivity of mammography was 82.5% versus 10% for US. In the dense breast, the sensitivity of mammography was 50% versus 42.6% for US. Sensitivity analysis by risk level was also performed. The US sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 22.7%, 24.5%, and 33.6%, respectively. The mammographic sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 25.0%, 66.4%, and 56.6%, respectively. Only the BRCA1/2 mutation carriers underwent MRI screening, which demonstrated a sensitivity of 93.7%. The authors did not analyze the efficacy of US screening based on both density and risk. Mammography With IV Contrast CEDM combines the techniques of conventional mammography with administration of IV contrast, thus leveraging functional imaging by assessing for lesion vascularity. A dual-energy technique is used to acquire the images in the conventional craniocaudal and mediolateral oblique projections. The acquired data are processed, producing a low- energy image and a diagnostic recombined image. There is limited but emerging literature regarding the use of CEDM in the screening setting. However, in the diagnostic setting, CEDM has been shown to demonstrate improved sensitivity and specificity over 2-D mammography [28-31]. The greatest improvement in the sensitivity and specificity is seen in women with dense breast tissue [29,30]. However, at this time, there is no relevant literature regarding the use of mammography with IV contrast for supplemental screening in intermediate-risk women with nondense breasts.

3158166

acrac_3158166_5

Supplemental Breast Cancer Screening Based on Breast Density

MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in intermediate-risk women with nondense breasts. MRI Breast Without and With IV Contrast Abbreviated AB-MRI performed with IV contrast is an abbreviated breast MRI examination. It is similar to a full MRI examination yet does not have a standard protocol; however, at minimum, it must include a precontrast and one postcontrast sequence. A T2-weighted sequence may also be included. There are currently limited data on screening women with nondense breasts at intermediate lifetime risk with AB-MRI. In 2 retrospective reader studies, in women recently diagnosed with unifocal breast cancer, the sensitivity of AB-MRI was comparable with the full protocol [36,37]. When the performance of AB-MRI was compared with screening US and mammography, there were 12 cancers in 12 women (CDR 15/1,000), 7 of which were not detected on WBUS and mammography [38]. In a prospective observational study of 443 women with mild to moderately elevated lifetime risk for breast cancer, AB-MRI had a similar diagnostic accuracy as the full MRI protocol [39]. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in intermediate-risk women with nondense breasts. MRI Breast Without and With IV Contrast There is some relevant literature supporting the use of MRI breast without and with IV contrast for supplemental screening in intermediate-risk women, specifically in women with a history of lobular carcinoma in situ or a personal history of breast cancer, although these studies included all breast densities [40-42].

Supplemental Breast Cancer Screening Based on Breast Density. MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in intermediate-risk women with nondense breasts. MRI Breast Without and With IV Contrast Abbreviated AB-MRI performed with IV contrast is an abbreviated breast MRI examination. It is similar to a full MRI examination yet does not have a standard protocol; however, at minimum, it must include a precontrast and one postcontrast sequence. A T2-weighted sequence may also be included. There are currently limited data on screening women with nondense breasts at intermediate lifetime risk with AB-MRI. In 2 retrospective reader studies, in women recently diagnosed with unifocal breast cancer, the sensitivity of AB-MRI was comparable with the full protocol [36,37]. When the performance of AB-MRI was compared with screening US and mammography, there were 12 cancers in 12 women (CDR 15/1,000), 7 of which were not detected on WBUS and mammography [38]. In a prospective observational study of 443 women with mild to moderately elevated lifetime risk for breast cancer, AB-MRI had a similar diagnostic accuracy as the full MRI protocol [39]. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in intermediate-risk women with nondense breasts. MRI Breast Without and With IV Contrast There is some relevant literature supporting the use of MRI breast without and with IV contrast for supplemental screening in intermediate-risk women, specifically in women with a history of lobular carcinoma in situ or a personal history of breast cancer, although these studies included all breast densities [40-42].

3158166

acrac_3158166_6

Supplemental Breast Cancer Screening Based on Breast Density

At the time of this writing, the American Cancer Society is currently re-reviewing the literature regarding intermediate-risk women; however, its current stance, last updated in 2007, states there is insufficient evidence to formulate a recommendation in this group [15]. As of 2018, ACR recommends annual surveillance MRI in women with dense breasts and a personal history of breast cancer as well as in women who were diagnosed before age 50 [43]. The ACR suggests that MRI should be considered in the following categories: in women with personal histories of breast cancer and who do not fit the 2 previously stated categories and in women with atypical ductal hyperplasia, atypical lobular hyperplasia, and lobular carcinoma in situ [43]. Supplemental Screening Based on Breast Density FDG-PET Breast Dedicated There is limited relevant literature regarding the use of FDG-PET breast dedicated for supplemental screening in intermediate-risk women with nondense breasts. This is not currently widely used in clinical practice. Sestamibi MBI There is limited relevant literature regarding the use of Tc-99m sestamibi MBI for supplemental screening in intermediate-risk women with nondense breasts. This is not currently widely used in clinical practice. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18]. Cortesi et al [35] evaluated the efficacy of biannual screening US examination in women who were BRCA mutation carriers, high-risk (non-BRCA1/2), and intermediate-risk patients.

Supplemental Breast Cancer Screening Based on Breast Density. At the time of this writing, the American Cancer Society is currently re-reviewing the literature regarding intermediate-risk women; however, its current stance, last updated in 2007, states there is insufficient evidence to formulate a recommendation in this group [15]. As of 2018, ACR recommends annual surveillance MRI in women with dense breasts and a personal history of breast cancer as well as in women who were diagnosed before age 50 [43]. The ACR suggests that MRI should be considered in the following categories: in women with personal histories of breast cancer and who do not fit the 2 previously stated categories and in women with atypical ductal hyperplasia, atypical lobular hyperplasia, and lobular carcinoma in situ [43]. Supplemental Screening Based on Breast Density FDG-PET Breast Dedicated There is limited relevant literature regarding the use of FDG-PET breast dedicated for supplemental screening in intermediate-risk women with nondense breasts. This is not currently widely used in clinical practice. Sestamibi MBI There is limited relevant literature regarding the use of Tc-99m sestamibi MBI for supplemental screening in intermediate-risk women with nondense breasts. This is not currently widely used in clinical practice. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18]. Cortesi et al [35] evaluated the efficacy of biannual screening US examination in women who were BRCA mutation carriers, high-risk (non-BRCA1/2), and intermediate-risk patients.

3158166

acrac_3158166_7

Supplemental Breast Cancer Screening Based on Breast Density

Overall, MRI had sensitivity of 93.7%, which was followed by mammography (55.0%), and US (29.4%). In the nondense breast, the sensitivity of mammography was 82.5% versus 10% for US. In the dense breast, the sensitivity of mammography was 50% versus 42.6% for US. Sensitivity analysis by risk level was also performed. The US sensitivitities for BRCA1/2, high-risk (non- BRCA1/2), and intermediate-risk patients were 22.7%, 24.5%, and 33.6%, respectively. The mammographic sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 25.0%, 66.4%, and 56.6%, respectively. Only the BRCA1/2 mutation carriers underwent MRI screening, which demonstrated a sensitivity of 93.7%. The authors did not analyze the efficacy of US screening based on both density and risk. The addition of WBUS to mammography increases the CDR in high-risk women. In a surveillance cohort study of 529 women with elevated lifetime risk, the sensitivity of mammography, WBUS, and the 2 modalities combined was 33%, 40%, and 49%, respectively [44]. However, subgroup analysis was not performed by breast density. In the same population, MRI had a sensitivity of 91% [44]. In the ACRIN 6666 trial, after 3 rounds of screening mammography and screening WBUS in women with dense breast tissue at intermediate to elevated lifetime risk, the sensitivity, specificity, and positive predictive value (PPV3) of mammography was 0.52, 0.91, and 0.38, Supplemental Screening Based on Breast Density respectively. The addition of US to mammography increased the sensitivity (0.76) but decreased the specificity (0.84) and the PPV3 (0.38) [45]. In a prospective cohort trial of 687 high-risk women, the cancer yield of mammography alone was 5.4/1,000 and increased to 7.7/1,000 with the addition of US [46].

Supplemental Breast Cancer Screening Based on Breast Density. Overall, MRI had sensitivity of 93.7%, which was followed by mammography (55.0%), and US (29.4%). In the nondense breast, the sensitivity of mammography was 82.5% versus 10% for US. In the dense breast, the sensitivity of mammography was 50% versus 42.6% for US. Sensitivity analysis by risk level was also performed. The US sensitivitities for BRCA1/2, high-risk (non- BRCA1/2), and intermediate-risk patients were 22.7%, 24.5%, and 33.6%, respectively. The mammographic sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 25.0%, 66.4%, and 56.6%, respectively. Only the BRCA1/2 mutation carriers underwent MRI screening, which demonstrated a sensitivity of 93.7%. The authors did not analyze the efficacy of US screening based on both density and risk. The addition of WBUS to mammography increases the CDR in high-risk women. In a surveillance cohort study of 529 women with elevated lifetime risk, the sensitivity of mammography, WBUS, and the 2 modalities combined was 33%, 40%, and 49%, respectively [44]. However, subgroup analysis was not performed by breast density. In the same population, MRI had a sensitivity of 91% [44]. In the ACRIN 6666 trial, after 3 rounds of screening mammography and screening WBUS in women with dense breast tissue at intermediate to elevated lifetime risk, the sensitivity, specificity, and positive predictive value (PPV3) of mammography was 0.52, 0.91, and 0.38, Supplemental Screening Based on Breast Density respectively. The addition of US to mammography increased the sensitivity (0.76) but decreased the specificity (0.84) and the PPV3 (0.38) [45]. In a prospective cohort trial of 687 high-risk women, the cancer yield of mammography alone was 5.4/1,000 and increased to 7.7/1,000 with the addition of US [46].

3158166

acrac_3158166_8

Supplemental Breast Cancer Screening Based on Breast Density

Mammography With IV Contrast CEDM combines the techniques of conventional mammography with administration of IV contrast, thus leveraging functional imaging by assessing for lesion vascularity. A dual-energy technique is utilized to acquire the images in the conventional craniocaudal and mediolateral oblique projections. The acquired data are processed, producing a low-energy image and a diagnostic recombined image. There is limited literature regarding the use of CEDM in the screening setting. However, in the diagnostic setting, CEDM has been shown to demonstrate improved sensitivity and specificity over 2-D mammography [28-31]. The greatest improvement in the sensitivity and specificity is seen in women with dense breast tissue [29,30]. At this time, there is limited but emerging literature regarding the use of mammography with IV contrast for supplemental screening in high-risk women with nondense breasts. Jochelson et al [47] screened 318 high-risk women using both CEDM and MRI [47]. Both techniques detected carcinomas not visualized on mammography, 2 using CEDM and 3 using MRI. MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in high-risk women with nondense breasts. MRI Breast Without and With IV Contrast Abbreviated AB-MRI performed with IV contrast is an abbreviated breast MRI examination. It is similar to a full MRI examination yet does not have a standard protocol; however, at minimum, it must include a precontrast and one postcontrast sequence. A T2-weighted sequence may also be included. There is limited relevant literature regarding the use of AB-MRI breast without and with IV contrast in high-risk women with nondense breasts. In 2 retrospective studies comparing the full diagnostic protocol with an abbreviated protocol in high-risk women, the authors found both protocols to have similar sensitivity [48,49].

Supplemental Breast Cancer Screening Based on Breast Density. Mammography With IV Contrast CEDM combines the techniques of conventional mammography with administration of IV contrast, thus leveraging functional imaging by assessing for lesion vascularity. A dual-energy technique is utilized to acquire the images in the conventional craniocaudal and mediolateral oblique projections. The acquired data are processed, producing a low-energy image and a diagnostic recombined image. There is limited literature regarding the use of CEDM in the screening setting. However, in the diagnostic setting, CEDM has been shown to demonstrate improved sensitivity and specificity over 2-D mammography [28-31]. The greatest improvement in the sensitivity and specificity is seen in women with dense breast tissue [29,30]. At this time, there is limited but emerging literature regarding the use of mammography with IV contrast for supplemental screening in high-risk women with nondense breasts. Jochelson et al [47] screened 318 high-risk women using both CEDM and MRI [47]. Both techniques detected carcinomas not visualized on mammography, 2 using CEDM and 3 using MRI. MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in high-risk women with nondense breasts. MRI Breast Without and With IV Contrast Abbreviated AB-MRI performed with IV contrast is an abbreviated breast MRI examination. It is similar to a full MRI examination yet does not have a standard protocol; however, at minimum, it must include a precontrast and one postcontrast sequence. A T2-weighted sequence may also be included. There is limited relevant literature regarding the use of AB-MRI breast without and with IV contrast in high-risk women with nondense breasts. In 2 retrospective studies comparing the full diagnostic protocol with an abbreviated protocol in high-risk women, the authors found both protocols to have similar sensitivity [48,49].

3158166

acrac_3158166_9

Supplemental Breast Cancer Screening Based on Breast Density

However, neither study evaluated the CDR by breast density. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in high-risk women with nondense breasts. FDG-PET Breast Dedicated There is limited relevant literature regarding the use of FDG-PET breast dedicated for supplemental screening in high-risk women with nondense breasts. This is not currently widely used in clinical practice. Sestamibi MBI There is limited relevant literature regarding the use of Tc-99m sestamibi MBI for supplemental screening in high- risk women with nondense breasts. This is not currently widely used in clinical practice. Variant 4: Supplemental breast cancer screening. Average-risk females with dense breasts. Mammography remains the only validated screening tool for breast cancer screening. Despite widespread mammographic screening, breast cancer represents the leading cause of cancer mortality in women. Although multiple studies have demonstrated improved survival and reduction in breast cancer mortality by up to 30% with regular mammographic screening, there continues to be approximately 40,000 breast cancer deaths annually [10- 13]. Women who have <15% lifetime risk are considered to be at average risk [15]. The greatest improvement in the CDR with DBT is seen in women with dense breast tissue [21,22,57,58]. Although the TOMMY trial did not reach statistical significance across all breast densities, in women with >50% breast density, statistical significance was achieved with the sensitivity of 2-D mammography plus DBT reaching 93% versus 86% for 2-D mammography alone [21,22]. In a meta-analysis of 16 studies evaluating women with dense breasts, addition of DBT improved the CDR, compared with 2-D mammography alone, in both diagnostic (relative risk [RR]: 1.16) and the screening (RR: 1.33) settings [58].

Supplemental Breast Cancer Screening Based on Breast Density. However, neither study evaluated the CDR by breast density. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in high-risk women with nondense breasts. FDG-PET Breast Dedicated There is limited relevant literature regarding the use of FDG-PET breast dedicated for supplemental screening in high-risk women with nondense breasts. This is not currently widely used in clinical practice. Sestamibi MBI There is limited relevant literature regarding the use of Tc-99m sestamibi MBI for supplemental screening in high- risk women with nondense breasts. This is not currently widely used in clinical practice. Variant 4: Supplemental breast cancer screening. Average-risk females with dense breasts. Mammography remains the only validated screening tool for breast cancer screening. Despite widespread mammographic screening, breast cancer represents the leading cause of cancer mortality in women. Although multiple studies have demonstrated improved survival and reduction in breast cancer mortality by up to 30% with regular mammographic screening, there continues to be approximately 40,000 breast cancer deaths annually [10- 13]. Women who have <15% lifetime risk are considered to be at average risk [15]. The greatest improvement in the CDR with DBT is seen in women with dense breast tissue [21,22,57,58]. Although the TOMMY trial did not reach statistical significance across all breast densities, in women with >50% breast density, statistical significance was achieved with the sensitivity of 2-D mammography plus DBT reaching 93% versus 86% for 2-D mammography alone [21,22]. In a meta-analysis of 16 studies evaluating women with dense breasts, addition of DBT improved the CDR, compared with 2-D mammography alone, in both diagnostic (relative risk [RR]: 1.16) and the screening (RR: 1.33) settings [58].

3158166

acrac_3158166_10

Supplemental Breast Cancer Screening Based on Breast Density

In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18]. At a high-volume screening center, when women with heterogeneously or extremely dense breast tissue were offered automated 3-D US screening in addition to full-field digital mammography (FFDM), the added CDR rose by 2.4/1,000 screened (6.6/1,000 automated 3-D US screening plus FFDM versus 4.2/1,000 for FFDM alone) [59]. However, the risk level of the population was not defined other than patients with a personal history of breast cancer were excluded, and 3.5% of the patients reported a family history of breast cancer. Buchberger et al [60] compared the performance of screening mammography with and without screening US in average-risk women. In the subgroup of women with dense breasts, the CDR increased from 1.8/1,000 to 2.4/1,000, with the addition of screening US. However, the PPV2 decreased from 52.7/1,000 with mammography alone to 37.7/1,000 with mammography plus US. For the entire population, there were 28 interval cancer rates within a 12- month period after screening (0.42/1,000): 18 in women with dense breasts and 10 in women with nondense breasts. In the Japan Strategic Anti-cancer Randomized Trial, 72,998 asymptomatic women 40 to 49 years of age were randomized to mammographic screening alone or mammographic plus US screening [61]. The sensitivity and specificity in the mammogram arm were 77.0% and 91.4%, respectively. In comparison, the mammogram plus US arm had a higher sensitivity (91.1%) but lower specificity (87.7%).

Supplemental Breast Cancer Screening Based on Breast Density. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18]. At a high-volume screening center, when women with heterogeneously or extremely dense breast tissue were offered automated 3-D US screening in addition to full-field digital mammography (FFDM), the added CDR rose by 2.4/1,000 screened (6.6/1,000 automated 3-D US screening plus FFDM versus 4.2/1,000 for FFDM alone) [59]. However, the risk level of the population was not defined other than patients with a personal history of breast cancer were excluded, and 3.5% of the patients reported a family history of breast cancer. Buchberger et al [60] compared the performance of screening mammography with and without screening US in average-risk women. In the subgroup of women with dense breasts, the CDR increased from 1.8/1,000 to 2.4/1,000, with the addition of screening US. However, the PPV2 decreased from 52.7/1,000 with mammography alone to 37.7/1,000 with mammography plus US. For the entire population, there were 28 interval cancer rates within a 12- month period after screening (0.42/1,000): 18 in women with dense breasts and 10 in women with nondense breasts. In the Japan Strategic Anti-cancer Randomized Trial, 72,998 asymptomatic women 40 to 49 years of age were randomized to mammographic screening alone or mammographic plus US screening [61]. The sensitivity and specificity in the mammogram arm were 77.0% and 91.4%, respectively. In comparison, the mammogram plus US arm had a higher sensitivity (91.1%) but lower specificity (87.7%).

3158166

acrac_3158166_11

Supplemental Breast Cancer Screening Based on Breast Density

The interval cancer rate was halved in the arm that received US screening, from the baseline level of 0.10% to 0.05%. Using Breast Cancer Surveillance Consortium data, Lee et al [62] assessed the performance of the addition of screening US in women with dense breasts in the community. The CDR of mammography plus US was 5.4/1,000 versus 5.5/1,000 for mammography alone. The false-positive biopsy rate and PPV of biopsy recommendations for Supplemental Screening Based on Breast Density mammography plus US were 52.0/1,000 and 9.5%, respectively, compared with 22.2/1,000 and 21.4%, respectively, for mammography alone. The interval cancer rate did not differ significantly in the 2 arms: the mammography plus US arm was 1.5/1,000 versus 1.9/1,000 for mammography alone. Utilizing the registry data and data from the literature, Sprague et al [63] used simulation models to assess the outcomes of supplemental US screening after negative screening mammography in women with dense breasts. Per 1,000 women screened, the authors concluded there would be 0.36 breast cancer deaths averted, 354 additional biopsy recommendations, and 1.7 quality-adjusted life-years gained at a cost $325,000 for each quality-adjusted life-year gained [63]. Although the supplemental screening debate was initiated partly because of the limitations of screening mammography in women with dense breast tissue, there is some evidence to suggest that breast density alone may not be sufficient reason to recommend supplemental screening. One way to assess the failure of mammography is by determining the interval cancer rate in the study population. Using the Breast Cancer Surveillance Consortium data, the women most likely to present with interval cancers were women with an elevated 5-year risk for breast cancer >1.67% and with dense breast tissue, representing approximately 24% of all women with dense breasts [64].

Supplemental Breast Cancer Screening Based on Breast Density. The interval cancer rate was halved in the arm that received US screening, from the baseline level of 0.10% to 0.05%. Using Breast Cancer Surveillance Consortium data, Lee et al [62] assessed the performance of the addition of screening US in women with dense breasts in the community. The CDR of mammography plus US was 5.4/1,000 versus 5.5/1,000 for mammography alone. The false-positive biopsy rate and PPV of biopsy recommendations for Supplemental Screening Based on Breast Density mammography plus US were 52.0/1,000 and 9.5%, respectively, compared with 22.2/1,000 and 21.4%, respectively, for mammography alone. The interval cancer rate did not differ significantly in the 2 arms: the mammography plus US arm was 1.5/1,000 versus 1.9/1,000 for mammography alone. Utilizing the registry data and data from the literature, Sprague et al [63] used simulation models to assess the outcomes of supplemental US screening after negative screening mammography in women with dense breasts. Per 1,000 women screened, the authors concluded there would be 0.36 breast cancer deaths averted, 354 additional biopsy recommendations, and 1.7 quality-adjusted life-years gained at a cost $325,000 for each quality-adjusted life-year gained [63]. Although the supplemental screening debate was initiated partly because of the limitations of screening mammography in women with dense breast tissue, there is some evidence to suggest that breast density alone may not be sufficient reason to recommend supplemental screening. One way to assess the failure of mammography is by determining the interval cancer rate in the study population. Using the Breast Cancer Surveillance Consortium data, the women most likely to present with interval cancers were women with an elevated 5-year risk for breast cancer >1.67% and with dense breast tissue, representing approximately 24% of all women with dense breasts [64].

3158166

acrac_3158166_12

Supplemental Breast Cancer Screening Based on Breast Density

Mammography With IV Contrast CEDM combines the techniques of conventional mammography with administration of IV contrast, thus leveraging functional imaging by assessing for lesion vascularity. A dual-energy technique is used to acquire the images in the conventional craniocaudal and mediolateral oblique projections. The acquired data are processed, producing a low- energy image and a diagnostic recombined image. There is limited but emerging literature regarding the use of CEDM in the screening setting. However, in the diagnostic setting, CEDM has been shown to demonstrate improved sensitivity and specificity over 2-D mammography [28-31]. The greatest improvement in the sensitivity and specificity is seen in women with dense breast tissue [29,30]. At this time, there is limited literature regarding the use of mammography with IV contrast for supplemental screening in average-risk women with dense breasts. MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in average-risk women with dense breasts. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in average-risk women with dense breasts. MRI Breast Without and With IV Contrast There are limited data regarding screening average-risk women with breast MRI with and without IV contrast. In a prospective observational trial, after negative mammographic screening, Kuhl et al [33] reported an additional CDR of 15.5/1,000 with MRI screening in average-risk women across all densities. However, the authors did not analyze the added CDR by breast density. FDG-PET Breast Dedicated There is limited relevant literature regarding the use of FDG-PET breast dedicated for supplemental screening in average-risk women with dense breasts [67]. This is not currently widely used in clinical practice.

Supplemental Breast Cancer Screening Based on Breast Density. Mammography With IV Contrast CEDM combines the techniques of conventional mammography with administration of IV contrast, thus leveraging functional imaging by assessing for lesion vascularity. A dual-energy technique is used to acquire the images in the conventional craniocaudal and mediolateral oblique projections. The acquired data are processed, producing a low- energy image and a diagnostic recombined image. There is limited but emerging literature regarding the use of CEDM in the screening setting. However, in the diagnostic setting, CEDM has been shown to demonstrate improved sensitivity and specificity over 2-D mammography [28-31]. The greatest improvement in the sensitivity and specificity is seen in women with dense breast tissue [29,30]. At this time, there is limited literature regarding the use of mammography with IV contrast for supplemental screening in average-risk women with dense breasts. MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in average-risk women with dense breasts. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in average-risk women with dense breasts. MRI Breast Without and With IV Contrast There are limited data regarding screening average-risk women with breast MRI with and without IV contrast. In a prospective observational trial, after negative mammographic screening, Kuhl et al [33] reported an additional CDR of 15.5/1,000 with MRI screening in average-risk women across all densities. However, the authors did not analyze the added CDR by breast density. FDG-PET Breast Dedicated There is limited relevant literature regarding the use of FDG-PET breast dedicated for supplemental screening in average-risk women with dense breasts [67]. This is not currently widely used in clinical practice.

3158166

acrac_3158166_13

Supplemental Breast Cancer Screening Based on Breast Density

Supplemental Screening Based on Breast Density Sestamibi MBI There is limited relevant literature regarding the use of Tc-99m sestamibi MBI for supplemental screening in average-risk women with dense breasts. This is not currently widely used in clinical practice. Variant 5: Supplemental breast cancer screening. Intermediate-risk females with dense breasts. Women at intermediate risk for breast cancer are defined as having a 15% to 20% lifetime risk [15]. Although there are clear screening guidelines for women with >20% lifetime risk, the screening guidelines have not been clearly defined for women who are at intermediate risk. Women in this category may include patients who have been diagnosed with lobular neoplasia, atypical ductal hyperplasia, previous history of breast cancer, or have a family history of breast cancer without known genetic mutations such as BRCA1/2. The greatest improvement in the CDR with DBT is seen in women with dense breast tissue [21,22,57,58]. Although the TOMMY trial did not reach statistical significance across all breast densities, in women with >50% breast density, statistical significance was achieved with the sensitivity of 2-D mammography plus DBT reaching 93% versus 86% for 2-D mammography alone [21,22]. In a meta-analysis of 16 studies evaluating women with dense breasts, DBT improved the CDR compared with 2-D mammography alone, in both the diagnostic (RR: 1.16) and the screening (RR: 1.33) settings [58]. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18].

Supplemental Breast Cancer Screening Based on Breast Density. Supplemental Screening Based on Breast Density Sestamibi MBI There is limited relevant literature regarding the use of Tc-99m sestamibi MBI for supplemental screening in average-risk women with dense breasts. This is not currently widely used in clinical practice. Variant 5: Supplemental breast cancer screening. Intermediate-risk females with dense breasts. Women at intermediate risk for breast cancer are defined as having a 15% to 20% lifetime risk [15]. Although there are clear screening guidelines for women with >20% lifetime risk, the screening guidelines have not been clearly defined for women who are at intermediate risk. Women in this category may include patients who have been diagnosed with lobular neoplasia, atypical ductal hyperplasia, previous history of breast cancer, or have a family history of breast cancer without known genetic mutations such as BRCA1/2. The greatest improvement in the CDR with DBT is seen in women with dense breast tissue [21,22,57,58]. Although the TOMMY trial did not reach statistical significance across all breast densities, in women with >50% breast density, statistical significance was achieved with the sensitivity of 2-D mammography plus DBT reaching 93% versus 86% for 2-D mammography alone [21,22]. In a meta-analysis of 16 studies evaluating women with dense breasts, DBT improved the CDR compared with 2-D mammography alone, in both the diagnostic (RR: 1.16) and the screening (RR: 1.33) settings [58]. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18].

3158166

acrac_3158166_14

Supplemental Breast Cancer Screening Based on Breast Density

In women with a personal history of breast cancer, the supplemental CDR of screening US has been reported to be 2.88/1,000 [34]. There was no difference in the CDR based on breast density or age. However, the authors reported an interval cancer rate of 1.5/1,000, which was higher in women who were <50 years of age and in those with dense breast tissue, indicating the failure of screening US in the 2 subgroups. Cortesi et al [35] evaluated the efficacy of biannual screening US examination in women who were BRCA mutation carriers, high-risk (non-BRCA1/2), and intermediate-risk patients. Overall, MRI had sensitivity of 93.7%, followed by mammography (55.0%) and US (29.4%). In the nondense breast, the sensitivity of mammography was 82.5% versus 10% for US. In the dense breast, the sensitivity of mammography was 50% versus 42.6% for US. Sensitivity analysis by risk level was also performed. The US sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and 12 Supplemental Screening Based on Breast Density intermediate-risk patients were 22.7%, 24.5%, and 33.6%, respectively. The mammographic sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 25.0%, 66.4%, and 56.6%, respectively. Only the BRCA1/2 mutation carriers underwent MRI screening, which demonstrated a sensitivity of 93.7%. The authors did not analyze the efficacy of US screening based on both density and risk. Although the supplemental screening debate was initiated partly because of the limitations of screening mammography in women with dense breast tissue, there is some evidence to suggest that breast density alone may not be sufficient reason to recommend supplemental screening. One way to assess the failure of mammography is by determining the interval cancer rate in the study population.

Supplemental Breast Cancer Screening Based on Breast Density. In women with a personal history of breast cancer, the supplemental CDR of screening US has been reported to be 2.88/1,000 [34]. There was no difference in the CDR based on breast density or age. However, the authors reported an interval cancer rate of 1.5/1,000, which was higher in women who were <50 years of age and in those with dense breast tissue, indicating the failure of screening US in the 2 subgroups. Cortesi et al [35] evaluated the efficacy of biannual screening US examination in women who were BRCA mutation carriers, high-risk (non-BRCA1/2), and intermediate-risk patients. Overall, MRI had sensitivity of 93.7%, followed by mammography (55.0%) and US (29.4%). In the nondense breast, the sensitivity of mammography was 82.5% versus 10% for US. In the dense breast, the sensitivity of mammography was 50% versus 42.6% for US. Sensitivity analysis by risk level was also performed. The US sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and 12 Supplemental Screening Based on Breast Density intermediate-risk patients were 22.7%, 24.5%, and 33.6%, respectively. The mammographic sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 25.0%, 66.4%, and 56.6%, respectively. Only the BRCA1/2 mutation carriers underwent MRI screening, which demonstrated a sensitivity of 93.7%. The authors did not analyze the efficacy of US screening based on both density and risk. Although the supplemental screening debate was initiated partly because of the limitations of screening mammography in women with dense breast tissue, there is some evidence to suggest that breast density alone may not be sufficient reason to recommend supplemental screening. One way to assess the failure of mammography is by determining the interval cancer rate in the study population.

3158166

acrac_3158166_15

Supplemental Breast Cancer Screening Based on Breast Density

Using Breast Cancer Surveillance Consortium data, the women most likely to present with interval cancers were women with an elevated 5-year risk for breast cancer >1.67% and with dense breast tissue, representing approximately 24% of all women with dense breasts [64]. Mammography With IV Contrast CEDM combines the techniques of conventional mammography with administration of IV contrast, thus leveraging functional imaging by assessing for lesion vascularity. A dual-energy technique is utilized to acquire the images in the conventional craniocaudal and mediolateral oblique projections. The acquired data are processed, producing a low-energy image and a diagnostic recombined image. There is limited literature regarding the use of CEDM in the screening setting. However, in the diagnostic setting, CEDM has been shown to demonstrate improved sensitivity and specificity over 2-D mammography [28-31]. The greatest improvement in the sensitivity and specificity is seen in women with dense breast tissue [29,30]. At this time, there is limited but emerging literature regarding the use of mammography with IV contrast for supplemental screening in intermediate-risk women with dense breasts. However, in women with dense breast tissue and given the limited sensitivity of mammography and the need for supplemental screening, CEDM may have a potential role; however, more data on CEDM in the screening setting in intermediate-risk women with dense breast tissue are needed. MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in intermediate-risk women with dense breasts. MRI Breast Without and With IV Contrast Abbreviated AB-MRI performed with IV contrast is an abbreviated breast MRI examination. It is similar to a full MRI examination yet does not have a standard protocol; however, at minimum, it must include a precontrast and one postcontrast sequence.

Supplemental Breast Cancer Screening Based on Breast Density. Using Breast Cancer Surveillance Consortium data, the women most likely to present with interval cancers were women with an elevated 5-year risk for breast cancer >1.67% and with dense breast tissue, representing approximately 24% of all women with dense breasts [64]. Mammography With IV Contrast CEDM combines the techniques of conventional mammography with administration of IV contrast, thus leveraging functional imaging by assessing for lesion vascularity. A dual-energy technique is utilized to acquire the images in the conventional craniocaudal and mediolateral oblique projections. The acquired data are processed, producing a low-energy image and a diagnostic recombined image. There is limited literature regarding the use of CEDM in the screening setting. However, in the diagnostic setting, CEDM has been shown to demonstrate improved sensitivity and specificity over 2-D mammography [28-31]. The greatest improvement in the sensitivity and specificity is seen in women with dense breast tissue [29,30]. At this time, there is limited but emerging literature regarding the use of mammography with IV contrast for supplemental screening in intermediate-risk women with dense breasts. However, in women with dense breast tissue and given the limited sensitivity of mammography and the need for supplemental screening, CEDM may have a potential role; however, more data on CEDM in the screening setting in intermediate-risk women with dense breast tissue are needed. MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in intermediate-risk women with dense breasts. MRI Breast Without and With IV Contrast Abbreviated AB-MRI performed with IV contrast is an abbreviated breast MRI examination. It is similar to a full MRI examination yet does not have a standard protocol; however, at minimum, it must include a precontrast and one postcontrast sequence.

3158166

acrac_3158166_16

Supplemental Breast Cancer Screening Based on Breast Density

A T2-weighted sequence may also be included. There are currently limited data on screening women with dense breasts at intermediate lifetime risk with AB-MRI. In 2 retrospective reader studies, in women recently diagnosed with unifocal breast cancer, the sensitivity of AB-MRI was comparable with the full protocol [36,37]. When the performance of AB-MRI was compared with screening US and mammography, 12 cancers in 12 women (CDR 15/1,000) were detected, 7 of which were not detected on WBUS and mammography [38]. In a prospective observational study of 443 women with mild to moderately elevated lifetime risk for breast cancer, AB- MRI had a similar diagnostic accuracy as the full MRI protocol [39]. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in intermediate-risk women with dense breasts. MRI Breast Without and With IV Contrast There is some relevant literature supporting the use of MRI breast without and with IV contrast for supplemental screening in intermediate-risk women, specifically in women with a history of lobular carcinoma in situ or a personal history of breast cancer, although these studies included all breast densities [40-42]. At the time of this writing, the American Cancer Society was re-reviewing the literature regarding intermediate-risk women; however, its current stance, last updated in 2007, states there is insufficient evidence to formulate a recommendation in this group [15]. As of 2018, the ACR recommends annual surveillance MRI in women with dense breasts and with a personal history of breast cancer, as well as in women who were diagnosed before age 50 [43].

Supplemental Breast Cancer Screening Based on Breast Density. A T2-weighted sequence may also be included. There are currently limited data on screening women with dense breasts at intermediate lifetime risk with AB-MRI. In 2 retrospective reader studies, in women recently diagnosed with unifocal breast cancer, the sensitivity of AB-MRI was comparable with the full protocol [36,37]. When the performance of AB-MRI was compared with screening US and mammography, 12 cancers in 12 women (CDR 15/1,000) were detected, 7 of which were not detected on WBUS and mammography [38]. In a prospective observational study of 443 women with mild to moderately elevated lifetime risk for breast cancer, AB- MRI had a similar diagnostic accuracy as the full MRI protocol [39]. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in intermediate-risk women with dense breasts. MRI Breast Without and With IV Contrast There is some relevant literature supporting the use of MRI breast without and with IV contrast for supplemental screening in intermediate-risk women, specifically in women with a history of lobular carcinoma in situ or a personal history of breast cancer, although these studies included all breast densities [40-42]. At the time of this writing, the American Cancer Society was re-reviewing the literature regarding intermediate-risk women; however, its current stance, last updated in 2007, states there is insufficient evidence to formulate a recommendation in this group [15]. As of 2018, the ACR recommends annual surveillance MRI in women with dense breasts and with a personal history of breast cancer, as well as in women who were diagnosed before age 50 [43].

3158166

acrac_3158166_17

Supplemental Breast Cancer Screening Based on Breast Density

The ACR suggests that MRI should be considered in the following categories: in women with personal histories of breast cancer and who do not fit the 2 previously stated categories as well as in women with atypical ductal hyperplasia, atypical lobular hyperplasia, and lobular carcinoma in situ [43]. FDG-PET Breast Dedicated There is limited relevant literature regarding the use of FDG-PET breast dedicated for supplemental screening in intermediate-risk women with dense breasts. This is not currently widely used in clinical practice. Supplemental Screening Based on Breast Density Sestamibi MBI There is limited relevant literature regarding the use of Tc-99m sestamibi MBI for supplemental screening in intermediate-risk women with dense breasts. This is not currently widely used in clinical practice. The greatest improvement in the CDR with DBT is seen in women with dense breast tissue [21,22,57,58]. Although the TOMMY trial did not reach statistical significance across all breast densities, in women with >50% breast density, statistical significance was achieved, with the sensitivity of 2-D mammography plus DBT reaching 93% versus 86% for 2-D mammography alone [21,22]. In a meta-analysis of 16 studies evaluating women with dense breasts, DBT improved the CDR compared with 2-D mammography alone in both the diagnostic (RR: 1.16) and the screening (RR: 1.33) settings [58]. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18].

Supplemental Breast Cancer Screening Based on Breast Density. The ACR suggests that MRI should be considered in the following categories: in women with personal histories of breast cancer and who do not fit the 2 previously stated categories as well as in women with atypical ductal hyperplasia, atypical lobular hyperplasia, and lobular carcinoma in situ [43]. FDG-PET Breast Dedicated There is limited relevant literature regarding the use of FDG-PET breast dedicated for supplemental screening in intermediate-risk women with dense breasts. This is not currently widely used in clinical practice. Supplemental Screening Based on Breast Density Sestamibi MBI There is limited relevant literature regarding the use of Tc-99m sestamibi MBI for supplemental screening in intermediate-risk women with dense breasts. This is not currently widely used in clinical practice. The greatest improvement in the CDR with DBT is seen in women with dense breast tissue [21,22,57,58]. Although the TOMMY trial did not reach statistical significance across all breast densities, in women with >50% breast density, statistical significance was achieved, with the sensitivity of 2-D mammography plus DBT reaching 93% versus 86% for 2-D mammography alone [21,22]. In a meta-analysis of 16 studies evaluating women with dense breasts, DBT improved the CDR compared with 2-D mammography alone in both the diagnostic (RR: 1.16) and the screening (RR: 1.33) settings [58]. In addition to the increase in the CDR, another benefit of adding DBT to 2-D mammography is the reduction in the recall rate [16-19]. In a single-center screening program, Sharpe et al [18] reported a reduction in the recall rate by 18.8%. In the prospective Oslo Tomosynthesis Screening Trial, the recall rate was reduced from 6.7/1,000 to 3.6/1,000 [19]. There is also evidence that the reduction in the recall rate is maintained over consecutive screening episodes [18].

3158166

acrac_3158166_18

Supplemental Breast Cancer Screening Based on Breast Density

Cortesi et al [35] evaluated the efficacy of biannual screening US examination in women who were BRCA mutation carriers, high-risk (non-BRCA1/2), and intermediate-risk patients. Overall, MRI had sensitivity of (93.7%), followed by mammography (55.0%), then US (29.4%). In the nondense breast, the sensitivity of mammography was 82.5% versus 10% for US. In the dense breast, the sensitivity of mammography was 50% versus 42.6% for US. Sensitivity analysis by risk level was also performed. The US sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 22.7%, 24.5%, and 33.6%, respectively. The mammographic sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 25.0%, 66.4%, and 56.6%, respectively. Only the BRCA1/2 mutation carriers underwent MRI screening, which demonstrated a sensitivity of 93.7%. The authors did not analyze the efficacy of US screening based on both density and risk. The addition of WBUS to mammography increases the CDR in high-risk women. In a surveillance cohort study of 529 women with elevated lifetime risk, the sensitivity of mammography, WBUS, and the 2 modalities combined was 33%, 40%, and 49%, respectively [44]. However, subgroup analysis was not performed by breast density. In Supplemental Screening Based on Breast Density the same population, MRI had a sensitivity of 91% [44]. In the ACRIN 6666 trial, after 3 rounds of screening mammography and screening WBUS in women with dense breast tissue at intermediate to elevated lifetime risk, the sensitivity, specificity, and PPV3 of mammography was 0.52, 0.91, and 0.38, respectively. The addition of US to mammography increased the sensitivity (0.76) but decreased the specificity (0.84) and PPV3 (0.38) [45]. In a prospective cohort trial of 687 high-risk women, the cancer yield of mammography alone was 5.4/1,000 and increased to 7.7/1,000 with the addition of US [46].

Supplemental Breast Cancer Screening Based on Breast Density. Cortesi et al [35] evaluated the efficacy of biannual screening US examination in women who were BRCA mutation carriers, high-risk (non-BRCA1/2), and intermediate-risk patients. Overall, MRI had sensitivity of (93.7%), followed by mammography (55.0%), then US (29.4%). In the nondense breast, the sensitivity of mammography was 82.5% versus 10% for US. In the dense breast, the sensitivity of mammography was 50% versus 42.6% for US. Sensitivity analysis by risk level was also performed. The US sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 22.7%, 24.5%, and 33.6%, respectively. The mammographic sensitivities for BRCA1/2, high-risk (non-BRCA1/2), and intermediate-risk patients were 25.0%, 66.4%, and 56.6%, respectively. Only the BRCA1/2 mutation carriers underwent MRI screening, which demonstrated a sensitivity of 93.7%. The authors did not analyze the efficacy of US screening based on both density and risk. The addition of WBUS to mammography increases the CDR in high-risk women. In a surveillance cohort study of 529 women with elevated lifetime risk, the sensitivity of mammography, WBUS, and the 2 modalities combined was 33%, 40%, and 49%, respectively [44]. However, subgroup analysis was not performed by breast density. In Supplemental Screening Based on Breast Density the same population, MRI had a sensitivity of 91% [44]. In the ACRIN 6666 trial, after 3 rounds of screening mammography and screening WBUS in women with dense breast tissue at intermediate to elevated lifetime risk, the sensitivity, specificity, and PPV3 of mammography was 0.52, 0.91, and 0.38, respectively. The addition of US to mammography increased the sensitivity (0.76) but decreased the specificity (0.84) and PPV3 (0.38) [45]. In a prospective cohort trial of 687 high-risk women, the cancer yield of mammography alone was 5.4/1,000 and increased to 7.7/1,000 with the addition of US [46].

3158166

acrac_3158166_19

Supplemental Breast Cancer Screening Based on Breast Density

MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in high-risk women with dense breasts. MRI Breast Without and With IV Contrast Abbreviated AB-MRI performed with IV contrast is an abbreviated breast MRI examination. It is similar to a full MRI examination yet does not have a standard protocol; however, at minimum, it must include a precontrast and one postcontrast sequence. A T2-weighted sequence may also be included. There is limited relevant literature regarding the use of AB-MRI breast without and with IV contrast in high-risk women with dense breasts. In 2 retrospective studies comparing the full diagnostic protocol to an abbreviated protocol in high-risk women, the authors found both protocols to have similar sensitivity [48,49]. However, neither study evaluated the CDR by breast density. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in high-risk women with dense breasts. MRI Breast Without and With IV Contrast The American Cancer Society advocates MRI screening in high-risk women regardless of breast density [15]. There is ample evidence in the literature supporting this recommendation [46,50-54]. In the high-risk population, sensitivity of mammography alone is approximately 31% to 33%, compared with the sensitivity of MRI alone which is 87% to 96% [45,46,55]. The combination of mammography and MRI yields 100% sensitivity compared with the 44% to 48% sensitivity of combined mammography and US [45,46]. In addition, the types of carcinoma detected on MRI compared with mammography may differ. Cancers detected on MRI are more likely to be invasive carcinomas (71%), whereas cancers detected on mammography were are likely to be ductal carcinoma in situ (65%), or associated with calcifications (88%) [56].

Supplemental Breast Cancer Screening Based on Breast Density. MRI Breast Without IV Contrast Abbreviated There is no relevant literature regarding the use of AB-MRI breast without IV contrast for supplemental screening in high-risk women with dense breasts. MRI Breast Without and With IV Contrast Abbreviated AB-MRI performed with IV contrast is an abbreviated breast MRI examination. It is similar to a full MRI examination yet does not have a standard protocol; however, at minimum, it must include a precontrast and one postcontrast sequence. A T2-weighted sequence may also be included. There is limited relevant literature regarding the use of AB-MRI breast without and with IV contrast in high-risk women with dense breasts. In 2 retrospective studies comparing the full diagnostic protocol to an abbreviated protocol in high-risk women, the authors found both protocols to have similar sensitivity [48,49]. However, neither study evaluated the CDR by breast density. MRI Breast Without IV Contrast There is no relevant literature regarding the use of MRI breast without IV contrast for supplemental screening in high-risk women with dense breasts. MRI Breast Without and With IV Contrast The American Cancer Society advocates MRI screening in high-risk women regardless of breast density [15]. There is ample evidence in the literature supporting this recommendation [46,50-54]. In the high-risk population, sensitivity of mammography alone is approximately 31% to 33%, compared with the sensitivity of MRI alone which is 87% to 96% [45,46,55]. The combination of mammography and MRI yields 100% sensitivity compared with the 44% to 48% sensitivity of combined mammography and US [45,46]. In addition, the types of carcinoma detected on MRI compared with mammography may differ. Cancers detected on MRI are more likely to be invasive carcinomas (71%), whereas cancers detected on mammography were are likely to be ductal carcinoma in situ (65%), or associated with calcifications (88%) [56].

3158166

acrac_3158177_0

Chronic Cough

Introduction/Background Chronic cough is defined by a duration lasting at least 8 weeks [1], often contributing to patient discomfort and altered psychosocial well-being [2]. The most common causes of chronic cough include smoking-related lung disease, upper airway cough syndrome (UACS), asthma, gastroesophageal reflux disease (GERD), and nonasthmatic eosinophilic bronchitis [1-3]. Cough can further be characterized by quality, that is productive or nonproductive. Conditions that may be associated with productive cough include bronchiectasis, chronic bronchitis, asthma, and immunodeficiencies [4-6]. There is varied opinion in the literature regarding the utility of cough productivity in the phenotypic stratification of chronic cough patients [5,7,8]. Cough is a protective reflex mediated by a complex network of afferent and efferent neuronal pathways [9]. Stimuli for cough can occur in the chest wall, airways, esophagus, larynx, and middle ear [10]. It is believed that a significant number of individuals with chronic cough may be suffering from cough hypersensitivity [11,12], in which the threshold for cough is lowered by repeated exposure to stimulus or inflammation [11-13]. There may be similarity in the neuronal hyperstimulation seen in chronic cough with that of chronic pain [14]. The etiology of chronic cough in some patients may be difficult to localize to an isolated source and is often multifactorial [15]. The complex pathophysiology, clinical presentation, and variable manifestations of chronic cough underscore the challenges faced by clinicians in the evaluation and management of these patients. Varying guidelines and algorithms exist for the evaluation and management of chronic cough, driven by a thorough history and physical examination, empiric treatment, and often diagnosis by exclusion [1,15,16]. Standardized clinical algorithms have shown efficacy in the diagnosis and treatment of chronic cough [7,10,16,17].

Chronic Cough. Introduction/Background Chronic cough is defined by a duration lasting at least 8 weeks [1], often contributing to patient discomfort and altered psychosocial well-being [2]. The most common causes of chronic cough include smoking-related lung disease, upper airway cough syndrome (UACS), asthma, gastroesophageal reflux disease (GERD), and nonasthmatic eosinophilic bronchitis [1-3]. Cough can further be characterized by quality, that is productive or nonproductive. Conditions that may be associated with productive cough include bronchiectasis, chronic bronchitis, asthma, and immunodeficiencies [4-6]. There is varied opinion in the literature regarding the utility of cough productivity in the phenotypic stratification of chronic cough patients [5,7,8]. Cough is a protective reflex mediated by a complex network of afferent and efferent neuronal pathways [9]. Stimuli for cough can occur in the chest wall, airways, esophagus, larynx, and middle ear [10]. It is believed that a significant number of individuals with chronic cough may be suffering from cough hypersensitivity [11,12], in which the threshold for cough is lowered by repeated exposure to stimulus or inflammation [11-13]. There may be similarity in the neuronal hyperstimulation seen in chronic cough with that of chronic pain [14]. The etiology of chronic cough in some patients may be difficult to localize to an isolated source and is often multifactorial [15]. The complex pathophysiology, clinical presentation, and variable manifestations of chronic cough underscore the challenges faced by clinicians in the evaluation and management of these patients. Varying guidelines and algorithms exist for the evaluation and management of chronic cough, driven by a thorough history and physical examination, empiric treatment, and often diagnosis by exclusion [1,15,16]. Standardized clinical algorithms have shown efficacy in the diagnosis and treatment of chronic cough [7,10,16,17].

3158177

acrac_3158177_1

Chronic Cough

Imaging plays a role in the evaluation, although there is lack of high-quality evidence guiding which modalities are useful and at what time point in the clinical evaluation they should be performed. Further research may be needed to evaluate which imaging modalities performed earlier in the clinical assessment are beneficial in long-term outcomes. Special Imaging Considerations Our literature review has identified several studies investigating thoracic ultrasound (US) in the evaluation of the noncardiac chest pain. Noncardiac thoracic US performed better than radiography in small prospective cohorts staged in the postoperative intensive care unit [18] and primary care setting [19], confidently identifying findings such as pneumothorax, pleural effusion, consolidation, and interstitial patterns of lung disease. US was shown to correlate well with high-resolution CT (HRCT) and pulmonary function test (PFT) abnormalities [20], and it performed well in several small studies of patients with known interstitial lung disease [21-23]. US may be prone to diminished specificity. For instance, Moazedi-Fuerst et al [24] showed that up to 12% of patients with normal HRCT studies had sonographic findings of an interstitial pattern based on B-lines and artifacts. Adequate US evaluation is operator dependent and requires experience to perform and interpret examination findings. To our knowledge, there is no relevant literature to support the use of thoracic US in the evaluation of chronic cough; however, it is included in this section for awareness and to promulgate future research into this particular topic. Reprint requests to: [email protected] Chronic Cough Low-dose chest CT (LDCT) has been clinically validated in the National Lung Screening Trial Research Team for early detection of lung cancer [25]. LDCT has been investigated for early detection of pulmonary infection in elderly patients with normal or equivocal chest radiographs.

Chronic Cough. Imaging plays a role in the evaluation, although there is lack of high-quality evidence guiding which modalities are useful and at what time point in the clinical evaluation they should be performed. Further research may be needed to evaluate which imaging modalities performed earlier in the clinical assessment are beneficial in long-term outcomes. Special Imaging Considerations Our literature review has identified several studies investigating thoracic ultrasound (US) in the evaluation of the noncardiac chest pain. Noncardiac thoracic US performed better than radiography in small prospective cohorts staged in the postoperative intensive care unit [18] and primary care setting [19], confidently identifying findings such as pneumothorax, pleural effusion, consolidation, and interstitial patterns of lung disease. US was shown to correlate well with high-resolution CT (HRCT) and pulmonary function test (PFT) abnormalities [20], and it performed well in several small studies of patients with known interstitial lung disease [21-23]. US may be prone to diminished specificity. For instance, Moazedi-Fuerst et al [24] showed that up to 12% of patients with normal HRCT studies had sonographic findings of an interstitial pattern based on B-lines and artifacts. Adequate US evaluation is operator dependent and requires experience to perform and interpret examination findings. To our knowledge, there is no relevant literature to support the use of thoracic US in the evaluation of chronic cough; however, it is included in this section for awareness and to promulgate future research into this particular topic. Reprint requests to: [email protected] Chronic Cough Low-dose chest CT (LDCT) has been clinically validated in the National Lung Screening Trial Research Team for early detection of lung cancer [25]. LDCT has been investigated for early detection of pulmonary infection in elderly patients with normal or equivocal chest radiographs.

3158177

acrac_3158177_2

Chronic Cough

Park et al [26] showed that 49 out of 166 elderly patients presenting to the emergency room who had normal chest radiographs were subsequently diagnosed with acute respiratory infection from findings on LDCT. In a prospective case comparison, Christe et al [27] showed good correlation between standard dose (150 mAs) and low-dose (40 mAs) CT protocols for evaluating the majority of pulmonary abnormalities, including bronchiectasis, air trapping, and pleural disease. Investigators did identify a drop in sensitivity in low-dose scans for ground-glass nodules, reticulation, and mucus plugging [27]. Ultra-low- dose CT demonstrated 91% sensitivity and 100% specificity in identifying asbestos-related lung abnormalities in a prospective cohort of 55 asymptomatic persons with relevant exposure history [28]. To our knowledge, there is no relevant literature to support the use of LDCT in the evaluation of chronic cough, and further research into this topic may be warranted. OR Discussion of Procedures by Variant Variant 1: Chronic cough lasting more than 8 weeks. No known risk factors for lung cancer. Initial imaging. Radiography Chest Chest radiography is recommended by numerous pulmonary and thoracic societies for the evaluation of chronic cough [1,10,15,29], although the exact timing of imaging has varied between groups. The American College of Chest Physicians (ACCP) includes a recommendation for an initial chest radiograph in their chronic cough algorithm [1]. The evidence supporting chest radiography at initial evaluation of chronic cough is limited to case series, observational studies, and retrospective analysis. Initial evaluation with posteroanterior chest radiography has been included as part of two investigative case series using standardized clinical protocols for chronic cough [7,17].

Chronic Cough. Park et al [26] showed that 49 out of 166 elderly patients presenting to the emergency room who had normal chest radiographs were subsequently diagnosed with acute respiratory infection from findings on LDCT. In a prospective case comparison, Christe et al [27] showed good correlation between standard dose (150 mAs) and low-dose (40 mAs) CT protocols for evaluating the majority of pulmonary abnormalities, including bronchiectasis, air trapping, and pleural disease. Investigators did identify a drop in sensitivity in low-dose scans for ground-glass nodules, reticulation, and mucus plugging [27]. Ultra-low- dose CT demonstrated 91% sensitivity and 100% specificity in identifying asbestos-related lung abnormalities in a prospective cohort of 55 asymptomatic persons with relevant exposure history [28]. To our knowledge, there is no relevant literature to support the use of LDCT in the evaluation of chronic cough, and further research into this topic may be warranted. OR Discussion of Procedures by Variant Variant 1: Chronic cough lasting more than 8 weeks. No known risk factors for lung cancer. Initial imaging. Radiography Chest Chest radiography is recommended by numerous pulmonary and thoracic societies for the evaluation of chronic cough [1,10,15,29], although the exact timing of imaging has varied between groups. The American College of Chest Physicians (ACCP) includes a recommendation for an initial chest radiograph in their chronic cough algorithm [1]. The evidence supporting chest radiography at initial evaluation of chronic cough is limited to case series, observational studies, and retrospective analysis. Initial evaluation with posteroanterior chest radiography has been included as part of two investigative case series using standardized clinical protocols for chronic cough [7,17].

3158177

acrac_3158177_3

Chronic Cough

Of the 131 patients evaluated, 49 chest radiographs were reported abnormal, with the final diagnosis of primary lung pathology other than asthma made in 29 patients, including 8 patients with bronchiectasis, 8 patients with interstitial lung disease, and 2 patients with neoplasm. The authors did not report the sensitivity or specificity of chest radiography in the setting of chronic cough; however, abnormal radiography was positively associated with underlying primary pulmonary pathology other than asthma (odds ratio 7.7) [7]. Ojoo et al [17] prospectively evaluated 112 patients with chronic cough in a similar standardized clinical investigative protocol. In this study, 7 radiographs (6.3%) were reported contributory to the final diagnosis. Of the 92 patients with diagnosis found at discharge, 12 (13%) patients had pathologies including bronchiectasis, interstitial lung disease, self-limiting cough, and 2 cases of lung cancer. Both of these studies reported clinical efficacy with standardized protocols utilizing initial chest radiography, with diagnosis achieved in 93% in the Kastelik et al study [7] and 82% in the Ojoo et al study [17]. The most commonly reported etiologies were reflux, asthma, postinfective or viral cough, and rhinitis, representing between 61% to 67% of final diagnosis, which were more common than bronchiectasis, interstitial lung disease, and neoplasm, which represents <13% to 18% of final diagnosis [7,17]. Both case series are limited by sample size, and both studies were performed out of referral centers, likely contributing to selection bias. Retrospective analyses have been performed specific to chronic cough and normal chest radiography. Turner et al [16] retrospectively reviewed the medical records of 404 patients with chronic cough referred to specialty care who either had or subsequently were found to have a normal chest radiograph.

Chronic Cough. Of the 131 patients evaluated, 49 chest radiographs were reported abnormal, with the final diagnosis of primary lung pathology other than asthma made in 29 patients, including 8 patients with bronchiectasis, 8 patients with interstitial lung disease, and 2 patients with neoplasm. The authors did not report the sensitivity or specificity of chest radiography in the setting of chronic cough; however, abnormal radiography was positively associated with underlying primary pulmonary pathology other than asthma (odds ratio 7.7) [7]. Ojoo et al [17] prospectively evaluated 112 patients with chronic cough in a similar standardized clinical investigative protocol. In this study, 7 radiographs (6.3%) were reported contributory to the final diagnosis. Of the 92 patients with diagnosis found at discharge, 12 (13%) patients had pathologies including bronchiectasis, interstitial lung disease, self-limiting cough, and 2 cases of lung cancer. Both of these studies reported clinical efficacy with standardized protocols utilizing initial chest radiography, with diagnosis achieved in 93% in the Kastelik et al study [7] and 82% in the Ojoo et al study [17]. The most commonly reported etiologies were reflux, asthma, postinfective or viral cough, and rhinitis, representing between 61% to 67% of final diagnosis, which were more common than bronchiectasis, interstitial lung disease, and neoplasm, which represents <13% to 18% of final diagnosis [7,17]. Both case series are limited by sample size, and both studies were performed out of referral centers, likely contributing to selection bias. Retrospective analyses have been performed specific to chronic cough and normal chest radiography. Turner et al [16] retrospectively reviewed the medical records of 404 patients with chronic cough referred to specialty care who either had or subsequently were found to have a normal chest radiograph.

3158177

acrac_3158177_4

Chronic Cough

Of the 266 patients who were given a diagnosis, 8 patients were found to have a diagnosis that could be made on imaging (4 with lower respiratory tract infection, 2 with malignancy, 1 with bronchiectasis, and 1 with pulmonary fibrosis). Although made explicit for the 2 cases with malignancy, it is only implied that the other 6 patients had normal chest radiography and that the diagnosis was made by other means (ie, chest CT). Additionally, the cases of malignancy were suspected on the Chronic Cough The diagnostic performance of chest radiography varies among the reviewed literature. In a systematic review of the literature, Piccazzo et al [30] reported a high negative predictive value for chest radiography in the evaluation of active and inactive tuberculosis. Abnormalities on chest radiography did have a lower specificity, and findings, which were equivalent the authors concluded, should prompt consideration for CT as this modality was shown to have better performance for both active and inactive tuberculosis. Colaci et al [31] prospectively evaluated 107 consecutive patients with systemic sclerosis with both radiography and HRCT. Their study showed strong correlation between radiographic and HRCT findings of interstitial lung disease, with strong association between cough symptoms and HRCT findings. Interstitial lung pattern and restrictive PFT was more commonly identified than airway involvement and obstructive disease. The authors, however, did not publish sensitivity or specificity for radiography [31]. The diagnostic performance of radiography for other patterns of lung disease may be worse. In a retrospective analysis of 236 patients with CT-proven bronchiectasis, Altenburg et al [32] reported that up to 34% of chest radiographs were reported unremarkable. As mentioned above, the most common CT findings in those patients with chronic cough with normal radiographs were in fact bronchiectasis (28%) and bronchial wall thickening (21%) [3].

Chronic Cough. Of the 266 patients who were given a diagnosis, 8 patients were found to have a diagnosis that could be made on imaging (4 with lower respiratory tract infection, 2 with malignancy, 1 with bronchiectasis, and 1 with pulmonary fibrosis). Although made explicit for the 2 cases with malignancy, it is only implied that the other 6 patients had normal chest radiography and that the diagnosis was made by other means (ie, chest CT). Additionally, the cases of malignancy were suspected on the Chronic Cough The diagnostic performance of chest radiography varies among the reviewed literature. In a systematic review of the literature, Piccazzo et al [30] reported a high negative predictive value for chest radiography in the evaluation of active and inactive tuberculosis. Abnormalities on chest radiography did have a lower specificity, and findings, which were equivalent the authors concluded, should prompt consideration for CT as this modality was shown to have better performance for both active and inactive tuberculosis. Colaci et al [31] prospectively evaluated 107 consecutive patients with systemic sclerosis with both radiography and HRCT. Their study showed strong correlation between radiographic and HRCT findings of interstitial lung disease, with strong association between cough symptoms and HRCT findings. Interstitial lung pattern and restrictive PFT was more commonly identified than airway involvement and obstructive disease. The authors, however, did not publish sensitivity or specificity for radiography [31]. The diagnostic performance of radiography for other patterns of lung disease may be worse. In a retrospective analysis of 236 patients with CT-proven bronchiectasis, Altenburg et al [32] reported that up to 34% of chest radiographs were reported unremarkable. As mentioned above, the most common CT findings in those patients with chronic cough with normal radiographs were in fact bronchiectasis (28%) and bronchial wall thickening (21%) [3].

3158177

acrac_3158177_5

Chronic Cough

The sensitivity of chest radiography for airway abnormalities was found to be 69% to 71% when referenced to helical CT in 70 patients with known airway abnormalities and lesions [33]. In a retrospective analysis of elderly emergency room patients who were evaluated for acute respiratory infection, Park et al [26] showed that chest radiographs were normal in 49 out of 166 confirmed cases on LDCT. Chest radiography was more often normal when LDCT identified ground-glass opacity (GGO), bronchial wall thickening, centrilobular nodules, and small and dependent consolidations. Chest radiography was also found to have poor correlation with chest CT for the presence of pulmonary opacities [34,35], with a positive predictive value of only 27% when compared with chest CT in a secondary analysis of a large retrospective cohort of 3,400 patients. Limitations of chest radiography reflect superimposed soft-tissue structures, radiographic findings related to comorbidities, and limited contrast resolution. Unfortunately, there are no high-quality studies evaluating the clinical efficacy of chest radiography in the early evaluation of patients with chronic cough, and this may be an area of future investigation. Based on the available literature, there is a suggestion that initial evaluation of chronic cough with chest radiography is beneficial in the clinical setting. There does remain question regarding the sensitivity of chest radiography in this group [3], which is likely skewed toward early disease and airway abnormalities. This could potentially result in delayed diagnosis for a small percent of patients with chronic cough. CT Chest There is no high-quality evidence to support the use of chest CT in the initial evaluation of patients presenting with chronic cough. No studies have directly compared the utility of contrast-enhanced versus noncontrast-enhanced CT imaging in regard to chronic cough.

Chronic Cough. The sensitivity of chest radiography for airway abnormalities was found to be 69% to 71% when referenced to helical CT in 70 patients with known airway abnormalities and lesions [33]. In a retrospective analysis of elderly emergency room patients who were evaluated for acute respiratory infection, Park et al [26] showed that chest radiographs were normal in 49 out of 166 confirmed cases on LDCT. Chest radiography was more often normal when LDCT identified ground-glass opacity (GGO), bronchial wall thickening, centrilobular nodules, and small and dependent consolidations. Chest radiography was also found to have poor correlation with chest CT for the presence of pulmonary opacities [34,35], with a positive predictive value of only 27% when compared with chest CT in a secondary analysis of a large retrospective cohort of 3,400 patients. Limitations of chest radiography reflect superimposed soft-tissue structures, radiographic findings related to comorbidities, and limited contrast resolution. Unfortunately, there are no high-quality studies evaluating the clinical efficacy of chest radiography in the early evaluation of patients with chronic cough, and this may be an area of future investigation. Based on the available literature, there is a suggestion that initial evaluation of chronic cough with chest radiography is beneficial in the clinical setting. There does remain question regarding the sensitivity of chest radiography in this group [3], which is likely skewed toward early disease and airway abnormalities. This could potentially result in delayed diagnosis for a small percent of patients with chronic cough. CT Chest There is no high-quality evidence to support the use of chest CT in the initial evaluation of patients presenting with chronic cough. No studies have directly compared the utility of contrast-enhanced versus noncontrast-enhanced CT imaging in regard to chronic cough.

3158177

acrac_3158177_6

Chronic Cough

Contrast-enhanced studies offer improved visualization of cardiopulmonary vasculature, mediastinal structures, and soft-tissue abnormalities [35], although MRI is developing a more consistent role for the latter. For the majority of studies, however, the noncontrast technique appears to be adequate. Chest CT is more sensitive than chest radiographs for the evaluation of most pulmonary abnormalities as well as mediastinal, cardiac, and chest wall findings, combining improved soft-tissue contrast and anatomical localization. This was demonstrated in a few small studies included in our literature review [3,32,36]. Chest CT is considered the reference standard for the noninvasive diagnosis of bronchiectasis [37] as well as of interstitial lung disease. CT abnormalities have been described in patients with chronic cough. Truba et al [3] performed a retrospective analysis of patients with chronic cough with negative initial chest radiograph findings, reporting 21 out of 59 patients with CT abnormalities believed to be relevant to chronic cough. The authors identified bronchiectasis and Chronic Cough Other studies showed a lack of association between cough symptoms and chest CT findings. Ooi et al [6] was unable to correlate respiratory symptom exacerbation (including cough symptoms) with HRCT findings of small airways abnormality, mosaic attenuation, and bronchial wall thickening in 60 patients with CT-proven bronchiectasis, although there was association between CT findings and sputum production and PFT abnormality. Wilsher et al [40], in a prospective analysis of 60 consecutive patients newly diagnosed with RA, demonstrated that bronchiectasis was prevalent in up to 48% of patients, despite only 30% of patients presenting with respiratory complaints. There was no association between the cough symptoms and bronchiectasis, although there was a statistically significant association for dyspnea symptoms.

Chronic Cough. Contrast-enhanced studies offer improved visualization of cardiopulmonary vasculature, mediastinal structures, and soft-tissue abnormalities [35], although MRI is developing a more consistent role for the latter. For the majority of studies, however, the noncontrast technique appears to be adequate. Chest CT is more sensitive than chest radiographs for the evaluation of most pulmonary abnormalities as well as mediastinal, cardiac, and chest wall findings, combining improved soft-tissue contrast and anatomical localization. This was demonstrated in a few small studies included in our literature review [3,32,36]. Chest CT is considered the reference standard for the noninvasive diagnosis of bronchiectasis [37] as well as of interstitial lung disease. CT abnormalities have been described in patients with chronic cough. Truba et al [3] performed a retrospective analysis of patients with chronic cough with negative initial chest radiograph findings, reporting 21 out of 59 patients with CT abnormalities believed to be relevant to chronic cough. The authors identified bronchiectasis and Chronic Cough Other studies showed a lack of association between cough symptoms and chest CT findings. Ooi et al [6] was unable to correlate respiratory symptom exacerbation (including cough symptoms) with HRCT findings of small airways abnormality, mosaic attenuation, and bronchial wall thickening in 60 patients with CT-proven bronchiectasis, although there was association between CT findings and sputum production and PFT abnormality. Wilsher et al [40], in a prospective analysis of 60 consecutive patients newly diagnosed with RA, demonstrated that bronchiectasis was prevalent in up to 48% of patients, despite only 30% of patients presenting with respiratory complaints. There was no association between the cough symptoms and bronchiectasis, although there was a statistically significant association for dyspnea symptoms.

3158177

acrac_3158177_7

Chronic Cough

The major pulmonary societies recommend noncontrast chest CT for the evaluation of chronic cough when the more common causes are excluded or empirically treated [1,10,15,29], and this is usually preceded by a chest radiograph. Of the studies evaluating the clinical management of patients with chronic cough, none utilized chest CT in the initial workup [3,7,16,17], and two studies specifically addressed a cohort of patients with normal chest radiography findings [3,16]. Kastelik et al [7] utilized a probability-based clinical algorithm in the prospective treatment of 131 patients presenting with cough lasting >8 weeks. All patients were initially evaluated with chest radiography and empirically treated based on the probability of underlying etiologies. Chest CT was performed in those patients with clinical suspicion of underlying disease (n = 29) as well as in those with indications or negative initial workup otherwise (n = 17). Of these, 26 out of 29 patients (positive predictive value of 90%) suspected of having underlying pulmonary disease were diagnosed and treated as appropriate; however, 17 patients in whom chest CT was performed without cause and as part of the algorithmic evaluation did not have findings relevant to their management. The authors concluded that chest CT should be performed only in selected patients and those with abnormal chest radiographs. In another prospective observational study, Ojoo et al [17] evaluated 112 consecutive patients with chronic cough. All patients were evaluated with chest radiography. Chest CT was performed only based on clinical suspicion. The authors reported that 74 out of 81 patients managed clinically without the need for chest CT examination, suggesting chest CT would not have made a difference in these patients. It should be noted that both studies did identify primary pulmonary etiologies felt to be associated with chronic cough, of which bronchiectasis and interstitial lung disease were the two most common.

Chronic Cough. The major pulmonary societies recommend noncontrast chest CT for the evaluation of chronic cough when the more common causes are excluded or empirically treated [1,10,15,29], and this is usually preceded by a chest radiograph. Of the studies evaluating the clinical management of patients with chronic cough, none utilized chest CT in the initial workup [3,7,16,17], and two studies specifically addressed a cohort of patients with normal chest radiography findings [3,16]. Kastelik et al [7] utilized a probability-based clinical algorithm in the prospective treatment of 131 patients presenting with cough lasting >8 weeks. All patients were initially evaluated with chest radiography and empirically treated based on the probability of underlying etiologies. Chest CT was performed in those patients with clinical suspicion of underlying disease (n = 29) as well as in those with indications or negative initial workup otherwise (n = 17). Of these, 26 out of 29 patients (positive predictive value of 90%) suspected of having underlying pulmonary disease were diagnosed and treated as appropriate; however, 17 patients in whom chest CT was performed without cause and as part of the algorithmic evaluation did not have findings relevant to their management. The authors concluded that chest CT should be performed only in selected patients and those with abnormal chest radiographs. In another prospective observational study, Ojoo et al [17] evaluated 112 consecutive patients with chronic cough. All patients were evaluated with chest radiography. Chest CT was performed only based on clinical suspicion. The authors reported that 74 out of 81 patients managed clinically without the need for chest CT examination, suggesting chest CT would not have made a difference in these patients. It should be noted that both studies did identify primary pulmonary etiologies felt to be associated with chronic cough, of which bronchiectasis and interstitial lung disease were the two most common.

3158177

acrac_3158177_8

Chronic Cough

In a retrospective cohort of patients with chronic cough referred to specialty care with normal chest radiographs, Turner et al [16] showed successful clinical management of 266 patients with minimal investigation and empiric treatment, requiring chest CT in 1 patient to diagnose bronchiectasis, 1 patient with pulmonary fibrosis, and 2 patients with malignancy. Future studies may be needed to better validate CT findings with clinical features in order to determine causation/association in the context of chronic cough. Lung patterns of disease and obvious abnormalities identified on CT will still be of clinical value, and an expected number of CT Chronic Cough abnormalities may be noncontributory to management. The role of chest CT in the initial evaluation of chronic cough remains indeterminate. The evidence suggests that wide application in all patients presenting with chronic cough may be of low clinical yield. Appropriate selection would likely improve the specificity of findings. FDG-PET/CT Skull Base to Mid-Thigh There is no relevant literature to support the use of fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG)-PET in the initial evaluation of chronic cough. FDG-PET/CT is not listed as an imaging option in the most recent ACCP guidelines on chronic cough [1]. FDG-PET utilizes a radiolabeled glucose analog to evaluate for upregulated metabolism in areas of infection, inflammation, and neoplasm [43]. MRI Chest There is no relevant literature to support the use of thoracic MRI in the initial evaluation of chronic cough. MRI of the chest is not included in the clinical algorithms recommended by various pulmonary and thoracic societies [1,10,15,29]. Thoracic MRI (nonbreast, noncardiac, nonmusculoskeletal) is currently utilized for the evaluation of indeterminate findings on other imaging modalities, most commonly mediastinal and thymic [44], less often utilized for vascular, airway, and pulmonary parenchymal abnormalities.

Chronic Cough. In a retrospective cohort of patients with chronic cough referred to specialty care with normal chest radiographs, Turner et al [16] showed successful clinical management of 266 patients with minimal investigation and empiric treatment, requiring chest CT in 1 patient to diagnose bronchiectasis, 1 patient with pulmonary fibrosis, and 2 patients with malignancy. Future studies may be needed to better validate CT findings with clinical features in order to determine causation/association in the context of chronic cough. Lung patterns of disease and obvious abnormalities identified on CT will still be of clinical value, and an expected number of CT Chronic Cough abnormalities may be noncontributory to management. The role of chest CT in the initial evaluation of chronic cough remains indeterminate. The evidence suggests that wide application in all patients presenting with chronic cough may be of low clinical yield. Appropriate selection would likely improve the specificity of findings. FDG-PET/CT Skull Base to Mid-Thigh There is no relevant literature to support the use of fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG)-PET in the initial evaluation of chronic cough. FDG-PET/CT is not listed as an imaging option in the most recent ACCP guidelines on chronic cough [1]. FDG-PET utilizes a radiolabeled glucose analog to evaluate for upregulated metabolism in areas of infection, inflammation, and neoplasm [43]. MRI Chest There is no relevant literature to support the use of thoracic MRI in the initial evaluation of chronic cough. MRI of the chest is not included in the clinical algorithms recommended by various pulmonary and thoracic societies [1,10,15,29]. Thoracic MRI (nonbreast, noncardiac, nonmusculoskeletal) is currently utilized for the evaluation of indeterminate findings on other imaging modalities, most commonly mediastinal and thymic [44], less often utilized for vascular, airway, and pulmonary parenchymal abnormalities.

3158177

acrac_3158177_9

Chronic Cough

MRI offers improved soft-tissue contrast, with advanced sequences able to identify soft-tissue characteristics to the cellular level (ie, diffusion-weighted imaging and spectroscopy). Advanced techniques allow for calculation of total lung water density, which has been shown to correlate with disease severity in adult patients with cystic fibrosis [45]. With the addition of intravenous (IV) contrast, the vascular nature of an abnormality is exquisitely imaged and has been shown to be as accurate as contrast-enhanced CT for the evaluation for pulmonary embolism [35]. IV contrast also offers the evaluation of pulmonary perfusion [46]. Limitations of pulmonary MRI include low signal-to-noise ratio (especially with a normally aerated lung), low spatial resolution, and need for adequate breath-hold technique or respiratory gating [46]. Parenchymal lung abnormalities naturally increase cellular proton density and/or water content while displacing low signal gas, providing contrast for visualization [35]. Despite limitations, pulmonary MRI has been shown to have noninferior diagnostic quality when compared with CT [46,47]. Ohno et al [47] analyzed 85 consecutive patients with high-resolution pulmonary MR using ultrashort time to echo and compared findings with both standard CT and LDCT. The authors found equivalent diagnostic efficacy for all observed findings, noting decreased sensitivity and image quality when evaluating for emphysema, bronchiectasis, and reticular opacity. The authors concluded that based on their assessment, ultrashort time to echo pulmonary MR is equivalent to LDCT imaging [47]. There is no relevant literature evaluating IV contrast versus noncontrast MRI in the setting of chronic cough, and decisions may need to be made on a case-by-case basis. Investigational contrast agents may prove useful in the future. Hyperpolarized gas MRI is an example, which is currently FDA approved for investigational use [44,48].

Chronic Cough. MRI offers improved soft-tissue contrast, with advanced sequences able to identify soft-tissue characteristics to the cellular level (ie, diffusion-weighted imaging and spectroscopy). Advanced techniques allow for calculation of total lung water density, which has been shown to correlate with disease severity in adult patients with cystic fibrosis [45]. With the addition of intravenous (IV) contrast, the vascular nature of an abnormality is exquisitely imaged and has been shown to be as accurate as contrast-enhanced CT for the evaluation for pulmonary embolism [35]. IV contrast also offers the evaluation of pulmonary perfusion [46]. Limitations of pulmonary MRI include low signal-to-noise ratio (especially with a normally aerated lung), low spatial resolution, and need for adequate breath-hold technique or respiratory gating [46]. Parenchymal lung abnormalities naturally increase cellular proton density and/or water content while displacing low signal gas, providing contrast for visualization [35]. Despite limitations, pulmonary MRI has been shown to have noninferior diagnostic quality when compared with CT [46,47]. Ohno et al [47] analyzed 85 consecutive patients with high-resolution pulmonary MR using ultrashort time to echo and compared findings with both standard CT and LDCT. The authors found equivalent diagnostic efficacy for all observed findings, noting decreased sensitivity and image quality when evaluating for emphysema, bronchiectasis, and reticular opacity. The authors concluded that based on their assessment, ultrashort time to echo pulmonary MR is equivalent to LDCT imaging [47]. There is no relevant literature evaluating IV contrast versus noncontrast MRI in the setting of chronic cough, and decisions may need to be made on a case-by-case basis. Investigational contrast agents may prove useful in the future. Hyperpolarized gas MRI is an example, which is currently FDA approved for investigational use [44,48].

3158177

acrac_3158177_10

Chronic Cough

Hyperpolarized Xe-129 gas MRI allows for improved contrast of the lung parenchyma with techniques allowing for the evaluation of anatomical and functional data [48]. These techniques are still in their infancy and will likely require future studies to evaluate their role in imaging in chronic cough. In 2014, Ackman et al [44] surveyed thoracic radiologists from the Society of Thoracic Radiology regarding utilization of nonvascular thoracic MRI. Respondents offered an insight into the underutilization of thoracic MRI. Challenges in implementing MRI into clinical practice were described, including ordering health care provider awareness, lack of training, and technical, as well as administration hurdle. Variant 2: Chronic cough lasting more than 8 weeks. Increased risk for lung cancer. Initial imaging. Radiography Chest Chest radiography is often performed early in the investigation of chronic cough and is recommended by the ACCP [1]. There was no literature that addressed this modality in the setting of chronic cough with risk factors for lung cancer. In each of the available prospective case series evaluating standardized clinical protocols for chronic cough, there was a diagnosis of malignancy in 2 of 131 [7] and 2 of 122 [17] patients assessed, all of whom had initial chest radiography. Both studies reported that clinical and radiographic findings led to increased suspicion of malignancy. In a retrospective analysis of patients with chronic cough who had normal chest radiographs, 2 out of 266 diagnosis were determined to be malignant. Both patients had suspicious clinical findings, which prompted further evaluation with chest CT [16]. Cough has been shown to be a prevalent symptom in patients with lung cancer [49,50], with 57% consecutive patients with lung cancer reporting cough as a symptom (n = 223). The exact prevalence of malignancy in a population with chronic cough is not well documented and likely <1% to 2% from the small collection of case series Chronic Cough reviewed.

Chronic Cough. Hyperpolarized Xe-129 gas MRI allows for improved contrast of the lung parenchyma with techniques allowing for the evaluation of anatomical and functional data [48]. These techniques are still in their infancy and will likely require future studies to evaluate their role in imaging in chronic cough. In 2014, Ackman et al [44] surveyed thoracic radiologists from the Society of Thoracic Radiology regarding utilization of nonvascular thoracic MRI. Respondents offered an insight into the underutilization of thoracic MRI. Challenges in implementing MRI into clinical practice were described, including ordering health care provider awareness, lack of training, and technical, as well as administration hurdle. Variant 2: Chronic cough lasting more than 8 weeks. Increased risk for lung cancer. Initial imaging. Radiography Chest Chest radiography is often performed early in the investigation of chronic cough and is recommended by the ACCP [1]. There was no literature that addressed this modality in the setting of chronic cough with risk factors for lung cancer. In each of the available prospective case series evaluating standardized clinical protocols for chronic cough, there was a diagnosis of malignancy in 2 of 131 [7] and 2 of 122 [17] patients assessed, all of whom had initial chest radiography. Both studies reported that clinical and radiographic findings led to increased suspicion of malignancy. In a retrospective analysis of patients with chronic cough who had normal chest radiographs, 2 out of 266 diagnosis were determined to be malignant. Both patients had suspicious clinical findings, which prompted further evaluation with chest CT [16]. Cough has been shown to be a prevalent symptom in patients with lung cancer [49,50], with 57% consecutive patients with lung cancer reporting cough as a symptom (n = 223). The exact prevalence of malignancy in a population with chronic cough is not well documented and likely <1% to 2% from the small collection of case series Chronic Cough reviewed.

3158177

acrac_3158177_11

Chronic Cough

Nevertheless, in the setting of elevated risk (ie, smoking and occupational exposure), the prevalence may be more significant. The role of chest radiography in the early evaluation of patients with chronic cough is largely founded in the exclusion of treatable and/or serious underlying pulmonary pathology. When extrapolating from the available evidence, there remains a question regarding the sensitivity of chest radiography for the detection of subclinical disease and early malignancy. In this setting, there may be a population of patients with chronic cough that may benefit from earlier use of advanced imaging. Future research into this topic may elucidate the specific features and clinical qualities that would provide risk stratification and entail appropriate imaging recommendation. CT Chest There is no relevant literature to support the use of chest CT in patients with chronic cough who have one or more high-risk factors for lung cancer. In nonsmoking adults, risk factors vary and include environmental pollutants, genetic variations, underlying chronic pulmonary disease, immunosuppression, and occupational exposures (ie, asbestos and silica) [52]. Cough has been shown to be a prevalent symptom in patients with lung cancer [49,50], 57% of a small cohort of consecutive patients with lung cancer (n = 223). However, the true prevalence of isolated cough (without any additional clinical features) in the setting of malignancy remains unknown. Some authors suggest this combination to be rare in the clinical setting of chronic cough [16]. In a retrospective analysis of patients with chronic cough who have normal chest radiographs, 2 out of 266 diagnoses were determined to be malignancy. Both patients had suspicious clinical findings, which prompted further evaluation with chest CT. It is also worth noting that both patients had a normal chest radiograph [16].

Chronic Cough. Nevertheless, in the setting of elevated risk (ie, smoking and occupational exposure), the prevalence may be more significant. The role of chest radiography in the early evaluation of patients with chronic cough is largely founded in the exclusion of treatable and/or serious underlying pulmonary pathology. When extrapolating from the available evidence, there remains a question regarding the sensitivity of chest radiography for the detection of subclinical disease and early malignancy. In this setting, there may be a population of patients with chronic cough that may benefit from earlier use of advanced imaging. Future research into this topic may elucidate the specific features and clinical qualities that would provide risk stratification and entail appropriate imaging recommendation. CT Chest There is no relevant literature to support the use of chest CT in patients with chronic cough who have one or more high-risk factors for lung cancer. In nonsmoking adults, risk factors vary and include environmental pollutants, genetic variations, underlying chronic pulmonary disease, immunosuppression, and occupational exposures (ie, asbestos and silica) [52]. Cough has been shown to be a prevalent symptom in patients with lung cancer [49,50], 57% of a small cohort of consecutive patients with lung cancer (n = 223). However, the true prevalence of isolated cough (without any additional clinical features) in the setting of malignancy remains unknown. Some authors suggest this combination to be rare in the clinical setting of chronic cough [16]. In a retrospective analysis of patients with chronic cough who have normal chest radiographs, 2 out of 266 diagnoses were determined to be malignancy. Both patients had suspicious clinical findings, which prompted further evaluation with chest CT. It is also worth noting that both patients had a normal chest radiograph [16].

3158177

acrac_3158177_12

Chronic Cough

In prospective case series evaluating standardized clinical protocols for chronic cough, the prevalence of malignancy was found to be between 1% and 2% [7,17]. CT of the chest is recommended for the evaluation of suspected pulmonary neoplasm [15,53] and has been recommended for patients with chronic cough who are smokers or when chronic lung disease is clinically suspected [16,53]. Chest CT has been shown to be more sensitive than chest radiography for the detection of lung cancer in the National Lung Cancer Screening Trial [25], with significant overall mortality benefit. There is limited data regarding the differential imaging evaluation of patients with chronic cough with occupational exposures. In a small sample of male welders (n = 74) with a normal chest radiograph, HRCT identified 27 patients with micronodular disease, 5 patients with emphysematous changes, 3 patients with GGO, and 1 patient with pleural thickening [51], questioning the sensitivity of radiography for occupational disease. Red flags and concerning features are discussed by various authors and larger pulmonary societies [1,13,29], although our literature review did not identify studies directly evaluating CT of the chest in patients with chronic cough with elevated risk. Additionally, red flags in this setting have not been validated and shown to be predictive of malignancy. In the setting of elevated lung cancer risk, there is no high-quality evidence supporting chest CT in patients with chronic cough. It is worth noting that in the few case series evaluating patients with chronic cough, malignancy was diagnosed in 1% to 2% of studied groups. This is a not an insignificant quantity and actually rivals case detection in both lung cancer and breast cancer screening. Furthermore, at least two of these reported malignancies were in the context of a normal chest radiograph. It is likely that the prevalence of malignancy in the setting of chronic cough is far lower, and these studies are likely biased by selection.

Chronic Cough. In prospective case series evaluating standardized clinical protocols for chronic cough, the prevalence of malignancy was found to be between 1% and 2% [7,17]. CT of the chest is recommended for the evaluation of suspected pulmonary neoplasm [15,53] and has been recommended for patients with chronic cough who are smokers or when chronic lung disease is clinically suspected [16,53]. Chest CT has been shown to be more sensitive than chest radiography for the detection of lung cancer in the National Lung Cancer Screening Trial [25], with significant overall mortality benefit. There is limited data regarding the differential imaging evaluation of patients with chronic cough with occupational exposures. In a small sample of male welders (n = 74) with a normal chest radiograph, HRCT identified 27 patients with micronodular disease, 5 patients with emphysematous changes, 3 patients with GGO, and 1 patient with pleural thickening [51], questioning the sensitivity of radiography for occupational disease. Red flags and concerning features are discussed by various authors and larger pulmonary societies [1,13,29], although our literature review did not identify studies directly evaluating CT of the chest in patients with chronic cough with elevated risk. Additionally, red flags in this setting have not been validated and shown to be predictive of malignancy. In the setting of elevated lung cancer risk, there is no high-quality evidence supporting chest CT in patients with chronic cough. It is worth noting that in the few case series evaluating patients with chronic cough, malignancy was diagnosed in 1% to 2% of studied groups. This is a not an insignificant quantity and actually rivals case detection in both lung cancer and breast cancer screening. Furthermore, at least two of these reported malignancies were in the context of a normal chest radiograph. It is likely that the prevalence of malignancy in the setting of chronic cough is far lower, and these studies are likely biased by selection.

3158177

acrac_3158177_13

Chronic Cough

Nevertheless, there may be a cohort of patients with chronic cough who would benefit from earlier chest CT, potentially leap-frogging weeks of sequential Chronic Cough therapeutic trials that may in fact delay diagnosis. Future research may be needed to evaluate which patients with chronic cough would gain from earlier evaluation with CT. FDG-PET/CT Skull Base to Mid-Thigh There is no relevant literature to support the use of FDG-PET/CT in the evaluation of chronic cough with increased risk factors for lung cancer. FDG-PET utilizes a radiolabeled glucose analog to evaluate for upregulated metabolism in areas of infection, inflammation, and neoplasm [43]. It is often utilized in the setting of solitary pulmonary nodules to allow for risk stratification. In the setting of known malignancy, FDG-PET is suitable for staging nodal and metastatic involvement. FDG-PET provides valuable input for further noninvasive testing and operative planning. In patients with chronic cough who have elevated risk factors for lung cancer, it is unlikely that FDG- PET would significantly add value over other imaging modalities, such as chest radiography or chest CT. MRI Chest There is no relevant literature to support the use of thoracic MRI in the evaluation of patients with chronic cough who have increased risk of lung cancer. No studies specific to chronic cough distinguished between contrast- enhanced and noncontrast MRI. In the setting of neoplasm of the chest wall or mediastinum, the addition of extracellular contrast agent would allow for differentiation of cystic and solid features and may add value in planning appropriate invasive testing. Additionally, vasculature is imaged well with the addition of contrast, and contrast-enhanced MRI has been shown to be nearly equivalent to contrast-enhanced CT for the evaluation for pulmonary embolism [29]. MRI was not evaluated in the two prospective clinical series evaluating the utility of a standardized clinical protocol for the evaluation of chronic cough [7,17].

Chronic Cough. Nevertheless, there may be a cohort of patients with chronic cough who would benefit from earlier chest CT, potentially leap-frogging weeks of sequential Chronic Cough therapeutic trials that may in fact delay diagnosis. Future research may be needed to evaluate which patients with chronic cough would gain from earlier evaluation with CT. FDG-PET/CT Skull Base to Mid-Thigh There is no relevant literature to support the use of FDG-PET/CT in the evaluation of chronic cough with increased risk factors for lung cancer. FDG-PET utilizes a radiolabeled glucose analog to evaluate for upregulated metabolism in areas of infection, inflammation, and neoplasm [43]. It is often utilized in the setting of solitary pulmonary nodules to allow for risk stratification. In the setting of known malignancy, FDG-PET is suitable for staging nodal and metastatic involvement. FDG-PET provides valuable input for further noninvasive testing and operative planning. In patients with chronic cough who have elevated risk factors for lung cancer, it is unlikely that FDG- PET would significantly add value over other imaging modalities, such as chest radiography or chest CT. MRI Chest There is no relevant literature to support the use of thoracic MRI in the evaluation of patients with chronic cough who have increased risk of lung cancer. No studies specific to chronic cough distinguished between contrast- enhanced and noncontrast MRI. In the setting of neoplasm of the chest wall or mediastinum, the addition of extracellular contrast agent would allow for differentiation of cystic and solid features and may add value in planning appropriate invasive testing. Additionally, vasculature is imaged well with the addition of contrast, and contrast-enhanced MRI has been shown to be nearly equivalent to contrast-enhanced CT for the evaluation for pulmonary embolism [29]. MRI was not evaluated in the two prospective clinical series evaluating the utility of a standardized clinical protocol for the evaluation of chronic cough [7,17].

3158177

acrac_3158177_14

Chronic Cough

MRI is also not explicitly recommended in the available practice guidelines for the evaluation of chronic cough by the ACCP [1]. Limitations of pulmonary MRI include low signal-to-noise ratio (especially with normally aerated lung), low spatial resolution, and need for adequate breath-hold technique or respiratory gating [46]. Based on our literature review, there is limited evidence evaluating the diagnostic quality of MRI for the evaluation of pulmonary abnormalities. Ohno et al [47] analyzed 85 consecutive patients with a high prevalence of pulmonary abnormality with high-resolution pulmonary MR using ultrashort time to echo and compared findings with both standard and LDCT acquisition. High-resolution MRI demonstrated excellent correlation with standard dose CT for mass or nodule, GGO, and lymphadenopathy. MRI demonstrated 89% sensitivity and 99% specificity for mass or nodule identification. The authors noted diminished imaging quality; however, this did not affect overall evaluation. The role of MRI in the setting of chronic cough and increased cancer risk remains unknown at this time. Future research in this setting may be warranted for certain populations. CT Chest The most commonly identified etiologies reported for chronic cough are asthma, GERD, and UACS [1]. Prospective case series have shown clinical efficacy in diagnosing and managing chronic cough using standard clinical algorithms with minimal investigation [7,17]. When this has failed to result in diagnosis, or when empiric sequential treatment for the most common etiologies has failed to resolve symptoms, further investigative modalities and additional specialty referrals have been recommended. The ACCP and German Respiratory Society both recommend HRCT after an appropriate clinical evaluation and empiric trial has been performed [1,29]. HRCT is typically performed with multidetector helical CT, and the standard for performance can be reviewed in the ACR practice parameter for HRCT [54].

Chronic Cough. MRI is also not explicitly recommended in the available practice guidelines for the evaluation of chronic cough by the ACCP [1]. Limitations of pulmonary MRI include low signal-to-noise ratio (especially with normally aerated lung), low spatial resolution, and need for adequate breath-hold technique or respiratory gating [46]. Based on our literature review, there is limited evidence evaluating the diagnostic quality of MRI for the evaluation of pulmonary abnormalities. Ohno et al [47] analyzed 85 consecutive patients with a high prevalence of pulmonary abnormality with high-resolution pulmonary MR using ultrashort time to echo and compared findings with both standard and LDCT acquisition. High-resolution MRI demonstrated excellent correlation with standard dose CT for mass or nodule, GGO, and lymphadenopathy. MRI demonstrated 89% sensitivity and 99% specificity for mass or nodule identification. The authors noted diminished imaging quality; however, this did not affect overall evaluation. The role of MRI in the setting of chronic cough and increased cancer risk remains unknown at this time. Future research in this setting may be warranted for certain populations. CT Chest The most commonly identified etiologies reported for chronic cough are asthma, GERD, and UACS [1]. Prospective case series have shown clinical efficacy in diagnosing and managing chronic cough using standard clinical algorithms with minimal investigation [7,17]. When this has failed to result in diagnosis, or when empiric sequential treatment for the most common etiologies has failed to resolve symptoms, further investigative modalities and additional specialty referrals have been recommended. The ACCP and German Respiratory Society both recommend HRCT after an appropriate clinical evaluation and empiric trial has been performed [1,29]. HRCT is typically performed with multidetector helical CT, and the standard for performance can be reviewed in the ACR practice parameter for HRCT [54].

3158177

acrac_3158177_15

Chronic Cough

Briefly, thin-slice acquisition and high spatial resolution reconstruction algorithms allow for near-isotropic voxels. This gives the reader additional tools for a thorough evaluation of the imaged field, including multiplanar reformats and both maximum and minimum intensity projection postprocessed images. In the setting of chronic cough, however, the various modalities and specific protocol designs have not been adequately investigated and compared. It is unclear how much the added resolving power of HRCT adds to the clinical evaluation of patients with chronic cough. Contrast- and noncontrast-enhanced studies have also not been compared. The addition of contrast would allow for improved visualization of the mediastinum, chest wall, and soft Chronic Cough HRCT is considered the reference standard for the evaluation of bronchiectasis. Bronchiectasis has been identified as an etiology for chronic cough in up to 8% of patients in case series [2,7]. Although suspicion for bronchiectasis can be aided by clinical features [5], chest radiography has shown poor sensitivity for evaluation. Altenburg et al [32] reported that radiographic findings did not detect ectatic airways in up to 34% of patients reviewed. Bronchiectasis and bronchial wall thickening were identified in 27% and 21% of patients, respectively, referred to chest CT for the evaluation of chronic cough in whom radiographs were reportedly normal [3]. There remains variable association between identification of bronchiectasis and clinical features. Bronchiectasis has been found by chest CT in asymptomatic persons enrolled in a health survey screening setting [4], as well as in asymptomatic elderly persons [37]. Wilsher et al [40] evaluated 60 consecutive patients with newly diagnosed RA and were unable to demonstrate a correlation between cough symptoms and bronchiectasis, although association reached significance for dyspnea.

Chronic Cough. Briefly, thin-slice acquisition and high spatial resolution reconstruction algorithms allow for near-isotropic voxels. This gives the reader additional tools for a thorough evaluation of the imaged field, including multiplanar reformats and both maximum and minimum intensity projection postprocessed images. In the setting of chronic cough, however, the various modalities and specific protocol designs have not been adequately investigated and compared. It is unclear how much the added resolving power of HRCT adds to the clinical evaluation of patients with chronic cough. Contrast- and noncontrast-enhanced studies have also not been compared. The addition of contrast would allow for improved visualization of the mediastinum, chest wall, and soft Chronic Cough HRCT is considered the reference standard for the evaluation of bronchiectasis. Bronchiectasis has been identified as an etiology for chronic cough in up to 8% of patients in case series [2,7]. Although suspicion for bronchiectasis can be aided by clinical features [5], chest radiography has shown poor sensitivity for evaluation. Altenburg et al [32] reported that radiographic findings did not detect ectatic airways in up to 34% of patients reviewed. Bronchiectasis and bronchial wall thickening were identified in 27% and 21% of patients, respectively, referred to chest CT for the evaluation of chronic cough in whom radiographs were reportedly normal [3]. There remains variable association between identification of bronchiectasis and clinical features. Bronchiectasis has been found by chest CT in asymptomatic persons enrolled in a health survey screening setting [4], as well as in asymptomatic elderly persons [37]. Wilsher et al [40] evaluated 60 consecutive patients with newly diagnosed RA and were unable to demonstrate a correlation between cough symptoms and bronchiectasis, although association reached significance for dyspnea.

3158177

acrac_3158177_16

Chronic Cough

Furthermore, bronchiectasis and bronchial wall thickening were identified in 48% and 58% of patients, respectively, with only 30% reporting any symptoms. Ooi et al [6] were unable to correlate the frequency of acute symptom exacerbation (including cough symptoms) and the degree of bronchial wall thickening in 60 consecutive patients with CT-proven bronchiectasis. In a large single-center prospective study, Grydeland et al [39] used quantitative analysis of bronchial wall thickening and compared findings between patients with COPD (n = 463) and those without COPD (n = 488). Airway wall thickening was found to be statistically associated with chronic cough in patients with COPD but did not reach significance in the non-COPD group. Interstitial lung disease and pulmonary fibrosis have been reported as etiologies of chronic cough, although prevalence of diagnosed cases is variable, ranging between 0.4% and 8% of the reviewed case series [2,7,16,17]. Chest CT is considered the reference standard in the evaluation of interstitial lung disease. Although at least one study reported good correlation between chest radiography and chest CT in this setting [46], chest CT has been shown to be more sensitive than both chest radiography [31] and LDCT [27]. In 97 patients with chronic cough who also had normal chest radiographs, Truba et al [3] reported interstitial abnormalities in 4 patients. Association between cough symptoms and CT findings of interstitial lung disease has been variable. In a small prospective cohort of 36 patients with RA, cough was not found to be statistically associated with restrictive findings on either PFT or HRCT analysis [31]. In 107 consecutive patients with systemic sclerosis, cough was associated with HRCT findings [31]. In the setting of chronic cough after failed initial evaluation and/or empiric treatment, the limited evidence suggests a role for chest CT to identify underlying pulmonary disease.

Chronic Cough. Furthermore, bronchiectasis and bronchial wall thickening were identified in 48% and 58% of patients, respectively, with only 30% reporting any symptoms. Ooi et al [6] were unable to correlate the frequency of acute symptom exacerbation (including cough symptoms) and the degree of bronchial wall thickening in 60 consecutive patients with CT-proven bronchiectasis. In a large single-center prospective study, Grydeland et al [39] used quantitative analysis of bronchial wall thickening and compared findings between patients with COPD (n = 463) and those without COPD (n = 488). Airway wall thickening was found to be statistically associated with chronic cough in patients with COPD but did not reach significance in the non-COPD group. Interstitial lung disease and pulmonary fibrosis have been reported as etiologies of chronic cough, although prevalence of diagnosed cases is variable, ranging between 0.4% and 8% of the reviewed case series [2,7,16,17]. Chest CT is considered the reference standard in the evaluation of interstitial lung disease. Although at least one study reported good correlation between chest radiography and chest CT in this setting [46], chest CT has been shown to be more sensitive than both chest radiography [31] and LDCT [27]. In 97 patients with chronic cough who also had normal chest radiographs, Truba et al [3] reported interstitial abnormalities in 4 patients. Association between cough symptoms and CT findings of interstitial lung disease has been variable. In a small prospective cohort of 36 patients with RA, cough was not found to be statistically associated with restrictive findings on either PFT or HRCT analysis [31]. In 107 consecutive patients with systemic sclerosis, cough was associated with HRCT findings [31]. In the setting of chronic cough after failed initial evaluation and/or empiric treatment, the limited evidence suggests a role for chest CT to identify underlying pulmonary disease.

3158177

acrac_3158177_17

Chronic Cough

Although it remains unclear if HRCT or LDCT play a role, clinical features may stratify patients for appropriate imaging evaluation in order to balance diagnostic yield and dose considerations. CT Maxillofacial CT of the sinuses and maxillofacial structures offers improved visualization of the sinonasal anatomy, allowing for localization, characterization, and grading of sinonasal pathology. UACS is largely a conglomerate terminology for cough related to inflammation or irritation of the upper airways, which includes postnasal drip, allergic and nonallergic CRS, as well as laryngeal reflux. UACS is considered one of the more common causes of chronic cough [1,53]. Based on a few case series of patients with chronic cough, UACS/rhinitis prevalence ranged from 6% to 65% [2,7,17]. Guidelines for the evaluation of chronic cough by the ACCP places sinus imaging under consideration after initial clinical examination and chest radiography [1]. The clinical evaluation of UACS is not sensitive and often not specific [53], and features may go unrecognized even by endoscopy [12,55,56], creating an integral role for noninvasive imaging. Symptoms alone were found to have a sensitivity between 37% and 73% in a retrospective analysis of 126 patients who underwent CT imaging for the evaluation of CRS [57]. Furthermore, up to 35% of patients who were diagnosed with CRS had normal endoscopic findings [57]. Abrass et al [55] reviewed 100 patients with clinical findings suspicious of CRS, all with negative endoscopic evaluations. The investigators reported between 20% and 49% of studied individuals had positive findings on point-of-care CT of the sinuses, and the utilization of imaging allowed for a decrease in unnecessary antibiotic prescription. In a small retrospective case control study evaluating patients with CRS, Conley et al [56] showed increased clinical accuracy and a lower rate of unnecessary antibiotic use when cone-beam CT was used to evaluate the sinuses at initial office visit. In a

Chronic Cough. Although it remains unclear if HRCT or LDCT play a role, clinical features may stratify patients for appropriate imaging evaluation in order to balance diagnostic yield and dose considerations. CT Maxillofacial CT of the sinuses and maxillofacial structures offers improved visualization of the sinonasal anatomy, allowing for localization, characterization, and grading of sinonasal pathology. UACS is largely a conglomerate terminology for cough related to inflammation or irritation of the upper airways, which includes postnasal drip, allergic and nonallergic CRS, as well as laryngeal reflux. UACS is considered one of the more common causes of chronic cough [1,53]. Based on a few case series of patients with chronic cough, UACS/rhinitis prevalence ranged from 6% to 65% [2,7,17]. Guidelines for the evaluation of chronic cough by the ACCP places sinus imaging under consideration after initial clinical examination and chest radiography [1]. The clinical evaluation of UACS is not sensitive and often not specific [53], and features may go unrecognized even by endoscopy [12,55,56], creating an integral role for noninvasive imaging. Symptoms alone were found to have a sensitivity between 37% and 73% in a retrospective analysis of 126 patients who underwent CT imaging for the evaluation of CRS [57]. Furthermore, up to 35% of patients who were diagnosed with CRS had normal endoscopic findings [57]. Abrass et al [55] reviewed 100 patients with clinical findings suspicious of CRS, all with negative endoscopic evaluations. The investigators reported between 20% and 49% of studied individuals had positive findings on point-of-care CT of the sinuses, and the utilization of imaging allowed for a decrease in unnecessary antibiotic prescription. In a small retrospective case control study evaluating patients with CRS, Conley et al [56] showed increased clinical accuracy and a lower rate of unnecessary antibiotic use when cone-beam CT was used to evaluate the sinuses at initial office visit. In a

3158177

acrac_70910_0

Female Breast Cancer Screening

Introduction/Background Breast cancer is the most common nonskin cancer diagnosis in women and is second only to lung cancer with respect to cancer deaths. Early detection of breast cancer from regular screening substantially reduces breast cancer mortality [1]. Because regular screening identifies tumors when they are smaller and with fewer nodal metastases, patients with screen-detected breast cancers are less likely to require mastectomy or chemotherapy, thereby also decreasing morbidity [2]. Breast cancer risk is frequently divided into 3 major categories: average, intermediate, and high risk. Numerous factors contribute to breast cancer risk, so no single method or definition is used to classify each woman into a specific risk category [3,4]. The use of validated statistical models based largely upon family history, which also incorporate additional risk factors, represents one mechanism to estimate risk. Currently, risk categories are most frequently defined by estimated lifetime risk; however, different time horizons, such as 5 or 10 year risk, may also be valuable for guideline development and informed decision-making [3]. Women at average risk are typically defined as those with <15% estimated lifetime risk for developing breast cancer, whereas intermediate-risk women are generally defined as those with a 15% to 20% estimated lifetime risk. The high-risk category typically includes women who have a >20 to 25% estimated lifetime risk: women who carry a deleterious genetic mutation that increases breast cancer risk, as well as untested first-degree relatives of patients with these mutations and women who have received radiation therapy to the thorax or upper abdomen at an early age (<30 years).

Female Breast Cancer Screening. Introduction/Background Breast cancer is the most common nonskin cancer diagnosis in women and is second only to lung cancer with respect to cancer deaths. Early detection of breast cancer from regular screening substantially reduces breast cancer mortality [1]. Because regular screening identifies tumors when they are smaller and with fewer nodal metastases, patients with screen-detected breast cancers are less likely to require mastectomy or chemotherapy, thereby also decreasing morbidity [2]. Breast cancer risk is frequently divided into 3 major categories: average, intermediate, and high risk. Numerous factors contribute to breast cancer risk, so no single method or definition is used to classify each woman into a specific risk category [3,4]. The use of validated statistical models based largely upon family history, which also incorporate additional risk factors, represents one mechanism to estimate risk. Currently, risk categories are most frequently defined by estimated lifetime risk; however, different time horizons, such as 5 or 10 year risk, may also be valuable for guideline development and informed decision-making [3]. Women at average risk are typically defined as those with <15% estimated lifetime risk for developing breast cancer, whereas intermediate-risk women are generally defined as those with a 15% to 20% estimated lifetime risk. The high-risk category typically includes women who have a >20 to 25% estimated lifetime risk: women who carry a deleterious genetic mutation that increases breast cancer risk, as well as untested first-degree relatives of patients with these mutations and women who have received radiation therapy to the thorax or upper abdomen at an early age (<30 years).

70910

acrac_70910_1

Female Breast Cancer Screening

Some women with a personal history of high-risk breast lesions, a personal history of breast cancer, dense breast tissue, or a family history of breast cancer may fit into the intermediate- or high-risk categories, depending upon their specific risk factors or combination of factors [3]. Elevated risk is sometimes used to refer to women in both the intermediate- and high-risk categories [3]. Breast cancer screening guidelines vary across medical professional organizations, although published guidelines agree that regular breast cancer screening decreases morbidity and breast cancer mortality [5-7]. Medical professional organizations may also define breast cancer risk categories using different methodologies. Although screening guidelines for high-risk patients have typically been similar, discrepant recommendations for average- and intermediate-risk women have sparked controversy and confusion. In part due to differences in screening guidelines, use of breast cancer screening modalities remains suboptimal in women of all risk categories. The ACR encourages patients to undergo breast cancer risk assessment by 25 years of age, so elevated-risk patients have the opportunity to benefit from earlier and more aggressive breast cancer screening regimens, when appropriate [3]. The ACR recommends that both the benefits and risks of breast cancer screening and supplemental screening be considered to assist patients in making informed decisions regarding their health care [8]. aPanel Chair, H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida. bMemorial Sloan Kettering Cancer Center, New York, New York. cMemorial Sloan Kettering Cancer Center, New York, New York. dPanel Vice-Chair, University of Cincinnati, Cincinnati, Ohio. eClinica Family Health, Lafayette, Colorado; American Academy of Family Physicians. fLenox Hill Hospital, Northwell Health, New York, New York; American College of Surgeons. gWashington University School of Medicine, Saint Louis, Missouri.

Female Breast Cancer Screening. Some women with a personal history of high-risk breast lesions, a personal history of breast cancer, dense breast tissue, or a family history of breast cancer may fit into the intermediate- or high-risk categories, depending upon their specific risk factors or combination of factors [3]. Elevated risk is sometimes used to refer to women in both the intermediate- and high-risk categories [3]. Breast cancer screening guidelines vary across medical professional organizations, although published guidelines agree that regular breast cancer screening decreases morbidity and breast cancer mortality [5-7]. Medical professional organizations may also define breast cancer risk categories using different methodologies. Although screening guidelines for high-risk patients have typically been similar, discrepant recommendations for average- and intermediate-risk women have sparked controversy and confusion. In part due to differences in screening guidelines, use of breast cancer screening modalities remains suboptimal in women of all risk categories. The ACR encourages patients to undergo breast cancer risk assessment by 25 years of age, so elevated-risk patients have the opportunity to benefit from earlier and more aggressive breast cancer screening regimens, when appropriate [3]. The ACR recommends that both the benefits and risks of breast cancer screening and supplemental screening be considered to assist patients in making informed decisions regarding their health care [8]. aPanel Chair, H. Lee Moffitt Cancer Center and Research Institute, Tampa, Florida. bMemorial Sloan Kettering Cancer Center, New York, New York. cMemorial Sloan Kettering Cancer Center, New York, New York. dPanel Vice-Chair, University of Cincinnati, Cincinnati, Ohio. eClinica Family Health, Lafayette, Colorado; American Academy of Family Physicians. fLenox Hill Hospital, Northwell Health, New York, New York; American College of Surgeons. gWashington University School of Medicine, Saint Louis, Missouri.

70910

acrac_70910_2

Female Breast Cancer Screening

hPenn State Health Hershey Medical Center, Hershey, Pennsylvania. iThomas Jefferson University Hospital, Philadelphia, Pennsylvania; American College of Obstetricians and Gynecologists. jUniversity of Utah, Salt Lake City, Utah. kKaiser Permanente, Atlanta, Georgia. lUniversity of Michigan, Ann Arbor, Michigan. mSt. Bernards Healthcare, Jonesboro, Arkansas. nUMass Memorial Medical Center/UMass Chan Medical School, Worcester, Massachusetts. oNYU Clinical Cancer Center, New York, New York. pProMedica Breast Care, Toledo, Ohio. qEmory University Hospital, Atlanta, Georgia; RADS Committee. rYale School of Medicine, New Haven, Connecticut; Society of General Internal Medicine. sHarvard Medical School, Boston, Massachusetts; American Geriatrics Society. tLoyola University Chicago, Stritch School of Medicine, Department of Radiation Oncology, Cardinal Bernardin Cancer Center, Maywood, Illinois; Commission on Radiation Oncology. uHoag Family Cancer Institute, Newport Beach, California and University of Southern California, Los Angeles, California; Commission on Nuclear Medicine and Molecular Imaging. vSpecialty Chair, Boston University School of Medicine, Boston, Massachusetts. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] Female Breast Cancer Screening Discussion of Procedures by Variant Variant 1: Adult female. Breast cancer screening. Average risk.

Female Breast Cancer Screening. hPenn State Health Hershey Medical Center, Hershey, Pennsylvania. iThomas Jefferson University Hospital, Philadelphia, Pennsylvania; American College of Obstetricians and Gynecologists. jUniversity of Utah, Salt Lake City, Utah. kKaiser Permanente, Atlanta, Georgia. lUniversity of Michigan, Ann Arbor, Michigan. mSt. Bernards Healthcare, Jonesboro, Arkansas. nUMass Memorial Medical Center/UMass Chan Medical School, Worcester, Massachusetts. oNYU Clinical Cancer Center, New York, New York. pProMedica Breast Care, Toledo, Ohio. qEmory University Hospital, Atlanta, Georgia; RADS Committee. rYale School of Medicine, New Haven, Connecticut; Society of General Internal Medicine. sHarvard Medical School, Boston, Massachusetts; American Geriatrics Society. tLoyola University Chicago, Stritch School of Medicine, Department of Radiation Oncology, Cardinal Bernardin Cancer Center, Maywood, Illinois; Commission on Radiation Oncology. uHoag Family Cancer Institute, Newport Beach, California and University of Southern California, Los Angeles, California; Commission on Nuclear Medicine and Molecular Imaging. vSpecialty Chair, Boston University School of Medicine, Boston, Massachusetts. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] Female Breast Cancer Screening Discussion of Procedures by Variant Variant 1: Adult female. Breast cancer screening. Average risk.

70910

acrac_70910_3

Female Breast Cancer Screening

Digital Breast Tomosynthesis Screening Digital breast tomosynthesis (DBT) displays reconstructed stacked images of the breast in combination with digital mammographic views, which may be synthetic mammograms reconstructed from the acquired tomosynthesis data set or full-field digital mammograms (FFDM). Compared to FFDM or synthetic mammograms alone, most studies demonstrate that DBT increases cancer detection rate (CDR) and decreases recall rate [14-22]; although some studies have not reached statistical significance [23] or have found less compelling results in subsets of women, such as those with extremely dense breasts [24,25]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27]. Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. Irrespective of risk category, meta-analyses have demonstrated an incremental increase in CDR of 1.6 to 3.2 per 1,000 screening DBT examinations and a 2.2% pooled decrease in recall rate compared to digital mammography [18,29,30]. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. In addition to mortality reduction, screening mammography decreases treatment morbidity, because screen-detected tumors are typically lower stage (eg, smaller and more likely to be node-negative), compared to breast cancers detected by palpation [2,8]. Despite these benefits, screening mammograms also have risks. The most common perceived risks include false-positive recalls and biopsies, overdiagnosis, and patient anxiety [5,7,32].

Female Breast Cancer Screening. Digital Breast Tomosynthesis Screening Digital breast tomosynthesis (DBT) displays reconstructed stacked images of the breast in combination with digital mammographic views, which may be synthetic mammograms reconstructed from the acquired tomosynthesis data set or full-field digital mammograms (FFDM). Compared to FFDM or synthetic mammograms alone, most studies demonstrate that DBT increases cancer detection rate (CDR) and decreases recall rate [14-22]; although some studies have not reached statistical significance [23] or have found less compelling results in subsets of women, such as those with extremely dense breasts [24,25]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27]. Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. Irrespective of risk category, meta-analyses have demonstrated an incremental increase in CDR of 1.6 to 3.2 per 1,000 screening DBT examinations and a 2.2% pooled decrease in recall rate compared to digital mammography [18,29,30]. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. In addition to mortality reduction, screening mammography decreases treatment morbidity, because screen-detected tumors are typically lower stage (eg, smaller and more likely to be node-negative), compared to breast cancers detected by palpation [2,8]. Despite these benefits, screening mammograms also have risks. The most common perceived risks include false-positive recalls and biopsies, overdiagnosis, and patient anxiety [5,7,32].

70910

acrac_70910_4

Female Breast Cancer Screening

Approximately 10% of screening mammograms result in a recall for additional imaging, although <2% result in a Female Breast Cancer Screening For women 40 to 49 years of age, randomized controlled trials and observational studies demonstrate that screening mammography decreases breast cancer mortality by 15% to 50% [1,8,32,33,39]. Results from the Cancer Intervention and Surveillance Modeling Network (CISNET) suggest that annual screening mammography in women 40 to 49 years of age saves 42% more lives and life-years than biennial screening due to faster growing tumors in younger women [31]. Women screened between 40 and 49 years of age are also less likely to require mastectomy or chemotherapy than women diagnosed with palpable tumors [2]. Non-Hispanic Black, Hispanic Black, and Hispanic White women have higher breast cancer mortality than non- Hispanic White women, and minority women often present at younger ages with more aggressive tumor subtypes [3,8]. Therefore, decreasing access to screening mammography, especially in women 40 to 49 years of age, may disproportionately impact minority women. Annual screening mammography results in a greater reduction in mortality compared to biennial screening [8]. In women 40 to 84 years of age, annual screening reduces mortality by 40%, compared to a 32% reduction for biennial screening [32]. With regular screening, interval breast cancers do occur with a higher frequency in women undergoing biennial or triennial screening compared to annual screening. The sensitivity of mammography is decreased in some groups of women, including those with dense breasts [40]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27].

Female Breast Cancer Screening. Approximately 10% of screening mammograms result in a recall for additional imaging, although <2% result in a Female Breast Cancer Screening For women 40 to 49 years of age, randomized controlled trials and observational studies demonstrate that screening mammography decreases breast cancer mortality by 15% to 50% [1,8,32,33,39]. Results from the Cancer Intervention and Surveillance Modeling Network (CISNET) suggest that annual screening mammography in women 40 to 49 years of age saves 42% more lives and life-years than biennial screening due to faster growing tumors in younger women [31]. Women screened between 40 and 49 years of age are also less likely to require mastectomy or chemotherapy than women diagnosed with palpable tumors [2]. Non-Hispanic Black, Hispanic Black, and Hispanic White women have higher breast cancer mortality than non- Hispanic White women, and minority women often present at younger ages with more aggressive tumor subtypes [3,8]. Therefore, decreasing access to screening mammography, especially in women 40 to 49 years of age, may disproportionately impact minority women. Annual screening mammography results in a greater reduction in mortality compared to biennial screening [8]. In women 40 to 84 years of age, annual screening reduces mortality by 40%, compared to a 32% reduction for biennial screening [32]. With regular screening, interval breast cancers do occur with a higher frequency in women undergoing biennial or triennial screening compared to annual screening. The sensitivity of mammography is decreased in some groups of women, including those with dense breasts [40]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27].

70910

acrac_70910_5

Female Breast Cancer Screening

Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. Given the limitations of mammography and to minimize interval cancers, supplemental screening modalities have been investigated in women at average risk. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. Rather than supplementing screening mammography with additional imaging modalities, some have suggested limiting women offered screening mammography based upon individual patient risk assessed by various risk models, breast density, or genetic information such as single-nucleotide polymorphism. However, the randomized controlled trials demonstrating mortality reduction and most large-scale observational studies enrolled women based upon geographic location and age, not other individual patient risk factors. In one observational study in women <50 Female Breast Cancer Screening years of age, restricting screening to women with a first-degree family history, extremely dense breast tissue, or both, would cause 66% of potentially screen-detected cancers to be missed [41]. To maximize the benefits, the ACR recommends screening mammography in average-risk women each year beginning at 40 years of age. Women should continue screening mammography as long as they remain in overall good health and are willing to undergo the examination and subsequent testing or biopsy, if an abnormality is identified [5,8].

Female Breast Cancer Screening. Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. Given the limitations of mammography and to minimize interval cancers, supplemental screening modalities have been investigated in women at average risk. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. Rather than supplementing screening mammography with additional imaging modalities, some have suggested limiting women offered screening mammography based upon individual patient risk assessed by various risk models, breast density, or genetic information such as single-nucleotide polymorphism. However, the randomized controlled trials demonstrating mortality reduction and most large-scale observational studies enrolled women based upon geographic location and age, not other individual patient risk factors. In one observational study in women <50 Female Breast Cancer Screening years of age, restricting screening to women with a first-degree family history, extremely dense breast tissue, or both, would cause 66% of potentially screen-detected cancers to be missed [41]. To maximize the benefits, the ACR recommends screening mammography in average-risk women each year beginning at 40 years of age. Women should continue screening mammography as long as they remain in overall good health and are willing to undergo the examination and subsequent testing or biopsy, if an abnormality is identified [5,8].

70910

acrac_70910_6

Female Breast Cancer Screening

MRI Breast Without and With IV Contrast Although data are limited regarding the use of breast MRI without and with IV contrast for screening women at average risk, one study has demonstrated that breast MRI demonstrates incremental cancer detection (15-16 cancers per 1,000 breast MRI examinations) over screening mammography with or without screening ultrasound (US) in average-risk women irrespective of breast density [42]. Breast MRI also decreases interval cancers [42,43]. In the DENSE trial, breast MRI significantly reduced interval cancers within women with extremely dense breast tissue and normal mammography, so the European Society of Breast Imaging now recommends screening breast MRI every 2 to 4 years in women 50 to 70 years of age with extremely dense breasts [43,44]. Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. MRI Breast Without and With IV Contrast Abbreviated Data are limited regarding the use of abbreviated breast MRI without and with IV contrast for screening women at average risk. The ECOG-ACRIN abbreviated MRI trial demonstrated a significantly higher CDR for abbreviated breast MRI without and with IV contrast (15 cancers per 1,000) compared with DBT (6 cancers per 1,000), although the study recruited women with dense breasts [45]. In addition to dense breasts, women enrolled in the trial had variable 5 and 10 year risk profiles based upon the Breast Cancer Surveillance Consortium risk calculator and 19% reported 1 or more first degree relatives with breast cancer [45].

Female Breast Cancer Screening. MRI Breast Without and With IV Contrast Although data are limited regarding the use of breast MRI without and with IV contrast for screening women at average risk, one study has demonstrated that breast MRI demonstrates incremental cancer detection (15-16 cancers per 1,000 breast MRI examinations) over screening mammography with or without screening ultrasound (US) in average-risk women irrespective of breast density [42]. Breast MRI also decreases interval cancers [42,43]. In the DENSE trial, breast MRI significantly reduced interval cancers within women with extremely dense breast tissue and normal mammography, so the European Society of Breast Imaging now recommends screening breast MRI every 2 to 4 years in women 50 to 70 years of age with extremely dense breasts [43,44]. Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. MRI Breast Without and With IV Contrast Abbreviated Data are limited regarding the use of abbreviated breast MRI without and with IV contrast for screening women at average risk. The ECOG-ACRIN abbreviated MRI trial demonstrated a significantly higher CDR for abbreviated breast MRI without and with IV contrast (15 cancers per 1,000) compared with DBT (6 cancers per 1,000), although the study recruited women with dense breasts [45]. In addition to dense breasts, women enrolled in the trial had variable 5 and 10 year risk profiles based upon the Breast Cancer Surveillance Consortium risk calculator and 19% reported 1 or more first degree relatives with breast cancer [45].

70910

acrac_70910_7

Female Breast Cancer Screening

Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. MRI Breast Without IV Contrast There is no relevant literature to support the use of MRI without IV contrast for screening women at average risk. MRI Breast Without IV Contrast Abbreviated There is no relevant literature to support the use of abbreviated MRI without IV contrast for screening women at average risk. Sestamibi MBI Data are limited regarding the use of sestamibi molecular breast imaging (MBI) for screening women at average risk. Most studies have focused upon women with dense breasts and variable risk profiles. One of the larger studies published to date of 1,696 women with recent negative or benign mammographic examinations showed that sestamibi MBI yielded an incremental CDR of 7.7 cancers per 1,000 examinations; however, all 13 cancers were detected in women with dense breasts [46]. Although 92% of the women within the study had <20% estimated lifetime risk, the estimates ranged from 6.1% to 17.2% [46]. Additional retrospective and prospective studies have demonstrated similar incremental CDR for sestamibi MBI of 6.5 to 9 per 1,000 over mammography [40,47]. Female Breast Cancer Screening Sestamibi MBI demonstrates similar sensitivity, better specificity, and lower recall rate compared to supplemental screening US in women with dense breasts [47,48]. US Breast Most studies evaluating the utility of screening with breast US have focused on women with dense breast tissue with or without other risk factors. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27].

Female Breast Cancer Screening. Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. MRI Breast Without IV Contrast There is no relevant literature to support the use of MRI without IV contrast for screening women at average risk. MRI Breast Without IV Contrast Abbreviated There is no relevant literature to support the use of abbreviated MRI without IV contrast for screening women at average risk. Sestamibi MBI Data are limited regarding the use of sestamibi molecular breast imaging (MBI) for screening women at average risk. Most studies have focused upon women with dense breasts and variable risk profiles. One of the larger studies published to date of 1,696 women with recent negative or benign mammographic examinations showed that sestamibi MBI yielded an incremental CDR of 7.7 cancers per 1,000 examinations; however, all 13 cancers were detected in women with dense breasts [46]. Although 92% of the women within the study had <20% estimated lifetime risk, the estimates ranged from 6.1% to 17.2% [46]. Additional retrospective and prospective studies have demonstrated similar incremental CDR for sestamibi MBI of 6.5 to 9 per 1,000 over mammography [40,47]. Female Breast Cancer Screening Sestamibi MBI demonstrates similar sensitivity, better specificity, and lower recall rate compared to supplemental screening US in women with dense breasts [47,48]. US Breast Most studies evaluating the utility of screening with breast US have focused on women with dense breast tissue with or without other risk factors. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27].

70910

acrac_70910_8

Female Breast Cancer Screening

Screening breast US in women with mammographically dense breasts, including those with risk factors placing them at increased breast cancer risk, identifies mammographically occult, small, node-negative invasive tumors with an increased CDR of 1.8 to 4.6 cancers per 1,000 women screened [40,49]. Although supplemental screening US in women with dense breasts results in an increased CDR, US also increases recall rate, false-positive examinations, and false-positive biopsies [26,49-55]. Data regarding supplemental screening US in average-risk women with nondense breasts is less compelling. In a study of 1,526 average-risk women without mammographic abnormalities, screening with US demonstrated an overall incremental CDR of 3.3 per 1,000, with 5.1 per 1,000 examinations in dense breasts and 0 per 1,000 in nondense breasts compared to digital mammography [56]. In another study of 1,003 average-risk women, US yielded an overall incremental CDR of 3.2 per 1,000 examinations, with 0 per 1,000 in nondense breasts, compared to DBT with or without digital mammography [53]. Digital Breast Tomosynthesis Screening DBT displays reconstructed stacked images of the breast in combination with digital mammographic views, which may be synthetic mammograms reconstructed from the acquired tomosynthesis dataset or FFDM. Compared to FFDM or synthetic mammograms alone, most studies demonstrate that DBT increases CDR and decreases recall rate [14-22]; although, some studies have not reached statistical significance [23] or have found less compelling results in subsets of women, such as those with extremely dense breasts [24,25]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27].

Female Breast Cancer Screening. Screening breast US in women with mammographically dense breasts, including those with risk factors placing them at increased breast cancer risk, identifies mammographically occult, small, node-negative invasive tumors with an increased CDR of 1.8 to 4.6 cancers per 1,000 women screened [40,49]. Although supplemental screening US in women with dense breasts results in an increased CDR, US also increases recall rate, false-positive examinations, and false-positive biopsies [26,49-55]. Data regarding supplemental screening US in average-risk women with nondense breasts is less compelling. In a study of 1,526 average-risk women without mammographic abnormalities, screening with US demonstrated an overall incremental CDR of 3.3 per 1,000, with 5.1 per 1,000 examinations in dense breasts and 0 per 1,000 in nondense breasts compared to digital mammography [56]. In another study of 1,003 average-risk women, US yielded an overall incremental CDR of 3.2 per 1,000 examinations, with 0 per 1,000 in nondense breasts, compared to DBT with or without digital mammography [53]. Digital Breast Tomosynthesis Screening DBT displays reconstructed stacked images of the breast in combination with digital mammographic views, which may be synthetic mammograms reconstructed from the acquired tomosynthesis dataset or FFDM. Compared to FFDM or synthetic mammograms alone, most studies demonstrate that DBT increases CDR and decreases recall rate [14-22]; although, some studies have not reached statistical significance [23] or have found less compelling results in subsets of women, such as those with extremely dense breasts [24,25]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27].

70910

acrac_70910_9

Female Breast Cancer Screening

Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. Irrespective of risk category, meta-analyses have demonstrated an incremental increase in CDR of 1.6 to 3.2 per 1,000 screening DBT examinations and a 2.2% pooled decrease in recall rate compared to digital mammography [18,29,30]. The degree of breast cancer mortality reduction from screening mammography varies with different screening regimens. Mortality reduction is greater when screening begins at 40 years of age rather than 45 or 50 years of age Within the limited studies of women at elevated risk due to personal and/or family history of breast cancer, DBT decreased recall rate without a significant increase in CDR compared to FFDM; however, small sample sizes restrict analyses [3,40]. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. The ACR recommends annual screening mammography beginning no later than 40 years of age for women at intermediate risk [3]. For those with a family history of breast cancer, mammography should begin earlier if familial breast cancer occurred at a young age, typically 10 years prior to the youngest age at presentation but generally not before age 30 [6]. For women who have lobular neoplasia or atypical hyperplasia diagnosed prior to 40 years of age, annual screening mammography should be performed from time of diagnosis but generally not prior to 30 years of age [38].

Female Breast Cancer Screening. Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. Irrespective of risk category, meta-analyses have demonstrated an incremental increase in CDR of 1.6 to 3.2 per 1,000 screening DBT examinations and a 2.2% pooled decrease in recall rate compared to digital mammography [18,29,30]. The degree of breast cancer mortality reduction from screening mammography varies with different screening regimens. Mortality reduction is greater when screening begins at 40 years of age rather than 45 or 50 years of age Within the limited studies of women at elevated risk due to personal and/or family history of breast cancer, DBT decreased recall rate without a significant increase in CDR compared to FFDM; however, small sample sizes restrict analyses [3,40]. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. The ACR recommends annual screening mammography beginning no later than 40 years of age for women at intermediate risk [3]. For those with a family history of breast cancer, mammography should begin earlier if familial breast cancer occurred at a young age, typically 10 years prior to the youngest age at presentation but generally not before age 30 [6]. For women who have lobular neoplasia or atypical hyperplasia diagnosed prior to 40 years of age, annual screening mammography should be performed from time of diagnosis but generally not prior to 30 years of age [38].

70910

acrac_70910_10

Female Breast Cancer Screening

Early detection of second breast cancers improves survival, so patients with a personal history of breast cancer should undergo annual mammography or DBT for surveillance following breast conservation therapy [3]. Despite the established mortality benefit, published guidelines differ in their recommendations for screening mammography due to variations in the perceptions of the relative risks and benefits [5,38]. The degree of breast cancer mortality reduction from screening mammography varies with different screening regimens. Mortality reduction is greater when screening begins 40 years of age rather than 45 or 50 years of age and when screening is done more frequently (annually rather than biennially) [8,31]. Annual screening mammography for women 40 to 84 years of age decreases mortality by 40% (12 lives per 1,000 women screened), whereas biennial screening mammography for women 50 to 74 years of age only decreases mortality by 23% (7 lives per 1,000 women screened) [32]. Earlier initiation of screening and more frequent screening result in a greater number of imaging For women 40 to 49 years of age, randomized controlled trials and observational studies demonstrate that screening mammography decreases breast cancer mortality by 15% to 50% [1,8,32,33,39]. Results from the CISNET suggest that annual screening mammography in women 40 to 49 years of age saves 42% more lives and life-years than biennial screening due to faster growing tumors in younger women [31]. Women screened between 40 and 49 years of age are also less likely to require mastectomy or chemotherapy than women diagnosed with palpable tumors [2]. Non-Hispanic black women, Hispanic black, and Hispanic white women have higher breast cancer mortality than non-Hispanic white women, and minority women often present at younger ages with more aggressive tumor subtypes [3,8]. Therefore, decreasing access to screening mammography, especially in women 40 to 49 years of age, may disproportionately impact minority women.

Female Breast Cancer Screening. Early detection of second breast cancers improves survival, so patients with a personal history of breast cancer should undergo annual mammography or DBT for surveillance following breast conservation therapy [3]. Despite the established mortality benefit, published guidelines differ in their recommendations for screening mammography due to variations in the perceptions of the relative risks and benefits [5,38]. The degree of breast cancer mortality reduction from screening mammography varies with different screening regimens. Mortality reduction is greater when screening begins 40 years of age rather than 45 or 50 years of age and when screening is done more frequently (annually rather than biennially) [8,31]. Annual screening mammography for women 40 to 84 years of age decreases mortality by 40% (12 lives per 1,000 women screened), whereas biennial screening mammography for women 50 to 74 years of age only decreases mortality by 23% (7 lives per 1,000 women screened) [32]. Earlier initiation of screening and more frequent screening result in a greater number of imaging For women 40 to 49 years of age, randomized controlled trials and observational studies demonstrate that screening mammography decreases breast cancer mortality by 15% to 50% [1,8,32,33,39]. Results from the CISNET suggest that annual screening mammography in women 40 to 49 years of age saves 42% more lives and life-years than biennial screening due to faster growing tumors in younger women [31]. Women screened between 40 and 49 years of age are also less likely to require mastectomy or chemotherapy than women diagnosed with palpable tumors [2]. Non-Hispanic black women, Hispanic black, and Hispanic white women have higher breast cancer mortality than non-Hispanic white women, and minority women often present at younger ages with more aggressive tumor subtypes [3,8]. Therefore, decreasing access to screening mammography, especially in women 40 to 49 years of age, may disproportionately impact minority women.

70910

acrac_70910_11

Female Breast Cancer Screening

Annual screening mammography results in a greater reduction in mortality compared to biennial screening [8]. In women 40 to 84 years of age, annual screening reduces mortality by 40%, compared to a 32% reduction for biennial screening [32]. With regular screening, interval breast cancers do occur with a higher frequency in women undergoing biennial or triennial screening compared to annual screening. The sensitivity of mammography is decreased in some groups of women, including those with dense breasts [40]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27]. Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. Given the limitations of mammography and to minimize interval cancers, supplemental screening modalities have been investigated in women at intermediate risk. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. Rather than supplementing screening mammography with additional imaging modalities, some have suggested limiting women offered screening mammography based upon individual patient risk assessed by various risk models, breast density, or genetic information such as single-nucleotide polymorphisms. However, the randomized controlled trials demonstrating mortality reduction and most large-scale observational studies enrolled women based upon age and geographic location, not individual patient risk factors.

Female Breast Cancer Screening. Annual screening mammography results in a greater reduction in mortality compared to biennial screening [8]. In women 40 to 84 years of age, annual screening reduces mortality by 40%, compared to a 32% reduction for biennial screening [32]. With regular screening, interval breast cancers do occur with a higher frequency in women undergoing biennial or triennial screening compared to annual screening. The sensitivity of mammography is decreased in some groups of women, including those with dense breasts [40]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27]. Compared to average breast density (near the threshold between heterogeneously dense and scattered areas of fibroglandular density), the relative risks for developing breast cancer are 1.2 for heterogeneously dense and 2.1 for extremely dense breasts [28]. Some health care providers may therefore consider women with extremely dense breasts to no longer be average risk. Given the limitations of mammography and to minimize interval cancers, supplemental screening modalities have been investigated in women at intermediate risk. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. Rather than supplementing screening mammography with additional imaging modalities, some have suggested limiting women offered screening mammography based upon individual patient risk assessed by various risk models, breast density, or genetic information such as single-nucleotide polymorphisms. However, the randomized controlled trials demonstrating mortality reduction and most large-scale observational studies enrolled women based upon age and geographic location, not individual patient risk factors.

70910

acrac_70910_12

Female Breast Cancer Screening

In one observational study in women <50 years of age, restricting screening to women with a first-degree family history, extremely dense breast tissue, or both, would cause 66% of potentially screen-detected cancers to be missed [41]. To maximize the benefits, the ACR recommends annual screening mammography beginning no later than 40 years of age for women at intermediate risk [3]. Women should continue screening mammography as long as they remain in overall good health and are willing to undergo the examination and subsequent testing or biopsy, if an abnormality is identified [5,8]. For those with a family history of breast cancer, mammography should begin earlier if familial breast cancer occurred at a young age, typically 10 years prior to the youngest age at presentation but generally not before 30 years of age [6]. For women who have lobular neoplasia or atypical hyperplasia diagnosed prior to 40 years of age, annual screening mammography should be performed from time of diagnosis but generally not prior to 30 years of age [38]. Early detection of second breast cancers improves survival, so patients with a personal history of breast cancer should undergo annual mammography or DBT for surveillance following breast conservation therapy [3]. Mammography With IV Contrast Data are limited regarding the use of mammography with IV contrast for breast cancer screening in intermediate- risk women. To date, published studies have predominantly included women with dense breasts and other risk factors resulting in intermediate- or high-risk profiles. Compared to mammography alone, mammography with IV contrast increases cancer detection (incremental CDR = 6.6-13 per 1,000) in women at elevated risk [57-60]. Female Breast Cancer Screening MRI Breast Without and With IV Contrast MRI has a higher CDR than mammography alone, DBT, or mammography/DBT combined with US [61-64].

Female Breast Cancer Screening. In one observational study in women <50 years of age, restricting screening to women with a first-degree family history, extremely dense breast tissue, or both, would cause 66% of potentially screen-detected cancers to be missed [41]. To maximize the benefits, the ACR recommends annual screening mammography beginning no later than 40 years of age for women at intermediate risk [3]. Women should continue screening mammography as long as they remain in overall good health and are willing to undergo the examination and subsequent testing or biopsy, if an abnormality is identified [5,8]. For those with a family history of breast cancer, mammography should begin earlier if familial breast cancer occurred at a young age, typically 10 years prior to the youngest age at presentation but generally not before 30 years of age [6]. For women who have lobular neoplasia or atypical hyperplasia diagnosed prior to 40 years of age, annual screening mammography should be performed from time of diagnosis but generally not prior to 30 years of age [38]. Early detection of second breast cancers improves survival, so patients with a personal history of breast cancer should undergo annual mammography or DBT for surveillance following breast conservation therapy [3]. Mammography With IV Contrast Data are limited regarding the use of mammography with IV contrast for breast cancer screening in intermediate- risk women. To date, published studies have predominantly included women with dense breasts and other risk factors resulting in intermediate- or high-risk profiles. Compared to mammography alone, mammography with IV contrast increases cancer detection (incremental CDR = 6.6-13 per 1,000) in women at elevated risk [57-60]. Female Breast Cancer Screening MRI Breast Without and With IV Contrast MRI has a higher CDR than mammography alone, DBT, or mammography/DBT combined with US [61-64].

70910

acrac_70910_13

Female Breast Cancer Screening

The incremental CDR of MRI in elevated-risk women ranges from 8 to 29 per 1,000 women, with lower CDR estimates in intermediate-risk women compared to high-risk BRCA mutation carriers [61-63,65,66]. In one study, breast MRI CDR was 15 per 1,000 in women with a prior biopsy demonstrating a high-risk lesion compared to 8 per 1,000 in women reporting a family history [65]. In women with a personal history of breast cancer, a meta-analysis estimated a CDR of 9 to 15 per 1,000 breast MRI [67]. Breast MRI detects small, node-negative invasive cancers at earlier tumor stages compared to mammography, as well as ductal carcinoma in situ [68,69]. Screening MRI also reduces interval cancers [69]. However, breast MRI has a higher recall rate than mammography (15.1% versus 6.4%) [70], higher frequency of BI-RADS category 3 assessment than mammography (14.8% versus 11.8%), and greater frequency of image-guided biopsies than mammography (11.8% versus 2.4%) [63]. MRI Breast Without IV Contrast There is no relevant literature to support the use of breast MRI without IV contrast for screening women at intermediate risk. MRI Breast Without IV Contrast Abbreviated There is no relevant literature to support the use of abbreviated breast MRI without IV contrast for screening women at intermediate risk. Sestamibi MBI Data are limited regarding the use of sestamibi MBI for screening women at intermediate risk. Most studies have focused upon women with dense breasts and variable risk profiles. Retrospective and prospective studies have demonstrated similar incremental CDR for sestamibi MBI of 6.5 to 9 over mammography, with a study demonstrating an incremental CDR of 16.5 per 1,000 in women at increased risk primarily due to family or personal history of breast cancer [40,47]. Sestamibi MBI demonstrates similar sensitivity, better specificity, and lower recall rate compared to supplemental screening US in women with dense breasts [47,48].

Female Breast Cancer Screening. The incremental CDR of MRI in elevated-risk women ranges from 8 to 29 per 1,000 women, with lower CDR estimates in intermediate-risk women compared to high-risk BRCA mutation carriers [61-63,65,66]. In one study, breast MRI CDR was 15 per 1,000 in women with a prior biopsy demonstrating a high-risk lesion compared to 8 per 1,000 in women reporting a family history [65]. In women with a personal history of breast cancer, a meta-analysis estimated a CDR of 9 to 15 per 1,000 breast MRI [67]. Breast MRI detects small, node-negative invasive cancers at earlier tumor stages compared to mammography, as well as ductal carcinoma in situ [68,69]. Screening MRI also reduces interval cancers [69]. However, breast MRI has a higher recall rate than mammography (15.1% versus 6.4%) [70], higher frequency of BI-RADS category 3 assessment than mammography (14.8% versus 11.8%), and greater frequency of image-guided biopsies than mammography (11.8% versus 2.4%) [63]. MRI Breast Without IV Contrast There is no relevant literature to support the use of breast MRI without IV contrast for screening women at intermediate risk. MRI Breast Without IV Contrast Abbreviated There is no relevant literature to support the use of abbreviated breast MRI without IV contrast for screening women at intermediate risk. Sestamibi MBI Data are limited regarding the use of sestamibi MBI for screening women at intermediate risk. Most studies have focused upon women with dense breasts and variable risk profiles. Retrospective and prospective studies have demonstrated similar incremental CDR for sestamibi MBI of 6.5 to 9 over mammography, with a study demonstrating an incremental CDR of 16.5 per 1,000 in women at increased risk primarily due to family or personal history of breast cancer [40,47]. Sestamibi MBI demonstrates similar sensitivity, better specificity, and lower recall rate compared to supplemental screening US in women with dense breasts [47,48].

70910

acrac_70910_14

Female Breast Cancer Screening

US Breast Most studies evaluating the utility of screening with breast US have focused on women with dense breast tissue with or without other risk factors. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27]. Screening breast US in women with mammographically dense breasts, including those with risk factors placing them at increased breast cancer risk, identifies predominantly mammographically occult, small, node-negative invasive tumors with an increased CDR of 1.8 to 4.6 cancers per 1,000 women screened [40,49]. Although supplemental screening US in women with dense breasts results in an increased CDR, US also increases recall rate, false-positive examinations, and false-positive biopsies [26,49-55]. In women undergoing annual mammography plus annual supplemental screening MRI, the addition of supplemental screening with US does not identify additional cancers and is therefore not routinely performed. The ACRIN 6666 trial enrolled women with dense breast tissue and at least one other breast cancer risk factor [61]. Compared to mammography alone, screening US detected 5.3 cancers per 1,000 in year 1, 3.7 cancers per 1,000 in years 2 and 3, and resulted in a larger number of false-positive examinations and false-positive biopsies each year [61]. In a prospective study limited to intermediate-risk women, sensitivity of mammography was 57%, US was 24.5%, and mammography combined with biannual US demonstrated 80.4% sensitivity [76]. In women with a Female Breast Cancer Screening personal history of breast cancer, supplemental US screening results in an incremental CDR of 2.4 to 2.9 cancers per 1,000 examinations over mammography alone; however, US screening has lower specificity [10,66].

Female Breast Cancer Screening. US Breast Most studies evaluating the utility of screening with breast US have focused on women with dense breast tissue with or without other risk factors. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27]. Screening breast US in women with mammographically dense breasts, including those with risk factors placing them at increased breast cancer risk, identifies predominantly mammographically occult, small, node-negative invasive tumors with an increased CDR of 1.8 to 4.6 cancers per 1,000 women screened [40,49]. Although supplemental screening US in women with dense breasts results in an increased CDR, US also increases recall rate, false-positive examinations, and false-positive biopsies [26,49-55]. In women undergoing annual mammography plus annual supplemental screening MRI, the addition of supplemental screening with US does not identify additional cancers and is therefore not routinely performed. The ACRIN 6666 trial enrolled women with dense breast tissue and at least one other breast cancer risk factor [61]. Compared to mammography alone, screening US detected 5.3 cancers per 1,000 in year 1, 3.7 cancers per 1,000 in years 2 and 3, and resulted in a larger number of false-positive examinations and false-positive biopsies each year [61]. In a prospective study limited to intermediate-risk women, sensitivity of mammography was 57%, US was 24.5%, and mammography combined with biannual US demonstrated 80.4% sensitivity [76]. In women with a Female Breast Cancer Screening personal history of breast cancer, supplemental US screening results in an incremental CDR of 2.4 to 2.9 cancers per 1,000 examinations over mammography alone; however, US screening has lower specificity [10,66].

70910

acrac_70910_15

Female Breast Cancer Screening

Digital Breast Tomosynthesis Screening DBT displays reconstructed stacked images of the breast in combination with digital mammographic views, which may be synthetic mammograms reconstructed from the acquired tomosynthesis data set or FFDM. Compared to FFDM or synthetic mammograms alone, most studies demonstrate that DBT increases CDR and decreases recall rate [14-22]; although, some studies have not reached statistical significance [23] or have found less compelling results in subsets of women, such as those with extremely dense breasts [24,25]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27]. Irrespective of risk category, meta-analyses have demonstrated an incremental increase in CDR of 1.6 to 3.2 per 1,000 screening DBT examinations and a 2.2% pooled decrease in recall rate compared to digital mammography [18,29,30]. Within the limited studies of women at elevated risk due to personal and/or family history of breast cancer, DBT decreased recall rate without a significant increase in CDR compared to FFDM; however, small sample sizes restrict analyses [3,40]. High-risk women should begin annual screening mammography at 30 years of age or 10 years prior to the youngest family member who had breast cancer, but generally not before 30 years of age [3]. Approximately one-third of breast cancers may only be detected on mammography in BRCA2 mutation carriers who are <40 years of age [79]. In some mutation carriers, some referring providers use mammography or DBT beginning at 40 years of age if patients undergo annual MRI [80]. Women who underwent thoracic or upper abdominal radiation therapy at an early age (<30 years) should begin screening mammography 8 years after radiation therapy but not before 25 years of age [3]. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35].

Female Breast Cancer Screening. Digital Breast Tomosynthesis Screening DBT displays reconstructed stacked images of the breast in combination with digital mammographic views, which may be synthetic mammograms reconstructed from the acquired tomosynthesis data set or FFDM. Compared to FFDM or synthetic mammograms alone, most studies demonstrate that DBT increases CDR and decreases recall rate [14-22]; although, some studies have not reached statistical significance [23] or have found less compelling results in subsets of women, such as those with extremely dense breasts [24,25]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27]. Irrespective of risk category, meta-analyses have demonstrated an incremental increase in CDR of 1.6 to 3.2 per 1,000 screening DBT examinations and a 2.2% pooled decrease in recall rate compared to digital mammography [18,29,30]. Within the limited studies of women at elevated risk due to personal and/or family history of breast cancer, DBT decreased recall rate without a significant increase in CDR compared to FFDM; however, small sample sizes restrict analyses [3,40]. High-risk women should begin annual screening mammography at 30 years of age or 10 years prior to the youngest family member who had breast cancer, but generally not before 30 years of age [3]. Approximately one-third of breast cancers may only be detected on mammography in BRCA2 mutation carriers who are <40 years of age [79]. In some mutation carriers, some referring providers use mammography or DBT beginning at 40 years of age if patients undergo annual MRI [80]. Women who underwent thoracic or upper abdominal radiation therapy at an early age (<30 years) should begin screening mammography 8 years after radiation therapy but not before 25 years of age [3]. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35].

70910

acrac_70910_16

Female Breast Cancer Screening

Mammography Screening To date, mammography is the only screening modality shown to decrease breast cancer mortality. Multiple randomized controlled trials demonstrate that invitation to screening mammography results in at least a 22% Female Breast Cancer Screening For women 40 to 49 years of age, randomized controlled trials and observational studies demonstrate that screening mammography decreases breast cancer mortality by 15% to 50% [1,8,32,33,39]. Results from the CISNET suggest that annual screening mammography in women 40 to 49 years of age saves 42% more lives and life-years than biennial screening due to faster growing tumors in younger women [31]. Women screened between 40 and 49 years of age are also less likely to require mastectomy or chemotherapy than women diagnosed with palpable tumors [2]. Non-Hispanic Black, Hispanic Black, and Hispanic White women have higher breast cancer mortality than non- Hispanic White women, and minority women often present at younger ages with more aggressive tumor subtypes [3,8]. Therefore, decreasing access to screening mammography, especially in women 40 to 49 years of age, may disproportionately impact minority women. Annual screening mammography results in a greater reduction in mortality compared to biennial screening [8]. In women 40 to 84 years of age, annual screening reduces mortality by 40%, compared to a 32% reduction for biennial screening [32]. With regular screening, interval breast cancers do occur with a higher frequency in women undergoing biennial or triennial screening compared to annual screening. The sensitivity of mammography is decreased in some groups of women, including those with dense breasts [40]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27]. Given the limitations of mammography and to minimize interval cancers, supplemental screening modalities have been investigated in women at high risk.

Female Breast Cancer Screening. Mammography Screening To date, mammography is the only screening modality shown to decrease breast cancer mortality. Multiple randomized controlled trials demonstrate that invitation to screening mammography results in at least a 22% Female Breast Cancer Screening For women 40 to 49 years of age, randomized controlled trials and observational studies demonstrate that screening mammography decreases breast cancer mortality by 15% to 50% [1,8,32,33,39]. Results from the CISNET suggest that annual screening mammography in women 40 to 49 years of age saves 42% more lives and life-years than biennial screening due to faster growing tumors in younger women [31]. Women screened between 40 and 49 years of age are also less likely to require mastectomy or chemotherapy than women diagnosed with palpable tumors [2]. Non-Hispanic Black, Hispanic Black, and Hispanic White women have higher breast cancer mortality than non- Hispanic White women, and minority women often present at younger ages with more aggressive tumor subtypes [3,8]. Therefore, decreasing access to screening mammography, especially in women 40 to 49 years of age, may disproportionately impact minority women. Annual screening mammography results in a greater reduction in mortality compared to biennial screening [8]. In women 40 to 84 years of age, annual screening reduces mortality by 40%, compared to a 32% reduction for biennial screening [32]. With regular screening, interval breast cancers do occur with a higher frequency in women undergoing biennial or triennial screening compared to annual screening. The sensitivity of mammography is decreased in some groups of women, including those with dense breasts [40]. Dense breast tissue decreases the sensitivity of mammography [26] and is an independent risk factor for developing breast cancer [27]. Given the limitations of mammography and to minimize interval cancers, supplemental screening modalities have been investigated in women at high risk.

70910

acrac_70910_17

Female Breast Cancer Screening

Because screening mammography decreases breast cancer mortality, screening Female Breast Cancer Screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. Rather than supplementing screening mammography with additional imaging modalities, some have suggested limiting women offered screening mammography based upon individual patient risk assessed by various risk models, breast density, or genetic information such as single-nucleotide polymorphism. However, the randomized controlled trials demonstrating mortality reduction and most large-scale observational studies enrolled women based upon age and geographic location, not individual patient risk factors. In one observational study in women <50 years of age, restricting screening to women with a first-degree family history, extremely dense breast tissue, or both, would cause 66% of potentially screen-detected cancers to be missed [41]. Numerous studies in high-risk women have evaluated the performance of mammography and supplemental screening modalities, such as US and MRI. Mammography consistently demonstrates lower sensitivity (25%-69%) than US or MRI, and high-risk women experience higher interval cancer rates than the general population [3,40]. The combination of mammography with MRI yields the highest sensitivity across high-risk groups of women (91%- 98%) [3,40,81]. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. High-risk women should begin annual screening mammography at 30 years of age or 10 years prior to the youngest family member who had breast cancer, but generally not before 30 years of age [3]. Approximately one-third of breast cancers may only be detected on mammography in BRCA2 mutation carriers who are <40 years of age [79].

Female Breast Cancer Screening. Because screening mammography decreases breast cancer mortality, screening Female Breast Cancer Screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. Rather than supplementing screening mammography with additional imaging modalities, some have suggested limiting women offered screening mammography based upon individual patient risk assessed by various risk models, breast density, or genetic information such as single-nucleotide polymorphism. However, the randomized controlled trials demonstrating mortality reduction and most large-scale observational studies enrolled women based upon age and geographic location, not individual patient risk factors. In one observational study in women <50 years of age, restricting screening to women with a first-degree family history, extremely dense breast tissue, or both, would cause 66% of potentially screen-detected cancers to be missed [41]. Numerous studies in high-risk women have evaluated the performance of mammography and supplemental screening modalities, such as US and MRI. Mammography consistently demonstrates lower sensitivity (25%-69%) than US or MRI, and high-risk women experience higher interval cancer rates than the general population [3,40]. The combination of mammography with MRI yields the highest sensitivity across high-risk groups of women (91%- 98%) [3,40,81]. Because screening mammography decreases breast cancer mortality, screening mammography or screening DBT is still performed in women undergoing supplemental screening studies [3,8,35]. High-risk women should begin annual screening mammography at 30 years of age or 10 years prior to the youngest family member who had breast cancer, but generally not before 30 years of age [3]. Approximately one-third of breast cancers may only be detected on mammography in BRCA2 mutation carriers who are <40 years of age [79].

70910

acrac_70910_18

Female Breast Cancer Screening

In some mutation carriers, some referring providers use mammography or DBT beginning at 40 years of age if patients undergo annual MRI [80]. Women who underwent thoracic or upper abdominal radiation therapy at an early age (<30 years) should begin screening mammography 8 years after radiation therapy but not before 25 years of age [3]. Mammography With IV Contrast Data are limited regarding the use of mammography with IV contrast for breast cancer screening in high-risk women. To date, published studies have predominantly included women with dense breasts and other risk factors resulting in intermediate or high-risk profiles. Compared to mammography alone, mammography with IV contrast increases sensitivity and cancer detection (incremental CDR = 6.6-13 per 1,000) in women at elevated risk [57-60]. Mammography with IV contrast may be useful in high-risk women as an alternative to MRI. MRI Breast Without and With IV Contrast MRI has a higher CDR than mammography alone, DBT, or mammography/DBT combined with US [61-64]. In high-risk women, supplemental screening MRI combined with mammography yields a 91% to 98% sensitivity, although the reported specificity of MRI is typically lower than mammography [40,81]. The incremental CDR of MRI in elevated-risk women ranges from 8 to 29 per 1,000 women, with higher CDR (26 per 1,000) in BRCA mutation carriers [61-63,65,66]. Breast MRI detects small, node-negative invasive cancers at earlier tumor stages compared to mammography, as well as ductal carcinoma in situ [68,69]. Screening MRI also reduces interval cancers [69]. However, breast MRI has a higher recall rate than mammography (15.1% versus 6.4%) [70], higher frequency of BI-RADS category 3 assessment than mammography (14.8% versus 11.8%), and a greater frequency of image-guided biopsies than mammography (11.8 versus 2.4%) [63]. Since 2007, the American Cancer Society has recommended annual breast MRI for breast cancer screening in high- risk women [4].

Female Breast Cancer Screening. In some mutation carriers, some referring providers use mammography or DBT beginning at 40 years of age if patients undergo annual MRI [80]. Women who underwent thoracic or upper abdominal radiation therapy at an early age (<30 years) should begin screening mammography 8 years after radiation therapy but not before 25 years of age [3]. Mammography With IV Contrast Data are limited regarding the use of mammography with IV contrast for breast cancer screening in high-risk women. To date, published studies have predominantly included women with dense breasts and other risk factors resulting in intermediate or high-risk profiles. Compared to mammography alone, mammography with IV contrast increases sensitivity and cancer detection (incremental CDR = 6.6-13 per 1,000) in women at elevated risk [57-60]. Mammography with IV contrast may be useful in high-risk women as an alternative to MRI. MRI Breast Without and With IV Contrast MRI has a higher CDR than mammography alone, DBT, or mammography/DBT combined with US [61-64]. In high-risk women, supplemental screening MRI combined with mammography yields a 91% to 98% sensitivity, although the reported specificity of MRI is typically lower than mammography [40,81]. The incremental CDR of MRI in elevated-risk women ranges from 8 to 29 per 1,000 women, with higher CDR (26 per 1,000) in BRCA mutation carriers [61-63,65,66]. Breast MRI detects small, node-negative invasive cancers at earlier tumor stages compared to mammography, as well as ductal carcinoma in situ [68,69]. Screening MRI also reduces interval cancers [69]. However, breast MRI has a higher recall rate than mammography (15.1% versus 6.4%) [70], higher frequency of BI-RADS category 3 assessment than mammography (14.8% versus 11.8%), and a greater frequency of image-guided biopsies than mammography (11.8 versus 2.4%) [63]. Since 2007, the American Cancer Society has recommended annual breast MRI for breast cancer screening in high- risk women [4].

70910

acrac_70910_19

Female Breast Cancer Screening

The ACR recommends annual breast MRI in high-risk women beginning as early as 25 years of age [3]. Female Breast Cancer Screening MRI Breast Without and With IV Contrast Abbreviated Data are limited regarding the use of abbreviated breast MRI without and with IV contrast for screening in high- risk women. Following the publication of the American Cancer Society guidelines for supplemental screening breast MRI in 2007, high-risk women have traditionally undergone conventional full protocol breast MRI without and with IV contrast [3,4]. However, multiple studies have demonstrated similar diagnostic accuracy for abbreviated protocol MRI compared to conventional full protocol breast MRI [73-75]. In a study evaluating 3,037 abbreviated breast MRI in 1,975 high-risk women, the CDR was 29 per 1,000, the interval cancer rate was 0.66 per 1,000, and all cancers missed by abbreviated breast MRI were node negative early-stage invasive malignancies [72]. MRI Breast Without IV Contrast There is no relevant literature to support the use of MRI without IV contrast for screening women at high risk. MRI Breast Without IV Contrast Abbreviated There is no relevant literature to support the use of abbreviated breast MRI without IV contrast for screening women at high risk. Sestamibi MBI Data are limited regarding the use of sestamibi MBI for screening women at high risk. Most studies have focused upon women with dense breasts and variable risk profiles. Retrospective and prospective studies have demonstrated similar incremental CDR for sestamibi MBI of 6.5 to 9 over mammography, with one study demonstrating an incremental CDR of 16.5 per 1,000 in women at increased risk primarily due to family or personal history of breast cancer [40,47]. Sestamibi MBI demonstrates similar sensitivity, better specificity, and lower recall rate compared to supplemental screening US in women with dense breasts [47,48].

Female Breast Cancer Screening. The ACR recommends annual breast MRI in high-risk women beginning as early as 25 years of age [3]. Female Breast Cancer Screening MRI Breast Without and With IV Contrast Abbreviated Data are limited regarding the use of abbreviated breast MRI without and with IV contrast for screening in high- risk women. Following the publication of the American Cancer Society guidelines for supplemental screening breast MRI in 2007, high-risk women have traditionally undergone conventional full protocol breast MRI without and with IV contrast [3,4]. However, multiple studies have demonstrated similar diagnostic accuracy for abbreviated protocol MRI compared to conventional full protocol breast MRI [73-75]. In a study evaluating 3,037 abbreviated breast MRI in 1,975 high-risk women, the CDR was 29 per 1,000, the interval cancer rate was 0.66 per 1,000, and all cancers missed by abbreviated breast MRI were node negative early-stage invasive malignancies [72]. MRI Breast Without IV Contrast There is no relevant literature to support the use of MRI without IV contrast for screening women at high risk. MRI Breast Without IV Contrast Abbreviated There is no relevant literature to support the use of abbreviated breast MRI without IV contrast for screening women at high risk. Sestamibi MBI Data are limited regarding the use of sestamibi MBI for screening women at high risk. Most studies have focused upon women with dense breasts and variable risk profiles. Retrospective and prospective studies have demonstrated similar incremental CDR for sestamibi MBI of 6.5 to 9 over mammography, with one study demonstrating an incremental CDR of 16.5 per 1,000 in women at increased risk primarily due to family or personal history of breast cancer [40,47]. Sestamibi MBI demonstrates similar sensitivity, better specificity, and lower recall rate compared to supplemental screening US in women with dense breasts [47,48].

70910

acrac_70910_20

Female Breast Cancer Screening

US Breast In high-risk women undergoing annual mammography plus annual supplemental screening MRI, the addition of supplemental screening with US does not identify additional cancers and is therefore not routinely performed. Screening US may be useful in high-risk patients as an alternative to MRI. However, high-risk women who do not undergo supplemental screening MRI should be counseled that the CDR of US is inferior to MRI. MRI has a higher CDR than mammography, DBT, or mammography/DBT combined with US [61-64]. The ACRIN 6666 trial enrolled women with elevated breast cancer risk [61]. Compared to mammography alone, screening US detected 5.3 cancers per 1,000 in year 1 and 3.7 cancers per 1,000 in years 2 and 3 and resulted in a larger number of false- positive examinations and false-positive biopsies each year [61]. After 3 consecutive rounds of mammography plus US, the incremental CDR of MRI was 14.7 per 1,000, although false-positive examinations also increased [61]. In a prospective multicenter study of 687 high-risk women who underwent clinical breast examination, mammography, US, and MRI for screening, the combination of MRI plus mammography maximized the breast cancers detected [62]. Mammography identified 5 cancers per 1,000 compared to 6 per 1,000 for US, 7.7 per 1,000 for mammography plus US, 14.9 per 1,000 for MRI, 14.9 per 1,000 for MRI plus US, 16 per 1,000 for mammography plus MRI, and 16 per 1,000 for mammography plus US plus MRI [62]. In a prospective study of BRCA mutation carriers and high-risk women, sensitivity of mammography was 25% and 66% whereas US was 23% and 34%, respectively [76]. In the high-risk group, mammography combined with biannual US demonstrated 100% sensitivity [76]; however, MRI was not performed. In a subset analysis of BRCA mutation carriers, MRI sensitivity was 94% [76]. In another study of 529 high-risk women suspected or proven to carry a deleterious BRCA mutation, the performance of US was also inferior to MRI [82].

Female Breast Cancer Screening. US Breast In high-risk women undergoing annual mammography plus annual supplemental screening MRI, the addition of supplemental screening with US does not identify additional cancers and is therefore not routinely performed. Screening US may be useful in high-risk patients as an alternative to MRI. However, high-risk women who do not undergo supplemental screening MRI should be counseled that the CDR of US is inferior to MRI. MRI has a higher CDR than mammography, DBT, or mammography/DBT combined with US [61-64]. The ACRIN 6666 trial enrolled women with elevated breast cancer risk [61]. Compared to mammography alone, screening US detected 5.3 cancers per 1,000 in year 1 and 3.7 cancers per 1,000 in years 2 and 3 and resulted in a larger number of false- positive examinations and false-positive biopsies each year [61]. After 3 consecutive rounds of mammography plus US, the incremental CDR of MRI was 14.7 per 1,000, although false-positive examinations also increased [61]. In a prospective multicenter study of 687 high-risk women who underwent clinical breast examination, mammography, US, and MRI for screening, the combination of MRI plus mammography maximized the breast cancers detected [62]. Mammography identified 5 cancers per 1,000 compared to 6 per 1,000 for US, 7.7 per 1,000 for mammography plus US, 14.9 per 1,000 for MRI, 14.9 per 1,000 for MRI plus US, 16 per 1,000 for mammography plus MRI, and 16 per 1,000 for mammography plus US plus MRI [62]. In a prospective study of BRCA mutation carriers and high-risk women, sensitivity of mammography was 25% and 66% whereas US was 23% and 34%, respectively [76]. In the high-risk group, mammography combined with biannual US demonstrated 100% sensitivity [76]; however, MRI was not performed. In a subset analysis of BRCA mutation carriers, MRI sensitivity was 94% [76]. In another study of 529 high-risk women suspected or proven to carry a deleterious BRCA mutation, the performance of US was also inferior to MRI [82].

70910

acrac_3102401_0

Staging and Follow up of Primary Vaginal Cancer

Introduction/Background Primary vaginal cancer is rare, comprising 1% to 2% of gynecologic malignancies and 20% of all malignancies involving the vagina [1,2]. More frequently, the vagina is involved secondarily either by direct invasion from malignancies originating in adjacent organs, most commonly the cervix or vulva, or by metastases from other pelvic or extrapelvic primary malignancies [1,2]. Additionally, any vaginal tumor involving the cervix or vulva, whether or not the lesion is centered in the vagina, is classified by the International Federation of Gynecology and Obstetrics (FIGO) system as a primary cervical or vulvar cancer, respectively. Squamous cell carcinoma is the most common underlying histology in primary vaginal cancer, representing 80% to 90% of primary vaginal cancer [3] and occurs most frequently in postmenopausal women, with adenocarcinoma representing around 5% to 10% of cases and even rarer histologies such as sarcoma, melanoma, and lymphoma accounting for the remainder [1,2]. Primary vaginal cancer is staged according to two systems, FIGO and the American Joint Committee on Cancer (AJCC). FIGO stipulates a clinical staging paradigm, whereby features derived from bimanual and/or rectovaginal examination, cystoscopy, proctoscopy, and radiography are permissible for incorporation into staging [4]. Although FIGO encourages the use of advanced imaging modalities such as CT, MRI, and PET to guide management, information derived from these examinations does not alter the formal clinical FIGO stage [4]. Given the rarity of primary vaginal cancer, treatment principles are derived from retrospective data in addition to extrapolation from more established management paradigms for cervical and anal squamous cell cancers. Surgical management for vaginal cancer is limited primarily to small (<2 cm) early stage lesions, with larger lesions posing greater difficultly for achieving negative surgical margins.

Staging and Follow up of Primary Vaginal Cancer. Introduction/Background Primary vaginal cancer is rare, comprising 1% to 2% of gynecologic malignancies and 20% of all malignancies involving the vagina [1,2]. More frequently, the vagina is involved secondarily either by direct invasion from malignancies originating in adjacent organs, most commonly the cervix or vulva, or by metastases from other pelvic or extrapelvic primary malignancies [1,2]. Additionally, any vaginal tumor involving the cervix or vulva, whether or not the lesion is centered in the vagina, is classified by the International Federation of Gynecology and Obstetrics (FIGO) system as a primary cervical or vulvar cancer, respectively. Squamous cell carcinoma is the most common underlying histology in primary vaginal cancer, representing 80% to 90% of primary vaginal cancer [3] and occurs most frequently in postmenopausal women, with adenocarcinoma representing around 5% to 10% of cases and even rarer histologies such as sarcoma, melanoma, and lymphoma accounting for the remainder [1,2]. Primary vaginal cancer is staged according to two systems, FIGO and the American Joint Committee on Cancer (AJCC). FIGO stipulates a clinical staging paradigm, whereby features derived from bimanual and/or rectovaginal examination, cystoscopy, proctoscopy, and radiography are permissible for incorporation into staging [4]. Although FIGO encourages the use of advanced imaging modalities such as CT, MRI, and PET to guide management, information derived from these examinations does not alter the formal clinical FIGO stage [4]. Given the rarity of primary vaginal cancer, treatment principles are derived from retrospective data in addition to extrapolation from more established management paradigms for cervical and anal squamous cell cancers. Surgical management for vaginal cancer is limited primarily to small (<2 cm) early stage lesions, with larger lesions posing greater difficultly for achieving negative surgical margins.

3102401

acrac_3102401_1

Staging and Follow up of Primary Vaginal Cancer

Although surgical options exist for locally advanced disease, they often involve a degree of pelvic exenteration and therefore confer substantial morbidity. For this reason, the management paradigm for locally advanced disease has largely trended toward definitive radiation therapy with concurrent chemotherapy [1,5]. Though data on the use of imaging in vaginal cancer are sparse, insights derived from the study of imaging in cervical cancer have reasonable generalizability to vaginal cancer because of similar tumor biology. Moreover, given the trend toward definitive chemoradiation for both cancers in all but early stage lesions, principles of postchemoradiation the recommendations outlined in this document are informed by principles translated from the literature on cervical cancer. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] Staging and Follow-up of Primary Vaginal Cancer imaging in initial and adaptive radiation planning for vaginal cancer is not specifically addressed in this document, and analogous principles for cervical cancer are covered in extensive detail elsewhere [6]. OR Discussion of Procedures by Variant Variant 1: Vaginal cancer. Pretreatment staging. Initial imaging. Although the 2009 FIGO staging system for vaginal cancer indicates that findings on advanced imaging (CT, MRI, PET/CT) should not modify stage designation [4], such imaging findings are routinely employed in clinical practice to prognosticate and guide management decisions in patients with vaginal cancer.

Staging and Follow up of Primary Vaginal Cancer. Although surgical options exist for locally advanced disease, they often involve a degree of pelvic exenteration and therefore confer substantial morbidity. For this reason, the management paradigm for locally advanced disease has largely trended toward definitive radiation therapy with concurrent chemotherapy [1,5]. Though data on the use of imaging in vaginal cancer are sparse, insights derived from the study of imaging in cervical cancer have reasonable generalizability to vaginal cancer because of similar tumor biology. Moreover, given the trend toward definitive chemoradiation for both cancers in all but early stage lesions, principles of postchemoradiation the recommendations outlined in this document are informed by principles translated from the literature on cervical cancer. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] Staging and Follow-up of Primary Vaginal Cancer imaging in initial and adaptive radiation planning for vaginal cancer is not specifically addressed in this document, and analogous principles for cervical cancer are covered in extensive detail elsewhere [6]. OR Discussion of Procedures by Variant Variant 1: Vaginal cancer. Pretreatment staging. Initial imaging. Although the 2009 FIGO staging system for vaginal cancer indicates that findings on advanced imaging (CT, MRI, PET/CT) should not modify stage designation [4], such imaging findings are routinely employed in clinical practice to prognosticate and guide management decisions in patients with vaginal cancer.

3102401

acrac_3102401_2

Staging and Follow up of Primary Vaginal Cancer

Recent updates to the FIGO staging system for cervical cancer, which incorporate advanced imaging results into staging [10], reflect the wide recognition that cross-sectional imaging provides actionable staging information not readily obtained by physical examination or conventional radiography. Moreover, the increasing use of definitive radiotherapy across all stages of vaginal cancer obligates the incorporation of advanced imaging into pretreatment evaluation, because it is essential for treatment planning. The rationale for optimizing staging accuracy in vaginal cancer, in part via the inclusion of cross-sectional imaging, is multifold. First, accurate initial staging is fundamental to prognostication [11], facilitating incorporation of expectations of treatment efficacy into goals of care. Second, proper initial staging permits selection of the most appropriate treatment based on extent of disease. Regarding local extent, for vaginal lesions deemed likely confined to the vaginal wall (stage I) based on clinical examination, exclusion of extravaginal invasion with further testing is essential for ensuring that planned definitive surgery is likely to achieve a disease-free margin or that a radiation field properly incorporates the tumor volume. Regional nodal metastases include pelvic nodal metastases, which are primarily detected with cross-sectional imaging, and inguinal nodes (in lower vaginal cancers), a subset of which can be identified on clinical examination. Pretreatment knowledge of suspicious nodes may impact the decision to pursue surgery versus radiation. In addition, the distribution of suspicious nodes has the potential to influence radiation-specific factors such as field and dose planning, including possible node-directed boost doses as employed in cervical cancer [12].

Staging and Follow up of Primary Vaginal Cancer. Recent updates to the FIGO staging system for cervical cancer, which incorporate advanced imaging results into staging [10], reflect the wide recognition that cross-sectional imaging provides actionable staging information not readily obtained by physical examination or conventional radiography. Moreover, the increasing use of definitive radiotherapy across all stages of vaginal cancer obligates the incorporation of advanced imaging into pretreatment evaluation, because it is essential for treatment planning. The rationale for optimizing staging accuracy in vaginal cancer, in part via the inclusion of cross-sectional imaging, is multifold. First, accurate initial staging is fundamental to prognostication [11], facilitating incorporation of expectations of treatment efficacy into goals of care. Second, proper initial staging permits selection of the most appropriate treatment based on extent of disease. Regarding local extent, for vaginal lesions deemed likely confined to the vaginal wall (stage I) based on clinical examination, exclusion of extravaginal invasion with further testing is essential for ensuring that planned definitive surgery is likely to achieve a disease-free margin or that a radiation field properly incorporates the tumor volume. Regional nodal metastases include pelvic nodal metastases, which are primarily detected with cross-sectional imaging, and inguinal nodes (in lower vaginal cancers), a subset of which can be identified on clinical examination. Pretreatment knowledge of suspicious nodes may impact the decision to pursue surgery versus radiation. In addition, the distribution of suspicious nodes has the potential to influence radiation-specific factors such as field and dose planning, including possible node-directed boost doses as employed in cervical cancer [12].

3102401

acrac_3102401_3

Staging and Follow up of Primary Vaginal Cancer

Regarding distant metastases, detection of extraregional nodal or solid organ lesions can obviate unnecessarily morbid radical pelvic surgery and instead direct care toward palliative regimens or radiotherapy with an extended field. Finally, the ability to accurately stage noninvasively can avoid the need for invasive staging procedures such as cystoscopy (for bladder mucosal invasion) and proctoscopy (for rectal mucosal invasion), both of which are historical components of the FIGO clinical staging system [1]. CT Chest Although thoracic metastases are known to occur in vaginal cancer, no studies specifically address their incidence or the incremental value of chest CT for initial staging. Pulmonary metastases have been studied to a limited degree in cervical cancer, occurring in approximately 5% to 10% of patients at diagnosis [13,14]. Pulmonary metastases appear to occur slightly more frequently as a site of recurrent disease, with one large study of recurrent cervical cancer indicating an overall incidence of 13%, and the lungs representing the only site of recurrence in 6% of cases [15]. In studies evaluating pulmonary metastases from cervical cancer, chest CT was the most frequent diagnostic modality employed, with the vast majority of patients asymptomatic at the time of imaging [16,17]. These findings support the use of chest CT with or without intravenous (IV) contrast in the early posttreatment evaluation of cervical cancer, as endorsed by the National Comprehensive Cancer Network (NCCN) guidelines, and suggest that a similar strategy would be useful for vaginal cancer. Staging and Follow-up of Primary Vaginal Cancer CT Abdomen and Pelvis Data on the diagnostic performance of CT in primary vaginal cancer staging are very limited.

Staging and Follow up of Primary Vaginal Cancer. Regarding distant metastases, detection of extraregional nodal or solid organ lesions can obviate unnecessarily morbid radical pelvic surgery and instead direct care toward palliative regimens or radiotherapy with an extended field. Finally, the ability to accurately stage noninvasively can avoid the need for invasive staging procedures such as cystoscopy (for bladder mucosal invasion) and proctoscopy (for rectal mucosal invasion), both of which are historical components of the FIGO clinical staging system [1]. CT Chest Although thoracic metastases are known to occur in vaginal cancer, no studies specifically address their incidence or the incremental value of chest CT for initial staging. Pulmonary metastases have been studied to a limited degree in cervical cancer, occurring in approximately 5% to 10% of patients at diagnosis [13,14]. Pulmonary metastases appear to occur slightly more frequently as a site of recurrent disease, with one large study of recurrent cervical cancer indicating an overall incidence of 13%, and the lungs representing the only site of recurrence in 6% of cases [15]. In studies evaluating pulmonary metastases from cervical cancer, chest CT was the most frequent diagnostic modality employed, with the vast majority of patients asymptomatic at the time of imaging [16,17]. These findings support the use of chest CT with or without intravenous (IV) contrast in the early posttreatment evaluation of cervical cancer, as endorsed by the National Comprehensive Cancer Network (NCCN) guidelines, and suggest that a similar strategy would be useful for vaginal cancer. Staging and Follow-up of Primary Vaginal Cancer CT Abdomen and Pelvis Data on the diagnostic performance of CT in primary vaginal cancer staging are very limited.

3102401

acrac_3102401_4

Staging and Follow up of Primary Vaginal Cancer

A small retrospective study evaluating fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG)-PET/CT in 23 patients with primary vaginal cancer found that CT and FDG-PET detected pelvic nodal metastases in 17% (4 of 23) and 35% (8 of 23) of patients, respectively, suggesting inferior sensitivity of CT alone [18]. CT has been studied more extensively in cervical cancer staging, with available data comparing CT to MRI for local staging and CT (with or without IV contrast) to PET for regional and distant staging. For local staging, the ACRIN 6651 study showed that CT and MRI had sensitivity of 42% and 53%, respectively, and specificity of 82% and 75%, respectively, for classifying disease as stage IIB (parametrial invasion) or higher, with none of these differences reaching statistical significance [19]. However, a more recent meta-analysis suggested improved performance of MRI for parametrial invasion with modern hardware (sensitivity 76%, specificity 94%), particularly when the field strength was 3T and diffusion-weighted imaging was included [20], whereas a recent study of multidetector CT showed only 50% sensitivity for parametrial invasion [21]. Although older literature suggested lower sensitivity of CT compared with FDG-PET/CT for nodal metastases [22], the more recent ACRIN 6671/Gynecology Oncology Group (GOG) 0233 trial demonstrated a more modest difference in sensitivity for abdominal nodes (42% versus 50%, respectively) [23]. Likewise, in a recent meta- analysis, CT had only modestly lower area under the curve (AUC) (0.83) compared with PET/CT (0.90) for detection of nodal metastases from cervical cancer [24]. For distant metastases from cervical cancer, CT is inferior in the detection of osseous metastases (sensitivity 66%) compared with FDG-PET/CT (sensitivity 96%) [25].

Staging and Follow up of Primary Vaginal Cancer. A small retrospective study evaluating fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG)-PET/CT in 23 patients with primary vaginal cancer found that CT and FDG-PET detected pelvic nodal metastases in 17% (4 of 23) and 35% (8 of 23) of patients, respectively, suggesting inferior sensitivity of CT alone [18]. CT has been studied more extensively in cervical cancer staging, with available data comparing CT to MRI for local staging and CT (with or without IV contrast) to PET for regional and distant staging. For local staging, the ACRIN 6651 study showed that CT and MRI had sensitivity of 42% and 53%, respectively, and specificity of 82% and 75%, respectively, for classifying disease as stage IIB (parametrial invasion) or higher, with none of these differences reaching statistical significance [19]. However, a more recent meta-analysis suggested improved performance of MRI for parametrial invasion with modern hardware (sensitivity 76%, specificity 94%), particularly when the field strength was 3T and diffusion-weighted imaging was included [20], whereas a recent study of multidetector CT showed only 50% sensitivity for parametrial invasion [21]. Although older literature suggested lower sensitivity of CT compared with FDG-PET/CT for nodal metastases [22], the more recent ACRIN 6671/Gynecology Oncology Group (GOG) 0233 trial demonstrated a more modest difference in sensitivity for abdominal nodes (42% versus 50%, respectively) [23]. Likewise, in a recent meta- analysis, CT had only modestly lower area under the curve (AUC) (0.83) compared with PET/CT (0.90) for detection of nodal metastases from cervical cancer [24]. For distant metastases from cervical cancer, CT is inferior in the detection of osseous metastases (sensitivity 66%) compared with FDG-PET/CT (sensitivity 96%) [25].

3102401

acrac_3102401_5

Staging and Follow up of Primary Vaginal Cancer

These findings, if applied to vaginal cancer, suggest that modern multidetector CT abdomen and pelvis is a reasonable staging tool for regional and distant metastases, although is likely inferior to MRI for local staging, modestly inferior to FDG-PET/CT for nodal metastases, and inferior to FDG-PET/CT for osseous metastases. The use of IV contrast is strongly encouraged when possible, because the improved tissue contrast likely benefits primary tumor evaluation, delineation of lymph nodes from adjacent vessels, and detection of hepatic metastases. No studies have specifically evaluated the performance of CT of the abdomen and pelvis without IV contrast for vaginal cancer staging. FDG-PET/CT Skull Base to Mid-Thigh Data regarding the diagnostic performance of PET/CT for initial staging in patients with vaginal cancer are limited. Lamoreaux et al [18], in a prospective study, evaluated the comparative performance of PET versus CT in 23 patients with primary vaginal cancer prior to treatment. PET identified suspicious pelvic and/or groin lymph nodes in 35% (8 of 23) of patients, whereas CT did so in only 17% (4 of 23) of patients, although a pathologic reference standard was present in only two sampled groin nodes. No patient had extrapelvic nodal or distant disease, limiting the applicability of this study to metastases outside of the pelvis. Although data are limited for primary vaginal cancer staging, a growing body of literature supports the role of FDG- PET/CT in the initial staging of cervical cancer. Prospective data from the ACRIN 6671/GOG 0233 trial suggested, with borderline statistical significance, that FDG-PET/CT is more sensitive than CT alone for extrapelvic nodal metastases in cervical cancer (50% versus 42%, respectively), with similar specificity (85% versus 89%, respectively) [23], supporting prior retrospective data [22].

Staging and Follow up of Primary Vaginal Cancer. These findings, if applied to vaginal cancer, suggest that modern multidetector CT abdomen and pelvis is a reasonable staging tool for regional and distant metastases, although is likely inferior to MRI for local staging, modestly inferior to FDG-PET/CT for nodal metastases, and inferior to FDG-PET/CT for osseous metastases. The use of IV contrast is strongly encouraged when possible, because the improved tissue contrast likely benefits primary tumor evaluation, delineation of lymph nodes from adjacent vessels, and detection of hepatic metastases. No studies have specifically evaluated the performance of CT of the abdomen and pelvis without IV contrast for vaginal cancer staging. FDG-PET/CT Skull Base to Mid-Thigh Data regarding the diagnostic performance of PET/CT for initial staging in patients with vaginal cancer are limited. Lamoreaux et al [18], in a prospective study, evaluated the comparative performance of PET versus CT in 23 patients with primary vaginal cancer prior to treatment. PET identified suspicious pelvic and/or groin lymph nodes in 35% (8 of 23) of patients, whereas CT did so in only 17% (4 of 23) of patients, although a pathologic reference standard was present in only two sampled groin nodes. No patient had extrapelvic nodal or distant disease, limiting the applicability of this study to metastases outside of the pelvis. Although data are limited for primary vaginal cancer staging, a growing body of literature supports the role of FDG- PET/CT in the initial staging of cervical cancer. Prospective data from the ACRIN 6671/GOG 0233 trial suggested, with borderline statistical significance, that FDG-PET/CT is more sensitive than CT alone for extrapelvic nodal metastases in cervical cancer (50% versus 42%, respectively), with similar specificity (85% versus 89%, respectively) [23], supporting prior retrospective data [22].

3102401

acrac_3102401_6

Staging and Follow up of Primary Vaginal Cancer

FDG-PET/CT is also more sensitive than conventional CT for osseous metastases [25], with sensitivity and specificity of 55% and 98%, respectively, for all distant metastases [13]. Accordingly, the NCCN guidelines endorse preference for whole-body FDG-PET/CT over conventional CT for initial staging of all cervical cancer designated stage II and above, with either FDG-PET/CT or conventional CT recommended in stage I disease [27]. Staging and Follow-up of Primary Vaginal Cancer Fluoroscopy Contrast Enema There is no relevant literature regarding the use of fluoroscopic contrast enema in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. MRI Pelvis Because of the rarity of vaginal cancer, the primary data regarding the use of MRI in initial staging of vaginal cancer are sparse. Taylor et al [28] retrospectively evaluated pelvic MRI for initial staging in 25 patients with primary vaginal cancer spanning all disease stages. MRI depicted the primary tumor in 96% (24 of 25) of patients, demonstrating hyperintense signal compared to muscle on T2-weighted images, and enabled assignment of a radiologic disease stage based on adaptation of FIGO clinical staging criteria. Because 80% (20 of 25) of patients received either radiation or palliative therapy, pathologic confirmation of imaging findings could be obtained in only 20% (5 of 25) of cases. Of these cases, MRI stage was concordant with pathologic stage in 40% (2 of 5) of the cases. More recent data in cervical cancer patients support the use of MRI for initial staging, with a meta-analysis suggesting high sensitivity (76%) and specificity (94%) of MRI for parametrial invasion [20]. Although MRI readily depicts lymph nodes, it has constraints similar to CT with regard to the limited sensitivity and specificity of size and morphologic criteria. No study has specifically evaluated the performance of MRI for pretreatment nodal staging in vaginal cancer.

Staging and Follow up of Primary Vaginal Cancer. FDG-PET/CT is also more sensitive than conventional CT for osseous metastases [25], with sensitivity and specificity of 55% and 98%, respectively, for all distant metastases [13]. Accordingly, the NCCN guidelines endorse preference for whole-body FDG-PET/CT over conventional CT for initial staging of all cervical cancer designated stage II and above, with either FDG-PET/CT or conventional CT recommended in stage I disease [27]. Staging and Follow-up of Primary Vaginal Cancer Fluoroscopy Contrast Enema There is no relevant literature regarding the use of fluoroscopic contrast enema in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. MRI Pelvis Because of the rarity of vaginal cancer, the primary data regarding the use of MRI in initial staging of vaginal cancer are sparse. Taylor et al [28] retrospectively evaluated pelvic MRI for initial staging in 25 patients with primary vaginal cancer spanning all disease stages. MRI depicted the primary tumor in 96% (24 of 25) of patients, demonstrating hyperintense signal compared to muscle on T2-weighted images, and enabled assignment of a radiologic disease stage based on adaptation of FIGO clinical staging criteria. Because 80% (20 of 25) of patients received either radiation or palliative therapy, pathologic confirmation of imaging findings could be obtained in only 20% (5 of 25) of cases. Of these cases, MRI stage was concordant with pathologic stage in 40% (2 of 5) of the cases. More recent data in cervical cancer patients support the use of MRI for initial staging, with a meta-analysis suggesting high sensitivity (76%) and specificity (94%) of MRI for parametrial invasion [20]. Although MRI readily depicts lymph nodes, it has constraints similar to CT with regard to the limited sensitivity and specificity of size and morphologic criteria. No study has specifically evaluated the performance of MRI for pretreatment nodal staging in vaginal cancer.

3102401

acrac_3102401_7

Staging and Follow up of Primary Vaginal Cancer

However, data from mixed cohorts of patients with recurrence of cervical, vaginal, and other gynecologic cancers have suggested superior sensitivity of FDG-PET/CT for pelvic nodal metastases compared with pelvic MRI and CT [29,30]. The use of IV contrast may improve tissue characterization but is not considered essential, with variable inclusion in published protocols for evaluation of vaginal [28,31] and cervical cancer [7,32-34]. No study has specifically compared the incremental utility of contrast-enhanced sequences over T2-weighted sequences for pelvic MRI in this context. Regarding the use of vaginal gel in MRI of the pelvis, there is insufficient primary data in the literature to support its routine use. MRI Abdomen and Pelvis Because of the rarity of vaginal cancer, the primary data regarding the use of MRI in initial staging of vaginal cancer are sparse. Taylor et al [28] retrospectively evaluated pelvic MRI for initial staging in 25 patients with primary vaginal cancer spanning all disease stages. MRI depicted the primary tumor in 96% (24 of 25) of patients, demonstrating hyperintense signal compared to muscle on T2-weighted images, and enabled assignment of a radiologic disease stage based on adaptation of FIGO clinical staging criteria. Because 80% (20 of 25) of patients received either radiation or palliative therapy, pathologic confirmation of imaging findings could be obtained in only 20% (5 of 25) of cases. Of these cases, MRI stage was concordant with pathologic stage in 40% (2 of 5). More recent data in cervical cancer patients support the use of MRI for initial staging, with a meta-analysis suggesting high sensitivity (76%) and specificity (94%) of MRI for parametrial invasion [20]. Although MRI readily depicts lymph nodes, it has constraints similar to CT with regard to the limited sensitivity and specificity of size and morphologic criteria.

Staging and Follow up of Primary Vaginal Cancer. However, data from mixed cohorts of patients with recurrence of cervical, vaginal, and other gynecologic cancers have suggested superior sensitivity of FDG-PET/CT for pelvic nodal metastases compared with pelvic MRI and CT [29,30]. The use of IV contrast may improve tissue characterization but is not considered essential, with variable inclusion in published protocols for evaluation of vaginal [28,31] and cervical cancer [7,32-34]. No study has specifically compared the incremental utility of contrast-enhanced sequences over T2-weighted sequences for pelvic MRI in this context. Regarding the use of vaginal gel in MRI of the pelvis, there is insufficient primary data in the literature to support its routine use. MRI Abdomen and Pelvis Because of the rarity of vaginal cancer, the primary data regarding the use of MRI in initial staging of vaginal cancer are sparse. Taylor et al [28] retrospectively evaluated pelvic MRI for initial staging in 25 patients with primary vaginal cancer spanning all disease stages. MRI depicted the primary tumor in 96% (24 of 25) of patients, demonstrating hyperintense signal compared to muscle on T2-weighted images, and enabled assignment of a radiologic disease stage based on adaptation of FIGO clinical staging criteria. Because 80% (20 of 25) of patients received either radiation or palliative therapy, pathologic confirmation of imaging findings could be obtained in only 20% (5 of 25) of cases. Of these cases, MRI stage was concordant with pathologic stage in 40% (2 of 5). More recent data in cervical cancer patients support the use of MRI for initial staging, with a meta-analysis suggesting high sensitivity (76%) and specificity (94%) of MRI for parametrial invasion [20]. Although MRI readily depicts lymph nodes, it has constraints similar to CT with regard to the limited sensitivity and specificity of size and morphologic criteria.

3102401

acrac_3102401_8

Staging and Follow up of Primary Vaginal Cancer

No study has specifically evaluated the performance of MRI for pretreatment nodal staging in vaginal cancer. However, data from mixed cohorts of patients with recurrence of cervical, vaginal, and other gynecologic cancers have suggested superior sensitivity of FDG-PET/CT for pelvic nodal metastases compared with pelvic MRI and CT [29,30]. If MRI of the abdomen and pelvis is used in place of CT of the abdomen and pelvis, the addition of chest CT is encouraged to evaluate for pulmonary metastases. The use of IV contrast may improve tissue characterization and is particularly beneficial when MRI of the abdomen is included, because it improves detection of hepatic metastases. Radiography Intravenous Urography There is no relevant literature regarding the use of radiographic IV urography in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. US Abdomen and Pelvis Transabdominal There is no relevant literature regarding the role of transabdominal abdominopelvic (TA) US in vaginal cancer staging. TAUS is inferior for visualizing the female genital tract compared with TVUS, and neither technique has a role in the evaluation of regional or distant disease. Variant 2: Posttreatment evaluation of vaginal cancer. No suspected recurrence. Initial imaging. As the use of definitive chemoradiation for the treatment of primary vaginal cancer has grown, so too has the role of cross-imaging for assessment of treatment response. In contrast to extirpative surgery, in which pathologic margin assessment can confirm removal of viable tumor, evaluation for tumor eradication following chemoradiation relies in part on imaging assessment. Much of the support for the value of early posttreatment imaging in primary vaginal cancer is extrapolated from the large body of literature on cervical cancer, for which the treatment paradigm and endpoints are analogous.

Staging and Follow up of Primary Vaginal Cancer. No study has specifically evaluated the performance of MRI for pretreatment nodal staging in vaginal cancer. However, data from mixed cohorts of patients with recurrence of cervical, vaginal, and other gynecologic cancers have suggested superior sensitivity of FDG-PET/CT for pelvic nodal metastases compared with pelvic MRI and CT [29,30]. If MRI of the abdomen and pelvis is used in place of CT of the abdomen and pelvis, the addition of chest CT is encouraged to evaluate for pulmonary metastases. The use of IV contrast may improve tissue characterization and is particularly beneficial when MRI of the abdomen is included, because it improves detection of hepatic metastases. Radiography Intravenous Urography There is no relevant literature regarding the use of radiographic IV urography in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. US Abdomen and Pelvis Transabdominal There is no relevant literature regarding the role of transabdominal abdominopelvic (TA) US in vaginal cancer staging. TAUS is inferior for visualizing the female genital tract compared with TVUS, and neither technique has a role in the evaluation of regional or distant disease. Variant 2: Posttreatment evaluation of vaginal cancer. No suspected recurrence. Initial imaging. As the use of definitive chemoradiation for the treatment of primary vaginal cancer has grown, so too has the role of cross-imaging for assessment of treatment response. In contrast to extirpative surgery, in which pathologic margin assessment can confirm removal of viable tumor, evaluation for tumor eradication following chemoradiation relies in part on imaging assessment. Much of the support for the value of early posttreatment imaging in primary vaginal cancer is extrapolated from the large body of literature on cervical cancer, for which the treatment paradigm and endpoints are analogous.

3102401

acrac_3102401_9

Staging and Follow up of Primary Vaginal Cancer

Early posttreatment imaging is performed most commonly following a period of approximately 3 to 6 months after the completion of chemoradiation. Some centers also image during therapy for early response assessment and/or adaptive radiation planning [7]. The goals of early posttreatment imaging are multiple. First, imaging response after chemoradiation is a potent predictor of oncologic outcome, therefore providing crucial prognostic data [40-42]. Second, the degree of imaging response directly informs therapeutic decision-making, because persistent or progressive disease following chemoradiation requires salvage therapy [40]. For persistent pelvic disease, options include salvage radical surgery or less commonly reirradiation. Detection of new distant disease following initial treatment obviates curative surgery and may direct therapy toward chemotherapeutic and/or palliative options. Finally, the degree of response can influence the frequency of subsequent surveillance, with complete response enabling more conservative follow- up testing [42]. Following complete response, there is no formally established role for routine surveillance imaging in asymptomatic patients treated for vaginal cancer nor has a role been established for cervical cancer. Guidelines generally advocate for routine clinical examination for surveillance in asymptomatic patients, with imaging suggested in the setting of symptoms or abnormal physical examination findings [43]. CT Chest Although thoracic metastases are known to occur in vaginal cancer, no studies specifically address their incidence or the incremental value of chest CT in early posttreatment evaluation. Pulmonary metastases have been studied to a limited degree in cervical cancer, occurring in approximately 5% to 10% of patients at diagnosis [13,14].

Staging and Follow up of Primary Vaginal Cancer. Early posttreatment imaging is performed most commonly following a period of approximately 3 to 6 months after the completion of chemoradiation. Some centers also image during therapy for early response assessment and/or adaptive radiation planning [7]. The goals of early posttreatment imaging are multiple. First, imaging response after chemoradiation is a potent predictor of oncologic outcome, therefore providing crucial prognostic data [40-42]. Second, the degree of imaging response directly informs therapeutic decision-making, because persistent or progressive disease following chemoradiation requires salvage therapy [40]. For persistent pelvic disease, options include salvage radical surgery or less commonly reirradiation. Detection of new distant disease following initial treatment obviates curative surgery and may direct therapy toward chemotherapeutic and/or palliative options. Finally, the degree of response can influence the frequency of subsequent surveillance, with complete response enabling more conservative follow- up testing [42]. Following complete response, there is no formally established role for routine surveillance imaging in asymptomatic patients treated for vaginal cancer nor has a role been established for cervical cancer. Guidelines generally advocate for routine clinical examination for surveillance in asymptomatic patients, with imaging suggested in the setting of symptoms or abnormal physical examination findings [43]. CT Chest Although thoracic metastases are known to occur in vaginal cancer, no studies specifically address their incidence or the incremental value of chest CT in early posttreatment evaluation. Pulmonary metastases have been studied to a limited degree in cervical cancer, occurring in approximately 5% to 10% of patients at diagnosis [13,14].

3102401

acrac_3102401_10

Staging and Follow up of Primary Vaginal Cancer

Pulmonary metastases appear to occur slightly more frequently as a site of recurrent disease, with one large study of recurrent cervical cancer indicating an overall incidence of 13%, and the lungs representing the only site of recurrence in 6% of cases [15]. Moreover, the lungs can uncommonly represent a site of distant disease that newly arises following definitive chemoradiation for disease that was initially locoregional [41]. In studies evaluating pulmonary metastases from cervical cancer, chest CT was the most frequent diagnostic modality employed, with the vast majority of patients asymptomatic at the time of imaging [16,17]. These findings support the use of chest CT with or without IV contrast in the early posttreatment evaluation of cervical cancer, as endorsed by the NCCN guidelines, and suggest that a similar strategy would be useful for vaginal cancer. CT Abdomen and Pelvis For detection of residual primary tumor after chemoradiation, CT alone is likely inferior compared with FDG- PET/CT and pelvic MRI based on extrapolation from data on comparative imaging performance in the pretreatment evaluation of cervical cancer [21,44]. CT lacks the tissue contrast of MRI and the metabolic data of FDG-PET, both of which are useful in deciphering posttreatment changes from residual disease. Because CT relies primarily on size criteria for nodal evaluation, it has limitations similar to MRI with respect to sensitivity and specificity for nodal metastases. Therefore, although CT may depict size regression of nodal metastases following therapy, it is likely at least modestly inferior for detecting new or residual disease in subcentimeter lymph nodes compared with FDG- PET/CT [22,23,29,30].

Staging and Follow up of Primary Vaginal Cancer. Pulmonary metastases appear to occur slightly more frequently as a site of recurrent disease, with one large study of recurrent cervical cancer indicating an overall incidence of 13%, and the lungs representing the only site of recurrence in 6% of cases [15]. Moreover, the lungs can uncommonly represent a site of distant disease that newly arises following definitive chemoradiation for disease that was initially locoregional [41]. In studies evaluating pulmonary metastases from cervical cancer, chest CT was the most frequent diagnostic modality employed, with the vast majority of patients asymptomatic at the time of imaging [16,17]. These findings support the use of chest CT with or without IV contrast in the early posttreatment evaluation of cervical cancer, as endorsed by the NCCN guidelines, and suggest that a similar strategy would be useful for vaginal cancer. CT Abdomen and Pelvis For detection of residual primary tumor after chemoradiation, CT alone is likely inferior compared with FDG- PET/CT and pelvic MRI based on extrapolation from data on comparative imaging performance in the pretreatment evaluation of cervical cancer [21,44]. CT lacks the tissue contrast of MRI and the metabolic data of FDG-PET, both of which are useful in deciphering posttreatment changes from residual disease. Because CT relies primarily on size criteria for nodal evaluation, it has limitations similar to MRI with respect to sensitivity and specificity for nodal metastases. Therefore, although CT may depict size regression of nodal metastases following therapy, it is likely at least modestly inferior for detecting new or residual disease in subcentimeter lymph nodes compared with FDG- PET/CT [22,23,29,30].

3102401

acrac_3102401_11

Staging and Follow up of Primary Vaginal Cancer

Staging and Follow-up of Primary Vaginal Cancer CT of the abdomen and pelvis is not commonly performed in the absence of chest CT, given that the lungs are a potential site of distant disease that may newly arise in patients who have undergone definitive chemoradiation for disease that was initially locoregional [41]. Importantly, CT alone is inferior to FDG/PET-CT for evaluation of distant disease in the bones [25] and modestly inferior for nodal assessment [23,41]. The use of IV contrast is strongly encouraged when possible, because the improved tissue contrast likely benefits primary tumor evaluation, delineation of lymph nodes from adjacent vessels, and detection of hepatic metastases. No studies have specifically assessed the performance of CT of the abdomen and pelvis without IV contrast for posttreatment evaluation of primary vaginal cancer. FDG-PET/CT Skull Base to Mid-Thigh Although data in primary vaginal cancer patients are limited, studies substantiating its treatment response assessment role in cervical cancer are numerous. In one prospective study in cervical cancer patients treated with definitive chemoradiation, FDG-PET/CT responses classified as complete metabolic response (absence of abnormal uptake at prior sites of disease), partial metabolic response and progressive disease at a mean of 3 months after therapy correlated closely with prognosis, with 3 year progression-free survival of 78%, 33%, and 0%, respectively [40]. In another prospective study, 9% (5 of 55) of patients developed new distant disease at the time of a posttreatment FDG-PET/CT scan, underscoring the value of whole-body imaging rather than pelvic-only imaging at the time of response evaluation [41]. Accordingly, the NCCN guidelines for cervical cancer recommend whole- body FDG-PET/CT at 3 to 6 months after completion of definitive therapy for disease stages II to IV, because it directly informs prognosis, therapy, and intensity of surveillance [27].

Staging and Follow up of Primary Vaginal Cancer. Staging and Follow-up of Primary Vaginal Cancer CT of the abdomen and pelvis is not commonly performed in the absence of chest CT, given that the lungs are a potential site of distant disease that may newly arise in patients who have undergone definitive chemoradiation for disease that was initially locoregional [41]. Importantly, CT alone is inferior to FDG/PET-CT for evaluation of distant disease in the bones [25] and modestly inferior for nodal assessment [23,41]. The use of IV contrast is strongly encouraged when possible, because the improved tissue contrast likely benefits primary tumor evaluation, delineation of lymph nodes from adjacent vessels, and detection of hepatic metastases. No studies have specifically assessed the performance of CT of the abdomen and pelvis without IV contrast for posttreatment evaluation of primary vaginal cancer. FDG-PET/CT Skull Base to Mid-Thigh Although data in primary vaginal cancer patients are limited, studies substantiating its treatment response assessment role in cervical cancer are numerous. In one prospective study in cervical cancer patients treated with definitive chemoradiation, FDG-PET/CT responses classified as complete metabolic response (absence of abnormal uptake at prior sites of disease), partial metabolic response and progressive disease at a mean of 3 months after therapy correlated closely with prognosis, with 3 year progression-free survival of 78%, 33%, and 0%, respectively [40]. In another prospective study, 9% (5 of 55) of patients developed new distant disease at the time of a posttreatment FDG-PET/CT scan, underscoring the value of whole-body imaging rather than pelvic-only imaging at the time of response evaluation [41]. Accordingly, the NCCN guidelines for cervical cancer recommend whole- body FDG-PET/CT at 3 to 6 months after completion of definitive therapy for disease stages II to IV, because it directly informs prognosis, therapy, and intensity of surveillance [27].

3102401

acrac_3102401_12

Staging and Follow up of Primary Vaginal Cancer

Fluoroscopy Contrast Enema There is no relevant literature regarding the use of fluoroscopic contrast enema in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. MRI Pelvis Although no study has specifically evaluated pelvic MRI for treatment response assessment in vaginal cancer patients, multiple studies support its potential value in cervical cancer to which analogous principles apply. Following successful therapy with chemoradiation, the initially intermediate to high-signal-intensity tumor on T2- weighted images decreases in both size and signal intensity, with eventual conversion to low-signal-intensity fibrotic tissue [7,31]. However, the main limitation of MRI in the very early posttreatment period (<2 months after completion) is its difficulty distinguishing early postradiation change from residual tumor, both of which can demonstrate intermediate- to high-signal T2-weighted intensity and avid gadolinium enhancement [33,34]. One retrospective study evaluating pelvic MRI at a median of 5 weeks after completion of chemoradiation for cervical cancer found that 37% (16 of 44) of MRI examinations were considered indeterminate for discriminating residual disease and fibrosis [34]. Despite diagnostic confidence in the remainder of cases, sensitivity and specificity for residual disease were 80% and 55%, respectively, indicating a high false-positive rate because of posttreatment change. A more recent retrospective study in cervical cancer patients found better performance of pelvic MRI at a later postchemoradiation time point (median 9 weeks) with strict objective diagnostic criteria, achieving sensitivity and specificity of 91% and 85%, respectively, for residual disease [33].

Staging and Follow up of Primary Vaginal Cancer. Fluoroscopy Contrast Enema There is no relevant literature regarding the use of fluoroscopic contrast enema in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. MRI Pelvis Although no study has specifically evaluated pelvic MRI for treatment response assessment in vaginal cancer patients, multiple studies support its potential value in cervical cancer to which analogous principles apply. Following successful therapy with chemoradiation, the initially intermediate to high-signal-intensity tumor on T2- weighted images decreases in both size and signal intensity, with eventual conversion to low-signal-intensity fibrotic tissue [7,31]. However, the main limitation of MRI in the very early posttreatment period (<2 months after completion) is its difficulty distinguishing early postradiation change from residual tumor, both of which can demonstrate intermediate- to high-signal T2-weighted intensity and avid gadolinium enhancement [33,34]. One retrospective study evaluating pelvic MRI at a median of 5 weeks after completion of chemoradiation for cervical cancer found that 37% (16 of 44) of MRI examinations were considered indeterminate for discriminating residual disease and fibrosis [34]. Despite diagnostic confidence in the remainder of cases, sensitivity and specificity for residual disease were 80% and 55%, respectively, indicating a high false-positive rate because of posttreatment change. A more recent retrospective study in cervical cancer patients found better performance of pelvic MRI at a later postchemoradiation time point (median 9 weeks) with strict objective diagnostic criteria, achieving sensitivity and specificity of 91% and 85%, respectively, for residual disease [33].

3102401

acrac_3102401_13

Staging and Follow up of Primary Vaginal Cancer

Therefore, for cervical cancer, the suggested time interval for determining posttherapy treatment response with pelvic MRI is 3 to 6 months after completion of therapy [27], although earlier imaging is sometimes used for interim assessment of tumor regression for prognostication and/or adaptive radiation planning. Because MRI relies primarily on size criteria for nodal evaluation, it has limitations similar to CT with respect to sensitivity and specificity for nodal metastases. Therefore, although MRI may depict size regression of nodal metastases following therapy, it is likely at least modestly inferior for detecting new or residual disease in subcentimeter lymph nodes compared to FDG-PET/CT [22,23,29,30]. The use of IV contrast may improve tissue characterization but is not considered essential, with variable inclusion in published protocols for evaluation of vaginal [28,31] and cervical cancer [7,32-34]. No study has specifically compared the incremental utility of gadolinium-enhanced sequences over T2-weighted sequences for pelvic MRI in this context. Regarding the use of vaginal gel in MRI of the pelvis, there is insufficient primary data in the literature to support its routine use. MRI Abdomen and Pelvis MRI of the abdomen and pelvis can be considered in the early posttreatment evaluation of primary vaginal cancer, although its main value is in the utility of pelvic MRI for primary tumor response assessment. MRI of the abdomen is not commonly included, given the availability of whole-body FDG-PET/CT or CT of the chest, abdomen, and Staging and Follow-up of Primary Vaginal Cancer pelvis for evaluation of distant disease. Although no study has specifically evaluated pelvic MRI for treatment response assessment in vaginal cancer patients, multiple studies support its potential value in cervical cancer to which analogous principles apply.

Staging and Follow up of Primary Vaginal Cancer. Therefore, for cervical cancer, the suggested time interval for determining posttherapy treatment response with pelvic MRI is 3 to 6 months after completion of therapy [27], although earlier imaging is sometimes used for interim assessment of tumor regression for prognostication and/or adaptive radiation planning. Because MRI relies primarily on size criteria for nodal evaluation, it has limitations similar to CT with respect to sensitivity and specificity for nodal metastases. Therefore, although MRI may depict size regression of nodal metastases following therapy, it is likely at least modestly inferior for detecting new or residual disease in subcentimeter lymph nodes compared to FDG-PET/CT [22,23,29,30]. The use of IV contrast may improve tissue characterization but is not considered essential, with variable inclusion in published protocols for evaluation of vaginal [28,31] and cervical cancer [7,32-34]. No study has specifically compared the incremental utility of gadolinium-enhanced sequences over T2-weighted sequences for pelvic MRI in this context. Regarding the use of vaginal gel in MRI of the pelvis, there is insufficient primary data in the literature to support its routine use. MRI Abdomen and Pelvis MRI of the abdomen and pelvis can be considered in the early posttreatment evaluation of primary vaginal cancer, although its main value is in the utility of pelvic MRI for primary tumor response assessment. MRI of the abdomen is not commonly included, given the availability of whole-body FDG-PET/CT or CT of the chest, abdomen, and Staging and Follow-up of Primary Vaginal Cancer pelvis for evaluation of distant disease. Although no study has specifically evaluated pelvic MRI for treatment response assessment in vaginal cancer patients, multiple studies support its potential value in cervical cancer to which analogous principles apply.

3102401

acrac_3102401_14

Staging and Follow up of Primary Vaginal Cancer

Following successful therapy with chemoradiation, the initially intermediate- to high-signal-intensity tumor on T2-weighted images decreases in both size and signal intensity, with eventual conversion to low-signal-intensity fibrotic tissue [7,31]. However, the main limitation of MRI in the very early posttreatment period (<2 months after completion) is its difficulty distinguishing early postradiation change from residual tumor, both of which can demonstrate intermediate- to high-signal T2-weighted intensity and avid gadolinium enhancement [33,34]. One retrospective study evaluating pelvic MRI at a median of 5 weeks after completion of chemoradiation for cervical cancer found that 37% (16 of 44) of MRI examinations were considered indeterminate for discriminating residual disease and fibrosis [34]. Despite diagnostic confidence in the remainder of cases, sensitivity and specificity for residual disease were 80% and 55%, respectively, indicating a high false-positive rate because of posttreatment change. A more recent retrospective study in cervical cancer patients found better performance of pelvic MRI at a later postchemoradiation time point (median 9 weeks) with strict objective diagnostic criteria, achieving sensitivity and specificity of 91% and 85%, respectively, for residual disease [33]. Therefore, for cervical cancer, the suggested time interval for determining posttherapy treatment response with pelvic MRI is 3 to 6 months after completion of therapy [27], although earlier imaging is sometimes used for interim assessment of tumor regression for prognostication and/or adaptive radiation planning. Because MRI relies primarily on size criteria for nodal evaluation, it has limitations similar to CT with respect to sensitivity and specificity for nodal metastases.

Staging and Follow up of Primary Vaginal Cancer. Following successful therapy with chemoradiation, the initially intermediate- to high-signal-intensity tumor on T2-weighted images decreases in both size and signal intensity, with eventual conversion to low-signal-intensity fibrotic tissue [7,31]. However, the main limitation of MRI in the very early posttreatment period (<2 months after completion) is its difficulty distinguishing early postradiation change from residual tumor, both of which can demonstrate intermediate- to high-signal T2-weighted intensity and avid gadolinium enhancement [33,34]. One retrospective study evaluating pelvic MRI at a median of 5 weeks after completion of chemoradiation for cervical cancer found that 37% (16 of 44) of MRI examinations were considered indeterminate for discriminating residual disease and fibrosis [34]. Despite diagnostic confidence in the remainder of cases, sensitivity and specificity for residual disease were 80% and 55%, respectively, indicating a high false-positive rate because of posttreatment change. A more recent retrospective study in cervical cancer patients found better performance of pelvic MRI at a later postchemoradiation time point (median 9 weeks) with strict objective diagnostic criteria, achieving sensitivity and specificity of 91% and 85%, respectively, for residual disease [33]. Therefore, for cervical cancer, the suggested time interval for determining posttherapy treatment response with pelvic MRI is 3 to 6 months after completion of therapy [27], although earlier imaging is sometimes used for interim assessment of tumor regression for prognostication and/or adaptive radiation planning. Because MRI relies primarily on size criteria for nodal evaluation, it has limitations similar to CT with respect to sensitivity and specificity for nodal metastases.

3102401

acrac_3102401_15

Staging and Follow up of Primary Vaginal Cancer

Therefore, although MRI may depict size regression of nodal metastases following therapy, it is likely at least modestly inferior for detecting new or residual disease in subcentimeter lymph nodes compared to FDG-PET/CT [22,23,29,30]. If MRI of the abdomen and pelvis is used in place of CT of the abdomen and pelvis, the addition of chest CT is encouraged to evaluate for pulmonary metastases. The use of IV contrast may improve tissue characterization and should be used especially when MRI of the abdomen is included, because it improves detection of hepatic metastases. Radiography Intravenous Urography There is no relevant literature regarding the use of radiographic IV urography in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. US Abdomen and Pelvis Transabdominal There is no relevant literature regarding the role of TAUS in vaginal cancer staging. TAUS is inferior for visualizing the female genital tract compared with TVUS, and neither technique has a role in nodal or distant evaluation. US Pelvis Transvaginal There is no relevant literature regarding the role of TVUS in the early posttreatment evaluation of primary vaginal cancer. Limited studies in cervical cancer patients have evaluated the use of color and/or power Doppler US for detecting changes in tumor vascularity as a marker of treatment response [45]. However, the applicability of these findings to clinical practice remains unclear. The NCCN guidelines do not currently endorse the use of TVUS for early posttreatment evaluation in cervical cancer, and its role in vaginal cancer remains undefined. Additionally, TVUS has limited utility for pelvic nodal evaluation [39]. Variant 3: Vaginal cancer. Suspected or known recurrence. Evaluate extent of disease. Initial imaging.

Staging and Follow up of Primary Vaginal Cancer. Therefore, although MRI may depict size regression of nodal metastases following therapy, it is likely at least modestly inferior for detecting new or residual disease in subcentimeter lymph nodes compared to FDG-PET/CT [22,23,29,30]. If MRI of the abdomen and pelvis is used in place of CT of the abdomen and pelvis, the addition of chest CT is encouraged to evaluate for pulmonary metastases. The use of IV contrast may improve tissue characterization and should be used especially when MRI of the abdomen is included, because it improves detection of hepatic metastases. Radiography Intravenous Urography There is no relevant literature regarding the use of radiographic IV urography in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. US Abdomen and Pelvis Transabdominal There is no relevant literature regarding the role of TAUS in vaginal cancer staging. TAUS is inferior for visualizing the female genital tract compared with TVUS, and neither technique has a role in nodal or distant evaluation. US Pelvis Transvaginal There is no relevant literature regarding the role of TVUS in the early posttreatment evaluation of primary vaginal cancer. Limited studies in cervical cancer patients have evaluated the use of color and/or power Doppler US for detecting changes in tumor vascularity as a marker of treatment response [45]. However, the applicability of these findings to clinical practice remains unclear. The NCCN guidelines do not currently endorse the use of TVUS for early posttreatment evaluation in cervical cancer, and its role in vaginal cancer remains undefined. Additionally, TVUS has limited utility for pelvic nodal evaluation [39]. Variant 3: Vaginal cancer. Suspected or known recurrence. Evaluate extent of disease. Initial imaging.

3102401

acrac_3102401_16

Staging and Follow up of Primary Vaginal Cancer

Cross-sectional imaging plays a crucial role in the evaluation of patients with known or suspected vaginal cancer recurrence, in which physical examination is of limited value in determining disease extent. In one retrospective study of patients with primary vaginal cancer who underwent definitive radiation and experienced recurrence, the mechanism of recurrence was locoregional alone in 56% for disease stages I and II and 71% for disease stages III to IVA, whereas the remainder of recurrences were distant [46]. Once locoregional recurrence is identified, the presence or absence of distant recurrence becomes a discriminating factor in eligibility for salvage pelvic exenteration. In the presence of distant recurrence, exenteration confers morbidity without significantly improving oncologic outcomes, whereas in the absence of distant recurrence, exenteration can potentially eradicate pelvic tumor burden. When distant disease has been excluded by imaging and a patient is deemed eligible for pelvic exenteration, the degree of local organ invasion determines whether partial (anterior or posterior) or total exenteration is indicated [32]. Therefore, imaging findings in patients with known or suspected vaginal cancer recurrence can influence both the appropriateness and type of salvage therapy, in addition to predicting prognosis. Staging and Follow-up of Primary Vaginal Cancer CT Chest Although thoracic metastases are known to occur in vaginal cancer, no studies specifically address their incidence or the incremental value of chest CT for suspected recurrence. Pulmonary metastases have been studied to a limited degree in cervical cancer, occurring in approximately 5% to 10% of patients at diagnosis [13,14]. Pulmonary metastases appear to occur slightly more frequently as a site of recurrent disease, with one large study of recurrent cervical cancer indicating an overall incidence of 13% and the lungs representing the only site of recurrence in 6% of cases [15].

Staging and Follow up of Primary Vaginal Cancer. Cross-sectional imaging plays a crucial role in the evaluation of patients with known or suspected vaginal cancer recurrence, in which physical examination is of limited value in determining disease extent. In one retrospective study of patients with primary vaginal cancer who underwent definitive radiation and experienced recurrence, the mechanism of recurrence was locoregional alone in 56% for disease stages I and II and 71% for disease stages III to IVA, whereas the remainder of recurrences were distant [46]. Once locoregional recurrence is identified, the presence or absence of distant recurrence becomes a discriminating factor in eligibility for salvage pelvic exenteration. In the presence of distant recurrence, exenteration confers morbidity without significantly improving oncologic outcomes, whereas in the absence of distant recurrence, exenteration can potentially eradicate pelvic tumor burden. When distant disease has been excluded by imaging and a patient is deemed eligible for pelvic exenteration, the degree of local organ invasion determines whether partial (anterior or posterior) or total exenteration is indicated [32]. Therefore, imaging findings in patients with known or suspected vaginal cancer recurrence can influence both the appropriateness and type of salvage therapy, in addition to predicting prognosis. Staging and Follow-up of Primary Vaginal Cancer CT Chest Although thoracic metastases are known to occur in vaginal cancer, no studies specifically address their incidence or the incremental value of chest CT for suspected recurrence. Pulmonary metastases have been studied to a limited degree in cervical cancer, occurring in approximately 5% to 10% of patients at diagnosis [13,14]. Pulmonary metastases appear to occur slightly more frequently as a site of recurrent disease, with one large study of recurrent cervical cancer indicating an overall incidence of 13% and the lungs representing the only site of recurrence in 6% of cases [15].

3102401

acrac_3102401_17

Staging and Follow up of Primary Vaginal Cancer

In studies evaluating pulmonary metastases from cervical cancer, chest CT was the most frequent diagnostic modality employed, with the vast majority of patients asymptomatic at the time of imaging [16,17]. These findings support the use of chest CT with or without IV contrast in the early posttreatment evaluation of cervical cancer, as endorsed by the NCCN guidelines, and suggest that a similar strategy would be useful for vaginal cancer. CT Abdomen and Pelvis Data on the diagnostic performance of CT in known or suspected recurrence of vaginal cancer are very limited, requiring extrapolation from pretreatment vaginal cancer cohorts as well as cohorts of patients with other gynecologic malignancies. Regarding local extent evaluation, the prospective ACRIN 6651 study of patients with cervical cancer prior to treatment, found that CT was insensitive for detection of rectal and bladder invasion, suggesting that performance would be similarly poor in the setting of recurrent disease prior to pelvic exenteration [19]. A small retrospective study evaluating FDG-PET/CT in 23 patients with primary vaginal cancer prior to treatment found that CT and FDG-PET detected pelvic nodal metastases in 17% (4 of 23) and 35% (8 of 23) of patients, respectively, suggesting inferior sensitivity of CT alone for nodal metastases. Although older literature suggested that CT is less sensitive than PET/CT for nodal metastases [22], the more recent ACRIN 6671/GOG 0233 trial in cervical cancer patients prior to treatment showed a more modest difference in sensitivity for abdominal nodes (42% versus 50%, respectively), and no significant difference in sensitivity for pelvic nodes (79% versus 83%, respectively) [23]. Likewise, CT had only modestly lower AUC (0.83) compared with PET/CT (0.90) for detection of nodal metastases from cervical cancer in a recent meta-analysis [24].

Staging and Follow up of Primary Vaginal Cancer. In studies evaluating pulmonary metastases from cervical cancer, chest CT was the most frequent diagnostic modality employed, with the vast majority of patients asymptomatic at the time of imaging [16,17]. These findings support the use of chest CT with or without IV contrast in the early posttreatment evaluation of cervical cancer, as endorsed by the NCCN guidelines, and suggest that a similar strategy would be useful for vaginal cancer. CT Abdomen and Pelvis Data on the diagnostic performance of CT in known or suspected recurrence of vaginal cancer are very limited, requiring extrapolation from pretreatment vaginal cancer cohorts as well as cohorts of patients with other gynecologic malignancies. Regarding local extent evaluation, the prospective ACRIN 6651 study of patients with cervical cancer prior to treatment, found that CT was insensitive for detection of rectal and bladder invasion, suggesting that performance would be similarly poor in the setting of recurrent disease prior to pelvic exenteration [19]. A small retrospective study evaluating FDG-PET/CT in 23 patients with primary vaginal cancer prior to treatment found that CT and FDG-PET detected pelvic nodal metastases in 17% (4 of 23) and 35% (8 of 23) of patients, respectively, suggesting inferior sensitivity of CT alone for nodal metastases. Although older literature suggested that CT is less sensitive than PET/CT for nodal metastases [22], the more recent ACRIN 6671/GOG 0233 trial in cervical cancer patients prior to treatment showed a more modest difference in sensitivity for abdominal nodes (42% versus 50%, respectively), and no significant difference in sensitivity for pelvic nodes (79% versus 83%, respectively) [23]. Likewise, CT had only modestly lower AUC (0.83) compared with PET/CT (0.90) for detection of nodal metastases from cervical cancer in a recent meta-analysis [24].

3102401

acrac_3102401_18

Staging and Follow up of Primary Vaginal Cancer

For distant metastases from cervical cancer, CT is inferior in the detection of osseous metastases (sensitivity 66%) compared with FDG-PET/CT (sensitivity 96%) [25]. These findings, if applied to vaginal cancer, suggest that CT is a reasonable staging tool for known or suspected tumor recurrence in the abdomen and pelvis, although it is likely inferior to MRI for evaluating local tumor extent, modestly inferior to FDG-PET/CT for nodal metastases, and inferior to FDG-PET/CT for osseous metastases. The use of IV contrast is strongly encouraged when possible, because the improved tissue contrast likely benefits primary tumor evaluation, delineation of lymph nodes from adjacent vessels, and detection of hepatic metastases. No studies have specifically assessed the performance of CT of the abdomen and pelvis without IV contrast for evaluation of known or suspected vaginal cancer recurrence. FDG-PET/CT Skull Base to Mid-Thigh No study has evaluated FDG-PET/CT in a cohort limited to patients with recurrent vaginal cancer. Data on the utility of FDG-PET/CT in this setting is limited to mixed cohorts of patients with various gynecologic malignancies, including vaginal cancer, with cervical squamous cell carcinoma generally comprising the majority of patients. One such cohort of 27 patients with recurrent gynecologic malignancies prior to pelvic exenteration was studied prospectively to compare FDG-PET and CT. FDG-PET was 100% sensitive and 73% specific for identifying extrapelvic metastases, most notably outperforming CT in the detection of pelvic and para-aortic nodal metastases [29]. A retrospective study of 85 patients with recurrent gynecologic malignancies reached similar conclusions, identifying findings suspicious for extraregional recurrence in 28% (24 of 85) of patients by PET versus 9% (8 of 85) of patients by conventional imaging (CT and pelvic MRI), with nodal metastases accounting for many of the discrepancies [30].

Staging and Follow up of Primary Vaginal Cancer. For distant metastases from cervical cancer, CT is inferior in the detection of osseous metastases (sensitivity 66%) compared with FDG-PET/CT (sensitivity 96%) [25]. These findings, if applied to vaginal cancer, suggest that CT is a reasonable staging tool for known or suspected tumor recurrence in the abdomen and pelvis, although it is likely inferior to MRI for evaluating local tumor extent, modestly inferior to FDG-PET/CT for nodal metastases, and inferior to FDG-PET/CT for osseous metastases. The use of IV contrast is strongly encouraged when possible, because the improved tissue contrast likely benefits primary tumor evaluation, delineation of lymph nodes from adjacent vessels, and detection of hepatic metastases. No studies have specifically assessed the performance of CT of the abdomen and pelvis without IV contrast for evaluation of known or suspected vaginal cancer recurrence. FDG-PET/CT Skull Base to Mid-Thigh No study has evaluated FDG-PET/CT in a cohort limited to patients with recurrent vaginal cancer. Data on the utility of FDG-PET/CT in this setting is limited to mixed cohorts of patients with various gynecologic malignancies, including vaginal cancer, with cervical squamous cell carcinoma generally comprising the majority of patients. One such cohort of 27 patients with recurrent gynecologic malignancies prior to pelvic exenteration was studied prospectively to compare FDG-PET and CT. FDG-PET was 100% sensitive and 73% specific for identifying extrapelvic metastases, most notably outperforming CT in the detection of pelvic and para-aortic nodal metastases [29]. A retrospective study of 85 patients with recurrent gynecologic malignancies reached similar conclusions, identifying findings suspicious for extraregional recurrence in 28% (24 of 85) of patients by PET versus 9% (8 of 85) of patients by conventional imaging (CT and pelvic MRI), with nodal metastases accounting for many of the discrepancies [30].

3102401

acrac_3102401_19

Staging and Follow up of Primary Vaginal Cancer

Concordant with these findings, the NCCN guidelines recommend whole-body FDG-PET/CT in patients with suspected recurrence of cervical cancer [27], although no such formal guidelines exist for vaginal cancer. Staging and Follow-up of Primary Vaginal Cancer Fluoroscopy Contrast Enema There is no relevant literature regarding the use of fluoroscopic contrast enema in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. Although MRI readily depicts lymph nodes, it has constraints similar to CT with regard to the limited sensitivity and specificity of size and morphologic criteria. No study has evaluated the diagnostic performance of MRI for nodal staging isolated to a cohort of primary vaginal cancer patients with disease recurrence. However, data from mixed cohorts of patients with recurrence of cervical, vaginal, and other gynecologic cancers have suggested superior sensitivity of FDG-PET/CT for pelvic nodal metastases compared with pelvic MRI and CT [29,30]. The use of IV contrast may improve tissue characterization but is not considered essential, with variable inclusion in published protocols for evaluation of vaginal [28,31] and cervical cancer [7,32-34]. No study has specifically compared the incremental utility of gadolinium-enhanced sequences over T2-weighted sequences for pelvic MRI in this context. Regarding the use of vaginal gel in MRI of the pelvis, there is insufficient primary data in the literature to support its routine use. Although MRI readily depicts lymph nodes, it has constraints similar to CT with regard to the limited sensitivity and specificity of size and morphologic criteria. No study has evaluated the diagnostic performance of MRI for nodal staging isolated to a cohort of primary vaginal cancer patients with disease recurrence.

Staging and Follow up of Primary Vaginal Cancer. Concordant with these findings, the NCCN guidelines recommend whole-body FDG-PET/CT in patients with suspected recurrence of cervical cancer [27], although no such formal guidelines exist for vaginal cancer. Staging and Follow-up of Primary Vaginal Cancer Fluoroscopy Contrast Enema There is no relevant literature regarding the use of fluoroscopic contrast enema in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. Although MRI readily depicts lymph nodes, it has constraints similar to CT with regard to the limited sensitivity and specificity of size and morphologic criteria. No study has evaluated the diagnostic performance of MRI for nodal staging isolated to a cohort of primary vaginal cancer patients with disease recurrence. However, data from mixed cohorts of patients with recurrence of cervical, vaginal, and other gynecologic cancers have suggested superior sensitivity of FDG-PET/CT for pelvic nodal metastases compared with pelvic MRI and CT [29,30]. The use of IV contrast may improve tissue characterization but is not considered essential, with variable inclusion in published protocols for evaluation of vaginal [28,31] and cervical cancer [7,32-34]. No study has specifically compared the incremental utility of gadolinium-enhanced sequences over T2-weighted sequences for pelvic MRI in this context. Regarding the use of vaginal gel in MRI of the pelvis, there is insufficient primary data in the literature to support its routine use. Although MRI readily depicts lymph nodes, it has constraints similar to CT with regard to the limited sensitivity and specificity of size and morphologic criteria. No study has evaluated the diagnostic performance of MRI for nodal staging isolated to a cohort of primary vaginal cancer patients with disease recurrence.

3102401

acrac_3102401_20

Staging and Follow up of Primary Vaginal Cancer

However, data from mixed cohorts of patients with recurrence of cervical, vaginal, and other gynecologic cancers have suggested superior sensitivity of FDG-PET/CT for pelvic nodal metastases compared with pelvic MRI and CT [29,30]. If MRI of the abdomen and pelvis is used in place of CT of the abdomen and pelvis, the addition of chest CT is encouraged to evaluate for pulmonary metastases. The use of IV contrast may improve tissue characterization and is particularly beneficial when MRI of the abdomen is included, because it improves detection of hepatic metastases. Radiography Intravenous Urography There is no relevant literature regarding the use of radiographic IV urography in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. US Pelvis Transvaginal There is no relevant literature regarding the role of TVUS in the evaluation of known or suspected vaginal cancer recurrence nor is there any such literature for cervical cancer recurrence. Additionally, the potential applicability of TVUS for recurrent vaginal cancer would be limited to local recurrence, because TVUS has little to no utility for pelvic nodal evaluation [39]. Supporting Documents The evidence table, literature search, and appendix for this topic are available at https://acsearch. acr.org/list. The appendix includes the strength of evidence assessment and the final rating round tabulations for each recommendation. For additional information on the Appropriateness Criteria methodology and other supporting documents go to www. acr.org/ac. Appropriateness Category Names and Definitions Relative Radiation Level Information Potential adverse health effects associated with radiation exposure are an important factor to consider when selecting the appropriate imaging procedure.

Staging and Follow up of Primary Vaginal Cancer. However, data from mixed cohorts of patients with recurrence of cervical, vaginal, and other gynecologic cancers have suggested superior sensitivity of FDG-PET/CT for pelvic nodal metastases compared with pelvic MRI and CT [29,30]. If MRI of the abdomen and pelvis is used in place of CT of the abdomen and pelvis, the addition of chest CT is encouraged to evaluate for pulmonary metastases. The use of IV contrast may improve tissue characterization and is particularly beneficial when MRI of the abdomen is included, because it improves detection of hepatic metastases. Radiography Intravenous Urography There is no relevant literature regarding the use of radiographic IV urography in the modern imaging workup of vaginal cancer, and its use has largely been replaced by cross-sectional imaging techniques. US Pelvis Transvaginal There is no relevant literature regarding the role of TVUS in the evaluation of known or suspected vaginal cancer recurrence nor is there any such literature for cervical cancer recurrence. Additionally, the potential applicability of TVUS for recurrent vaginal cancer would be limited to local recurrence, because TVUS has little to no utility for pelvic nodal evaluation [39]. Supporting Documents The evidence table, literature search, and appendix for this topic are available at https://acsearch. acr.org/list. The appendix includes the strength of evidence assessment and the final rating round tabulations for each recommendation. For additional information on the Appropriateness Criteria methodology and other supporting documents go to www. acr.org/ac. Appropriateness Category Names and Definitions Relative Radiation Level Information Potential adverse health effects associated with radiation exposure are an important factor to consider when selecting the appropriate imaging procedure.

3102401

acrac_3195150_0

Endometriosis

Introduction/Background Endometriosis is a common condition impacting approximately 10% of individuals assigned female at birth [1]. The disorder is caused by endometrial-like tissue located outside of the endometrial cavity, associated with inflammation and fibrosis, on or extending below the peritoneal surface [2]. Endometriosis that extends below the peritoneum is often referred to as deep endometriosis (DE). Endometriosis is usually multifocal and typically occurs in predictable locations in the pelvis. The diagnosis of endometriosis is challenging due to variable presenting symptoms and nonspecific physical examination findings [5]. Historically, the diagnosis of endometriosis was made by diagnostic laparoscopy with histologic inspection. Studies have shown that preoperative imaging is associated with decreased morbidity and mortality and reduces the need for repeat surgeries by reducing the number of incomplete surgeries. The literature now supports the use of imaging before surgery, because information gained from imaging studies helps inform patient decision making, is important for surgical planning, and impacts management [1,6,7]. Expert consensus groups advise using an MRI protocol tailored for detection of DE. Moderate bladder distention and vaginal contrast are recommended to help improve lesion conspicuity involving these structures [29]. There is aCleveland Clinic, Cleveland, Ohio. bPanel Chair, University of Michigan, Ann Arbor, Michigan. cClinica Family Health, Lafayette, Colorado; American Academy of Family Physicians. dUniversity of Kentucky, Lexington, Kentucky. eAlpert Medical School of Brown University, Providence, Rhode Island; Commission on Nuclear Medicine and Molecular Imaging. fHospital of the University of Pennsylvania, Philadelphia, Pennsylvania. gEmory University School of Medicine, Atlanta, Georgia; Committee on Emergency Radiology-GSER. hNew York University Langone Health, New York, New York. iVanderbilt University Medical Center, Nashville, Tennessee.

Endometriosis. Introduction/Background Endometriosis is a common condition impacting approximately 10% of individuals assigned female at birth [1]. The disorder is caused by endometrial-like tissue located outside of the endometrial cavity, associated with inflammation and fibrosis, on or extending below the peritoneal surface [2]. Endometriosis that extends below the peritoneum is often referred to as deep endometriosis (DE). Endometriosis is usually multifocal and typically occurs in predictable locations in the pelvis. The diagnosis of endometriosis is challenging due to variable presenting symptoms and nonspecific physical examination findings [5]. Historically, the diagnosis of endometriosis was made by diagnostic laparoscopy with histologic inspection. Studies have shown that preoperative imaging is associated with decreased morbidity and mortality and reduces the need for repeat surgeries by reducing the number of incomplete surgeries. The literature now supports the use of imaging before surgery, because information gained from imaging studies helps inform patient decision making, is important for surgical planning, and impacts management [1,6,7]. Expert consensus groups advise using an MRI protocol tailored for detection of DE. Moderate bladder distention and vaginal contrast are recommended to help improve lesion conspicuity involving these structures [29]. There is aCleveland Clinic, Cleveland, Ohio. bPanel Chair, University of Michigan, Ann Arbor, Michigan. cClinica Family Health, Lafayette, Colorado; American Academy of Family Physicians. dUniversity of Kentucky, Lexington, Kentucky. eAlpert Medical School of Brown University, Providence, Rhode Island; Commission on Nuclear Medicine and Molecular Imaging. fHospital of the University of Pennsylvania, Philadelphia, Pennsylvania. gEmory University School of Medicine, Atlanta, Georgia; Committee on Emergency Radiology-GSER. hNew York University Langone Health, New York, New York. iVanderbilt University Medical Center, Nashville, Tennessee.

3195150

acrac_3195150_1

Endometriosis

jUniversity of Michigan, Ann Arbor, Michigan. kMayo Clinic, Rochester, Minnesota. lSpecialty Chair, New York University Medical Center, New York, New York. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] Endometriosis less agreement surrounding maneuvers to help improve detection of bowel lesions, including bowel preparation, rectal contrast, fasting, and administration of anti-peristaltic agents [30-34]. Though fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG)-PET/CT has not been studied for the clinical variants described in this paper, a retrospective study showed that endometriosis can be detected on FDG-PET/CT [35]. The radiopharmaceutical fluoroestradiol, an estrogen analog PET agent currently approved for use in patients with metastatic breast cancer, has shown promise as an agent that can be used to detect endometriosis in early clinical trials [36]. OR Discussion of Procedures by Variant Variant 1: Adult. Clinically suspected pelvic endometriosis. Initial imaging. CT Pelvis With IV Contrast There is no relevant literature to support the use of pelvic CT with intravenous (IV) contrast as the initial imaging modality for clinically suspected endometriosis. CT Pelvis Without and With IV Contrast There is no relevant literature to support the use of pelvic CT without and with IV contrast as the initial imaging modality for clinically suspected endometriosis. CT Pelvis Without IV Contrast There is no relevant literature to support the use of pelvic CT without IV contrast as the initial imaging modality for clinically suspected endometriosis.

Endometriosis. jUniversity of Michigan, Ann Arbor, Michigan. kMayo Clinic, Rochester, Minnesota. lSpecialty Chair, New York University Medical Center, New York, New York. The American College of Radiology seeks and encourages collaboration with other organizations on the development of the ACR Appropriateness Criteria through representation of such organizations on expert panels. Participation on the expert panel does not necessarily imply endorsement of the final document by individual contributors or their respective organization. Reprint requests to: [email protected] Endometriosis less agreement surrounding maneuvers to help improve detection of bowel lesions, including bowel preparation, rectal contrast, fasting, and administration of anti-peristaltic agents [30-34]. Though fluorine-18-2-fluoro-2-deoxy-D-glucose (FDG)-PET/CT has not been studied for the clinical variants described in this paper, a retrospective study showed that endometriosis can be detected on FDG-PET/CT [35]. The radiopharmaceutical fluoroestradiol, an estrogen analog PET agent currently approved for use in patients with metastatic breast cancer, has shown promise as an agent that can be used to detect endometriosis in early clinical trials [36]. OR Discussion of Procedures by Variant Variant 1: Adult. Clinically suspected pelvic endometriosis. Initial imaging. CT Pelvis With IV Contrast There is no relevant literature to support the use of pelvic CT with intravenous (IV) contrast as the initial imaging modality for clinically suspected endometriosis. CT Pelvis Without and With IV Contrast There is no relevant literature to support the use of pelvic CT without and with IV contrast as the initial imaging modality for clinically suspected endometriosis. CT Pelvis Without IV Contrast There is no relevant literature to support the use of pelvic CT without IV contrast as the initial imaging modality for clinically suspected endometriosis.

3195150