Faculty Bibliography

PURPOSE/OBJECTIVE:To explore variation in downstream relative costs associated with ovarian cysts incidentally detected on ultrasound. METHODS:For 200 consecutive incidental ovarian cysts on ultrasound, ultrasound reports were classified in terms of presence of a radiologist recommendation for additional imaging. All downstream events (imaging, office visits, and surgery) associated with the cysts were identified from the electronic health record. Medical costs associated with these downstream events were estimated using national Medicare rates. Average cost per cyst was stratified by various factors; cost ratios were computed among subgroups. RESULTS:Average costs per cyst were 1.9 times greater in postmenopausal than premenopausal women. Relative to when follow-up imaging was neither recommended nor obtained, costs were 1.1 times greater when follow-up imaging was recommended but not obtained, 5.1 times greater when follow-up imaging was both recommended and obtained, and 8.1 times greater when follow-up imaging was obtained despite not being recommended. Costs were 2.5 times greater when the radiologist underrecommended follow-up compared with Society of Radiologists in Ultrasound (SRU) guidelines for management of ovarian cysts, 3.0 times greater when the ordering physician overmanaged compared with the radiologist's recommendation, as well as 1.7 times and 3.8 times greater when the ordering physician undermanaged and overmanaged compared with SRU guidelines, respectively. Four ovarian neoplasms, although no ovarian malignancy, were diagnosed in the cohort. CONCLUSION/CONCLUSIONS:Follow-up costs for incidental ovarian cysts are highly variable based on a range of factors. Radiologist recommendations may contribute to lower costs among patients receiving follow-up imaging. Such recommendations should reflect best practices and support the follow-up that will be of likely greatest value for patient care.

PURPOSE/OBJECTIVE:To assess changing utilization of pre-biopsy prostate MRI in Medicare beneficiaries. MATERIALS AND METHODS/METHODS:Men undergoing prostate biopsy were identified in 5% Medicare Research Identifiable Files from October 2010 through September 2015. Rates of pre-biopsy prostate MRI (any pelvic MRI ≤6 months before biopsy with a prostate indication diagnosis code) were assessed. Temporal changes and variation by geography and among populations were determined. RESULTS:In male Medicare beneficiaries, pre-biopsy MRI utilization rates increased from 0.1% in 2010 to 0.7% in 2011, 1.2% in 2012, 2.9% in 2013, 4.7% in 2014, and 10.3% in 2015. In 2015, pre-biopsy prostate MRI rates varied significantly by patient age (5.7% for >80 years vs. 8.4%-9.3% for other age ranges, p=0.040), race (5.8% in blacks vs. 10.1% in whites, p=0.009), and geographic region (from 6.3% in the Midwest to 12.5% in the Northeast; p<0.001). Rates were highest in Wyoming (25.0%), New York (23.7%), and Minnesota (20.5%), but <1% in ten states. CONCLUSION/CONCLUSIONS:Historic Medicare claims provide novel insights into dramatically increasing adoption of MRI prior to prostate biopsy. Following earlier minimal usage, utilization increased sharply beginning in 2013, exceeding 10% in 2015. However, substantial racial and geographic variation in adoption exists. Continued educational, research, and policy efforts are warranted to optimize the role of pre-biopsy MRI and minimize sociodemographic and geographic disparities.

PURPOSE: To evaluate radiologist preferences regarding specific content that warrants inclusion in prostate MRI reports. METHODS: Sixty-one members of the Society of Abdominal Radiology responded to a 74-item survey regarding specific content warranted in prostate MRI reports, conducted in August 2016. RESULTS: General items deemed essential report content by >/= 50% of respondents were prostate volume (80%), extent of prostate hemorrhage (74%), TURP defects (69%), coil type (64%), BPH (61%), contrast dose (61%), contrast agent (59%), medications administered (59%), and magnet strength (54%). Details regarding lesion description deemed essential by >/= 50% were overall PI-RADS category (88%), DCE (+/-) (82%), subjective degree of diffusion restriction (72%), T2WI intensity (72%), T2WI margins (65%), T2WI shape (52%), DWI 1-5 score (50%), and T2WI 1-5 score (50%). Details deemed essential to include in the report Impression by >/= 50% of respondents were lymphadenopathy and metastases (100%), EPE (98%), SVI (98%), neurovascular bundle involvement (93%), index lesion location (93%), PI-RADS category of index lesion (82%), number of suspicious lesions (78%), significance of index lesion PI-RADS category (53%), and PI-RADS category of non-index lesions (52%). Preferred methods for lesion localization were slice/image number (68%), 3-part craniocaudal level (68%), zonal location (65%), anterior vs. posterior location (57%), and medial vs. lateral position (56%). Least preferred methods for localization were numeric sector from the PI-RADS sector map (8%), annotated screen capture (10%), and graphical schematic of PI-RADS sector map (11%). CONCLUSION: Radiologists generally deemed a high level of detail warranted in prostate MRI reports. The PI-RADS v2 sector map was disliked for lesion localization.

Imaging is critically important for the diagnosis, staging, and management of men with high-risk prostate cancer. Conventional imaging modalities, including computed tomography and radionuclide bone scan have been employed for local and metastatic staging, but their performance has generally been poor. Sodium fluoride positron emission tomography is recommended when there is high suspicion for bone metastases despite a negative or indeterminate bone scan. Magnetic resonance imaging has advantages in local staging but its value depends on the extent of disease. Whole body positron emission tomography/magnetic resonance imaging could provide both local and distant staging although the technology is not yet widely disseminated. None of the existing positron emission tomography agents are recommended in practice guidelines, however, among them, prostate specific membrane antigen-based tracers seem to hold the most promise based on sensitivity and specificity.

PURPOSE/OBJECTIVE:The aim of this study was to assess radiologists' performance on Medicare quality measures and identify physician characteristics potentially influencing such scores. METHODS:Medicare quality scores reported by US radiologists in 2015 were obtained from CMS. Associations were explored with publicly available physician characteristic data. RESULTS:Overall, 15,045 radiologists reported 40,427 Medicare quality scores encompassing 25 claims measures, 18 registry measures, and 2 qualified clinical data registry (QCDR) measures. Claims measures included reporting fluoroscopic times (n = 10,152; mean score, 80.3 ± 27.6), carotid ultrasound stenosis (n = 8,940; mean score, 86.8 ± 20.6), inappropriate mammography use of "probably benign" (n = 8,083; mean score, 0.4 ± 3.3), mammography reminders (n = 7,229; mean score, 86.6 ± 29.0), bone scintigraphy correlation (n = 2,712; mean score, 76.0 ± 27.0), and line-related infection prevention (n = 2,226; mean score, 83.3 ± 27.4). Registry measures were reported by ≤17 radiologists. The two QCDR measures were dose index registry participation (n = 246; mean score, 99.5 ± 1.4) and mammography recall rate (n = 77; mean score, 9.0 ± 5.6). Higher scores were observed for radiologists in larger practices (strongest independent predictor), in subspecialized practices, in academic practices, in the South and West, and with fewer years in practice. The fluoroscopic exposure times measure had the best performance scores by musculoskeletal and interventional radiologists, carotid Doppler measure by abdominal radiologists, mammography measures by breast radiologists, bone scintigraphy measure by musculoskeletal and nuclear medicine radiologists, and line infection measure by interventionalists. The dose registry participation QCDR measure had near perfect performance across generalists and subspecialists. CONCLUSIONS:Current Medicare performance metrics favor radiologists in larger practices and subspecialized radiologists, possibly reflecting support infrastructures and the narrow focus of most metrics, respectively. These findings may assist targeted data-driven reporting by radiologists and guide efforts to refine existing and develop new metrics.

OBJECTIVE:The purpose of this study is to assess the utility of texture analysis of multiple MRI sequences for the differentiation of uterine leiomyomas and leiomyosarcomas. MATERIALS AND METHODS/METHODS:Seventeen leiomyosarcomas and 51 leiomyomas undergoing MRI before resection were included. Whole-lesion volumes of interest were placed on T2-weighted images, contrast-enhanced T1-weighted images, and apparent diffusion coefficient (ADC) maps. The diagnostic performance of histogram metrics was assessed. RESULTS:achieved sensitivity of 82.4% and specificity of 74.5%. CONCLUSION/CONCLUSIONS:For whole-lesion histogram metrics obtained on various MRI sequences, T2-weighted images provided the highest, and ADC maps the lowest, performance for differentiating uterine leiomyomas and leiomyosarcomas. Metrics reflecting percentiles from the bottom half of the histogram distribution outperformed the standard mean. Models combining the T2-weighted imaging whole-lesion metrics and patient age achieved particularly high diagnostic performance. Although these findings require validation in larger studies, they have implications for facilitating improved treatment selection for these two entities.

PURPOSE/OBJECTIVE:To study opioid prescribing behavior of US interventional radiologists (IRs). METHODS:Using Medicare Physician and Other Supplier Public Use File claims, we identified 2,133 radiologists whose practice in 2015 comprised predominantly interventional radiology. Cross-linking the Medicare Part D Prescriber File, their opioid prescription writing behavior was characterized. RESULTS:Most (52.2%) IRs wrote 10 or fewer prescriptions total for Medicare beneficiaries. Of the 47.8% who wrote >10 prescriptions, 87.4% prescribed an opioid, most commonly hydrocodone with acetaminophen, at least once (71.3%, 1-10 opioid prescriptions; 27.4%, 11-100; 1.3%, ≥101). Overall, 23.0% of all prescriptions by those IRs writing >10 were for opioids, with an average 8.0-day prescription. Average opioid prescriptions per IR were significantly (P ≤ .015) independently associated with their providing clinical evaluation and management (E&M) services (9.7 opioid prescriptions per IR with demonstrable E&M encounters versus 2.2 if not), practice size (12.6 for practices with ≤ 10 members versus 3.7-4.8 for larger groups), and geography (8.3 in the South versus 3.6-4.0 elsewhere). Rates were highest in Georgia (39.5) and lowest in Delaware (2.0). Higher opioid prescribing rates showed additional univariable associations with more years in practice and nonacademic practices. CONCLUSION/CONCLUSIONS:Most IRs write few, if any, prescriptions for Medicare beneficiaries. Of those who do, the large majority writes for opioids, at rates higher than national physician benchmarks. IRs' opioid prescribing varies significantly based on physician and practice characteristics and particularly whether the IR provides clinical E&M services. In light of the nation's opioid epidemic, these observations may guide education, practice improvement, and policy efforts to optimize opioid prescribing.