Faculty Bibliography

PURPOSE: A 2008 federal report expressed concern regarding substantial regional variation in imaging expenditures. The aims of this study were to evaluate trends in regional variation in Medicare imaging utilization and expenditures from 2007 to 2011 and to compare these trends with regional variation in other health service categories. METHODS: Data were based on CMS's Chronic Condition Data Warehouse and organized on the basis of 306 US health referral regions (HRRs). Imaging costs per beneficiary, standardized for regional differences in reimbursement rates, and imaging utilization per beneficiary were recorded per HRR from 2007 through 2011. Costs and utilization were also recorded for other service categories in 2011. Regional variation was assessed via relative risk (RR; the ratio between the highest and lowest HRRs) and coefficient of variation (CV; the standard deviation divided by the mean among all HRRs). Correlations between imaging and other service categories were assessed using Pearson's correlation coefficient. RESULTS: There was minimal change in regional variation in imaging costs or utilization between 2007 and 2011. Regional variation in imaging costs (RR, 5.70-5.88; CV, 33.0%-33.3%) was considerably greater than variation in imaging utilization (RR, 2.11%-2.25%; CV, 14.2%-14.6%). Imaging costs and utilization showed moderate to strong correlations with those of other service categories (r = 0.572-0.869). In 2011, regional variation in imaging utilization (RR, 2.25; CV, 14.2%) was considerably lower than variation in utilization of other service categories (RR, 2.80-10.78; CV, 20.9%-33.3%). CONCLUSIONS: Regional variation in imaging utilization is considerably lower than both variation in imaging costs and variation in utilization of other major service categories. It is unclear whether variation in imaging utilization provides an optimal individual target for major policy decisions.

Compared with earlier prostate MRI techniques that suffered from relatively poor sensitivity and specificity for detecting prostate cancer because of their reliance predominantly on morphology, multiparametric prostate MRI (mpMRI) in 2013 provides a wealth of functional information that has made possible vastly improved detection and characterization of prostate cancer. Our aims in this article are to describe the various imaging sequences that comprise the mpMRI exam, as well as to review current literature on the strengths/weaknesses of these sequences; to delineate strategies for standardizing interpretation and reporting of MRI results; and finally, to expound on the role that prostate MRI currently does and potentially can play in clinical practice.

PURPOSE: The purpose of this study was to compare the diagnostic performance of diffusion-weighted imaging (DWI) and dynamic contrast-enhanced (DCE) imaging for prostate cancer (PCa) detection by performing a meta-analysis of studies evaluating these techniques within the same patient cohort. METHODS: Evidence-based online databases were searched for studies reporting the performance of DWI and DCE in PCa detection in the same patient cohorts using histopathology as reference standard and providing sufficient data to construct 2 x 2 contingency tables. Pooled estimates of diagnostic performance were computed across included studies. RESULTS: Of 80 initial studies identified, 5 studies (total of 265 patients and 1730 prostatic regions) met criteria for inclusion in the meta-analysis. Pooled sensitivity was 58.4% (95% confidence interval [CI], 53.5%-63.1%) for DWI and 55.3% (95% CI, 50.4%-60.1%) for DCE. Pooled specificity was 89.0% (95% CI, 87.2%-0.7%) for DWI and 87.9% (95% CI, 86.0%-89.6%) for DCE. At summary receiver-operating-characteristic analysis, area-under-the-curve was 0.810 (0.059) for DWI and 0.786 (0.079) for DCE. Heterogeneity across studies was high for sensitivity and specificity [inconsistency index (I), >90%], although heterogeneity of specificity was substantially improved after excluding an outlier study in terms of diagnostic threshold (I = 0.0%-68.8%). Relative performance of DWI and DCE remained similar after this exclusion CONCLUSIONS: There was a paucity of studies comparing DWI and DCE in the same patient cohorts, and heterogeneity among these studies was substantial. Nevertheless, performance of DWI and DCE was similar across identified studies, with both techniques showing substantially better specificity than sensitivity. Larger studies with uniform methodology are warranted to further understand relative merits of the 2 techniques.

PURPOSE: The purpose of this study was to compare computed tomography (CT) and magnetic resonance imaging (MRI) in terms of likelihood of providing a definitive diagnosis (DD) and a recommendation for additional imaging (RAI), when performed to evaluate indeterminate liver and renal lesions detected on ultrasound as well as in terms of impact on imaging costs. METHODS: This retrospective study was Health Insurance Portability an Accountability Act (HIPAA)-compliant and institutional review board-approved, with waiver of informed consent. We identified consecutive indeterminate liver and renal lesions detected on ultrasound that underwent contrast-enhanced CT or MRI for further characterization. Reports from follow-up studies were reviewed for whether the impression provided DD and RAI. Frequency of DD and RAI was compared between CT and MRI using the Fisher exact test. On the basis of the observed frequency of DD, anticipated imaging costs were compared in a hypothetical sample of 100 patients with indeterminate lesions between first obtaining multiphase CT for all lesions and a subsequent MRI for those lesions indeterminate on CT versus directly obtaining a multiphase MRI for all lesions. RESULTS: A total of 143 renal lesions were included, of which 77 and 66 underwent CT and MRI, respectively. Magnetic resonance imaging was significantly more likely than CT to provide DD (95.5% vs 77.9%; P = 0.003) and significantly less likely to provide RAI (1.5% vs 10.4%; P = 0.038). A total of 221 liver lesions were included, of which 76 and 145 underwent CT and MRI, respectively. Magnetic resonance imaging was significantly more likely than CT to provide DD (95.2% vs 71.1%; P < 0.001) and significantly less likely to provide RAI (0% vs 10.5%; P < 0.001). Across the entire study cohort, there were 13 instances of MRI recommended after an indeterminate CT and 1 case of CT recommended after an indeterminate MRI. A DD was provided in 8 of 9 instances in which MRI was performed after an indeterminate CT. However, anticipated imaging costs were higher when directly obtaining MRI for all indeterminate lesions, compared with initially obtaining multiphase CT, for both kidney ($64,739 vs $49,759) and liver ($64,739 vs. $56,975) lesions, respectively. CONCLUSIONS: For indeterminate liver and renal lesions detected on ultrasound, MRI is more likely to provide DD and less likely to provide RAI in comparison with CT, although these differences did not result in lower anticipated imaging costs.