Quality and value of subspecialty reinterpretation of outside thoracic CT scans of patients referred to a tertiary cancer center.

Abstract

265 Background: At tertiary cancer centers, physicians frequently request reinterpretation of imaging studies performed at outside institutions. The purposes of this study were to determine the quality of outside computed tomography (CT) scans of the chest and compare the accuracy of accompanying radiology reports from outside institutions and our multidisciplinary cancer center. Methods: Two thoracic radiologists graded the quality of 59 outside chest CT scans and generated independent reports for 52 of the scans. A third thoracic radiologist scored the outside reports and reinterpretations for quality. Fisher’s exact tests were used to compare the frequency with which crucial items appeared in outside reports and reinterpretations. Next, two outside thoracic radiologists identified discrepancies between outside reports and reinterpretations (first radiologist) and determined whether the outside report or reinterpretation was more accurate in each case (second radiologist). Finally, the impact of discrepancies on management was evaluated, largely based on NCCN guidelines. Results: Of the 59 outside CT scans, 35 (59%) were of poor quality. Reinterpretations were more likely than outside reports to include information about lymph nodes, adrenal and liver metastasis, tumor nodules, and tumor texture. In 19 of 52 cases (37%), discrepancies were identified between outside reports and reinterpretations. In 17 of these cases, the reinterpretation was superior; in 2 cases, the reinterpretation and outside report were of equal quality. Among these 17 cases, reinterpretation allowed staging in nine cases that could not be staged with information from the outside reports; resulted in upstaging without management change in one case and upstaging with management change in four cases among the five cases with staging information present in both sets of reports; and revealed a significant omission (2 cases) or error (1 case) that changed management in three cases. In total, reinterpretation resulted in significant changes to 16 of 52 (31%) of CT scans. Conclusions: Subspecialty reinterpretation of chest CT scans can substantially improve clinical management.