Abdominal and pelvic computed tomography (CT) interpretation: discrepancy rates among experienced radiologists

To assess the discrepancy rate for the interpretation of abdominal and pelvic computed tomography (CT) examinations among experienced radiologists. Ninety abdominal and pelvic CT examinations reported by three experienced radiologists who specialize in abdominal imaging were randomly selected from the radiological database. The same radiologists, blinded to previous interpretation, were asked to re-interpret 60 examinations: 30 of their previous interpretations and 30 interpreted by others. All reports were assessed for the degree of discrepancy between initial and repeat interpretations according to a three-level scoring system: no discrepancy, minor, or major discrepancy. Inter- and intrareader discrepancy rates and causes were evaluated. CT examinations included in the investigation were performed on 90 patients (43 men, mean age 59 years, SD 14, range 19-88) for the following indications: follow-up/evaluation of malignancy (69/90, 77%), pancreatitis (5/90, 6%), urinary tract stone (4/90, 4%) or other (12/90, 13%). Interobserver and intraobserver major discrepancy rates were 26 and 32%, respectively. Major discrepancies were due to missed findings, different opinions regarding interval change of clinically significant findings, and the presence of recommendation. Major discrepancy of between 26 and 32% was observed in the interpretation of abdominal and pelvic CT examinations.