ObjectiveTo assess the impact that examiner experience has on the reproducibility and accuracy of computed tomography (CT) enterography in the detection of radiological signs in patients with Crohn's disease.Materials and MethodsThis was a retrospective, cross-sectional observational study involving the analysis of CT enterography scans of 20 patients with Crohn's disease. The exams were analyzed independently by two radiologists in their last year of residence (duo I) and by two abdominal imaging specialists (duo II). The interobserver agreement of each pair of examiners in identifying the main radiological signs was calculated with the kappa test. The accuracy of the examiners with less experience was quantified by using the consensus among three experienced examiners as a reference.ResultsDuo I and duo II obtained a similar interobserver agreement, with a moderate to good correlation, for mural hyperenhancement, parietal thickening, mural stratification, fat densification, and comb sign (kappa: 0.45-0.64). The less experienced examiners showed an accuracy > 80% for all signs, except for lymph nodes and fistula, for which it ranged from 60% to 75%.ConclusionLess experienced examiners have a tendency to present a level of interobserver agreement similar to that of experienced examiners in evaluating Crohn's disease through CT enterography, as well as showing satisfactory accuracy in identifying most radiological signs of the disease.

Background The most widely accepted computed tomography (CT) protocol for diagnosis of acute pyelonephritis (APN) includes at least a pre- and post-contrast scan, which may expose patients to higher doses of ionizing radiation. Purpose To establish the accuracy, reproducibility, and degree of confidence in CT diagnosis of acute pyelonephritis (APN) and urolithiasis using only images obtained during the nephrographic phase. Material and Methods A retrospective study of 100 consecutive patients (88 women; age range, 19-70 years) with clinical and laboratory suspicion of APN and who underwent triphasic abdominal CT scans (non-contrast, nephrographic, and excretory phases) was performed. Two readers first evaluated independently only the nephrographic phase of scans, and, in a second session, the entire study. The diagnostic reference standard was settled by a third experienced radiologist who reviewed all triphasic scans and clinical data. Results The accuracy of only nephrographic phase for diagnosis of APN and urolithiasis was in the range of 90.3-91.78% and 96.27-99.25%, respectively. There was no significant difference in comparison with the triphasic reading (z: -0.4 - 0.2; P = 0.34-0.83). The average degree of confidence for APN also showed no significant variation for both readers ( P = 0.4 and 0.08). Almost perfect inter-observer agreements for the diagnosis of APN (k = 0.86, P < 0.001) and for urolithiasis (k = 0.84, P < 0.001) were observed when considering only the nephrographic phase. Conclusion CT assessment of APN and urolithiasis can be accurately performed using only the late nephrographic phase, with consequent dose reduction.