Low-dose unenhanced multidetector CT of patients with suspected renal colic

OBJECTIVE. This study is designed to assess the intraobserver and interobserver agreements and the diagnostic performances of low-dose unenhanced multidetector CT (MDCT) in patients with suspected renal colic. SUBJECTS AND METHODS. The study included 106 patients who underwent unenhanced MDCT with 4 x 2.5 mm collimation, 120 kVp, 30 mAs, and, if necessary, additional focused acquisitions at 60 or 120 mAs on areas with an equivocal ureteral stone or with significant image noise. The effective radiation dose was computer-simulated with software based on the Monte Carlo model and International Commission on Radiological Protection recommendations. CT scans were archived and independently reviewed by three radiologists during two interpretation sessions on a workstation with three dimensions functions. Intraobserver and interobserver agreements were calculated with the kappa statistics. Accuracy for detection of ureteral stone on low-dose MDCT was calculated by comparison with combined clinical (stone passage), surgical (stone retrieval, extracorporeal shock wave lithottipsy), biologic (urinalysis, urine culture), and other imaging (excretory urography, standard-dose MDCT, follow-up sonography, and abdominal radiography) findings or by evidence for an alternative diagnosis. RESULTS. Ureteral stones were present in 38 (36%) of 106 patients. Thirty-six of 38 ureteral stones were detected by low-dose MDCT. From reviewer to reviewer, the number of true-positive, false-positive, true-negative, and false-negative findings ranged, respectively, from 34 to 36, 1 to 4, 64 to 68, and 2 to 4. The corresponding sensitivity, specificity, and accuracy ranged from 89.5% to 94.7%, from 94.1% to 100%, and from 93.4% to 98.1%, respectively. The intraobserver and interobserver agreements were excellent, with kappa values ranging from 0.87 to 0.98. In 13 patients, an alternative diagnosis explaining the patient's symptoms was proposed by all reviewers using images obtained at 30 mAs. No additional or alternative diagnosis was found at standard dose. At 30 mAs, the mean effective dose was 1.2 mSv in men and 1.9 mSv in women. Additional acquisitions at 60 mAs, all focused on the lower pelvis, were acquired in 20 patients, but the corresponding images were needed by the reviewers for only six of them. The acquisitions at 60 mAs were responsible for an additional mean effective dose of 0.5 in men and 0.8 mSv in women. CONCLUSION. Our study shows that low-dose unenhanced MDCT is appropriate for the diagnosis of ureteral stones, and that it provides excellent intraobserver and interobserver agreements and does not obscure alternative diagnoses.