Sensitivity of Non-Contrast Computed Tomography for Small Renal Calculi with Endoscopy as the Gold Standard

Abstract

Objectives To compare the sensitivity of non-contrast CT to endoscopy for detection of renal calculi. Imaging modalities for detection of nephrolithiasis have centered on abdominal x-ray (KUB), ultrasound (US), and non-contrast computed tomography (CT). Sensitivities of 58-62% (KUB), 45% (US), and 95-100% (CT) have been previously reported. However, these results have never been correlated with endoscopic findings. Methods Idiopathic calcium oxalate stone formers with symptomatic calculi requiring ureteroscopy (URS) were studied. At the time of surgery, the number and location of all calculi within the kidney were recorded followed by basket retrieval. Each calculus was measured and sent for micro CT and infrared spectrophotometry. All CT scans were reviewed by the same genitourinary radiologist who was blinded to the endoscopic findings. The radiologist reported on the number, location, and size of each calculus. Results 18 renal units were studied in 11 patients. Average time from CT scan to URS was 28.6 days. The mean number of calculi identified per kidney was 9.2±6.1 for endoscopy and 5.9±4.1 for CT (p<0.004). The mean size of total renal calculi (sum of longest stone diameters) per kidney was 22.4±17.1 mm and 18.2±13.2 mm for endoscopy and CT, respectively (p=0.06). Conclusions CT scan underreports the number of renal calculi, probably missing some small stones and unable to distinguish those lying in close proximity to one another. However, the total stone burden seen by CT is, on average, accurate when compared to that found on endoscopic examination.