Sonographic diagnosis of hydronephrosis in patients with spinal cord injury: influence of bladder fullness.

Objective: To investigate and compare the diagnostic accuracy of prevoid and postvoid renal sonography in detecting hydronephrosis in patients with spinal cord injury.

Study Design: A prospective, blind comparison of renal sonography and excretory urography in 67 spinal cord injury patients who underwent periodic urologic examinations. Renal sonography was performed twice, once when the patient's bladder was physiologically full (prevoid) and again when it was just emptied (postvoid).

Results: Of 140 kidneys, 24 from 16 patients were found to have hydronephrosis by excretory urography; 116 kidneys had normal urogram findings. Prevoid sonography missed the diagnosis of hydronephrosis in one kidney and showed hydronephrosis in 18 kidneys that had normal results on excretory urography (sensitivity, 95.8%; specificity, 84.5%; negative predictive value, 99.0%). The postvoid sonograms did not detect hydronephrosis in four kidneys and showed hydronephrosis in six kidneys that had normal results on excretory urography (sensitivity, 83.3%; specificity, 94.8%; negative predictive value, 96.5%). Compared to excretory urography, renal sonography detected eight more upper urinary tract abnormalities, which were confirmed by cystograms or radioisotopic renograms.

Conclusion: Performing renal sonography while the bladder is full can increase the sensitivity in detecting hydronephrosis in asymptomatic spinal cord injured patients.