Evaluating and grading cystographic leakage: correlation with clinical outcomes in patients undergoing robotic prostatectomy

To classify cystographically detected urinary leaks in patients undergoing computer-assisted (robotic) radical prostatectomy (RP) and to evaluate its effect on postoperative outcomes. Between October 2001 and October 2007, 3327 patients had a RP using a technique described previously. The data were entered prospectively into an approved database. Before catheter removal, all patients had a gravity cystogram taken 7 days after RP. All patients who had a detectable urinary leak on cystography were stratified into three groups by two independent radiologists using a previously described grading system. Patients were evaluated with a validated International Prostate Symptom Score at 3-, 6-, 9- and 12-month intervals after RP. The continence status was determined based on a patient-reported questionnaire. Medical records in these patients were reviewed for the presence of complications requiring secondary interventions. In all, 287 patients (8.6%) had a detectable leak on cystography, of which 179 (62.4%), 84 (29.3%) and 24 (8.4%) were grades I, II and III, respectively. Of the patients with a detectable leak 70% were continent within 3 months and 94% had no involuntary urinary leakage at 1 year. Eight of 287 (2.8%) patients required a secondary intervention to correct bladder neck contracture. All eight of these patients had a grade II or III leak on cystography. The presence of a urinary leak might delay the time to continence, but has no adverse effect on long-term urinary control. Quantifying the gradation of leakage according to the described classification might provide the clinician with prognostic information about patients at risk for future interventions.