BACKGROUND Continent urinary diversion (CUD) is performed on patients after total cystourethrectomy due to invasive bladder cancer or other diseases requiring urinary diversion. Since 1984, we have performed CUD using either the Kock pouch procedure or a modified Indiana pouch method. During the long-term follow-up, increasing numbers of late complications occurred, especially after the Kock pouch procedure. We summarized the frequency of postoperative complications, their management, and outcome of CUD using these 2 methods. METHODS Between 1984 and 1995, CUD using the Kock pouch method was performed on 102 patients, and on 61 patients using the Indiana pouch method, with a mean follow-up period of 83 and 50 months, respectively. The patients were between 24 and 82 years old, with a mean age of 59 years. In most patients, the Kock pouch was constructed according to the original method, except for the use of a Dacron fabric collar, and the Indiana pouch was made using the ileal patch method. RESULTS Of 95 patients followed up for more than 3 months after CUD using the Kock pouch method, efferent and afferent nipple valve malfunction occurred in 21 (22%) and 26 (27%) patients, respectively. Efferent complications occurred during the first year after surgery, whereas afferent complications occurred several years after surgery. Open repair surgery (excluding ileus) was necessary for 36 (38%) patients. Of 59 patients followed for more than 3 months after CUD using the Indiana pouch method, 7 patients with stomal stenosis, 2 patients with stomal prolapse, and 1 patient with stricture at the ureterocolonic anastomosis were treated surgically. An hourglass-like deformity of the pouch was seen in 2 patients, on whom the original Indiana pouch method was performed. As a whole, 13 patients (22%) were retreated with open surgery. Urinary calculi occurred in 42 (44%) patients after CUD using the Kock pouch method and in 11 patients (19%) using the Indiana pouch method. The outcome was excellent, good, fair, and poor in 31, 37, 24, and 8% of the patients after CUD using the Kock pouch method, and in 36, 17, 42, and 5% of the patients after CUD using the Indiana pouch method. CONCLUSIONS We performed CUD using either the Kock pouch or the Indiana pouch method with a success rate of approximately 90%. The rates of revision, reoperation and stone formation were higher with the Kock pouch procedure than with the Indiana pouch procedure, while varying degrees of incomplete incontinence and difficulty in catheterization were seen more often in the latter.