- Email: [email protected]
Continent ileocolonic urinary reservoir (Miami pouch): the University of Miami experience over 15 years
American Journal of Obstetrics and Gynecology (2004) 190, 994e1003
www.elsevier.com/locate/ajog
Continent ileocolonic urinary reservoir (Miami pouch): The University of Miami experience over 15 years Emery M. Salom, MD,* Luis E. Mendez, MD, Dana Schey, MD, Nicholas Lambrou, MD, Noor Kassira, MS, Orlando Go´mez-Marı´n, MSc, PhD, Hervy Averette, MD, Manuel Pen˜alver, MD University of Miami, School of Medicine, Jackson Memorial Hospital /Sylvester Comprehensive Cancer Center, Division of Gynecologic Oncology, Miami, Fla
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS Continent ileocolonic urinary reservoir Miami pouch Pelvic exenteration Urinary diversion Reconstruction
Objective: A patient with a recurrent central pelvic malignancy after radiation will require urinary diversion as part of the reconstructive phase of the pelvic exenteration. The aim of our study was to assess the result of our 15-year experience with a continent ileocolonic urinary reservoir, which is known as the Miami pouch. Study design: Since 1988, all patients who received a continent ileocolonic urinary reservoir in the Division of Gynecologic Oncology, University of Miami School of Medicine, were included in the study. Parameters that were evaluated during the study period include functional outcomes, early and late perioperative complications, and their treatment. Results: A total of 90 patients were identified from February 1988 to December 2002. Seventyeight patients (87%) had a recurrent central pelvic malignancy, and 82 patients (91%) received radiation before the Miami pouch procedure. The nonereservoir-related morbidities were fever (76%), wound complication (30%), pelvic collection (12%), ileus/small bowel obstruction (12%), and postoperative death (11%). The most common reservoir-related complications were urinary infection (40%), ureteral stricture (20%), and difficulty with self-catheterization (18%). In our study, the overall complication rate that was related directly to the Miami pouch was 53%. Conservative treatment resolvedO80% of these cases. The rate of urinary continence that was achieved in our patients was 93% during our 15-year experience with the Miami pouch. Conclusion: The Miami pouch is a good alternative for continent urinary diversion during exenteration or radiation-induced damage. The rate of major complications that require aggressive surgical intervention is acceptable. Most postoperative complications (80%) can be corrected with the use of conservative techniques that are associated with fewer deaths than reoperation and thus should be used first. The technique is simple and effective in women who are at high risk, who have undergone previous radiation therapy, and who have a high rate of functional success and is a profound advantage for a woman’s psychosocial well-being. Ó 2004 Elsevier Inc. All rights reserved.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Presented at the Twenty-Second Annual Meeting of the American Gynecological and Obstetrical Society, Napa, California, September 18-20, 2003. * Reprint requests: Emery M. Salom, MD, Division of Gynecologic Oncology, University of Miami School of Medicine, Jackson 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2004.01.023
Memorial Hospital/ Sylvester Cancer Center, 1611 NW 12th Ave, Holtz Building East Tower Room 3003, Miami, FL 33136. E-mail: [email protected]
995
Salom et al Pelvic exenteration is the most radical and destructive elective procedure described in pelvic surgery. For many women with central pelvic recurrences after radiotherapy, an exenteration is the only procedure that can be offered with curative intent. As a consequence, the reconstruction of the pelvic floor, gastrointestinal tract, sexual organs, and urinary system remains a great challenge for many pelvic surgeons. Much advancement has been made over the past 150 years in this arena, especially with respect to urinary diversion. The urinary diversion technique ureterosigmoidostomy, described by Lloyd1 in the 1850s, remained relatively unchanged until Bricker’s2 publication in 1950. Bricker popularized the use of the incontinent ileal conduit that has remained the most popular and extensively studied form of urinary diversion. The long-term outcome studies have revealed that the postoperative complication rates that are associated with the ileal conduits range between 25% and 49%.3 In an attempt to avoid such long-term complications, Gilchrist et al4 were the first to describe the use of the ileocecal valve for a continent urinary diversion. Many techniques of continent urinary diversion have been developed that used the terminal ileum and the ascending, transverse, or descending colon, which include the Miami pouch,5 Mainz pouch,6 Indiana pouch,7 and Florida pouch.8 Many of these various techniques came about in an effort to decrease the high initial failure rate of 10% to 20%, urine leakage rate of 37%, and reoperation rate of 20% with the original continent ileal reservoir, the Kock pouch. This pouch uses the principle of an intussuscepted nipple valve system for continence.9,10 Because of all the previously described complications, a number of different alternatives have been explored, which includes the Miami pouch. In 1987, Bejany and Politano5 described the use of a low-pressure detubularized colonic reservoir with a tapered ileum and pursestring suture around the ileocecal valve as its continent mechanism that is simple and effective. In 1988, Penalver et al11 were the first to describe the use of the continent ileocolonic urinary reservoir in recurrent gynecologic malignancies. Since that time the Division of Gynecologic Oncology at the University of Miami School of Medicine has been using the Miami pouch as the preferred method of urinary diversion. We previously published our experience and the number of evolutionary changes of the technique and trends towards conservative treatment of complications in hopes of improving the functional outcome and decreasing the postoperative morbidity that are associated with the Miami pouch.11-15 We report the largest experience with continent ileocolonic reservoirs in patients who have undergone radiation therapy in the literature to date. We review our experience over the last 15 years that involved the early and long-term postoperative complications, conserva-
tive and surgical treatment, and overall continence of the Miami pouch.
Material and methods We evaluated all patients who underwent an ileocolonic continent urinary reservoir (Miami pouch) from February 1988 to December 2002 at the University of Miami, Division of Gynecologic Oncology. These patients have been followed prospectively for the recognition and treatment of complications and functionality of the procedure. All details as to the patient’s progress are recorded in the Division’s Miami Database System with the use of the FileMaker Pro (File Maker, Inc, Santa Clara, Calif) software. We reviewed the last 15 years since the creation and evolution of the Miami pouch as a continent urinary reservoir for patients with gynecologic malignancies and its complications. The operative technique used for the creation of the Miami pouch is identical to those described previously at our institution in 1988.11 The reservoir consists of the distal 12 cm of the ileum and the ascending and transverse colon transected distal to the middle colonic artery. The isolated intestinal segment that averages 30 cm in length is incised along the tenia and thus detubulerized. The opened ascending and transverse portions of the colon are anastomosed in a U-shaped fashion to create a low-pressure system. The distal ileum is catheterized with a 14F silicone rubber Foley catheter and tapered to this diameter with a gastrointestinal automatic stapler. The ileocecal valve is then reinforced with 3 20 silk sutures placed 0.5 cm apart. As a result, the ileum provides a high-pressure outlet to maintain continence. The ureters are spatulated and anastomosed to the transverse colon, which is the nonirradiated portion, and are stented. During the early evolution of the Miami pouch, the ureters were anastomosed with the use of a submucosal tunnel, which has been abandoned subsequently in an attempt to prevent ureteral stricture. The pouch is anchored to the anterior abdominal wall fascia with 4 to 5 permanent sutures. The ileal portion is brought out to the anterior abdominal wall. A Jackson Pratt drain is placed posterior to the pouch for the collection of any extravasated urine, and the Foley catheter is left in the pouch for 14 days and flushed with 50 mL of saline solution every 4 hours. The nephroureteral stents are removed on days 14 through 21 along with the Foley catheter after a contrast study of the urinary reservoir and the ureters is performed to assure the absence of any urinary leakage, reflux, or upper tract obstruction. The data for all patients who underwent the creation of the continent ileocolonic urinary reservoir were analyzed. The information was abstracted from the computerized database and medical records with respect to age, body mass index, primary disease site, indications for
996 Table I
Salom et al Indications for surgical procedure
Procedure
Patients (n)
Recurrent cervical cancer Recurrent endometrial cancer Uterine leiomyosarcoma Recurrent ovarian cancer Recurrent vaginal cancer Recurrent vulvar cancer Primary vulvar cancer Primary vulvar/vaginal melanoma Vesical vaginal fistula Bladder atony Urethral cancer Mons pubis perineural sarcoma Rectal sarcoma Total
68 2 2 1 2 1 1 2 5 3 1 1 1 90
urinary diversion, history of radiation therapy, and types of procedures performed in conjunction with the Miami Pouch. We included all early postoperative complications and long-term complications with the respective treatment strategies. Early postoperative complications were defined as all those complications that occurred !60 days after the day of surgery; long-term complications were defined as those complications that occurredO60 days after the surgery date. The functional outcomes that were specific to the urinary reservoir were evaluated and included difficulty with catheterization, urinary infection, pyelonephritis, early reservoir leak, ureteral obstruction, stomal stricture, fistula formation, incontinence, and renal function. The data were evaluated routinely in the clinic 2 weeks after discharge and every 3 months thereafter. The patients were followed for the development of any signs of tumor recurrence and postoperative complications. Follow-up and survival were measured from the day of surgery to the date of last contact or death. Serum electrolytes, blood urea nitrogen level, creatinine level, and urine cultures are evaluated routinely with each clinic visit. Renal function was analyzed by the determination of blood urea nitrogen and creatinine levels, which were defined as the normal values of 7 to 22 mg/ dL and 0.8 to 1.5 mg/dL, respectively. In 1993, we reported the result of multichannel urodynamics in the first 16 patients, which confirmed the low-pressure reservoir and the high-pressure ileal outlet.2 We no longer obtain routine urodynamics on our patients. Statistical analysis consisted of chi-squared analysis and Fisher exact test.
Results The ileocolonic continent urinary reservoir (Miami pouch) has been the only method used for the creation
Table II Additional surgical procedures at the time of ileocolonic continent urinary conduit (Miami pouch) Procedure Exenteration Total Anterior Low colorectal anastomosis Colostomy End Diverting Pelvic floor reconstruction Omental patch and dura mater Omental path Dura mater Omental patch and fascia lata Omental patch and alloderm graft Omental patch and saline implant Dura mater and saline implant Neovagina Transverse rectus abdominus myocutaneous flaps/vertical rectus abdominis myocutaneous flaps (TRAM/VRAM) Gracilus myocutaneous flap Sigmoid/Ileal Martius flap augmentation Dura mater
Patients (n) 71 10 27 43 17 15 13 5 2 2 1 2 9
2 6 3 2
of a continent urinary diversion since 1988 at the University of Miami, Division of Gynecologic Oncology. A total of 90 patients have been identified who have undergone construction of a continent ileocolonic urinary reservoir at the University of Miami/Jackson Hospital Systems and Sylvester Comprehensive Cancer Center. As shown in Table I, a urinary diversion was performed in most of the cases for primary advanced or recurrent pelvic malignancies in 82 patients (91%), for vesicovaginal fistulas in 5 patients (6%) after pelvic radiation, and for bladder atony in 3 patients (3%). The most common additional surgical procedure was a total pelvic exenteration in 71 patients (79%). Table II shows all the other procedures, which included low colorectal resection with end-to-end anastomosis in 27 patients (30%), the creation of a neovagina in 22 patients (24%), and pelvic floor reconstruction in 40 patients (44%). The age of the patients ranged from 23 to 80 years, with a mean of 51 years. The mean height was 1.6 meters (range, 1.4-1.8 meters), and the mean weight was 70 kg (range, 14-23 kg). The average body mass index was 27 kg/m2 (range, 14-43 kg/m2). A total of 82 patients (91%) received previous radiation therapy to the pelvis with doses ofO4500 cGy. The average blood loss was 2048 ml (range, 350-12,000 ml). The average blood transfusion was 7 units of packed red blood cells. A total of 29 patients (32%) were alive at the time of last contact. A total of 6 patients were lost during the follow-up
997
Salom et al Table III
General perioperative complications
Complication Febrile morbidity Pneumonia Thromboembolic disease Wound complications Pelvic collection Paralytic ileus/partial bowel obstruction Gastrointestinal fistulas (excluding Miami pouch) Postoperative death
No. of patients (n = 90) 68 9 7 27 11 11 6
(76%) (10%) (8%) (30%) (12%) (12%) (7%)
10 (11%)
period, and the cases could not be evaluated for longterm complications. Twenty-nine patients were alive at last date of follow-up. During the immediate postoperative period, 10 patients died of complications as the result of the surgery (mortality rate, 11%). Of these, 5 patients died of sepsis; 2 patients died of adult respiratory distress syndrome; 2 patients died of coagulopathy, and 1 patient died of a myocardial infarction. We evaluated the general perioperative factors, not necessarily associated with the creation of the urinary reservoir, to have a comprehensive understanding of the patients perioperative risk (Table III). Febrile morbidity was the most common perioperative finding, occurring in three quarters of our patients. Fever was defined as a temperature elevation of R100.4(F that occurred twice during a 24-hour period that were separated by at least 4 hours or 1 temperature elevation R101.5(F. Nine patients (10%) were given a diagnosis of pneumonia. Wound complications occurred in 27 patients (30%) during their hospital stay. We defined wound complication as any simple wound abnormality that involved the abdominal incision, stomas, or perineum (including simply erythema, cellulitis, seroma, abscess and dehiscence). Most wound infections were treated conservatively, except in 4 patients (4%) who had fascial dehiscence that required surgical revision and closure. Eleven patients (12%) were found to have a pelvic collection, most of which were treated conservatively with the use of computerized tomographyeguided drainage; only 2 patients underwent laparotomy for organized hematomas. All 12 cases (12%) of paralytic ileus and partial small bowel obstruction were treated conservatively with intravenous hydration, bowel rest, and total parenteral nutrition, if indicated. Six rectovaginal fistulas were encountered exclusively in the patients who had undergone low colorectal resection with end-to-end anastomosis and in 1 patient with a colotomy at the time of an anterior exenteration. We found that patients with low colorectal resection with end-to-end anastomosis without a protective colostomy experienced a rectovaginal fistula during the immediate postoperative period statistically more often than those
Table IV Ileocolonic continent conduit (Miami pouch) related complications Complication Early: !60 d (n = 90) UTI Pouch leak Ureteral stricture Difficult catheterization Incontinence Miami pouch fistula Late:O60 d (n = 74) UTI/pyelonephritis Ureteral stricture Incontinence Miami pouch fistula Difficult catheterization Urinary stones
Patients (n) 36 13 10 10 5 2
(40%) (14%) (11%) (11%) (5%) (2%)
31 8 7 6 6 5
(42%) (11%) (9%) (8%) (7%) (7%)
patients who had a protective colostomy (40% vs 6%, respectively), with a probability value of .043. We evaluated the condition of the patients for both early and late complications that were associated with the continent urinary diversion (Table IV). Forty-eight patients (53%) experienced complications that were related to the urinary diversion; the most common complication was urinary tract infection that was culture proven in 40% of the patients. The most common organism that was isolated during the early postoperative period was a fungal organism. During the long-term follow-up period, 19 patients (25%) had pyelonephritis, which resolved with conservative treatment with the use of intravenous antibiotics and hydration in all patients. The most common complication that was related directly to the surgical reconstruction of the pouch was an anastomotic leak in 13 patients (14%). The site of anastomotic breakdown was identified at the level of the ureterocolonic anastomosis in 6 patients and the suture line of the reservoir in 7 patients. The identification of a urine leak was made after high output persisted from the Jackson Pratt drains or pelvic drainage catheter and subsequently confirmed on a radiographic contrast study. Seven patients (53%) underwent surgical correction of the anastomotic leak that involved ureteral reimplantation (5 patients), revision of ileocolonic conduit perforation (1 patient), and conversion to an ileal conduit (1 patient) after infarction of the original ileocolonic reservoir. We encountered an equal number of both early and late ureteral obstructions, which occurred in 18 patients (20%) in this case series. The detection of hydronephrosis was determined on the basis of an elevation in creatinine level at the time of the clinic visit or symptoms of flank pain. Routine imaging was not performed in the late postoperative period. All cases were treated initially conservatively with percutaneous nephrostomies with
998 subsequent internalization of a nephroureteral stent. The ureters were imaged and, if adequate flow by contrast studies was present, then the stents were removed. If an obstruction persisted, then a transluminal ureteral dilation was performed with a balloon angioplasty catheter. Of the 10 patients who were found to have highgrade ureteral obstructions during the early postoperative period, only 1 patient needed surgical correction. The conditions of 2 of the 8 patients (25%) with highgrade obstruction that was detected after 2 months did not respond to conservative treatment and required ureterocolonic reimplantation. Fourteen patients complained of difficulty catheterizing the ileocolonic reservoir. Difficulty with catheterization was defined by the patient’s inability to insert the Foley without evidence of stomal stricture at the level of the skin. Eight of the 9 patients who complained of difficulty catheterizing the reservoir achieved resolution with conservative treatment that consisted of aggressive education and training after prolonged continuous catheterization for 4 to 6 weeks. Many of these patients who experienced trouble with catheterization in the early postoperative period were found to have deviation of the ileal segment from over distension of the reservoir that formed an acute angle from the skin to the reservoir. Seven patients (77%) required fluoroscopic insertion of the Foley catheter. Of the 5 patients who experienced difficulty self-catheterizing during the late postoperative period, 2 patients were found to have stones that obstructed the pouch, and 1 patient was found to have a deviated ileal stomal outlet that required surgical correction. Stomal skin stricture was reported only in 3 patients, all of whom were treated by outpatient revision of the skin site. A fistulous tract that involved the ileocolonic reservoir was detected in 8 patients (9%), with 2 detections occurring in the early and 6 detections occurring in the late postoperative periods. Four patients had fistulas from the pouch to the skin, 3 of which involved the bowel and 1 of which involved the common iliac vessel. Three of the reservoir-cutaneous fistulas resolved with conservative treatment, which consisted of prolonged catheterization and nephrostomy tube drainage. One fistula was treated surgically. Two of the enteroreservoir fistulas were treated surgically, and one fistula was treated conservatively with prolonged parenteral nutrition, percutaneous nephrostomy drainage, and somatostatin. The ureterovascular fistula was treated conservatively with a nephroureteral stent and fluoroscopic placement of a vascular stent into the left common iliac vessels, thus bypassing the tract. Five of our patients (7%) were diagnosed with nephrolithiasis during the past 15 years. Three of the 5 patients were treated with lithotripsy and nephroureteral stents. Two patients required revision of the ileocolonic reservoir as a result of a large volume and number of
Salom et al stones filling the pouch, which caused low urine capacity, difficulty with catheterization, and erosion into the colon. Urinary incontinence was reported by 8 patients (11%) with a Miami pouch. Most of the patients were treated conservatively, which included decreased intervals of self-catheterization from every 4 to 6 hours to every 2 to 3 hours or continuous catheter drainage. Three patients were treated surgically with revision of the ileal segment of the ileocolonic reservoir, with successful correction in 2 patients and persistent, although significantly reduced, incontinence in 1 patient. Overall, 6 patients continued to experience some form of urinary incontinence that was minimal in nature and did not require surgical intervention. At the time of the last follow-up examination, 92% of the patients experienced normal continent function of the Miami pouch. With regards to abnormal renal function that was associated with the Miami pouch, 2 patients (3%) were found to have persistently elevated creatinine levels of 1.8 and 2.5 mg/dL as a result of ureteral stricture that was treated with percutaneous nephrostomies. Ninety-seven percent of patients retained normal renal function, which was determined by a serum creatinine level of !1.6 mg/dL.
Comment During the 1950s, when the pelvic exenteration was first described by Brunschwig,16 the technique of urinary diversion was simply to anastomose bilateral ureters into the end colostomy, also known as a wet colostomy. In 1951, Bricker2 developed the ileal conduit that soon became the standard of care for urinary diversion at the time of pelvic exenteration. The advantage the ileal conduit posed at that time was to decrease reflux of fecal fluid into the ureter and kidney. Although the ileal and colonic conduits have been used the most frequently and reported in the literature, they fail to provide a continent mechanism and require dependence on an external appliance, decreased volume capacity, increased peristomal irritation, and alterations in body image. Because of the need for an external appliance, Boyd et al17 confirmed the improvement in body image in patients with a continent reservoir over ileal conduits. In 1989, Penalver et al11 first reported our experience in patients who underwent a radical pelvic resection after radiation therapy with the use of the continent ileocolonic reservoir (Miami pouch) as an option to an ileal or colonic conduit with good success. In our study, the overall complication rate that is related directly to the Miami pouch is 53%. Most of these patients (91%) had received radiation therapy before construction of the continent ileocolonic reservoir. The detrimental effects of radiation therapy on postoperative
999
Salom et al morbidity are well documented in the literature. In the study by Wilson et al18 of the Indiana pouch, they report the increase in late complications as a result of radiation therapy. Ramirez et al19 reported a similar overall complication rate of 65% in 40 patients who had been treated previously with radiotherapy who had undergone a continent ileocolonic reservoir. The functional outcomes of urinary diversion that had been conducted in patients who had undergone previous radiation therapy with gynecologic malignancies cannot be compared with numerous reports of urinary diversion in healthy patients who had not undergone radiation because the physiologic condition of the disease and healing process are unique. The effects of radiation are most evident in the gastrointestinal tract. All our patients who had a gastrointestinal fistula occurred in patients who had undergone a low colorectal resection, except in 1 patient whose fistula developed from a colotomy. Many authors have reported the high incidence of complications and wound breakdown with low colorectal anastomosis in irradiated fields. In 1990, Hatch et al20 found that 50% of patients with a low colorectal anastomosis did not experience healing; all of the patients required colostomies. In 1990 Berek et al21 reported on 11 patients with exenteration who underwent a low colorectal anastomosis with protective colostomies; only approximately 60% of the patients had successful reversal of their colostomy. Recently, Husain et al22 found an incidence of 50% anastomotic breakdown rate in patients who were treated with colostomies. In our present report, we found that, of the patients with low colorectal anastomosis without a protective colostomy, 4 of 7 patients (40%) experienced an anastomotic breakdown and a rectovaginal fistula that necessitated a reoperation and a colostomy. For this reason, we feel it is essential to perform a protective colostomy at the time of low colorectal anastomosis in patients who have previously undergone radiation therapy. The most common complication that we found in our series of patients was urinary tract infections (40%) and pyelonephritis (21%; Table V). The most common organism that was isolated in urine cultures was a fungal organism, specifically Candida albicans. The incidence of urinary infection in the immediate postoperative period is not surprising because the reservoir is created from the ascending and transverse colon. Preoperative bowel preparation fails to eradicate the bacterial load that becomes apparent after the creation of the pouch. The long-term urinary infections are a result of a number of factors: increased colonization from multiple nonsterile catheterizations every 2 to 6 hours, ureteral reflux, and partial ureteral obstruction in some patients. We have found that the routine administration of prophylactic antibiotic immediately after surgery or suppressive therapy does not reduce the incidence of infection. Few
Table V Treatment of complications that were associated directly with continent ileocolonic reservoir (Miami pouch) Patients (n) Complication UTI/pyelonephritis Ureteral stricture Difficult selfcatheterization* Pouch leak Incontinence Miami pouch fistula Urinary stones Stomal stricture
Treatement (n) Conservative
Surgical
36 (40%) 18 (20%) 16 (18%)
52 (100%) 15 (83%) 12 (75%)
0 3 (17%) 3 (25%)
13 (14%) 12 (13%) 8 (9%)
6 (46%) 9 (75%) 5 (63%)
7 (54%) 3 (25%) 3 (37%)
5 (6%) 3 (3%)
3 (60%) 3 (100%)
2 (40%) 0
* Difficult self-catheterization was caused by urinary stone in 2 cases, which were corrected with the same surgical procedure.
studies have evaluated the rate of overall urinary infection after diversion. Ramirez et al19 and Ordorica et al23 reported similar rates of urinary infection of 35% and 38%, respectively. The group at M.D. Anderson found that 52% of patients experience a urinary infection while receiving prophylactic antibiotics.19 The most common urinary complication that is associated directly with the continent ileocolonic urinary reservoir is ureteral stricture in 18 patients (20%). The incidence of ureteral obstruction in the literature in patients who have undergone radiation therapy ranges from 3% to 28%.19,23-26 Most of our patients (84%) were treated conservatively with prolonged percutaneous nephrostomies, nephroureteral stenting, and balloon dilation. We did find that of the 8 patients who did not receive antecedent radiotherapy, not 1 patient had a ureteral complication. Clearly, radiation has a profound impact on functional outcome, which was confirmed by Webster et al,26 who reviewedO179 patients and found a 28.4% rate of ureteral obstruction compared with only 6.3% in patients who had not undergone radiation therapy. Stomal stricture was encountered in 3% of our patients, who were found to have minimal scarring that would prevent the passage of a 14F catheter. Two patients were instructed to use a 10F to 12F catheter; the other patient required a simple revision under local anesthetic in the clinic with complete resolution. Difficult self-catheterizing of the urinary reservoir was noted in 16 patients (18%). The incidence of difficulty with self-catheterization as reported in the literature ranges from 5% to 50%.11-13,18-19,23-25,27 We found that overdistension of the pouch resulted in a 90degree angle of deviation of the neourethra, which thus prevented insertion of the Foley catheter. We have modified the technique, as previously described, by anchoring the pouch to the anterior abdominal fascia and placing 4 to 5 permanent sutures to decrease the stoma-to-pouch deviation. The aim is to allow for
1000 a straight course from the stoma to the reservoir. Ninety-two percent of the patients with difficulty selfcatheterizing that was not related to stone formation was treated with prolonged catheterization. One patient with an ileal occlusion that resulted from a peristomal abscess that fibrosed required surgical revision of the ileal portion. Two of the patients who had difficulty catheterizing had a large volume of stones within the reservoir, which resulted in complete obliteration of the pouch and required surgical removal and pouch revision. Urinary stone formation has been found in 6% of our patients. The aforementioned cases of urinary stones were a result of the use of permanent automatic gastrointestinal anastomatic stapler with permanent titanium to close the pouch, which has been abandoned in our institution. The permanent gastrointestinal anastomosis titanium staples are used currently on the ileal segment without any incidence of stone formation. The 3 additional patients who had stones were treated with lithotripsy and nephroureteral stents, and the condition resolved completely. Another complication of the urinary reservoir is the presence of an anastomotic urinary leak (Table V). The site of the leak was approximately equal between the ureter (46%) and the reservoir (54%). When an anastomotic leak is encountered, we found that most leaks did not resolve with conservative treatment, but rather 54% of the leaks required surgical intervention with revision of the pouch or ureterocolonic reimplantation. Two patients were found to have a perforation of the pouch that was the result of over distension from an obstruction of the Foley catheter with a mucus plug. Proper and timely flushing of the reservoir every 2 to 3 hours immediately postoperatively is paramount. Fistulas that involved the Miami pouch were found in 8 patients (9%), of which 62% of the complications resolved completely with conservative therapy. Two cases of the pouch-to-colon fistulas were a result of stone formation eroding from the pouch to the colon. In a total of 6 patients (7%), a spontaneous fistula occurred in the late postoperative period. Fistulas are rare complications of urinary diversions in the patients who have not received radiation therapy. Wammack et al28 recently reported the high rate of strictures, fistulas, and leaks of 22% in patients who had undergone radiation therapy compared with 6.5% for patients who had not undergone radiation therapy. Similarly in our series, the patients who did not receive radiotherapy never had a fistula. Twelve patients experienced some form of incontinence that was resolved completely with surgical revision in 2 patients and prolonged catheterization along with decreased frequency of emptying in 4 patients. The remaining 6 patients of the 74 evaluable patients during long-term follow-up either declined surgical correction or experienced minimal symptoms that de-
Salom et al manded no further intervention. This translates into an overall continence rate of 93%, which is compatible with previous reports. The literature reports the rate of continence for ileocolonic reservoirs between 75% and 100% in a relatively small number of patients over a short follow-up period.19,24,27 One must not consider only the incidence of continence of this procedure but also the ability to maintain renal function. Hancock et al29 and Segreti et al30 showed a 12% to 17% rate of loss of renal function in patients with ileal, sigmoid, and transverse colonic reservoirs. We have found that 3% of our patient had permanent renal loss of function from long-standing ureteral stricture that was related directly to the Miami pouch. It appears that the continent ileocolonic urinary reservoir maintains and preserves adequate renal function over the course of time. In conclusion, the Miami pouch is a safe and effective technique that can be offered to patients who are undergoing radical pelvic resection, with a continence rate of 93%. Diligent postoperative surveillance is paramount in all patients who have undergone a urinary diversion with a history of previous radiotherapy because of the moderate incidence of perioperative complications. The detection of complications in a timely fashion with the use of advanced radiographic imaging can prevent many of the long-term consequences that are experienced by these patients. Over the past 15 years, most of the postoperative complications (80%) could be corrected with the use of conservative techniques that were associated with fewer deaths and thus should be used first. Our experience has shown that the Miami pouch is a reasonable alternative after bladder resection over incontinent diversion that would have profound advantages to a woman’s psychosocial well-being.
Acknowledgments We thank all the faculty members and fellows from the University of Miami over the past 2 decades who have contributed to this endeavor.
References 1. Lloyd M. Ectopia vescae (absence of the anterior wall of the bladder) operation and subsequent death. Lancet 1851;2: 370-2. 2. Bricker EM. Bladder substitution after pelvic evisceration. Surg Clin North Am 1950;30:1511-21. 3. Nurmi M, Puntala P, Alanen A. Evaluation of 144 cases of ileal conduits in adults. Eur Urol 1988;15:89-93. 4. Gilchrist RK, Merricks JW, Hamlin HH, Rieger IT. Construction of a substitute bladder and urethra. Surg Gynecol Obstet 1950;90: 752-5. 5. Bejany DE, Politano VA. Stapled and non-stapled tapered distal ileum for the construction of a continent colonic urinary reservoir. J Urol 1988;140:491-4.
1001
Salom et al 6. Thuroff JW, Alkeran P, Engelman U, Riedmiller H, Jadcobi GH, Hohenfellner R. The Mainz pouch (mixed augmentation ileum and cecum) applicable for bladder augmentation and continent urinary diversion. Eur Urol 1985;11:150-60. 7. Rowland GR, Mitchell ME, Bihrle R, Kahnoski RJ, Piser JE. Indiana continent urinary reservoir. J Urol 1987;137:1136-9. 8. Lockhart JL, Pow-Sang J, Kahn HM, Sanford E. A continent colonic urinary reservoir: the Florida pouch. J Urol 1990;144:864-7. 9. Montie JE, MacGregor PS, Fazio VM, Lavery L. Continent ileal urinary reservoir (Kock pouch). Urol Clin North Am 1986;13:251-4. 10. Ahlering TA, Kanellos AW, Boyd SD, Lieskovshy G, Skinner DG, Bernstein L. A comparative study of perioperative complications with Kock pouch urinary diversion in highly irradiated versus nonirradiated patients. J Urol 1988;139:1202-4. 11. Penalver MA, Bejany DE, Averette HE, Donato DM, Sevin BU, Suarez G. Continent urinary diversion in gynecologic oncology. Gynecol Oncol 1988;34:274-88. 12. Penalver MA, Bejany D, Donalto D, Seven BU, Averette HE. Functional characteristics and follow-up of the continent ileal colonic urinary reservoir: Miami pouch. Can Suppl 1993;71:1667-72. 13. Penalver MA, Donato DM, Sevin BU, Bloch WE, Alvarez WE, Averette HE. Complications of the ileocolonic continent urinary reservoir (Miami pouch). Gynecol Oncol 1994;52:360-4. 14. Penalver MA, Angioli R, Mirhashemi R, Malik R. Management of early and late complications of the ileocolonic continent urinary reservoir (Miami pouch). Gynecol Oncol 1998;68:185-91. 15. Angioli R, Estape R, Cantuaria G, Mirhashemi R, Williams H, Martin J, et al. Urinary complications of the Miami pouch: trend of conservative management. Am J Obstet Gynecol 1998;179:343-8. 16. Brunschwig A. Complete excision of pelvic viscera for abdominal carcinoma. Cancer 1948;1:177-88. 17. Boyd SD, Feinberg SM, Skinner DG, Lieskovshy G, Baron D, Richardson J. Quality of life survey of urinary diversion patients: comparison of ileal conduits versus continent Kock ileal reservoirs. J Urol 1987;138:1386-9. 18. Wilson TG, Moreno JG, Ahlering TE. Late complications of modified Indiana pouch. J Urol 1994;151:331-4. 19. Ramirez PT, Modesitt SC, Morris M, Edwards C, Bevers MW, Warton JT, et al. Functional outcomes and complications of continent urinary diversions in patients with gynecologic malignancies. Gynecol Oncol 2002;85:285-91. 20. Hatch KD, Gelder MS, Soong S, Baker VV, Shingleton HM. Pelvic exenteration with low rectal anastomosis: survival, complications and prognostic factors. Gynecol Oncol 1990;38:462-7. 21. Berek JS, Hacker NF, Lagasse LD. Rectosigmoid colectomy and reanastomosis to facilitate resection of primary and recurrent gynecologic cancer. Obstet Gynecol 1984;64:715-20. 22. Husain A, Curtin J, Brown C, Chi D, Hoskins W, Poynor K, et al. Continent urinary diversion and low-rectal anastomosis in patients undergoing exenterative procedures for recurrent gynecologic malignancies. Gynecol Oncol 2000;78:208-11. 23. Ordorica RC, Masel J, Seigne J, Persky L, Lockhart JL. Evaluation and management of mechanical dysfunction in continent colonic urinary reservoir. J Urol 2000;163:1679-84. 24. Mannel RS, Manetta A, Buller R, Braly PS, Walker JL, Archer JS. Use of ileocecal continent urinary reservoir in patients with previous pelvic radiation. Gynecol Oncol 1995;59:376-8. 25. Hartenbach EM, Saltzman AK, Carter JR, Fowler JM, Hunter DW, Carlson JW, et al. Nonsurgical management strategies for the functional complications of ileocolonic continent urinary reservoirs. Gynecol Oncol 1995;59:358-63. 26. Webster C, Bukkapatnam R, Seigne JD, Pow-Sang J, Hoffman M, Helal M, et al. Continent colonic urinary reservoir (Florida pouch): long-term surgical complications (greater than 11 years). J Urol 2003;169:174-6. 27. Dottino PR, Segna RA, Jennings TS, Beddoe AM, Cohen CJ. The stapled continent ileocecal urinary reservoir in the sur-
gical management of gynecologic malignancy. Gynecol Oncol 1994;55:185-9. 28. Wammack R, Wricke C, Hohenfellner R. Long-term results of ileocecal continent urinary diversion in patients treated with and without previous pelvic radiation. J Urol 2002;167:2058-62. 29. Hancock KC, Copeland LJ, Gershenson DM, Saul PB, Wharton JT, Rutledge FN. Urinary conduits in gynecologic oncology. Obstet Gynecol 1986;67:680-4. 30. Segreti EM, Morris M, Levenback C, Lucas KR, Gershenson DM, Burke TW. Transverse colonic urinary diversions in gynecologic oncology. Gynecol Oncol 1996;63:66-70.
Discussion DR PETER SCHWARTZ, New Haven. The successful treatment of recurrent lower reproductive tract cancers, especially cervical cancers, requires a dramatic surgical approach to achieve cure. Pelvic exenterations have the ability to cure recurrent disease. Dr Salom et al have been leaders in the field of exenterative surgery and the creation of a continent ileocolonic urinary reservoir, the so-called Miami pouch.1-3 Continent urinary reservoirs allow patients to be free of an external appliance to collect urine and may be more effective in preserving renal function than a noncontinent conduit. The surgery is more complex than for noncontinent conduits (ie, ileal or transverse colon conduits). Complication rates after this surgery are high, and reoperations are associated with excessive mortality rates. Conservative treatment of postoperative complications should reduce the perioperative mortality rate. In this study, Dr Salom et al describe the treatment of complications that have occurred perioperatively and subsequent to the formation of Miami pouches in a series of 90 patients, 82 of whom (91%) have undergone previous pelvic radiation therapy. The study starts by describing perioperative complications. The patients have experienced the usual postoperative exenteration complications, including febrile morbidity, wound complications, pelvic collections, and ileus and small bowel obstruction. The experience is insightful regarding rectovaginal fistulas, because 4 of 10 patients (40%) who did not undergo a protective colostomy after a colon reanastomosis experienced fistula formation, whereas only 2 patients (6%) who underwent a protective colostomy experienced a fistula. The difference in fistulas was significant at the level of .043. Avoidance and treatment of long-term complications was a second issue of this article. The authors concentrated on conservative treatment of long-term complications. Long-term complications were divided into those that occurred within 60 days of the surgery and those than occurred atO60 days from the time of surgery. Interestingly, 5 of the 7 types of complications that were reported occurred in relatively the same frequency, both early after surgery and atO60 days after surgery. For example, the rate of urinary tract infections was 40%
www.elsevier.com/locate/ajog
Continent ileocolonic urinary reservoir (Miami pouch): The University of Miami experience over 15 years Emery M. Salom, MD,* Luis E. Mendez, MD, Dana Schey, MD, Nicholas Lambrou, MD, Noor Kassira, MS, Orlando Go´mez-Marı´n, MSc, PhD, Hervy Averette, MD, Manuel Pen˜alver, MD University of Miami, School of Medicine, Jackson Memorial Hospital /Sylvester Comprehensive Cancer Center, Division of Gynecologic Oncology, Miami, Fla
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS Continent ileocolonic urinary reservoir Miami pouch Pelvic exenteration Urinary diversion Reconstruction
Objective: A patient with a recurrent central pelvic malignancy after radiation will require urinary diversion as part of the reconstructive phase of the pelvic exenteration. The aim of our study was to assess the result of our 15-year experience with a continent ileocolonic urinary reservoir, which is known as the Miami pouch. Study design: Since 1988, all patients who received a continent ileocolonic urinary reservoir in the Division of Gynecologic Oncology, University of Miami School of Medicine, were included in the study. Parameters that were evaluated during the study period include functional outcomes, early and late perioperative complications, and their treatment. Results: A total of 90 patients were identified from February 1988 to December 2002. Seventyeight patients (87%) had a recurrent central pelvic malignancy, and 82 patients (91%) received radiation before the Miami pouch procedure. The nonereservoir-related morbidities were fever (76%), wound complication (30%), pelvic collection (12%), ileus/small bowel obstruction (12%), and postoperative death (11%). The most common reservoir-related complications were urinary infection (40%), ureteral stricture (20%), and difficulty with self-catheterization (18%). In our study, the overall complication rate that was related directly to the Miami pouch was 53%. Conservative treatment resolvedO80% of these cases. The rate of urinary continence that was achieved in our patients was 93% during our 15-year experience with the Miami pouch. Conclusion: The Miami pouch is a good alternative for continent urinary diversion during exenteration or radiation-induced damage. The rate of major complications that require aggressive surgical intervention is acceptable. Most postoperative complications (80%) can be corrected with the use of conservative techniques that are associated with fewer deaths than reoperation and thus should be used first. The technique is simple and effective in women who are at high risk, who have undergone previous radiation therapy, and who have a high rate of functional success and is a profound advantage for a woman’s psychosocial well-being. Ó 2004 Elsevier Inc. All rights reserved.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Presented at the Twenty-Second Annual Meeting of the American Gynecological and Obstetrical Society, Napa, California, September 18-20, 2003. * Reprint requests: Emery M. Salom, MD, Division of Gynecologic Oncology, University of Miami School of Medicine, Jackson 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/j.ajog.2004.01.023
Memorial Hospital/ Sylvester Cancer Center, 1611 NW 12th Ave, Holtz Building East Tower Room 3003, Miami, FL 33136. E-mail: [email protected]
995
Salom et al Pelvic exenteration is the most radical and destructive elective procedure described in pelvic surgery. For many women with central pelvic recurrences after radiotherapy, an exenteration is the only procedure that can be offered with curative intent. As a consequence, the reconstruction of the pelvic floor, gastrointestinal tract, sexual organs, and urinary system remains a great challenge for many pelvic surgeons. Much advancement has been made over the past 150 years in this arena, especially with respect to urinary diversion. The urinary diversion technique ureterosigmoidostomy, described by Lloyd1 in the 1850s, remained relatively unchanged until Bricker’s2 publication in 1950. Bricker popularized the use of the incontinent ileal conduit that has remained the most popular and extensively studied form of urinary diversion. The long-term outcome studies have revealed that the postoperative complication rates that are associated with the ileal conduits range between 25% and 49%.3 In an attempt to avoid such long-term complications, Gilchrist et al4 were the first to describe the use of the ileocecal valve for a continent urinary diversion. Many techniques of continent urinary diversion have been developed that used the terminal ileum and the ascending, transverse, or descending colon, which include the Miami pouch,5 Mainz pouch,6 Indiana pouch,7 and Florida pouch.8 Many of these various techniques came about in an effort to decrease the high initial failure rate of 10% to 20%, urine leakage rate of 37%, and reoperation rate of 20% with the original continent ileal reservoir, the Kock pouch. This pouch uses the principle of an intussuscepted nipple valve system for continence.9,10 Because of all the previously described complications, a number of different alternatives have been explored, which includes the Miami pouch. In 1987, Bejany and Politano5 described the use of a low-pressure detubularized colonic reservoir with a tapered ileum and pursestring suture around the ileocecal valve as its continent mechanism that is simple and effective. In 1988, Penalver et al11 were the first to describe the use of the continent ileocolonic urinary reservoir in recurrent gynecologic malignancies. Since that time the Division of Gynecologic Oncology at the University of Miami School of Medicine has been using the Miami pouch as the preferred method of urinary diversion. We previously published our experience and the number of evolutionary changes of the technique and trends towards conservative treatment of complications in hopes of improving the functional outcome and decreasing the postoperative morbidity that are associated with the Miami pouch.11-15 We report the largest experience with continent ileocolonic reservoirs in patients who have undergone radiation therapy in the literature to date. We review our experience over the last 15 years that involved the early and long-term postoperative complications, conserva-
tive and surgical treatment, and overall continence of the Miami pouch.
Material and methods We evaluated all patients who underwent an ileocolonic continent urinary reservoir (Miami pouch) from February 1988 to December 2002 at the University of Miami, Division of Gynecologic Oncology. These patients have been followed prospectively for the recognition and treatment of complications and functionality of the procedure. All details as to the patient’s progress are recorded in the Division’s Miami Database System with the use of the FileMaker Pro (File Maker, Inc, Santa Clara, Calif) software. We reviewed the last 15 years since the creation and evolution of the Miami pouch as a continent urinary reservoir for patients with gynecologic malignancies and its complications. The operative technique used for the creation of the Miami pouch is identical to those described previously at our institution in 1988.11 The reservoir consists of the distal 12 cm of the ileum and the ascending and transverse colon transected distal to the middle colonic artery. The isolated intestinal segment that averages 30 cm in length is incised along the tenia and thus detubulerized. The opened ascending and transverse portions of the colon are anastomosed in a U-shaped fashion to create a low-pressure system. The distal ileum is catheterized with a 14F silicone rubber Foley catheter and tapered to this diameter with a gastrointestinal automatic stapler. The ileocecal valve is then reinforced with 3 20 silk sutures placed 0.5 cm apart. As a result, the ileum provides a high-pressure outlet to maintain continence. The ureters are spatulated and anastomosed to the transverse colon, which is the nonirradiated portion, and are stented. During the early evolution of the Miami pouch, the ureters were anastomosed with the use of a submucosal tunnel, which has been abandoned subsequently in an attempt to prevent ureteral stricture. The pouch is anchored to the anterior abdominal wall fascia with 4 to 5 permanent sutures. The ileal portion is brought out to the anterior abdominal wall. A Jackson Pratt drain is placed posterior to the pouch for the collection of any extravasated urine, and the Foley catheter is left in the pouch for 14 days and flushed with 50 mL of saline solution every 4 hours. The nephroureteral stents are removed on days 14 through 21 along with the Foley catheter after a contrast study of the urinary reservoir and the ureters is performed to assure the absence of any urinary leakage, reflux, or upper tract obstruction. The data for all patients who underwent the creation of the continent ileocolonic urinary reservoir were analyzed. The information was abstracted from the computerized database and medical records with respect to age, body mass index, primary disease site, indications for
996 Table I
Salom et al Indications for surgical procedure
Procedure
Patients (n)
Recurrent cervical cancer Recurrent endometrial cancer Uterine leiomyosarcoma Recurrent ovarian cancer Recurrent vaginal cancer Recurrent vulvar cancer Primary vulvar cancer Primary vulvar/vaginal melanoma Vesical vaginal fistula Bladder atony Urethral cancer Mons pubis perineural sarcoma Rectal sarcoma Total
68 2 2 1 2 1 1 2 5 3 1 1 1 90
urinary diversion, history of radiation therapy, and types of procedures performed in conjunction with the Miami Pouch. We included all early postoperative complications and long-term complications with the respective treatment strategies. Early postoperative complications were defined as all those complications that occurred !60 days after the day of surgery; long-term complications were defined as those complications that occurredO60 days after the surgery date. The functional outcomes that were specific to the urinary reservoir were evaluated and included difficulty with catheterization, urinary infection, pyelonephritis, early reservoir leak, ureteral obstruction, stomal stricture, fistula formation, incontinence, and renal function. The data were evaluated routinely in the clinic 2 weeks after discharge and every 3 months thereafter. The patients were followed for the development of any signs of tumor recurrence and postoperative complications. Follow-up and survival were measured from the day of surgery to the date of last contact or death. Serum electrolytes, blood urea nitrogen level, creatinine level, and urine cultures are evaluated routinely with each clinic visit. Renal function was analyzed by the determination of blood urea nitrogen and creatinine levels, which were defined as the normal values of 7 to 22 mg/ dL and 0.8 to 1.5 mg/dL, respectively. In 1993, we reported the result of multichannel urodynamics in the first 16 patients, which confirmed the low-pressure reservoir and the high-pressure ileal outlet.2 We no longer obtain routine urodynamics on our patients. Statistical analysis consisted of chi-squared analysis and Fisher exact test.
Results The ileocolonic continent urinary reservoir (Miami pouch) has been the only method used for the creation
Table II Additional surgical procedures at the time of ileocolonic continent urinary conduit (Miami pouch) Procedure Exenteration Total Anterior Low colorectal anastomosis Colostomy End Diverting Pelvic floor reconstruction Omental patch and dura mater Omental path Dura mater Omental patch and fascia lata Omental patch and alloderm graft Omental patch and saline implant Dura mater and saline implant Neovagina Transverse rectus abdominus myocutaneous flaps/vertical rectus abdominis myocutaneous flaps (TRAM/VRAM) Gracilus myocutaneous flap Sigmoid/Ileal Martius flap augmentation Dura mater
Patients (n) 71 10 27 43 17 15 13 5 2 2 1 2 9
2 6 3 2
of a continent urinary diversion since 1988 at the University of Miami, Division of Gynecologic Oncology. A total of 90 patients have been identified who have undergone construction of a continent ileocolonic urinary reservoir at the University of Miami/Jackson Hospital Systems and Sylvester Comprehensive Cancer Center. As shown in Table I, a urinary diversion was performed in most of the cases for primary advanced or recurrent pelvic malignancies in 82 patients (91%), for vesicovaginal fistulas in 5 patients (6%) after pelvic radiation, and for bladder atony in 3 patients (3%). The most common additional surgical procedure was a total pelvic exenteration in 71 patients (79%). Table II shows all the other procedures, which included low colorectal resection with end-to-end anastomosis in 27 patients (30%), the creation of a neovagina in 22 patients (24%), and pelvic floor reconstruction in 40 patients (44%). The age of the patients ranged from 23 to 80 years, with a mean of 51 years. The mean height was 1.6 meters (range, 1.4-1.8 meters), and the mean weight was 70 kg (range, 14-23 kg). The average body mass index was 27 kg/m2 (range, 14-43 kg/m2). A total of 82 patients (91%) received previous radiation therapy to the pelvis with doses ofO4500 cGy. The average blood loss was 2048 ml (range, 350-12,000 ml). The average blood transfusion was 7 units of packed red blood cells. A total of 29 patients (32%) were alive at the time of last contact. A total of 6 patients were lost during the follow-up
997
Salom et al Table III
General perioperative complications
Complication Febrile morbidity Pneumonia Thromboembolic disease Wound complications Pelvic collection Paralytic ileus/partial bowel obstruction Gastrointestinal fistulas (excluding Miami pouch) Postoperative death
No. of patients (n = 90) 68 9 7 27 11 11 6
(76%) (10%) (8%) (30%) (12%) (12%) (7%)
10 (11%)
period, and the cases could not be evaluated for longterm complications. Twenty-nine patients were alive at last date of follow-up. During the immediate postoperative period, 10 patients died of complications as the result of the surgery (mortality rate, 11%). Of these, 5 patients died of sepsis; 2 patients died of adult respiratory distress syndrome; 2 patients died of coagulopathy, and 1 patient died of a myocardial infarction. We evaluated the general perioperative factors, not necessarily associated with the creation of the urinary reservoir, to have a comprehensive understanding of the patients perioperative risk (Table III). Febrile morbidity was the most common perioperative finding, occurring in three quarters of our patients. Fever was defined as a temperature elevation of R100.4(F that occurred twice during a 24-hour period that were separated by at least 4 hours or 1 temperature elevation R101.5(F. Nine patients (10%) were given a diagnosis of pneumonia. Wound complications occurred in 27 patients (30%) during their hospital stay. We defined wound complication as any simple wound abnormality that involved the abdominal incision, stomas, or perineum (including simply erythema, cellulitis, seroma, abscess and dehiscence). Most wound infections were treated conservatively, except in 4 patients (4%) who had fascial dehiscence that required surgical revision and closure. Eleven patients (12%) were found to have a pelvic collection, most of which were treated conservatively with the use of computerized tomographyeguided drainage; only 2 patients underwent laparotomy for organized hematomas. All 12 cases (12%) of paralytic ileus and partial small bowel obstruction were treated conservatively with intravenous hydration, bowel rest, and total parenteral nutrition, if indicated. Six rectovaginal fistulas were encountered exclusively in the patients who had undergone low colorectal resection with end-to-end anastomosis and in 1 patient with a colotomy at the time of an anterior exenteration. We found that patients with low colorectal resection with end-to-end anastomosis without a protective colostomy experienced a rectovaginal fistula during the immediate postoperative period statistically more often than those
Table IV Ileocolonic continent conduit (Miami pouch) related complications Complication Early: !60 d (n = 90) UTI Pouch leak Ureteral stricture Difficult catheterization Incontinence Miami pouch fistula Late:O60 d (n = 74) UTI/pyelonephritis Ureteral stricture Incontinence Miami pouch fistula Difficult catheterization Urinary stones
Patients (n) 36 13 10 10 5 2
(40%) (14%) (11%) (11%) (5%) (2%)
31 8 7 6 6 5
(42%) (11%) (9%) (8%) (7%) (7%)
patients who had a protective colostomy (40% vs 6%, respectively), with a probability value of .043. We evaluated the condition of the patients for both early and late complications that were associated with the continent urinary diversion (Table IV). Forty-eight patients (53%) experienced complications that were related to the urinary diversion; the most common complication was urinary tract infection that was culture proven in 40% of the patients. The most common organism that was isolated during the early postoperative period was a fungal organism. During the long-term follow-up period, 19 patients (25%) had pyelonephritis, which resolved with conservative treatment with the use of intravenous antibiotics and hydration in all patients. The most common complication that was related directly to the surgical reconstruction of the pouch was an anastomotic leak in 13 patients (14%). The site of anastomotic breakdown was identified at the level of the ureterocolonic anastomosis in 6 patients and the suture line of the reservoir in 7 patients. The identification of a urine leak was made after high output persisted from the Jackson Pratt drains or pelvic drainage catheter and subsequently confirmed on a radiographic contrast study. Seven patients (53%) underwent surgical correction of the anastomotic leak that involved ureteral reimplantation (5 patients), revision of ileocolonic conduit perforation (1 patient), and conversion to an ileal conduit (1 patient) after infarction of the original ileocolonic reservoir. We encountered an equal number of both early and late ureteral obstructions, which occurred in 18 patients (20%) in this case series. The detection of hydronephrosis was determined on the basis of an elevation in creatinine level at the time of the clinic visit or symptoms of flank pain. Routine imaging was not performed in the late postoperative period. All cases were treated initially conservatively with percutaneous nephrostomies with
998 subsequent internalization of a nephroureteral stent. The ureters were imaged and, if adequate flow by contrast studies was present, then the stents were removed. If an obstruction persisted, then a transluminal ureteral dilation was performed with a balloon angioplasty catheter. Of the 10 patients who were found to have highgrade ureteral obstructions during the early postoperative period, only 1 patient needed surgical correction. The conditions of 2 of the 8 patients (25%) with highgrade obstruction that was detected after 2 months did not respond to conservative treatment and required ureterocolonic reimplantation. Fourteen patients complained of difficulty catheterizing the ileocolonic reservoir. Difficulty with catheterization was defined by the patient’s inability to insert the Foley without evidence of stomal stricture at the level of the skin. Eight of the 9 patients who complained of difficulty catheterizing the reservoir achieved resolution with conservative treatment that consisted of aggressive education and training after prolonged continuous catheterization for 4 to 6 weeks. Many of these patients who experienced trouble with catheterization in the early postoperative period were found to have deviation of the ileal segment from over distension of the reservoir that formed an acute angle from the skin to the reservoir. Seven patients (77%) required fluoroscopic insertion of the Foley catheter. Of the 5 patients who experienced difficulty self-catheterizing during the late postoperative period, 2 patients were found to have stones that obstructed the pouch, and 1 patient was found to have a deviated ileal stomal outlet that required surgical correction. Stomal skin stricture was reported only in 3 patients, all of whom were treated by outpatient revision of the skin site. A fistulous tract that involved the ileocolonic reservoir was detected in 8 patients (9%), with 2 detections occurring in the early and 6 detections occurring in the late postoperative periods. Four patients had fistulas from the pouch to the skin, 3 of which involved the bowel and 1 of which involved the common iliac vessel. Three of the reservoir-cutaneous fistulas resolved with conservative treatment, which consisted of prolonged catheterization and nephrostomy tube drainage. One fistula was treated surgically. Two of the enteroreservoir fistulas were treated surgically, and one fistula was treated conservatively with prolonged parenteral nutrition, percutaneous nephrostomy drainage, and somatostatin. The ureterovascular fistula was treated conservatively with a nephroureteral stent and fluoroscopic placement of a vascular stent into the left common iliac vessels, thus bypassing the tract. Five of our patients (7%) were diagnosed with nephrolithiasis during the past 15 years. Three of the 5 patients were treated with lithotripsy and nephroureteral stents. Two patients required revision of the ileocolonic reservoir as a result of a large volume and number of
Salom et al stones filling the pouch, which caused low urine capacity, difficulty with catheterization, and erosion into the colon. Urinary incontinence was reported by 8 patients (11%) with a Miami pouch. Most of the patients were treated conservatively, which included decreased intervals of self-catheterization from every 4 to 6 hours to every 2 to 3 hours or continuous catheter drainage. Three patients were treated surgically with revision of the ileal segment of the ileocolonic reservoir, with successful correction in 2 patients and persistent, although significantly reduced, incontinence in 1 patient. Overall, 6 patients continued to experience some form of urinary incontinence that was minimal in nature and did not require surgical intervention. At the time of the last follow-up examination, 92% of the patients experienced normal continent function of the Miami pouch. With regards to abnormal renal function that was associated with the Miami pouch, 2 patients (3%) were found to have persistently elevated creatinine levels of 1.8 and 2.5 mg/dL as a result of ureteral stricture that was treated with percutaneous nephrostomies. Ninety-seven percent of patients retained normal renal function, which was determined by a serum creatinine level of !1.6 mg/dL.
Comment During the 1950s, when the pelvic exenteration was first described by Brunschwig,16 the technique of urinary diversion was simply to anastomose bilateral ureters into the end colostomy, also known as a wet colostomy. In 1951, Bricker2 developed the ileal conduit that soon became the standard of care for urinary diversion at the time of pelvic exenteration. The advantage the ileal conduit posed at that time was to decrease reflux of fecal fluid into the ureter and kidney. Although the ileal and colonic conduits have been used the most frequently and reported in the literature, they fail to provide a continent mechanism and require dependence on an external appliance, decreased volume capacity, increased peristomal irritation, and alterations in body image. Because of the need for an external appliance, Boyd et al17 confirmed the improvement in body image in patients with a continent reservoir over ileal conduits. In 1989, Penalver et al11 first reported our experience in patients who underwent a radical pelvic resection after radiation therapy with the use of the continent ileocolonic reservoir (Miami pouch) as an option to an ileal or colonic conduit with good success. In our study, the overall complication rate that is related directly to the Miami pouch is 53%. Most of these patients (91%) had received radiation therapy before construction of the continent ileocolonic reservoir. The detrimental effects of radiation therapy on postoperative
999
Salom et al morbidity are well documented in the literature. In the study by Wilson et al18 of the Indiana pouch, they report the increase in late complications as a result of radiation therapy. Ramirez et al19 reported a similar overall complication rate of 65% in 40 patients who had been treated previously with radiotherapy who had undergone a continent ileocolonic reservoir. The functional outcomes of urinary diversion that had been conducted in patients who had undergone previous radiation therapy with gynecologic malignancies cannot be compared with numerous reports of urinary diversion in healthy patients who had not undergone radiation because the physiologic condition of the disease and healing process are unique. The effects of radiation are most evident in the gastrointestinal tract. All our patients who had a gastrointestinal fistula occurred in patients who had undergone a low colorectal resection, except in 1 patient whose fistula developed from a colotomy. Many authors have reported the high incidence of complications and wound breakdown with low colorectal anastomosis in irradiated fields. In 1990, Hatch et al20 found that 50% of patients with a low colorectal anastomosis did not experience healing; all of the patients required colostomies. In 1990 Berek et al21 reported on 11 patients with exenteration who underwent a low colorectal anastomosis with protective colostomies; only approximately 60% of the patients had successful reversal of their colostomy. Recently, Husain et al22 found an incidence of 50% anastomotic breakdown rate in patients who were treated with colostomies. In our present report, we found that, of the patients with low colorectal anastomosis without a protective colostomy, 4 of 7 patients (40%) experienced an anastomotic breakdown and a rectovaginal fistula that necessitated a reoperation and a colostomy. For this reason, we feel it is essential to perform a protective colostomy at the time of low colorectal anastomosis in patients who have previously undergone radiation therapy. The most common complication that we found in our series of patients was urinary tract infections (40%) and pyelonephritis (21%; Table V). The most common organism that was isolated in urine cultures was a fungal organism, specifically Candida albicans. The incidence of urinary infection in the immediate postoperative period is not surprising because the reservoir is created from the ascending and transverse colon. Preoperative bowel preparation fails to eradicate the bacterial load that becomes apparent after the creation of the pouch. The long-term urinary infections are a result of a number of factors: increased colonization from multiple nonsterile catheterizations every 2 to 6 hours, ureteral reflux, and partial ureteral obstruction in some patients. We have found that the routine administration of prophylactic antibiotic immediately after surgery or suppressive therapy does not reduce the incidence of infection. Few
Table V Treatment of complications that were associated directly with continent ileocolonic reservoir (Miami pouch) Patients (n) Complication UTI/pyelonephritis Ureteral stricture Difficult selfcatheterization* Pouch leak Incontinence Miami pouch fistula Urinary stones Stomal stricture
Treatement (n) Conservative
Surgical
36 (40%) 18 (20%) 16 (18%)
52 (100%) 15 (83%) 12 (75%)
0 3 (17%) 3 (25%)
13 (14%) 12 (13%) 8 (9%)
6 (46%) 9 (75%) 5 (63%)
7 (54%) 3 (25%) 3 (37%)
5 (6%) 3 (3%)
3 (60%) 3 (100%)
2 (40%) 0
* Difficult self-catheterization was caused by urinary stone in 2 cases, which were corrected with the same surgical procedure.
studies have evaluated the rate of overall urinary infection after diversion. Ramirez et al19 and Ordorica et al23 reported similar rates of urinary infection of 35% and 38%, respectively. The group at M.D. Anderson found that 52% of patients experience a urinary infection while receiving prophylactic antibiotics.19 The most common urinary complication that is associated directly with the continent ileocolonic urinary reservoir is ureteral stricture in 18 patients (20%). The incidence of ureteral obstruction in the literature in patients who have undergone radiation therapy ranges from 3% to 28%.19,23-26 Most of our patients (84%) were treated conservatively with prolonged percutaneous nephrostomies, nephroureteral stenting, and balloon dilation. We did find that of the 8 patients who did not receive antecedent radiotherapy, not 1 patient had a ureteral complication. Clearly, radiation has a profound impact on functional outcome, which was confirmed by Webster et al,26 who reviewedO179 patients and found a 28.4% rate of ureteral obstruction compared with only 6.3% in patients who had not undergone radiation therapy. Stomal stricture was encountered in 3% of our patients, who were found to have minimal scarring that would prevent the passage of a 14F catheter. Two patients were instructed to use a 10F to 12F catheter; the other patient required a simple revision under local anesthetic in the clinic with complete resolution. Difficult self-catheterizing of the urinary reservoir was noted in 16 patients (18%). The incidence of difficulty with self-catheterization as reported in the literature ranges from 5% to 50%.11-13,18-19,23-25,27 We found that overdistension of the pouch resulted in a 90degree angle of deviation of the neourethra, which thus prevented insertion of the Foley catheter. We have modified the technique, as previously described, by anchoring the pouch to the anterior abdominal fascia and placing 4 to 5 permanent sutures to decrease the stoma-to-pouch deviation. The aim is to allow for
1000 a straight course from the stoma to the reservoir. Ninety-two percent of the patients with difficulty selfcatheterizing that was not related to stone formation was treated with prolonged catheterization. One patient with an ileal occlusion that resulted from a peristomal abscess that fibrosed required surgical revision of the ileal portion. Two of the patients who had difficulty catheterizing had a large volume of stones within the reservoir, which resulted in complete obliteration of the pouch and required surgical removal and pouch revision. Urinary stone formation has been found in 6% of our patients. The aforementioned cases of urinary stones were a result of the use of permanent automatic gastrointestinal anastomatic stapler with permanent titanium to close the pouch, which has been abandoned in our institution. The permanent gastrointestinal anastomosis titanium staples are used currently on the ileal segment without any incidence of stone formation. The 3 additional patients who had stones were treated with lithotripsy and nephroureteral stents, and the condition resolved completely. Another complication of the urinary reservoir is the presence of an anastomotic urinary leak (Table V). The site of the leak was approximately equal between the ureter (46%) and the reservoir (54%). When an anastomotic leak is encountered, we found that most leaks did not resolve with conservative treatment, but rather 54% of the leaks required surgical intervention with revision of the pouch or ureterocolonic reimplantation. Two patients were found to have a perforation of the pouch that was the result of over distension from an obstruction of the Foley catheter with a mucus plug. Proper and timely flushing of the reservoir every 2 to 3 hours immediately postoperatively is paramount. Fistulas that involved the Miami pouch were found in 8 patients (9%), of which 62% of the complications resolved completely with conservative therapy. Two cases of the pouch-to-colon fistulas were a result of stone formation eroding from the pouch to the colon. In a total of 6 patients (7%), a spontaneous fistula occurred in the late postoperative period. Fistulas are rare complications of urinary diversions in the patients who have not received radiation therapy. Wammack et al28 recently reported the high rate of strictures, fistulas, and leaks of 22% in patients who had undergone radiation therapy compared with 6.5% for patients who had not undergone radiation therapy. Similarly in our series, the patients who did not receive radiotherapy never had a fistula. Twelve patients experienced some form of incontinence that was resolved completely with surgical revision in 2 patients and prolonged catheterization along with decreased frequency of emptying in 4 patients. The remaining 6 patients of the 74 evaluable patients during long-term follow-up either declined surgical correction or experienced minimal symptoms that de-
Salom et al manded no further intervention. This translates into an overall continence rate of 93%, which is compatible with previous reports. The literature reports the rate of continence for ileocolonic reservoirs between 75% and 100% in a relatively small number of patients over a short follow-up period.19,24,27 One must not consider only the incidence of continence of this procedure but also the ability to maintain renal function. Hancock et al29 and Segreti et al30 showed a 12% to 17% rate of loss of renal function in patients with ileal, sigmoid, and transverse colonic reservoirs. We have found that 3% of our patient had permanent renal loss of function from long-standing ureteral stricture that was related directly to the Miami pouch. It appears that the continent ileocolonic urinary reservoir maintains and preserves adequate renal function over the course of time. In conclusion, the Miami pouch is a safe and effective technique that can be offered to patients who are undergoing radical pelvic resection, with a continence rate of 93%. Diligent postoperative surveillance is paramount in all patients who have undergone a urinary diversion with a history of previous radiotherapy because of the moderate incidence of perioperative complications. The detection of complications in a timely fashion with the use of advanced radiographic imaging can prevent many of the long-term consequences that are experienced by these patients. Over the past 15 years, most of the postoperative complications (80%) could be corrected with the use of conservative techniques that were associated with fewer deaths and thus should be used first. Our experience has shown that the Miami pouch is a reasonable alternative after bladder resection over incontinent diversion that would have profound advantages to a woman’s psychosocial well-being.
Acknowledgments We thank all the faculty members and fellows from the University of Miami over the past 2 decades who have contributed to this endeavor.
References 1. Lloyd M. Ectopia vescae (absence of the anterior wall of the bladder) operation and subsequent death. Lancet 1851;2: 370-2. 2. Bricker EM. Bladder substitution after pelvic evisceration. Surg Clin North Am 1950;30:1511-21. 3. Nurmi M, Puntala P, Alanen A. Evaluation of 144 cases of ileal conduits in adults. Eur Urol 1988;15:89-93. 4. Gilchrist RK, Merricks JW, Hamlin HH, Rieger IT. Construction of a substitute bladder and urethra. Surg Gynecol Obstet 1950;90: 752-5. 5. Bejany DE, Politano VA. Stapled and non-stapled tapered distal ileum for the construction of a continent colonic urinary reservoir. J Urol 1988;140:491-4.
1001
Salom et al 6. Thuroff JW, Alkeran P, Engelman U, Riedmiller H, Jadcobi GH, Hohenfellner R. The Mainz pouch (mixed augmentation ileum and cecum) applicable for bladder augmentation and continent urinary diversion. Eur Urol 1985;11:150-60. 7. Rowland GR, Mitchell ME, Bihrle R, Kahnoski RJ, Piser JE. Indiana continent urinary reservoir. J Urol 1987;137:1136-9. 8. Lockhart JL, Pow-Sang J, Kahn HM, Sanford E. A continent colonic urinary reservoir: the Florida pouch. J Urol 1990;144:864-7. 9. Montie JE, MacGregor PS, Fazio VM, Lavery L. Continent ileal urinary reservoir (Kock pouch). Urol Clin North Am 1986;13:251-4. 10. Ahlering TA, Kanellos AW, Boyd SD, Lieskovshy G, Skinner DG, Bernstein L. A comparative study of perioperative complications with Kock pouch urinary diversion in highly irradiated versus nonirradiated patients. J Urol 1988;139:1202-4. 11. Penalver MA, Bejany DE, Averette HE, Donato DM, Sevin BU, Suarez G. Continent urinary diversion in gynecologic oncology. Gynecol Oncol 1988;34:274-88. 12. Penalver MA, Bejany D, Donalto D, Seven BU, Averette HE. Functional characteristics and follow-up of the continent ileal colonic urinary reservoir: Miami pouch. Can Suppl 1993;71:1667-72. 13. Penalver MA, Donato DM, Sevin BU, Bloch WE, Alvarez WE, Averette HE. Complications of the ileocolonic continent urinary reservoir (Miami pouch). Gynecol Oncol 1994;52:360-4. 14. Penalver MA, Angioli R, Mirhashemi R, Malik R. Management of early and late complications of the ileocolonic continent urinary reservoir (Miami pouch). Gynecol Oncol 1998;68:185-91. 15. Angioli R, Estape R, Cantuaria G, Mirhashemi R, Williams H, Martin J, et al. Urinary complications of the Miami pouch: trend of conservative management. Am J Obstet Gynecol 1998;179:343-8. 16. Brunschwig A. Complete excision of pelvic viscera for abdominal carcinoma. Cancer 1948;1:177-88. 17. Boyd SD, Feinberg SM, Skinner DG, Lieskovshy G, Baron D, Richardson J. Quality of life survey of urinary diversion patients: comparison of ileal conduits versus continent Kock ileal reservoirs. J Urol 1987;138:1386-9. 18. Wilson TG, Moreno JG, Ahlering TE. Late complications of modified Indiana pouch. J Urol 1994;151:331-4. 19. Ramirez PT, Modesitt SC, Morris M, Edwards C, Bevers MW, Warton JT, et al. Functional outcomes and complications of continent urinary diversions in patients with gynecologic malignancies. Gynecol Oncol 2002;85:285-91. 20. Hatch KD, Gelder MS, Soong S, Baker VV, Shingleton HM. Pelvic exenteration with low rectal anastomosis: survival, complications and prognostic factors. Gynecol Oncol 1990;38:462-7. 21. Berek JS, Hacker NF, Lagasse LD. Rectosigmoid colectomy and reanastomosis to facilitate resection of primary and recurrent gynecologic cancer. Obstet Gynecol 1984;64:715-20. 22. Husain A, Curtin J, Brown C, Chi D, Hoskins W, Poynor K, et al. Continent urinary diversion and low-rectal anastomosis in patients undergoing exenterative procedures for recurrent gynecologic malignancies. Gynecol Oncol 2000;78:208-11. 23. Ordorica RC, Masel J, Seigne J, Persky L, Lockhart JL. Evaluation and management of mechanical dysfunction in continent colonic urinary reservoir. J Urol 2000;163:1679-84. 24. Mannel RS, Manetta A, Buller R, Braly PS, Walker JL, Archer JS. Use of ileocecal continent urinary reservoir in patients with previous pelvic radiation. Gynecol Oncol 1995;59:376-8. 25. Hartenbach EM, Saltzman AK, Carter JR, Fowler JM, Hunter DW, Carlson JW, et al. Nonsurgical management strategies for the functional complications of ileocolonic continent urinary reservoirs. Gynecol Oncol 1995;59:358-63. 26. Webster C, Bukkapatnam R, Seigne JD, Pow-Sang J, Hoffman M, Helal M, et al. Continent colonic urinary reservoir (Florida pouch): long-term surgical complications (greater than 11 years). J Urol 2003;169:174-6. 27. Dottino PR, Segna RA, Jennings TS, Beddoe AM, Cohen CJ. The stapled continent ileocecal urinary reservoir in the sur-
gical management of gynecologic malignancy. Gynecol Oncol 1994;55:185-9. 28. Wammack R, Wricke C, Hohenfellner R. Long-term results of ileocecal continent urinary diversion in patients treated with and without previous pelvic radiation. J Urol 2002;167:2058-62. 29. Hancock KC, Copeland LJ, Gershenson DM, Saul PB, Wharton JT, Rutledge FN. Urinary conduits in gynecologic oncology. Obstet Gynecol 1986;67:680-4. 30. Segreti EM, Morris M, Levenback C, Lucas KR, Gershenson DM, Burke TW. Transverse colonic urinary diversions in gynecologic oncology. Gynecol Oncol 1996;63:66-70.
Discussion DR PETER SCHWARTZ, New Haven. The successful treatment of recurrent lower reproductive tract cancers, especially cervical cancers, requires a dramatic surgical approach to achieve cure. Pelvic exenterations have the ability to cure recurrent disease. Dr Salom et al have been leaders in the field of exenterative surgery and the creation of a continent ileocolonic urinary reservoir, the so-called Miami pouch.1-3 Continent urinary reservoirs allow patients to be free of an external appliance to collect urine and may be more effective in preserving renal function than a noncontinent conduit. The surgery is more complex than for noncontinent conduits (ie, ileal or transverse colon conduits). Complication rates after this surgery are high, and reoperations are associated with excessive mortality rates. Conservative treatment of postoperative complications should reduce the perioperative mortality rate. In this study, Dr Salom et al describe the treatment of complications that have occurred perioperatively and subsequent to the formation of Miami pouches in a series of 90 patients, 82 of whom (91%) have undergone previous pelvic radiation therapy. The study starts by describing perioperative complications. The patients have experienced the usual postoperative exenteration complications, including febrile morbidity, wound complications, pelvic collections, and ileus and small bowel obstruction. The experience is insightful regarding rectovaginal fistulas, because 4 of 10 patients (40%) who did not undergo a protective colostomy after a colon reanastomosis experienced fistula formation, whereas only 2 patients (6%) who underwent a protective colostomy experienced a fistula. The difference in fistulas was significant at the level of .043. Avoidance and treatment of long-term complications was a second issue of this article. The authors concentrated on conservative treatment of long-term complications. Long-term complications were divided into those that occurred within 60 days of the surgery and those than occurred atO60 days from the time of surgery. Interestingly, 5 of the 7 types of complications that were reported occurred in relatively the same frequency, both early after surgery and atO60 days after surgery. For example, the rate of urinary tract infections was 40%