Delayed Ileal Pouch-Anal Anastomosis Complications and Functional Results Susan Galandiuk, M.D., John H. Pemberton, M.D., Jane Tsao, M.D., Duane M. Ilstrup, M.S., Bruce G. Wolff, M.D. From the Department of Surgery, Section of Colon a n d Rectal Surgery, Mayo Clinic, Rochester, Minnesota In patients with chronic ulcerative colitis (CUC), ileal pouch-anal anastomosis (IPAA) can be performed either at the time of colectomy or as a delayed procedure after total abdominal colectomy and ileostomy. There has been debate as to whether delayed IPAA results in superior functional results, since patients are frequently steroidfree and have little evidence of active disease. To assess this, we analyzed 95 patients who had undergone total abdominal colectomy, either with ileostomy and Hartmann's procedure or with ileorectostomy, 2-183 months prior to IPAA. Postoperative complications and functional results were compared with those of 776 CUC patients who underwent IPAAat the time of abdominal colectomy. Indications for prior colectomy included toxic megacoIon (40 percent), failed medical therapy (36 percent), other reasons (e.g., iatrogenic perforation, cancer) (6 percent), and reasons unclear (18 percent). Nineteen percent of delayed-IPAA patients were taking steroids at the time of pouch construction. Follow-ups were similar in the two groups. The incidence of septic and obstructive complications after delayed IPAA vs. IPAA at the time of colectomy were 10.5 percent vs. 5.4 percent and 6.5 percent vs. 14.5 percent, respectively. There were no significant differences in postoperative functional results between the two groups. Delayed IPAA confers no advantage over IPAA performed at the time of colectomy in terms of functional outcome. Delayed IPAAwas associated with a significantly higher rate of septic complications but a lower incidence of postoperative obstruction. [Key words: Ileal pouch-anal anastomosis; Three-stage procedure; Colectomy; Postoperative complications] Galandiuk S, Pemberton JH, Tsao J, Ilstrup DM, Wolff BG. Delayed ileal pouch-anal anastomosis: complications and functional results. Dis Colon Rectum 1991;34:755-758. otal abdominal c o l e c t o m y and ileal pouch-anal anastomosis (IPAA) has b e c o m e the operation of choice in m a n y patients with chronic ulcerative colitis (CUC). This p r o c e d u r e is usually p e r f o r m e d in two stages: a t e m p o r a r y diverting loop i l e o s t o m y is closed 8 weeks after the c o l e c t o m y and IPAA. Many CUC patients, however, require e m e r g e n c y surgery for toxic m e g a c o l o n or hemorrhage; subtotal c o l e c t o m y with an e n d i l e o s t o m y and Hartmann's p o u c h is f r e q u e n t l y p e r f o r m e d . This two-
T
Read at the meeting of The American Society of Colon and Rectal Surgeons, St. Louis, Missouri, April 29 to May 4, 1990. Address reprint requests to Dr. Galandiuk: Department of Surgery, University of Louisville, Louisville, Kentucky 40292.
stage t e c h n i q u e is the p r e f e r r e d approach because many of these patients are on high doses of steroids, nutritionally d e p l e t e d , and in generally ill health at the time of colectomy. Importantly, most of these patients w h o u n d e r g o IPAA at a later date have acceptable functional results. 1' 2 When the IPAA is s u b s e q u e n t l y p e r f o r m e d , patients are healthier and are often on low doses of steroids. These factors w o u l d p r e s u m a b l y lead to fewer postoperative complications and, perhaps, improved functional results following ileostomy closure. Three-stage p r o c e d u r e s in these patients are, however, associated with an additional operation, concomitantly longer hospitalizations, p r o l o n g e d recovery times, and increased costs. We, therefore, wanted to c o m p a r e postoperative complications and functional results in patients with three-stage and two-stage IPAA. PATIENTS AND METHODS Eight h u n d r e d seventy-one patients with CUC u n d e r w e n t IPAA at the Mayo Clinic from January 1981 to S e p t e m b e r 1989. Our surgical t e c h n i q u e and postoperative results have b e e n previously described. 3'4 O f these 871 patients, 95 had u n d e r g o n e a previous partial or subtotal c o l e c t o m y ("delayedIPAA" group). Data regarding steroid intake, indication for colectomy, type and f r e q u e n c y of postoperative complications, and postoperative functional results in this subgroup w e r e analyzed retrospectively and c o m p a r e d with those of patients w h o u n d e r w e n t IPAA at the time of c o l e c t o m y ("concurrent-IPAA" group). Follow-up data w e r e regularly collated by an i n d e p e n d e n t registered nurse data clerk. Postoperative follow-up was c o n s i d e r e d c o m p l e t e if information regarding p o u c h function had b e e n obtained either in p e r s o n or via t e l e p h o n e interview within the last 12 months. For those patients with a functioning IPAA, the f r e q u e n c y of daily and nocturnal b o w e l m o v e m e n t s was d o c u m e n t e d , and 755
756
GALANDIUK E T A L
daytime and nocturnal continence was categorized as no incontinence, spotting (up to two episodes per week of minor staining of undergarments), or gross incontinence (more than two episodes per week or frank incontinence). The two postoperative complications that we focused on were postoperative small bowel obstruction and pelvic sepsis. The category of small bowel obstruction included patients who developed symptoms of small bowel obstruction postoperatively, regardless of whether they required subsequent surgery. The diagnosis of pelvic sepsis was made in the presence of leukocytosis, pyrexia with or without peritoneal signs, and demonstration of a localized fluid collection on CT or demonstration of a pouch or IPAA leak on radiologic contrast studies.
Analysis Comparison of proportions was made with chisquare tests. The distributions of ordinal variables, such as incontinence, were compared in two or more groups using rank-sum tests. P values less than 0.05 were considered statistically significant. RESULTS IPAA was performed at the time of colectomy in 89 percent of CUC patients undergoing this procedure. In the remaining 95 patients, IPAA was performed 2-183 months (mean, 24 months) after a prior partial or subtotal colectomy. These delayed-IPAA patients ranged in age from 14 to 60 years (mean, 31 years) and included 43 w o m e n a n d 52 men. Indications for previous colectomy were as follows: toxic megacolon, 40 percent; failed medical therapy, 36 percent; and various other reasons, e.g., perforation, carcinoma, 6 percent. In 18 percent of patients, the initial procedure had been performed at another institution and the indication for surgery was not known. Eighty-six percent of patients with prior colectomy had undergone a subtotal colectomy, Hartmann's procedure, and end ileostomy. Nine percent had had an ileorectostomy or ileosigmoidostomy. In the remaining patients, other procedures such as right or left hemicolectomy or segmental resection had been performed. At the time of IPAA, 19 percent of these patients still required oral steroid therapy. Postoperatively, 6 percent of all 871 CUC patients developed pelvic sepsis and 13 percent had
Dis Colon Rectum, September 1991
small bowel obstruction. The frequency of these complications in the delayed- and concurrent-IPAA groups is shown in Figure 1. The delayed-IPAA group had a significantly lower incidence of obstructive complications (6 patients, or 7 percent, vs. 106 patients, or 15 percent; P < 0.05) but a significantly higher rate of pelvic septic complications than the concurrent-IPAA group (10 patients, or 11 percent, vs. 42 patients, or 5 percent; P < 0.05). However, there was no statistical difference between the two groups in the rate of wound infection (2 percent in each group) or in the frequency of a l l postoperative complications (23 patients, or 23 percent, in the delayed-IPAA group vs. 230 patients, or 30 percent, in the concurrent-IPAA group). There was also no difference in the frequency of reoperation for pelvic sepsis or in the frequency of reoperation for obstruction. Postoperative functional results in terms of daytime and nighttime stool frequency, pad use, and the presence of gross fecal incontinence were similar in the two groups (Table 1). DISCUSSION Supporters of three-stage IPAA cite the fact that there is less pelvic inflammation, and, therefore, a more pliable pelvis with improved pouch distensibility and superior functional results, after prior colectomy.1.5 Since most active disease has been removed in these patients, steroid doses can often be reduced and in many cases discontinued. In addition, nutritional status and general health improve. All of these factors presumably result in decreased rates of postoperative complications and improved function. Indeed, in their paper reintro30.
P~ Delo~ed IPAA 22~ Concurrent IPAA r"-1
2520a.
15e
o D.
V/A
105
o
[--1 Y'~ Wound Infection
Obstructlon
Pelvic Sepsis
All C o r r
,lication8
Figure 1. Postoperative complications in patients (N = 871) with delayed IPAA and concurrent IPAA. *Statistically significant (P < 0.05).
DELAYED ILEAL POUCH-ANALANASTOMOSIS
Vol. 34, No. 9
Table 1. Postoperative Function Parameter Stool (median) Day Night Pad worn (%) Severe incontinence (%) Day Night
Delayed IPAA
Concurrent IPAA
5 1 34
5 1 34
3 8
3 10
ducing ileoanal anastomosis, Martin e t al. 5 proposed that subtotal colectomy and ileostomy should be performed to clear the rectum of gross disease prior to mucosectomy and ileoanal anastomosis, thus minimizing the risk of postoperative pelvic infection. Skarsgard e t al. 1 reported an 8 percent incidence of peripouch sepsis in their series of 75 S-pouch patients, 83 percent of whom had undergone previous colectomy. This is similar to the 11 percent in our series. Small bowel obstruction was noted in 6 percent of patients. The authors argued that three-stage procedures were preferable in patients with dysplasia, since an unsuspected carcinoma was found in one of their patients undergoing a two-stage procedure, necessitating pouch removal. In addition, three-stage procedures were recommended in patients undergoing emergency surgery, those in poor general health, and patients on steroids at the time of the colectomy. In Oresland e t a l . ' s 6 series, a 5 percent rate of pelvic sepsis and an 8 percent rate of small bowel obstruction (requiring laparotomy) were noted among 100 patients, 56 percent of whom had IPAA performed after previous colectomy. Fleshman e t al. 2 analyzed postoperative complications and functional results in 179 patients and found that previous subtotal colectomy had no effect on either of these factors. They noted, however, that these results were not representative, because the more severely ill patients and those on high doses of steroid were generally treated with three-stage procedures. In the only prospective study comparing twoand three-stage procedures, Nicholls e t a17 found that the incidence of pelvic sepsis (18 percent and 17 percent, respectively) and small bowel obstruction (19 percent and 15 percent, respectively) did not differ between groups. They also found that a
757
serum albumin level less than 4 g/dl and oral steroids had no effect on postoperative complications. A three-stage procedure was, however, reco m m e n d e d for those patients in whom malignancy or Crohn's disease could not be ruled out. In our study, the incidence of pelvic sepsis was significantly higher in patients with delayed IPAA compared with patients with IPAA at the time of colectomy. Conversely, small bowel obstruction occurred more frequently in patients with IPAA at the time of colectomy. The higher incidence of pelvic sepsis could possibly be due to intraoperative contamination from the ileostomy or to increased IPAA tension secondary to tethering of the small bowel mesentery from the previous surgery. On the other hand, this same tethering of the small bowel in patients with prior colectomy may be responsible for the lower incidence of small bowel obstruction. We do not propose that all patients undergoing emergency surgery for ulcerative colitis undergo IPAA at the time of colectomy, regardless of the experience of the surgeon. We have demonstrated, however, that prior colectomy d o e s n o t result in a decrease in postoperative complications overall and confers no functional advantage compared with two-stage procedures. Therefore, in experienced hands, two-stage and not three-stage IPAA is the preferred approach in most patients with CUC who can be operated upon electively.
REFERENCES 1. Skarsgard ED, Atkinson KG, Bell GA, Pezim ME, Seal AM, Sharp FR. Function and quality of life results after ileal pouch surgery for chronic ulcerative colitis and familial polyposis. Am J Surg 1989;157:467-71. 2. Fleshman JW, Cohen Z, McLeod RS, Stern H, Blair J. The ileal reservoir and ileoanal anastomosis procedure: factors affecting technical and functional outcome. Dis Colon Rectum 1988;31:10-6. 3. Ballantyne GH, Pemberton JH, Beart RW Jr, Wolff BG, Dozois RR. Ileal J pouch-anal anastomosis: current techniques. Dis Colon Rectum 1985;28: 197-202. 4. Pemberton JH, Kelly KA, Beart RW Jr, Dozois RR, Wolff BG, Ilstrup DM. Ileal pouch-anal anastomosis for chronic ulcerative colitis: long-term results. Ann Surg 1987;206:504-13. 5. Martin LW, LeCoultre C, Schubert WK. Total colectomy and mucosal proctectomy with preservation of continence in ulcerative colitis. Ann Surg 1977;
758
GALANDIUK E T AL
186:477-80. 6. Oresland T, Fasth S, Nordgren S, Hulten L. The clinical and functional outcome after restorative proctocolectomy. A prospective study in 100 patients. Int J Color Dis 1989;4:50-6.
Dis Colon Rectum, September 1991
7. Nicholls RJ, Holt SD, Lubowski DZ. Restorative proctocolectomy with ileal reservoir: comparison of twostage vs. three-stage procedures and analysis of factors that might affect outcome. Dis Colon Rectum 1989;32:323-6.
T
Read at the meeting of The American Society of Colon and Rectal Surgeons, St. Louis, Missouri, April 29 to May 4, 1990. Address reprint requests to Dr. Galandiuk: Department of Surgery, University of Louisville, Louisville, Kentucky 40292.
stage t e c h n i q u e is the p r e f e r r e d approach because many of these patients are on high doses of steroids, nutritionally d e p l e t e d , and in generally ill health at the time of colectomy. Importantly, most of these patients w h o u n d e r g o IPAA at a later date have acceptable functional results. 1' 2 When the IPAA is s u b s e q u e n t l y p e r f o r m e d , patients are healthier and are often on low doses of steroids. These factors w o u l d p r e s u m a b l y lead to fewer postoperative complications and, perhaps, improved functional results following ileostomy closure. Three-stage p r o c e d u r e s in these patients are, however, associated with an additional operation, concomitantly longer hospitalizations, p r o l o n g e d recovery times, and increased costs. We, therefore, wanted to c o m p a r e postoperative complications and functional results in patients with three-stage and two-stage IPAA. PATIENTS AND METHODS Eight h u n d r e d seventy-one patients with CUC u n d e r w e n t IPAA at the Mayo Clinic from January 1981 to S e p t e m b e r 1989. Our surgical t e c h n i q u e and postoperative results have b e e n previously described. 3'4 O f these 871 patients, 95 had u n d e r g o n e a previous partial or subtotal c o l e c t o m y ("delayedIPAA" group). Data regarding steroid intake, indication for colectomy, type and f r e q u e n c y of postoperative complications, and postoperative functional results in this subgroup w e r e analyzed retrospectively and c o m p a r e d with those of patients w h o u n d e r w e n t IPAA at the time of c o l e c t o m y ("concurrent-IPAA" group). Follow-up data w e r e regularly collated by an i n d e p e n d e n t registered nurse data clerk. Postoperative follow-up was c o n s i d e r e d c o m p l e t e if information regarding p o u c h function had b e e n obtained either in p e r s o n or via t e l e p h o n e interview within the last 12 months. For those patients with a functioning IPAA, the f r e q u e n c y of daily and nocturnal b o w e l m o v e m e n t s was d o c u m e n t e d , and 755
756
GALANDIUK E T A L
daytime and nocturnal continence was categorized as no incontinence, spotting (up to two episodes per week of minor staining of undergarments), or gross incontinence (more than two episodes per week or frank incontinence). The two postoperative complications that we focused on were postoperative small bowel obstruction and pelvic sepsis. The category of small bowel obstruction included patients who developed symptoms of small bowel obstruction postoperatively, regardless of whether they required subsequent surgery. The diagnosis of pelvic sepsis was made in the presence of leukocytosis, pyrexia with or without peritoneal signs, and demonstration of a localized fluid collection on CT or demonstration of a pouch or IPAA leak on radiologic contrast studies.
Analysis Comparison of proportions was made with chisquare tests. The distributions of ordinal variables, such as incontinence, were compared in two or more groups using rank-sum tests. P values less than 0.05 were considered statistically significant. RESULTS IPAA was performed at the time of colectomy in 89 percent of CUC patients undergoing this procedure. In the remaining 95 patients, IPAA was performed 2-183 months (mean, 24 months) after a prior partial or subtotal colectomy. These delayed-IPAA patients ranged in age from 14 to 60 years (mean, 31 years) and included 43 w o m e n a n d 52 men. Indications for previous colectomy were as follows: toxic megacolon, 40 percent; failed medical therapy, 36 percent; and various other reasons, e.g., perforation, carcinoma, 6 percent. In 18 percent of patients, the initial procedure had been performed at another institution and the indication for surgery was not known. Eighty-six percent of patients with prior colectomy had undergone a subtotal colectomy, Hartmann's procedure, and end ileostomy. Nine percent had had an ileorectostomy or ileosigmoidostomy. In the remaining patients, other procedures such as right or left hemicolectomy or segmental resection had been performed. At the time of IPAA, 19 percent of these patients still required oral steroid therapy. Postoperatively, 6 percent of all 871 CUC patients developed pelvic sepsis and 13 percent had
Dis Colon Rectum, September 1991
small bowel obstruction. The frequency of these complications in the delayed- and concurrent-IPAA groups is shown in Figure 1. The delayed-IPAA group had a significantly lower incidence of obstructive complications (6 patients, or 7 percent, vs. 106 patients, or 15 percent; P < 0.05) but a significantly higher rate of pelvic septic complications than the concurrent-IPAA group (10 patients, or 11 percent, vs. 42 patients, or 5 percent; P < 0.05). However, there was no statistical difference between the two groups in the rate of wound infection (2 percent in each group) or in the frequency of a l l postoperative complications (23 patients, or 23 percent, in the delayed-IPAA group vs. 230 patients, or 30 percent, in the concurrent-IPAA group). There was also no difference in the frequency of reoperation for pelvic sepsis or in the frequency of reoperation for obstruction. Postoperative functional results in terms of daytime and nighttime stool frequency, pad use, and the presence of gross fecal incontinence were similar in the two groups (Table 1). DISCUSSION Supporters of three-stage IPAA cite the fact that there is less pelvic inflammation, and, therefore, a more pliable pelvis with improved pouch distensibility and superior functional results, after prior colectomy.1.5 Since most active disease has been removed in these patients, steroid doses can often be reduced and in many cases discontinued. In addition, nutritional status and general health improve. All of these factors presumably result in decreased rates of postoperative complications and improved function. Indeed, in their paper reintro30.
P~ Delo~ed IPAA 22~ Concurrent IPAA r"-1
2520a.
15e
o D.
V/A
105
o
[--1 Y'~ Wound Infection
Obstructlon
Pelvic Sepsis
All C o r r
,lication8
Figure 1. Postoperative complications in patients (N = 871) with delayed IPAA and concurrent IPAA. *Statistically significant (P < 0.05).
DELAYED ILEAL POUCH-ANALANASTOMOSIS
Vol. 34, No. 9
Table 1. Postoperative Function Parameter Stool (median) Day Night Pad worn (%) Severe incontinence (%) Day Night
Delayed IPAA
Concurrent IPAA
5 1 34
5 1 34
3 8
3 10
ducing ileoanal anastomosis, Martin e t al. 5 proposed that subtotal colectomy and ileostomy should be performed to clear the rectum of gross disease prior to mucosectomy and ileoanal anastomosis, thus minimizing the risk of postoperative pelvic infection. Skarsgard e t al. 1 reported an 8 percent incidence of peripouch sepsis in their series of 75 S-pouch patients, 83 percent of whom had undergone previous colectomy. This is similar to the 11 percent in our series. Small bowel obstruction was noted in 6 percent of patients. The authors argued that three-stage procedures were preferable in patients with dysplasia, since an unsuspected carcinoma was found in one of their patients undergoing a two-stage procedure, necessitating pouch removal. In addition, three-stage procedures were recommended in patients undergoing emergency surgery, those in poor general health, and patients on steroids at the time of the colectomy. In Oresland e t a l . ' s 6 series, a 5 percent rate of pelvic sepsis and an 8 percent rate of small bowel obstruction (requiring laparotomy) were noted among 100 patients, 56 percent of whom had IPAA performed after previous colectomy. Fleshman e t al. 2 analyzed postoperative complications and functional results in 179 patients and found that previous subtotal colectomy had no effect on either of these factors. They noted, however, that these results were not representative, because the more severely ill patients and those on high doses of steroid were generally treated with three-stage procedures. In the only prospective study comparing twoand three-stage procedures, Nicholls e t a17 found that the incidence of pelvic sepsis (18 percent and 17 percent, respectively) and small bowel obstruction (19 percent and 15 percent, respectively) did not differ between groups. They also found that a
757
serum albumin level less than 4 g/dl and oral steroids had no effect on postoperative complications. A three-stage procedure was, however, reco m m e n d e d for those patients in whom malignancy or Crohn's disease could not be ruled out. In our study, the incidence of pelvic sepsis was significantly higher in patients with delayed IPAA compared with patients with IPAA at the time of colectomy. Conversely, small bowel obstruction occurred more frequently in patients with IPAA at the time of colectomy. The higher incidence of pelvic sepsis could possibly be due to intraoperative contamination from the ileostomy or to increased IPAA tension secondary to tethering of the small bowel mesentery from the previous surgery. On the other hand, this same tethering of the small bowel in patients with prior colectomy may be responsible for the lower incidence of small bowel obstruction. We do not propose that all patients undergoing emergency surgery for ulcerative colitis undergo IPAA at the time of colectomy, regardless of the experience of the surgeon. We have demonstrated, however, that prior colectomy d o e s n o t result in a decrease in postoperative complications overall and confers no functional advantage compared with two-stage procedures. Therefore, in experienced hands, two-stage and not three-stage IPAA is the preferred approach in most patients with CUC who can be operated upon electively.
REFERENCES 1. Skarsgard ED, Atkinson KG, Bell GA, Pezim ME, Seal AM, Sharp FR. Function and quality of life results after ileal pouch surgery for chronic ulcerative colitis and familial polyposis. Am J Surg 1989;157:467-71. 2. Fleshman JW, Cohen Z, McLeod RS, Stern H, Blair J. The ileal reservoir and ileoanal anastomosis procedure: factors affecting technical and functional outcome. Dis Colon Rectum 1988;31:10-6. 3. Ballantyne GH, Pemberton JH, Beart RW Jr, Wolff BG, Dozois RR. Ileal J pouch-anal anastomosis: current techniques. Dis Colon Rectum 1985;28: 197-202. 4. Pemberton JH, Kelly KA, Beart RW Jr, Dozois RR, Wolff BG, Ilstrup DM. Ileal pouch-anal anastomosis for chronic ulcerative colitis: long-term results. Ann Surg 1987;206:504-13. 5. Martin LW, LeCoultre C, Schubert WK. Total colectomy and mucosal proctectomy with preservation of continence in ulcerative colitis. Ann Surg 1977;
758
GALANDIUK E T AL
186:477-80. 6. Oresland T, Fasth S, Nordgren S, Hulten L. The clinical and functional outcome after restorative proctocolectomy. A prospective study in 100 patients. Int J Color Dis 1989;4:50-6.
Dis Colon Rectum, September 1991
7. Nicholls RJ, Holt SD, Lubowski DZ. Restorative proctocolectomy with ileal reservoir: comparison of twostage vs. three-stage procedures and analysis of factors that might affect outcome. Dis Colon Rectum 1989;32:323-6.
