Ileal pouchyanal anastomosis (IPAA) is the procedure

Transcription

1 ORIGINAL CONTRIBUTION Proximal Diversion at the Time of Ileal Pouch Anal Anastomosis for Ulcerative Colitis: Current Practices of North American Colorectal Surgeons Sandra L. de Montbrun, M.D. & Paul M. Johnson, M.D. Department of Surgery, Queen Elizabeth II Health Sciences Centre, Dalhousie University, Halifax, Nova Scotia, Canada PURPOSE: Pelvic sepsis is a serious complication after ileal pouchyanal anastomosis for ulcerative colitis that may lead to pouch failure or poor function. Although a temporary loop ileostomy may be created at the time of ileal pouchyanal anastomosis to prevent or minimize the consequences of an anastomotic leak, research has suggested that an ileostomy can be safely omitted in selected patients. The purpose of this study was to examine the use of proximal diversion by colorectal surgeons at the time of ileal pouchyanal anastomosis for ulcerative colitis. METHODS: A questionnaire was mailed to all practicing fellows of The American Society of Colon and Rectal Surgeons in North America. Surgeons were asked to describe their typical practice for a number of clinical scenarios. RESULTS: Questionnaires were mailed to 913 American Society of Colon and Rectal Surgeons fellows, and 63 percent responded. For a patient who has had a prior colectomy and is not taking steroids, 27 percent of surgeons would perform ileal pouchyanal anastomosis alone, and 73 percent would perform ileal pouchyanal anastomosis with a loop ileostomy. For a patient who has not had previous surgery and is taking prednisone 40 mg/day, 16 percent of surgeons would perform a subtotal colectomy with an end ileostomy, 82 percent would perform a total proctocolectomy and ileal pouchyanal anastomosis with a loop ileostomy, and 2 percent would Supported by a grant from the Capital Health Research Fund. Poster presentation at the meeting of The American Society of Colon and Rectal Surgeons, St. Louis, Missouri, June 2 to 6, Address of correspondence: Paul M. Johnson, M.D., QEII Health Sciences Centre, 1278 Tower Road, Room 8-025, Halifax, NS, Canada B3H 2Y9. paul.johnson@dal.ca Dis Colon Rectum 2009; 52: 1178Y1183 DOI: /DCR.0b013e31819f24fc BThe ASCRS 2009 perform a total proctocolectomy and ileal pouchyanal anastomosis without an ileostomy. There was no relationship between practice setting, annual ileal pouchyanal anastomosis volume, or years in practice and surgeon response for either scenario. CONCLUSIONS: The majority of surgeons create a temporary loop ileostomy at the time of ileal pouchyanal anastomosis for ulcerative colitis. KEY WORDS: Surgery; Ulcerative colitis; Ileal pouchyanal anastomosis; Ileostomy. Ileal pouchyanal anastomosis (IPAA) is the procedure of choice for most patients with ulcerative colitis (UC) who require surgery. Although this procedure is associated with a high quality of life and long-term outcomes are favorable, 1Y3 complications after IPAA are common. The most significant complication is pelvic sepsis, which is reported to occur in up to 23 percent of patients. 4Y6 Patients who experience pelvic sepsis have an increased risk of pouch failure and may have poor pouch function. 7,8 Several techniques may be used to prevent or minimize the consequences of an anastomotic leak after IPAA. These include the creation of a diverting loop ileostomy, placement of a pelvic drain, and placement of a rectal tube. Two small, nonrandomized studies have suggested that a diverting loop ileostomy at the time of IPAA may protect against an anastomotic leak and subsequent pelvic sepsis. 9,10 However, the creation of a loop ileostomy is associated with a high complication rate, and closure of an ileostomy requires a second operation and hospitalization. This second operation increases the total length of hospital stay and the overall costs associated with the procedure. 11 Furthermore, several studies have suggested that a loop ileostomy can be safely omitted at the time of IPAA for UC. 12Y19 Given these conflicting data, the role of proximal diversion at the time of IPAA for UC remains unclear. The purpose of this study was to determine the 1178 DISEASES OF THE COLON & RECTUM VOLUME 52: 6 (2009)

2 Diseases of the Colon & Rectum Volume 52: 6 (2009) 1179 current practices of North American colorectal surgeons with regard to the use of a temporary diverting loop ileostomy at the time of IPAA for UC. The secondary objective was to examine the use of rectal tubes and pelvic drains at the time of IPAA for UC. METHODS A questionnaire was mailed to all fellows of The American Society of Colon and Rectal Surgeons within Canada and the United States. A second questionnaire was sent to all nonresponders after six weeks. Surgeons were asked to provide information about years in surgical practice, annual IPAA volume, and practice setting (academic vs. community). An academic center was defined as a center with a general surgery residency program and/or a colorectal surgery fellowship program. Annual IPAA volume was classified as low (1Y5 cases/year), medium (6Y15 cases/year), or high (915 cases/year). Surgeons were asked to describe their typical practice for three different clinical scenarios. Information about the use of pelvic drains and rectal tubes at the time of IPAA surgery for UC was also collected. Data were entered into a computerized database. Categorical variables were compared using the chisquared test, and continuous variables were examined using Student s t-test. The relationship between years in surgical practice and the use of a diverting ileostomy, pelvic drains, and rectal tubes was examined by quartiles. Statistical significance was set at Approval for this study was obtained from the Research Ethics Board of the Queen Elizabeth II Health Sciences Centre, Dalhousie University (Halifax, NS, Canada). involving patients with UC (Figs. 1Y3). For all three scenarios, the majority of surgeons reported that they create a loop ileostomy at the time of IPAA for UC. There was no association between the use of a diverting loop ileostomy and surgeon years in practice, annual IPAA volume, or practice setting for any of the clinical scenarios. The majority of surgeons routinely place a drain in the pelvis at the time of IPAA for UC (Table 1). The average time that a pelvic drain is left in place is 4 T 2 days, and the most common type of drain used is a Jackson-Pratt (67 percent). There was no association between the use of a pelvic drain and surgeon years in practice, annual IPAA volume, or practice setting. Rectal tubes are not commonly used at the time of IPAA (Table 2). When used, the average time that a rectal tube is left in place is 5 T 2.8 days, and the most common type of rectal tube used is a Foley catheter (40 percent). There was no association between the placement of a rectal tube at the time of IPAA and surgeon years in practice or practice setting. However, there was an association between the use of a rectal tube and annual IPAA volume. Low-volume surgeons were significantly less likely to use a rectal tube at the time of IPAA than medium-volume or high-volume surgeons (P G 0.001). DISCUSSION Pelvic sepsis is a serious complication after IPAA for UC that may occur in up to 23 percent of patients. 4Y6 A leak at the ileal-anal anastomosis is responsible for most cases of pelvic sepsis. 7,20 This may present as peritonitis, a pelvic abscess, or a fistula to the vagina or perineum. RESULTS Questionnaires were mailed to 913 fellows of The American Society of Colon and Rectal Surgeons in Canada and the United States, and 575 (63 percent) were returned. Of surgeons who responded, 466 (81 percent) perform IPAA for UC. Compared with surgeons who do not perform IPAA, surgeons who perform this procedure have been in clinical practice for fewer years (16.4 vs years, P G 0.001) and are more likely to work in an academic center (65 vs. 47 percent, P G 0.001). Of surgeons who perform IPAA, 48.4 percent reported that they perform 1 to 5 procedures per year, 37.3 percent perform 6 to 15 procedures per year, and 14.3 percent perform more than 15 procedures per year. Surgeons who work in an academic setting were likely to be highvolume surgeons (P G 0.001) as compared with surgeons who practice in a community hospital. Surgeons who perform IPAA provided information about their typical practice for three clinical scenarios FIGURE 1. Surgical practice of North American colorectal surgeons in the setting of a 30-year-old patient with ulcerative colitis who had a previous subtotal colectomy with end ileostomy six months ago. The patient has been off steroids for three months. At the time of ileal pouchyanal anastomosis (IPAA), there are no intraoperative complications, and there is no tension on the anastomosis.

3 1180 DE MONTBRUN AND JOHNSON: PROXIMAL DIVERSION AND IPAA FOR UC TABLE 1. Use of pelvic drains by North American colorectal surgeons at the time of ileal pouchyanal anastomosis for ulcerative colitis Indication for use of a pelvic drain Always 64% Only if I m concerned about the anastomosis 21% Never 15% FIGURE 2. Surgical practice of North American colorectal surgeons in the setting of a 30-year-old patient with ulcerative colitis who has not had previous surgery and is presently taking prednisone 10 mg/day. At the time of surgery, the tissues are of reasonable quality, and there are no intraoperative complications. IPAA = ileal pouchyanal anastomosis; STC = subtotal colectomy; TPC = total proctocolectomy. Patients who experience pelvic sepsis are at an increased risk of pouch failure 7,8 and poor pouch function. 8 Furthermore, additional surgery or interventional radiologic procedures may be required to treat the infection. One approach that surgeons may use to prevent or minimize the consequences of an anastomotic leak is the creation of a proximal diverting loop ileostomy at the time of IPAA. However, previous studies have reported conflicting results regarding the benefits associated with a FIGURE 3. Surgical practice of North American colorectal surgeons in the setting of a 30-year-old patient with ulcerative colitis who has not had previous surgery and is presently taking prednisone 40 mg/day. At the time of surgery, the tissues are of reasonable quality, and there are no intraoperative complications. IPAA = ileal pouchyanal anastomosis; STC = subtotal colectomy; TPC = total proctocolectomy. loop ileostomy. Thus, the role of proximal diversion at the time of IPAA is unclear. The present study has demonstrated that the majority of North American colorectal surgeons create a loop ileostomy at the time of IPAA for UC. Although practice parameters from The American Society of Colon and Rectal Surgeons 21 suggest that an ileostomy may be reasonably omitted in a patient with a tension-free IPAA who is not taking high-dose steroids, 73 percent of surgeons in the present study indicated that they would create a loop ileostomy given a similar scenario. Surgeons reported higher rates of proximal diversion given scenarios with patients taking prednisone 10 mg/day and 40 mg/ day. Interestingly, there was no association between the use of proximal diversion and annual surgeon IPAA volume, practice setting (academic vs. community), or years in surgical practice. Although the majority of colorectal surgeons create a loop ileostomy at the time of IPAA, there is little evidence to show that proximal diversion improves outcomes after IPAA. Tjandra et al. 9 retrospectively compared complications after IPAA between 50 patients who did not have an ileostomy and 50 matched patients who did have an ileostomy. The rate of anastomotic leak and pelvic sepsis was significantly higher among patients without an ileostomy (14 vs. 4 percent). A second retrospective study by Cohen et al. 10 also demonstrated a significantly higher rate of pelvic sepsis in patients who were not diverted compared with those with an ileostomy (18 vs. 7 percent, respectively). Williamson et al. 20 compared 50 patients who had IPAA with an ileostomy with 50 patients who did not have an ileostomy. There was no difference in the reported rate of pelvic sepsis between groups. However, among all patients who developed pelvic sepsis, those without an ileostomy were more likely to require reoperation to control the infection as compared with patients who did have an ileostomy (65 vs. 0 percent, respectively). TABLE 2. Use of rectal tubes by North American colorectal surgeons at the time of ileal pouchyanal anastomosis for ulcerative colitis Indication for use of a rectal tube Always 9% Only if I don t make an ileostomy 10% Only if I m worried about the anastomosis 8% Never 73%

4 Diseases of the Colon & Rectum Volume 52: 6 (2009) 1181 In contrast to these findings, a number of studies have suggested that IPAA can be safely performed without proximal diversion. 12Y19 However, the results of these studies must be interpreted cautiously because they are all nonrandomized and are inherently biased since patients were selected for omission of an ileostomy. Accordingly, the outcomes of these studies may be influenced by differences between patients who had an ileostomy and those who did not. This is illustrated in the study by Remzi et al. 19 which compared 1,725 patients who had IPAA with an ileostomy and 277 patients who had IPAA without an ileostomy. Although there were no differences in the rates of anastomotic leak and pelvic sepsis between groups, the patients who were not given an ileostomy were younger, they weighed less, they were taking lower doses of steroids, and their operations were less difficult. Only one randomized trial has studied proximal diversion at the time of IPAA. 22 Forty-five patients who met the inclusion criteria (no steroids, water-tight pouch, intact ileoanal anastomosis) were randomly assigned to IPAA with a loop ileostomy or IPAA without an ileostomy. Although there was no difference in the rate of septic complications between groups, the small sample size of this study limits generalization of these results. Although the benefit of proximal diversion at the time of IPAA remains unclear, there is considerable evidence that creation and closure of a loop ileostomy is associated with a high complication rate. Patients with a loop ileostomy may experience stoma retraction, stoma prolapse, parastomal hernia, bowel obstruction, high stoma output, and pouching problems. 23 High ileostomy output has been reported in 10 percent of patients after IPAA with an ileostomy, and these patients may require hospitalization or early closure of the ileostomy to control dehydration. 14,24 Closure of a loop ileostomy is associated with a morbidity rate of up to 33 percent, 23 including a risk of anastomotic leak and abdominal sepsis. 25 Bowel obstruction has been reported to occur in 6.4 to 15 percent of patients after loop ileostomy closure, and 25 to 45 percent of these patients require surgery. 24,25 Patients who have a loop ileostomy created at the time of IPAA are at an increased risk of both early and late postoperative small-bowel obstruction compared with patients who did not have a loop ileostomy. 26 Creation of a loop ileostomy at the time of IPAA is also associated with an increased overall length of stay in hospital and increased treatment costs as compared with IPAA without an ileostomy. 11Y15 These differences reflect the second operation and hospitalization that are required for closure of the ileostomy. Although numerous studies have evaluated proximal diversion at the time of IPAA for UC, there are no data available in the literature regarding the safety and effectiveness of pelvic drains. Studies from the colorectal cancer literature have investigated the role of pelvic drains and have produced conflicting results. Although most studies have suggested that pelvic drains provide no benefit and are associated with few complications, some studies have reported that they may increase the risk of an anastomotic leak. 27Y31 Other studies have suggested that pelvic drains may prevent or minimize pelvic sepsis. 32,33 The present study found that the majority of colorectal surgeons routinely place a pelvic drain at the time of IPAA for UC. Some surgeons place a drain selectively, and 15 percent never use a pelvic drain. In contrast, the majority of colorectal surgeons never place a rectal tube at the time of IPAA. Although the use of rectal tubes at the time of IPAA for UC has previously been reported in the literature, 10,12,14,22,34 there is no evidence from the UC or colorectal cancer literature to show that rectal tubes provide any benefit, nor is there evidence to show that they cause harm. In theory, a rectal tube left in the pouch during the early postoperative period may decompress the pouch and minimize anastomotic tension, thereby preventing anastomotic leak. Despite a complete lack of data to support their use, 28 percent of surgeons routinely or selectively place a rectal tube. In the present study, 81 percent of surgeons indicated that they would perform total proctocolectomy with IPAA and proximal diversion in a patient with steroiddependent UC taking prednisone 40 mg/day. The safety of performing IPAA in a patient taking prednisone 40 mg/day with or without a diverting ileostomy is unclear. Heuschen et al. 5 observed that preoperative steroid use was an independent risk factor for septic complications after IPAA for UC. Increasing steroid dose was associated with an increased risk of pelvic sepsis, and patients taking prednisone 940 mg/day had higher rates of septic complications than patients taking prednisone 1 to 40 mg/day. A recent study by Lim et al. 6 also demonstrated that steroid use was an independent risk factor for pelvic sepsis after IPAA. The rate of septic complications was evaluated according to the preoperative steroid dose (G20 mg/day, 20Y40 mg/day, and 940mg/day). There was a significant stepwise increase in the rate of septic complications as the dose of steroids increased. These investigators recommended that patients taking doses of steroids equivalent to 920 mg prednisone per day should undergo subtotal colectomy with end ileostomy and deferral of IPAA. Annual surgeon IPAA volume was examined as a factor that might be associated with the use of proximal diversion. The a priori categorization of IPAA volume as low (1Y5 procedures), medium (6Y15 procedures), and high (915 procedures) was based on discussions with experienced colorectal surgeons because there was no precedent in the IPAA literature. Interestingly, almost half of all colorectal surgeons who perform IPAA reported that they only perform one to five procedures per year. It is unknown if there is a volume-outcomes relationship

5 1182 DE MONTBRUN AND JOHNSON: PROXIMAL DIVERSION AND IPAA FOR UC after IPAA for UC. There is evidence from the rectal cancer literature to suggest that patients treated by highvolume surgeons have lower rates of anastomotic leak than patients treated by low-volume surgeons. 35 However, there is little agreement about the threshold that defines high volume, and it has been suggested that subspecialty training in colorectal surgery may have a more significant impact on patient outcomes than procedure volume. 36,37 Kennedy et al. 38 examined outcomes after IPAA according to hospital procedure volume. Institutions with a high volume of IPAA surgeries, defined as more than 100 procedures during the study period of 6.5 years, had lower rates of reoperation and pouch excision than hospitals that had lower procedure volumes. The present study conducted a mail survey to evaluate the use of proximal diversion at the time of IPAA for UC, and response bias is a potential limitation of this research technique. No information is available about the nonresponders. However, the response rate of 63 percent exceeds the accepted standard of 60 percent. 39 Furthermore, there is evidence in the medical literature that systematic differences may not exist between survey responders and nonresponders. 40 Thus, there is not necessarily a relationship between response rates and bias, and even surveys with a very low response rate may provide a representative sample of the target population. 41 Another potential limitation of this research is that data regarding the type of anastomosis (i.e., mucosectomy with handsewn anastomosis vs. stapled anastomosis) were not collected. This could be a factor that is associated with the use of proximal diversion. However, a recent meta-analysis that compared outcomes after IPAA between 2,699 patients who had a handsewn anastomosis and 1,484 patients who had a stapled anastomosis found there was no difference in the rate of anastomotic leak, pelvic sepsis, or fistulas between the two groups. 42 Hence, there is no clear evidence to suggest that a handsewn anastomosis is associated with a higher rate of pelvic septic complications that would justify more liberal use of proximal diversion. Furthermore, there is no evidence that proximal diversion is associated with improved outcomes after IPAA according to the type of ileoanal anastomosis created (handsewn vs. stapled). Finally, other factors (e.g., patient nutrition and use of biologic agents) may also influence the decision to create a proximal ileostomy at the time of IPAA. A final limitation of this research is the use of a short and simple questionnaire. This was done to maximize the response rate to the survey, but it limited the clinical parameters and scenarios that could be assessed. CONCLUSIONS The majority of North American colorectal surgeons create a temporary loop ileostomy at the time of IPAA for UC regardless of patient steroid use or previous subtotal colectomy. Although there is no evidence to support the use of proximal diversion in patients undergoing IPAA who are taking steroids, the increased risk of anastomotic leak in this patient population may justify the use of a loop ileostomy. Further research regarding the role of proximal diversion and the risks of performing IPAA in patients taking steroids is required. There is evidence to suggest that a loop ileostomy may be safely omitted in selected patients who are not taking steroids and undergo a tension-free anastomosis without intraoperative complications. Given the lack of evidence that proximal diversion improves outcomes after IPAA and the high rate of complications and increased costs associated with a loop ileostomy, the role of proximal diversion needs to be clearly defined in this patient population and should be examined in a randomized, controlled clinical trial. REFERENCES 1. Fazio VW, Ziv Y, Church JM, et al. Ileal pouchyanal anastomoses complications and function in 1005 patients. Ann Surg 1995;222:120Y7. 2. Muir AJ, Edwards LJ, Sanders LL, et al. A prospective evaluation of health-related quality of life after ileal pouch anal anastomosis for ulcerative colitis. Am J Gastroenterol 2001; 96:1480Y5. 3. Fazio VW, O Riodain MG, Lavery IC, et al. Long-term functional outcome and quality of life after stapled restorative proctocolectomy. Ann Surg 1999;230:575Y Kariv Y, Delaney CP, Senagore AJ, et al. Clinical outcomes and cost analysis of a Bfast track[ postoperative care pathway for ileal pouch-anal anastomosis: a case control study. Dis Colon Rectum 2007;50:137Y Heuschen UA, Hinz U, Allemeyer EH, et al. Risk factors for ileoanal J pouchyrelated septic complications in ulcerative colitis and familial adenomatous polyposis. Ann Surg 2002; 235:207Y Lim M, Sagar P, Abdulgader A, Thekkinkattil D, Burke Dermot. The impact of preoperative immunomodulation on pouch-related septic complications after ileal pouchyanal anastomosis. Dis Colon Rectum 2007;50:943Y Heuschen UA, Allemeyer EH, Hinz U, Lucas M, Herfarth C, Heuschen G. Outcome after septic complications in J pouch procedures. Br J Surg 2002;89:194Y Farouk R, Dozois RR, Pemberton JH. Incidence and subsequent impact of pelvic abscess after ileal pouchyanal anastomosis for chronic ulcerative colitis. Dis Colon Rectum 1998;41:1239Y Tjandra JJ, Fazio VW, Milsom JW, Lavery IC, Oakley JR, Fabre JM. Omission of temporary diversion in restorative proctocolectomyvis it safe? Dis Colon Rectum 1993;36: 1007Y Cohen Z, McLeod RS, Stephen W, et al. Continuing evolution of the pelvic pouch procedure. Ann Surg 1992;216:506Y Swenson BR, Hollenbeak CS, Poritz LS, et al. Modified two stage ileal pouchyanal anastomosis: equivalent outcomes with less resource utilization. Dis Colon Rectum 2005;48:256Y61.

6 Diseases of the Colon & Rectum Volume 52: 6 (2009) Matikainen M, Santavirta J, Hiltunen KM. Ileoanal anastomosis without covering ileostomy. Dis Colon Rectum 1990;33:384Y Heuschen UA, Hinz U, Allemeyer EH, Lucas M, Heuschen G, Herfarth C. One or two stage procedure for restorative proctocolectomy: rationale for a surgical strategy in ulcerative colitis. Ann Surg 2001;234:788Y Galandiuk S, Wolff BG, Dozois RR, Beart RW. Ileal pouchy anal anastomosis without ileostomy. Dis Colon Rectum 1991;34:870Y Sagar PM, Lewis W, Holdsworth PJ, Johnston D. One-stage restorative proctocolectomy without temporary defunctioning ileostomy. Dis Colon Rectum 1992;35:582Y Gorfine SR, Gelernt IM, Bauer JJ, Harris MT, Kreel I. Restorative proctocolectomy without diverting ileostomy. Dis Colon Rectum 1995;38:188Y Ikeuchi H, Nakano H, Uchino M et al. Safety of one-stage restorative proctocolectomy for ulcerative colitis. Dis Colon Rectum 2005;48:1550Y Hainsworth PJ, Bartolo DC. Selective omission of loop ileostomy in restorative proctocolectomy. Int J Colorectal Dis 1998;13:119Y Remzi FH, Fazio VW, Gorgun E, et al. The outcome after restorative proctocolectomy with or without defunctioning ileostomy. Dis Colon Rectum 2006;49:470Y Williamson ME, Lewis WG, Sagar PM, Holdsworth PJ, Johnston D. One-stage restorative proctocolectomy without temporary ileostomy for ulcerative colitis: a note of caution. Dis Colon Rectum 1997;40:1019Y Cohen JL, Strong SA, Hyman NH, et al. Practice parameters for the surgical treatment of ulcerative colitis. Dis Colon Rectum 2005;48:1997Y Grobler SP, Hosie KB, Keighley MR. Randomized trial of loop ileostomy in restorative proctocolectomy. Br J Surg 79:903Y Kaidar-Person O, Person B, Wexner SD. Complications of construction and closure of temporary loop ileostomy. J Am Coll Surg 2005;201:759Y Gunnarsson U, Karlbom U, Docker M, Raab Y, Pahlman L. Proctocolectomy and pelvic pouchvis a diverting stoma dangerous for the patient? Colorectal Dis 2004;6:23Y Wong KS, Remzi FH, Gorgun E, et al. Loop ileostomy closure after restorative proctocolectomy: outcome in 1,504 patients. Dis Colon Rectum 2005;48:243Y MacLean AR, Cohen Z, MacRae HM, et al. Risk of small bowel obstruction after the ileal pouchyanal anastomosis. Ann Surg 2002;235:200Y Merad F, Hay JM, Fingerhut A, et al. Is prophylactic pelvic drainage useful after elective rectal or anal anastomosis? A multicenter controlled randomized trial. French Association for Surgical Research. Surgery 1999;125:529Y Vignali A, Fazio VW, Lavery IC, et al. Factors associated with the occurrence of leaks in stapled rectal anastomosis: a review of 1,014 patients. J Am Coll Surg 1997;185:105Y Brown SR, Seow-Choen F, Eu KW, et al. A prospective randomized study of drains in infra-peritoneal rectal anastomoses. Tech Coloproctol 2001;5:89Y Yeh CY, Chanchien CR, Wang JY, et al. Pelvic drainage and other risk factors for leakage after elective anterior resection in rectal cancer patients: a prospective study of 978 patients. Ann Surg 2005;241:9Y Jesus EC, Karliczek A, Matos D, Castro AA, Atallah AN. Prophylactic anastomotic drainage for colorectal surgery. Cochrane Database Syst Rev 2004;(4):CD Peeters KC, Tollenaar RA, Marijnen CA, et al. Risk factors for anastomotic failure after total mesorectal excision of rectal cancer. Br J Surg 2005;92:211Y Gingold BS, Jagelman DG. Value of pelvic suction-irrigation in reducing morbidity of low anterior resection of the rectumva ten-year experience. Surgery 1982;91:394Y Arai K, Koganei K, Kimura H, et al. Incidence and outcome of complications following restorative proctocolectomy. Am J Surg 2005;190:39Y Boroski DW, Kelly SB, Bradburn DM, Wilson RG, Gunn A, Ratcliffe AA. Impact of surgeon volume and specialization on short-term outcomes in colorectal surgery. Br J Surg 2007;94:880Y Martling A, Cedermark B, Johansson H, Rutqvist LE, Holm T. The surgeon as a prognostic factor after the introduction of total mesorectal excision in the treatment of rectal cancer. Br J Surg 2002;89:1008Y Smith JAE, King PM, Lane HS, Thompson MR. Evidence of the effect of Fspecialization on the management, surgical outcome and survival from colorectal cancer in Wessex. Br J Surg 2003;90:583Y Kennedy ED, Rothwell DM, Cohen Z, McLeod RS. Increased experience and surgical technique lead to improved outcome after ileal pouch-anal anastomosis: a population-based study. Dis Colon Rectum 2006;49:958Y Groves RM. Survey errors and survey costs. New York: John Wiley and Sons Inc., Robertson J, Walkom EJ, McGettigan P. Response rates and representativeness: a lottery incentive improves physician survey return rates. Pharmacoepidemiol Drug Saf 2005;8: 571Y Asch DA, Jedrziewski MK, Christakis NA. Response rates to mail surveys published in medical journals. J Clin Epidemiol 1997;50:1129Y Lovegrove RE, Constantinides VA, Heriot AG, et al. A comparison of hand-sewn versus stapled ileal pouch anal anastomosis (IPAA) following proctocolectomy: a metaanalysis of 4183 patients. Ann Surg 2006;244:18Y26.