- Email: [email protected]
Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience
CROHNS-00892; No of Pages 8 Journal of Crohn's and Colitis (2013) xx, xxx–xxx
Available online at www.sciencedirect.com
ScienceDirect
Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience ☆ Peter Andersson ⁎, Rickard Norblad, Johan D. Söderholm, Pär Myrelid Department of Surgery, County Council of Östergötland, Linköping, Sweden Division of Surgery, Department of Clinical and Experimental Medicine, Faculty of Health Sciences, Linköping University, Sweden
Received 1 October 2013; received in revised form 12 November 2013; accepted 13 November 2013 KEYWORDS Ulcerative colitis; IRA; IPAA; Complications; Failure
Abstract Introduction: Ileal pouch anal anastomosis (IPAA) is the standard procedure for reconstruction after colectomy for ulcerative colitis (UC). However, ileorectal anastomosis (IRA) as an alternative has, recently experienced a revival. This study from a single center compares the clinical outcomes of these procedures. Methods: From 1992 to 2006, 253 patients consecutively underwent either IRA (n=105) or IPAA (n=148). Selection to either procedure was determined on the basis of rectal inflammation, presence of dysplasia/cancer or patient preferences. Patient-records were retrospectively evaluated. Mean follow-up time was 5.4 and 6.3 years respectively. Results: Major postoperative complications occurred in 12.4% of patients after IRA and in 12.8% after IPAA (ns). Complications of any kind after IRA or IPAA, even including subsequent stoma-closure, occurred in 23.8% and 39.9% respectively (pb 0.01). Estimated cumulative failure rates after 5 and 10 years were 10.1% and 24.1% for IRA and 6.1% and 18.6% for IPAA respectively (ns). The most common cause for failure was intractable proctitis (4.8%) and unspecified dysfunction (4.8%) respectively. At follow-up 76.9% of patients with IRA had proctitis and 34.1% with IPAA had pouchitis. Estimated cumulative cancer-risk after 10, 20 and 25 year duration of disease was 0.0%, 2.1% and 8.7% for IRA. Figures for IPAA were 0.7%, 1.8% and 1.8% (ns).
☆ Conference presentations: Swedish surgical society; Gothenburg, 2010. ECCO annual meeting; Vienna, 2013. ⁎ Corresponding author at: Dept. of Surgery, Linköping University Hospital, S-581 85 Linköping, Sweden. Tel.: + 46 10 1033022; fax: + 46 101033570. E-mail address: [email protected] (P. Andersson). 1873-9946/$ - see front matter © 2013 European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.crohns.2013.11.014 Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
2
P. Andersson et al. Conclusion: Failure-rates did not significantly differ between patients operated with IRA or IPAA. Patients operated with IPAA had a higher cumulative number of postoperative complications. The high long-term cancer-risk after IRA indicates that this procedure should be an interim solution in younger patients. © 2013 European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved.
1. Introduction Reconstructive surgery after colectomy for ulcerative colitis (UC) was launched by Aylett and colleagues around 1950 by introduction of ileorectal anastomosis (IRA).1 Concerns about the method were raised, however, after reports on a substantial risk for the development of cancer in the remaining rectum.2 Due to the advent of the ileal pouch anal anastomosis (IPAA) in the 1970s, the use of IRA was abandoned in most cases of UC.3 Since then IPAA has become the standard method of reconstruction after surgery for UC.4,5 Long-term experience with IPAA has, however, revealed some morbidity associated with the procedure as regards impaired continence and sexual function including fecundability as well as pouchitis.6–9 This together with the knowledge that topical anti-inflammatory treatment and meticulous surveillance of the rectum might reduce the risk for cancer has led to the advocacy of IRA in selected cases.10–12 Surgeons in Scandinavia have been particularly willing to adopt this approach. We report the failure rates, cancer risks, functional outcome and need for medication in patients operated on with IRA at our institution in comparison with those operated on with IPAA during the same time period.
2. Patients and Methods Linköping University Hospital is a tertiary referral hospital serving a population of 1.1 million. From 1992 to 2006, 253 consecutive patients with ulcerative colitis (UC) were operated at the surgical department with either ileorectal anastomosis (IRA) or ileal pouch anal anastomosis (IPAA) as the primary reconstructive procedure after colectomy. The diagnosis of UC was based on clinical history, endoscopy and microscopic examination of the colectomy specimen according to criteria of Lennard-Jones.13 Patient selection to either procedure was determined by the degree of inflammation in the rectal mucosa, presence of dysplasia/cancer or patient preferences. Inflammatory activity at rectoscopy confined to a modified Baron−Ginsburg (BG) 0-1 after topical treatment with 5-aminosalicylic acid (5-ASA) and absence of dysplasia or cancer made patients eligible for IRA.14 Some patients were accepted for IRA, however, in spite of the presence of dysplasia or even cancer according to their personal preferences. With the exception of these patients, patients not suitable for IRA due to medical reasons or personal preferences were chosen for IPAA. IRA was performed in 105 patients (age 10–75, 34 women) and IPAA in 148 (age 13–65, 52 women). Background data on the patients are presented in Table 1. Patients were regularly seen at least once at the out-patient clinic during the follow-up period until the end of 2008, the return to their county hospital or death. Follow-up, which was recorded in
our quality control program, consisted of a clinical examination including an evaluation of function and, when symptoms demanded or the duration of disease was longer than ten years in patients with IRA, endoscopy with biopsies. Follow-up after IRA was scheduled yearly and after IPAA at least every second year. All records have been reviewed retrospectively to collect peri- and postoperative data including 30-day complication rate as well as failure rates, development of dysplasia or cancer, functional outcome, presence of proctitis or pouchitis and medication at follow-up. Complications after surgery were graded according to Clavien–Dindo where complications graded three or more were considered major.15 Patients were classified on the basis of the most serious complication in the case of more than one complication. Failure was defined as either one of the following events — proctectomy, excision of the ileal pouch, and permanent defunctioning with an ileostomy — or the occurrence of cancer where linkage to the previous surgical procedure could not be excluded. The study was approved by the regional ethics committee in Linköping, Sweden.
2.1. Statistical Methods Values are given as mean and standard-deviations (SD) unless otherwise stated. The Mann–Whitney-U test and Chi-square test were used when appropriate. Survival
Table 1 Background data on 253 patients undergoing colectomy for ulcerative colitis followed by reconstruction with either ileorectal anastomosis (IRA) or ileal pouch anal anastomosis (IPAA).
Sex Male Female Age at diagnosis (years) Age at colectomy (years) Age at reconstructive surgery (years) Duration of disease at reconstruction (years) Age at follow-up (years) Follow-up time (years)
IRA
IPAA
n = 105
n = 148
p-Value
0.64 71 (68) 96 (65) 34 (32) 52 (35) 26.9 ± 11.4 26.8 ± 10.3 0.93 33.3 ± 14.7 34.2 ± 11.4 0.23 34.2 ± 14.7 35.4 ± 11.3 0.16 7.2 ± 7.5
8.6 ± 7.7
0.06
39.5 ± 15.1 41.7 ± 12.2 0.07 5.4 ± 3.8 6.3 ± 4.8 0.25
Data are given as mean values ± SD or as numbers (%).
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
IRA compared to IPAA in ulcerative colitis
3
analysis according to Kaplan–Meier using log-rank test was used for comparing failure-rates as well as for calculating cumulative cancer-incidence. All p-values were two-tailed and p-values less than 0.05 were considered significant. Calculations were performed with Statistica® statistical package (version 9, Stat-soft Inc., Tulsa, OK 74104, USA).
3. Results 3.1. Colectomy and Reconstructive Surgery The most common indications for colectomy were severe colitis and steroid dependency regardless of whether the latter mode of reconstruction was IRA or IPAA (Table 2). The annual distribution of reconstructive surgical procedures is shown in Fig. 1. Normally, two consultants jointly performed the surgeries. In total nine different consultants participated in the procedures, and of these nine, seven performed all of the IPAA and 99 of the IRA procedures. The number of IPAA procedures for each consultant ranged from nine to 84; corresponding figures for IRA were three to 42. Most procedures were open — thirty-nine were started as laparoscopies of which nine IRA were converted to open surgery. Operative data are summarized in Table 3. During follow-up time the diagnosis of UC was changed to Crohn's disease in five (4.8%) patients with IRA and in three (2.0%) with IPAA.
3.2. Complications Major postoperative complications occurred in 13 (12.4%) of patients and exclusively minor in 11 (10.4%) after IRA. Corresponding figures for IPAA were 19 (12.8%) and 31 (20.9%) (p = 0.91 and p = 0.02 respectively). After also including the closure of a defunctioning loop-ileostomy figures were 13 (12.4%) and 12 (11.4%) for IRA and 26 (17.6%) and 33 (22.3%) for IPAA (p = 0.26 and p = 0.01). Any complication after IRA or IPAA including stoma closure occurred in 23.8% and
Table 2 Indication for surgery in 253 patients undergoing colectomy for ulcerative colitis followed by reconstruction with either ileorectal anastomosis (IRA) or ileal pouch anal anastomosis (IPAA). Values are numbers (%).
Severe colitis Profuse hemorrhage Steroid dependency Cancer prevention without dysplasia Dysplasia including cancer in situ a Cancer Unclear Total a
IRA
IPAA
(n = 105)
(n = 148)
56 (53.3) 1 (0.9) 36 (34.3) 5 (4.8)
72 (48.6) 1 (0.7) 62 (41.8) – (0)
ns ns ns 0.02
3 (2.9)
7 (4.7)
ns
3 (2.9) 1 (0.9) 105
6 (4.0) – (0) 148
ns –
One patient with cancer in situ in each group.
p-Value
39.9% of patients respectively (p b 0.01). Types and numbers of patients with complications are shown in Table 4.
3.3. Failures Sixteen patients with IPAA and 14 with IRA were recorded as experiencing failures. Estimated cumulative failure rates after five and ten years were 10.1% and 24.1% for IRA and 6.1% and 18.6% for IPAA (p = 0.26) (Fig. 2). No difference in failure rates could be detected between the first part and the second part of the study-period. Separate analysis of patients fulfilling the criteria for IRA but receiving IPAA due to personal preferences (n = 35) revealed corresponding estimated cumulative failure rates of 3.2% and 16.3% after 5 and 10 years (p = 0.22). Reasons for failures are listed in Table 5. Of the patients with IRA six had a proctectomy with an ileostomy, five had a proctectomy with an IPAA, two had their rectum defunctioned with an ileostomy and one died from livermetastases without further surgery shortly after being diagnosed with rectal cancer. Seven patients with IPAA had an ileostomy after excision of their pelvic pouch, six had their pouch defunctioned with an ileostomy and three had their pouch converted into a continent Kock-pouch.
3.4. Cancer or Dysplasia Two cases (ages 40 and 34) of adenocarcinoma, both in the rectum, occurred in patients with IRA. Time from diagnosis and reconstructive surgery was 21/9 years and 12/11 years respectively. In patients with IPAA one case (age 25) of adenocarcinoma occurred in the ileoanal anastomosis and another (age 50) was found as a disseminated adenocarcinoma in the abdomen where the primary tumor was unidentified but the bowel as origin could not be excluded. Time from diagnosis and reconstructive surgery was 11/9 years and 2/1 year respectively. Out of the four patients diagnosed with cancer at follow-up three had neither cancer nor dysplasia in the colectomy specimen; one patient with IRA done as a one-stage procedure had cancer in the colectomy specimen as an unexpected finding. Five more patients with cancer or dysplasia in their colectomy specimens were reconstructed with IRA, one of whom developed dysplasia in the rectal mucosa requiring proctectomy. One case of low-grade dysplasia in a locally removed tubular adenoma was found at follow-up among patients with IRA; the patient was reluctant to have surgery and had not developed any further dysplasia at follow-up three years later. Thirteen patients reconstructed with IPAA had cancer or dysplasia in their colectomy specimens but none of those had dysplasia or cancer during follow-up. Estimated cumulative cancer-risk after 10, 20 and 25 year duration of disease was 0.0%, 2.1% and 8.7% for IRA and 0.7%, 1.8% and 1.8% for IPAA (p = 0.59). One patient with rectal carcinoma was alive with no signs of recurrence at follow-up (2.5 years); the others died from their cancer.
3.5. Functional Outcome Patients with an intact IRA at follow-up had no proctitis in 21 (23.1%) of cases, proctitis classified to BG 1 in 38 (41.8%), BG 2 in 23(25.3%), BG 3 in one case (1.1%) and in eight (8.8%) instances rectoscopy had not been performed. Patients with
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
4
P. Andersson et al.
Figure 1
Distribution of reconstructive operations after colectomy for ulcerative colitis 1992–2006
an intact IPAA had a history of episodic pouchitis defined as an episodic need for anti- or probiotic medication due to an increased stool-rate with or without endoscopic changes in 27 (20.5%) of cases. Chronic pouchitis defined as a need for continuous anti-, probiotic or anti-inflammatory medication or having had at least three relapses per year was present in 18 (13.6%) patients and unspecified dysfunction in eight (6.1%). Stool-rate daytime (median, range when recorded (IRA, n = 89 and IPAA, n = 108)) was 4 (1–11) and 5 (2–13) respectively (p = 0.03). There was no association between
IPAA
IRA.
stool-rate or unspecified dysfunction and postoperative complications neither major nor minor.
3.6. Medication At the time of follow-up, after exclusion of failures, 83 (91.2%) patients with IRA and nine (6.9) patients with IPAA were on some kind of inflammatory-moderating medication (p b 0.00001). Medication for patients with IRA was
Table 3 Operative data on 253 patients undergoing reconstructive surgery between 1992 and 2006 after colectomy for ulcerative colitis.
Laparoscopic approach (converted) Type of anastomosis Stapled (n) Hand sewn (n) Anastomotic height (cm) Blood loss (ml) Reconstruction only (n = 84/127) Concurrent colectomy (n = 18/21) a Operating-time (min) Previous colectomy (n = 85/127) Concurrent colectomy (n = 18/21) a Defunctioning ileostomy (n) Time to closure of ileostomy (days)
IRA
IPAA
p
(n = 105)
(n = 148)
29 (9)
1 (–)
102 (97.1) 3 (2.9) 17.2 ± 3.2
143 (96.6) 5 (3.4) –
ns
90.2 ± 86.6 227.8 ± 228.8
367.9 ± 312.4 316.7 ± 227.7
b 0.0001 ns
131.7 ± 44.5 186.7 ± 96.0 5 120.4 ± 19.6
194.1 ± 47.6 211.2 ± 60.8 142 113.2 ± 55.1
b 0.0001 0.04 – ns
–
Data are given as mean values ± SD or as n (%). a Data missing on 3 cases with IRA. Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
Complication
IRA primary reconstruction (all severities of complications) (n = 105)
IPAA primary reconstruction (all severities of complications) (n = 148)
IRA loop-closure (all severities of complications) (n = 5)
IPAA loop-closure (all severities of complications) (n = 142)
IRA reconstruction and loop-closure together (Clavien–Dindo ≥ 3)
IPAA reconstruction and loop-closure together (Clavien–Dindo ≥ 3)
Fluid-balance disturbance/dehydration Small bowel obstruction (complete/partial) Anastomotic leak Abdominal infection (deep/superficial) Hemorrhage/hematoma Prolonged ileus Small bowel perforation Urinary tract infection Bladder retention Compartment syndrome (leg) Enterocutaneous fistula Wound dehiscence Rectovaginal fistula Anal stenosis Thrombo-embolic episodes Pneumonia Dural puncture Pancreatitis Total number of complications p-Value
– 7 3 3 5 2 1 1 1 – – 1 – – – – – – 24
11 5 7 7 4 6 – 2 – 2 – 1 – – 2 1 1 1 50
– – – – – 1 – – – – – – – – – – – – 1
– 4 1 8 – 2 – – 1 – 2 – 1 1 1 – – – 21
– 1 3 2 5 – 1 – – – – 1 – – – – – – 13
1 5 7 1 4 1 – – – 1 – 1 1 1 2 1 – – 26
ns
ns
IRA compared to IPAA in ulcerative colitis
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
Table 4 Any kind of complication within 30 days of operation with ileorectal anastomosis (IRA) or ileal pouch anal anastomosis (IPAA) as well as severe complications (Clavien– Dindo ≥ 3) after primary reconstruction including the closure of an optional temporary diverting loop ileostomy. Data are given as number (n) of patients. One patient with IPAA had complications classified as Clavien–Dindo ≥ 3 on two occasions.
b 0.01
5
6
P. Andersson et al.
Fig. 2 Survival (months) of IRA or IPAA without proctectomy, excision of the pouch, permanent defunctioning or occurrence of cancer.
distributed as follows: 5-ASA as single medication (n = 62) or combined with azathioprine (n = 1) or topical corticosteroids as single medication (n = 5), combined with 5-ASA (n = 12), combined with 5-ASA and azathioprine (n = 1), combined with azathioprine only (n = 1) or combined with anti-TNF only (n = 1). Patients with IPAA were on antibiotics (n = 5),
Table 5 Causes of failure in patients primarily reconstructed with ileorectal anastomosis (IRA) or ileal pouch anal anastomosis (IPAA) after colectomy for ulcerative colitis. Values are numbers (%).
Dysfunction/frequent defecation Dysfunction/evacuation difficulties Dysfunction/short bowel after anastomotic leak Proctitis/intractable Proctitis/fistulas Incontinence/sphincter lesion Chronic pouchitis Anal stenosis Chronic pelvic infection Failed attempt for ileostomy closure Dysplasia Cancer/ileoanal anastomosis Cancer/rectal Cancer abdomen/unknown primary Unclear Total failures
IRA
IPAA
(n = 105)
(n = 148)
– – 1 (1.0)
5 (3.4) 2 (1.4) –
5 (4.8) 4 (3.8) – – – – – 1 (1.0) – 2 (1.9) – 1 (1.0) 14 (13.3)
– – 1 (0.7) 2 (1.4) 1 (0.7) 2 (1.4) 1 (0.7) – 1 (0.7) 1 (0.7) – 16 (10.8)
corticosteroids (n = 1) or azathioprine (n = 3). Furthermore IPAA-patients had a history of episodic antibiotics (n = 21) or probiotics (n = 12).
4. Discussion Here we report the hitherto largest case series comparing the long-term outcomes of IRA and IPAA for reconstructive surgery after colectomy for UC. To the best of our knowledge, no comparative study of IRA and IPAA reconstruction has been published since Parc and coworkers published their study in 1989.16 Parc compared patients operated with IPAA between 1983 and 1988 to a historical cohort operated with IRA at the same single institution between 1961 and 1982. In 2001 we launched a national randomized trial comparing IRA and IPAA but this trial was not possible to carry through due to the unwillingness among surgeons to include patients and also the unwillingness of patients to be enrolled in the study. Therefore we designed and carried out this retrospective study in order to elucidate still important surgical aspects of the two techniques, focusing particularly on IRA. Our data represent the experience from a single center as did the data from Parc's study, but in contrast with Parc, our data are based on operations performed by a limited number of surgeons during a defined time-period comprising both types of reconstruction. Baseline data between the two groups did not differ; there was a preponderance of men in both groups, a gender distribution in colectomy patients previously noted by others.17,18 The most common indications for colectomy in both groups were severe colitis and steroid-dependency. Cancer or dysplasia as indications was evenly distributed between the groups and, surprisingly, cancer prevention was significantly more common in the group later chosen for IRA. It
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
IRA compared to IPAA in ulcerative colitis could be anticipated that patients with dysplasia or cancer who were later reconstructed with IRA would be elderly but this was not always the case, and this might have contributed to one of the deaths. Several authors have claimed that one of the advantages with IRA is that the surgery is less complex than for IPAA.11,12,19 Operating time and bleeding were truly significantly less for IRA in our cohort although, particularly concerning operating time, the difference was not as marked as could have been expected, as the standard IRA procedure still required more than 2 h to complete. As concerns the primary procedure, major postoperative complications were equally frequent after IRA and IPAA but minor complications (Clavien–Dindo b 3) were more common after IPAA. These findings indicate that IRA is not as simple a procedure as is commonly believed. After the inclusion of loop-ileostomy closure the complication rate after IRA, however, was clearly lower than after IPAA but this was mainly due to a notably high rate of minor complications after IPAA. Ten year cumulative failure-rates for IRA and IPAA after surgery for UC are reported to vary between 26–31% and 6– 13% respectively.12,17,20,21 As it was not possible to make a randomized study this retrospective study is not based on the study of fully comparable groups of patients. A majority of patients could not be allocated to IRA because of the state of their rectum. The study reflects the results of a clinical feasibility praxis where patients considered suitable for IRA were selected in many cases for IRA and the rest were offered the standard procedure IPAA. Comparison of failure-rates, bearing these circumstances in mind, therefore has to be made with caution. In this study the estimated cumulative failure-rates after five and ten years for IRA and IPAA did not differ suggesting that with proper selection criteria and topical maintenance treatment of the rectum the failure rate after IRA can be kept to the same level as after IPAA. Our 10-year cumulative failure-rate of 18.6% for IPAA was, however, higher than previously reported.20 It might be speculated that this can be explained by the condition that our patient-population selected for this procedure had more extensive colorectal inflammation than patients in other studies, since most of our patients with no or only mild rectal inflammation were allocated to IRA. Modern reports on cumulative failure-rates after IRA for UC are sparse but the rates reported vary between 16–19% after five years and up to 31% after ten; figures in accordance with ours.12,21,22 Separate failure-analysis of 35 patients receiving an IPAA where the rectum was free of inflammation and where a true choice between IRA and IPAA existed did not deviate from either the rest of the IPAA-patients or from those receiving an IRA. Ever since Baker et al.2 in 1978 reported a cumulative cancer-risk of 6% after 20 years and rising to 18% after 35 years in patients operated with IRA, surgeons have had concerns regarding the method particularly with the advent of IPAA where the risk for malignant transformation seems to be very low.23,24 Recent figures from the Cleveland Clinic confirm a cancer-risk after IRA on an even higher level of 14% after 20 years.12 Two rectal cancers were found in our cohort resulting in a cumulative cancer-risk of 8.6% after 25 year duration of disease. The lower, but still substantial, risk in our cohort compared to the Cleveland Clinic cohort might at least partly be explained by another selection of patients as our patients were operated at a later date and also possibly by a
7 more consequent use of 5-ASA preparations.10 Even though no cancers were found among our patients with IRA before 10 years of disease duration, all in line with previous findings,2,22,25 this reinforces the belief that IRA has to be used with caution. Part of the high cancer-risk in our study is actually the effect of poor selection of patients and in addition insufficient surveillance at the home county hospital, which in one case led to death. The risk for poor or, as in the current case, non-compliance with follow-up protocols from both patients and health-care institutions is consequently a crucial factor that must be taken into account when selecting patients to IRA. However, IPAA seems to carry a risk for developing cancer as well; the cumulative cancer risk after 25 years in our limited cohort was 1.8%. It might be claimed that the patient with the disseminated abdominal adenocarcinoma should not have been included as the origin of the tumor could not definitely be determined, but even after exclusion of this patient a risk of 1.2% persisted. Unfortunately no true epidemiological registry-based studies on the risk for cancer after IPAA exist. Functional outcome regarding the number of loose stools was significantly better among patients with IRA; the median number of four bowel movements during daytime was less than those reported by others but might be due to retrospective data extracted from patient-records.12,22,26 As expected a vast majority of patients with IRA required topical maintenance treatment mainly of 5-ASA but steroids were also commonly used in order to control proctitis whereas for patients with IPAA only a small percentage needed regular medication. Although not part of this study, sexual function and fecundability are important issues regarding surgery for UC as the disease often affects young people. In familial adenomatous polyposis (FAP) female fecundability has been shown to stay close to normal after IRA but to be impaired after IPAA; the latter relationship is even more true for UC.7,27,28 Data are unfortunately lacking as regards fecundability after IRA in UC but there is reason to believe that the results for FAP also are applicable for UC. In our opinion, IRA is justified under certain conditions provided that no or only mild rectal inflammation is present or that this can be controlled with topical 5-ASA. Strict absence of dysplasia or cancer is mandatory if IRA is to be used except possibly in a few elderly patients with a limited life-expectancy. To young patients having a short duration of disease and with concerns regarding fecundability and sexual function IRA might be offered as an interim procedure, but they should be informed that an IPAA must be planned for in the not too distant future due to the established cancer-risk after more than 10 years of disease duration. Patients should tolerate anti-inflammatory topic medication in order to not only control pending proctitis but also reduce the cancer-risk to some extent. Finally and very importantly, patients have to fully understand the need for meticulous surveillance and agree to comply with a plan for at least yearly visits for rectal biopsies. Unless these conditions are fulfilled patients should be offered an IPAA.
Conflict of Interest Statement The authors declare no conflicts of interests. This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
8
P. Andersson et al.
Acknowledgments Mats Fredriksson, Linköping Academic Research Center, for statistical advice.
Statement of authorship PA ([email protected]) designed the study, acquired and analyzed the data as well as drafted, wrote and revised the manuscript. RN ([email protected]) acquired the data and revised the manuscript. JDS (johan.d.soderholm@ liu.se) designed the study and revised the manuscript. PM ([email protected]) designed the study, drafted and revised the manuscript. All authors approved the final manuscript.
References 1. Aylett SO. Conservative surgery in the treatment of ulcerative colitis. Br Med J 1953;2(4850):1348–51. 2. Baker WN, Glass RE, Ritchie JK, Aylett SO. Cancer of the rectum following colectomy and ileorectal anastomosis for ulcerative colitis. Br J Surg 1978;65(12):862–8. 3. Parks AG, Nicholls RJ. Proctocolectomy without ileostomy for ulcerative colitis. Br Med J 1978;2(6130):85–8. 4. McGuire BB, Brannigan AE, O'Connell PR. Ileal pouch-anal anastomosis. Br J Surg 2007;94(7):812–23. 5. McLaughlin SD, Clark SK, Tekkis PP, Ciclitira PJ, Nicholls RJ. Review article: restorative proctocolectomy, indications, management of complications and follow-up—a guide for gastroenterologists. Aliment Pharmacol Ther 2008;27(10):895–909. 6. Berndtsson I, Oresland T, Hulten L. Sexuality in patients with ulcerative colitis before and after restorative proctocolectomy: a prospective study. Scand J Gastroenterol 2004;39(4):374–9. 7. Ording Olsen K, Juul S, Berndtsson I, Oresland T, Laurberg S. Ulcerative colitis: female fecundity before diagnosis, during disease, and after surgery compared with a population sample. Gastroenterology 2002;122(1):15–9. 8. Landy J, Al-Hassi HO, McLaughlin SD, Knight SC, Ciclitira PJ, Nicholls RJ, et al. Etiology of pouchitis. Inflamm Bowel Dis 2012;18(6):1146–55. 9. Wasmuth HH, Trano G, Midtgard TM, Wibe A, Endreseth BH, Myrvold HE. Long-term function after ileal pouch-anal anastomosis — function does not deteriorate with time. Colorectal Dis 2010;12(10 Online):e283–90. 10. Pinczowski D, Ekbom A, Baron J, Yuen J, Adami HO. Risk factors for colorectal cancer in patients with ulcerative colitis: a case– control study. Gastroenterology 1994;107(1):117–20. 11. Börjesson L, Lundstam U, Öresland T, Brevinge H, Hultén L. The place for colectomy and ileorectal anastomosis: a valid surgical option for ulcerative colitis? Tech Coloproctol 2006;10(3):237–41. 12. da Luz Moreira A, Kiran RP, Lavery I. Clinical outcomes of ileorectal anastomosis for ulcerative colitis. Br J Surg 2010;97(1): 65–9.
13. Lennard-Jones JE. Classification of inflammatory bowel disease. Scand J Gastroenterol Suppl 1989;170:2–6. 14. Baron JH, Connell AM, Lennard-Jones JE. Variation between observers in describing mucosal appearances in proctocolitis. Br Med J 1964;1(5375):89–92. 15. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240(2): 205–13. 16. Parc R, Legrand M, Frileux P, Tiret E, Ratelle R. Comparative clinical results of ileal-pouch anal anastomosis and ileorectal anastomosis in ulcerative colitis. Hepatogastroenterology 1989;36(4):235–9. 17. Tulchinsky H, Hawley PR, Nicholls J. Long-term failure after restorative proctocolectomy for ulcerative colitis. Ann Surg 2003;238(2):229–34. 18. Wasmuth HH, Trano G, Endreseth B, Rydning A, Wibe A, Myrvold HE. Long-term surgical load in patients with ileal pouch-anal anastomosis. Colorectal Dis 2009;11(7):711–8. 19. Elton C, Makin G, Hitos K, Cohen CR. Mortality, morbidity and functional outcome after ileorectal anastomosis. Br J Surg 2003;90(1):59–65. 20. Hahnloser D, Pemberton JH, Wolff BG, Larson DR, Crownhart BS, Dozois RR. Results at up to 20 years after ileal pouch-anal anastomosis for chronic ulcerative colitis. Br J Surg 2007;94(3): 333–40. 21. Lepistö A, Järvinen HJ. Fate of the rectum after colectomy with ileorectal anastomosis in ulcerative colitis. Scand J Surg 2005;94(1):40–2. 22. Pastore RL, Wolff BG, Hodge D. Total abdominal colectomy and ileorectal anastomosis for inflammatory bowel disease. Dis Colon Rectum 1997;40(12):1455–64. 23. O'Riordan JM, Kirsch R, Mohseni M, McLeod RS, Cohen Z. Long-term risk of adenocarcinoma post-ileal pouch-anal anastomosis for ulcerative colitis: report of two cases and review of the literature. Int J Colorectal Dis 2012;27(3): 405–10. 24. Egan L, D'Inca R, Jess T, Pellino G, Carbonnel F, Bokemeyer B, et al. Non-colorectal intestinal tract carcinomas in inflammatory bowel disease: results of the 3rd ECCO Pathogenesis Scientific Workshop (II). J Crohns Colitis 2013. http: //dx.doi.org/10.1016/j.crohns.2013.04.009 [published Online First: Epub Date]. 25. Mann CV. Total colectomy and ileorectal anastomosis for ulcerative colitis. World J Surg 1988;12(2):155–9. 26. Saito Y, Sawada T, Tsuno N, Watanabe Y, Higuchi M, Shinozaki M, et al. Total colectomy and ileorectal anastomosis in ulcerative colitis. J Gastroenterol 1995;30(Suppl 8):131–4. 27. Olsen KO, Juul S, Bulow S, Järvinen HJ, Bakka A, Björk J, et al. Female fecundity before and after operation for familial adenomatous polyposis. Br J Surg 2003;90(2):227–31. 28. Nieuwenhuis MH, Douma KF, Bleiker EM, Bemelman WA, Aaronson NK, Vasen HF. Female fertility after colorectal surgery for familial adenomatous polyposis: a nationwide cross-sectional study. Ann Surg 2010;252(2):341–4.
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
Available online at www.sciencedirect.com
ScienceDirect
Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience ☆ Peter Andersson ⁎, Rickard Norblad, Johan D. Söderholm, Pär Myrelid Department of Surgery, County Council of Östergötland, Linköping, Sweden Division of Surgery, Department of Clinical and Experimental Medicine, Faculty of Health Sciences, Linköping University, Sweden
Received 1 October 2013; received in revised form 12 November 2013; accepted 13 November 2013 KEYWORDS Ulcerative colitis; IRA; IPAA; Complications; Failure
Abstract Introduction: Ileal pouch anal anastomosis (IPAA) is the standard procedure for reconstruction after colectomy for ulcerative colitis (UC). However, ileorectal anastomosis (IRA) as an alternative has, recently experienced a revival. This study from a single center compares the clinical outcomes of these procedures. Methods: From 1992 to 2006, 253 patients consecutively underwent either IRA (n=105) or IPAA (n=148). Selection to either procedure was determined on the basis of rectal inflammation, presence of dysplasia/cancer or patient preferences. Patient-records were retrospectively evaluated. Mean follow-up time was 5.4 and 6.3 years respectively. Results: Major postoperative complications occurred in 12.4% of patients after IRA and in 12.8% after IPAA (ns). Complications of any kind after IRA or IPAA, even including subsequent stoma-closure, occurred in 23.8% and 39.9% respectively (pb 0.01). Estimated cumulative failure rates after 5 and 10 years were 10.1% and 24.1% for IRA and 6.1% and 18.6% for IPAA respectively (ns). The most common cause for failure was intractable proctitis (4.8%) and unspecified dysfunction (4.8%) respectively. At follow-up 76.9% of patients with IRA had proctitis and 34.1% with IPAA had pouchitis. Estimated cumulative cancer-risk after 10, 20 and 25 year duration of disease was 0.0%, 2.1% and 8.7% for IRA. Figures for IPAA were 0.7%, 1.8% and 1.8% (ns).
☆ Conference presentations: Swedish surgical society; Gothenburg, 2010. ECCO annual meeting; Vienna, 2013. ⁎ Corresponding author at: Dept. of Surgery, Linköping University Hospital, S-581 85 Linköping, Sweden. Tel.: + 46 10 1033022; fax: + 46 101033570. E-mail address: [email protected] (P. Andersson). 1873-9946/$ - see front matter © 2013 European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.crohns.2013.11.014 Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
2
P. Andersson et al. Conclusion: Failure-rates did not significantly differ between patients operated with IRA or IPAA. Patients operated with IPAA had a higher cumulative number of postoperative complications. The high long-term cancer-risk after IRA indicates that this procedure should be an interim solution in younger patients. © 2013 European Crohn's and Colitis Organisation. Published by Elsevier B.V. All rights reserved.
1. Introduction Reconstructive surgery after colectomy for ulcerative colitis (UC) was launched by Aylett and colleagues around 1950 by introduction of ileorectal anastomosis (IRA).1 Concerns about the method were raised, however, after reports on a substantial risk for the development of cancer in the remaining rectum.2 Due to the advent of the ileal pouch anal anastomosis (IPAA) in the 1970s, the use of IRA was abandoned in most cases of UC.3 Since then IPAA has become the standard method of reconstruction after surgery for UC.4,5 Long-term experience with IPAA has, however, revealed some morbidity associated with the procedure as regards impaired continence and sexual function including fecundability as well as pouchitis.6–9 This together with the knowledge that topical anti-inflammatory treatment and meticulous surveillance of the rectum might reduce the risk for cancer has led to the advocacy of IRA in selected cases.10–12 Surgeons in Scandinavia have been particularly willing to adopt this approach. We report the failure rates, cancer risks, functional outcome and need for medication in patients operated on with IRA at our institution in comparison with those operated on with IPAA during the same time period.
2. Patients and Methods Linköping University Hospital is a tertiary referral hospital serving a population of 1.1 million. From 1992 to 2006, 253 consecutive patients with ulcerative colitis (UC) were operated at the surgical department with either ileorectal anastomosis (IRA) or ileal pouch anal anastomosis (IPAA) as the primary reconstructive procedure after colectomy. The diagnosis of UC was based on clinical history, endoscopy and microscopic examination of the colectomy specimen according to criteria of Lennard-Jones.13 Patient selection to either procedure was determined by the degree of inflammation in the rectal mucosa, presence of dysplasia/cancer or patient preferences. Inflammatory activity at rectoscopy confined to a modified Baron−Ginsburg (BG) 0-1 after topical treatment with 5-aminosalicylic acid (5-ASA) and absence of dysplasia or cancer made patients eligible for IRA.14 Some patients were accepted for IRA, however, in spite of the presence of dysplasia or even cancer according to their personal preferences. With the exception of these patients, patients not suitable for IRA due to medical reasons or personal preferences were chosen for IPAA. IRA was performed in 105 patients (age 10–75, 34 women) and IPAA in 148 (age 13–65, 52 women). Background data on the patients are presented in Table 1. Patients were regularly seen at least once at the out-patient clinic during the follow-up period until the end of 2008, the return to their county hospital or death. Follow-up, which was recorded in
our quality control program, consisted of a clinical examination including an evaluation of function and, when symptoms demanded or the duration of disease was longer than ten years in patients with IRA, endoscopy with biopsies. Follow-up after IRA was scheduled yearly and after IPAA at least every second year. All records have been reviewed retrospectively to collect peri- and postoperative data including 30-day complication rate as well as failure rates, development of dysplasia or cancer, functional outcome, presence of proctitis or pouchitis and medication at follow-up. Complications after surgery were graded according to Clavien–Dindo where complications graded three or more were considered major.15 Patients were classified on the basis of the most serious complication in the case of more than one complication. Failure was defined as either one of the following events — proctectomy, excision of the ileal pouch, and permanent defunctioning with an ileostomy — or the occurrence of cancer where linkage to the previous surgical procedure could not be excluded. The study was approved by the regional ethics committee in Linköping, Sweden.
2.1. Statistical Methods Values are given as mean and standard-deviations (SD) unless otherwise stated. The Mann–Whitney-U test and Chi-square test were used when appropriate. Survival
Table 1 Background data on 253 patients undergoing colectomy for ulcerative colitis followed by reconstruction with either ileorectal anastomosis (IRA) or ileal pouch anal anastomosis (IPAA).
Sex Male Female Age at diagnosis (years) Age at colectomy (years) Age at reconstructive surgery (years) Duration of disease at reconstruction (years) Age at follow-up (years) Follow-up time (years)
IRA
IPAA
n = 105
n = 148
p-Value
0.64 71 (68) 96 (65) 34 (32) 52 (35) 26.9 ± 11.4 26.8 ± 10.3 0.93 33.3 ± 14.7 34.2 ± 11.4 0.23 34.2 ± 14.7 35.4 ± 11.3 0.16 7.2 ± 7.5
8.6 ± 7.7
0.06
39.5 ± 15.1 41.7 ± 12.2 0.07 5.4 ± 3.8 6.3 ± 4.8 0.25
Data are given as mean values ± SD or as numbers (%).
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
IRA compared to IPAA in ulcerative colitis
3
analysis according to Kaplan–Meier using log-rank test was used for comparing failure-rates as well as for calculating cumulative cancer-incidence. All p-values were two-tailed and p-values less than 0.05 were considered significant. Calculations were performed with Statistica® statistical package (version 9, Stat-soft Inc., Tulsa, OK 74104, USA).
3. Results 3.1. Colectomy and Reconstructive Surgery The most common indications for colectomy were severe colitis and steroid dependency regardless of whether the latter mode of reconstruction was IRA or IPAA (Table 2). The annual distribution of reconstructive surgical procedures is shown in Fig. 1. Normally, two consultants jointly performed the surgeries. In total nine different consultants participated in the procedures, and of these nine, seven performed all of the IPAA and 99 of the IRA procedures. The number of IPAA procedures for each consultant ranged from nine to 84; corresponding figures for IRA were three to 42. Most procedures were open — thirty-nine were started as laparoscopies of which nine IRA were converted to open surgery. Operative data are summarized in Table 3. During follow-up time the diagnosis of UC was changed to Crohn's disease in five (4.8%) patients with IRA and in three (2.0%) with IPAA.
3.2. Complications Major postoperative complications occurred in 13 (12.4%) of patients and exclusively minor in 11 (10.4%) after IRA. Corresponding figures for IPAA were 19 (12.8%) and 31 (20.9%) (p = 0.91 and p = 0.02 respectively). After also including the closure of a defunctioning loop-ileostomy figures were 13 (12.4%) and 12 (11.4%) for IRA and 26 (17.6%) and 33 (22.3%) for IPAA (p = 0.26 and p = 0.01). Any complication after IRA or IPAA including stoma closure occurred in 23.8% and
Table 2 Indication for surgery in 253 patients undergoing colectomy for ulcerative colitis followed by reconstruction with either ileorectal anastomosis (IRA) or ileal pouch anal anastomosis (IPAA). Values are numbers (%).
Severe colitis Profuse hemorrhage Steroid dependency Cancer prevention without dysplasia Dysplasia including cancer in situ a Cancer Unclear Total a
IRA
IPAA
(n = 105)
(n = 148)
56 (53.3) 1 (0.9) 36 (34.3) 5 (4.8)
72 (48.6) 1 (0.7) 62 (41.8) – (0)
ns ns ns 0.02
3 (2.9)
7 (4.7)
ns
3 (2.9) 1 (0.9) 105
6 (4.0) – (0) 148
ns –
One patient with cancer in situ in each group.
p-Value
39.9% of patients respectively (p b 0.01). Types and numbers of patients with complications are shown in Table 4.
3.3. Failures Sixteen patients with IPAA and 14 with IRA were recorded as experiencing failures. Estimated cumulative failure rates after five and ten years were 10.1% and 24.1% for IRA and 6.1% and 18.6% for IPAA (p = 0.26) (Fig. 2). No difference in failure rates could be detected between the first part and the second part of the study-period. Separate analysis of patients fulfilling the criteria for IRA but receiving IPAA due to personal preferences (n = 35) revealed corresponding estimated cumulative failure rates of 3.2% and 16.3% after 5 and 10 years (p = 0.22). Reasons for failures are listed in Table 5. Of the patients with IRA six had a proctectomy with an ileostomy, five had a proctectomy with an IPAA, two had their rectum defunctioned with an ileostomy and one died from livermetastases without further surgery shortly after being diagnosed with rectal cancer. Seven patients with IPAA had an ileostomy after excision of their pelvic pouch, six had their pouch defunctioned with an ileostomy and three had their pouch converted into a continent Kock-pouch.
3.4. Cancer or Dysplasia Two cases (ages 40 and 34) of adenocarcinoma, both in the rectum, occurred in patients with IRA. Time from diagnosis and reconstructive surgery was 21/9 years and 12/11 years respectively. In patients with IPAA one case (age 25) of adenocarcinoma occurred in the ileoanal anastomosis and another (age 50) was found as a disseminated adenocarcinoma in the abdomen where the primary tumor was unidentified but the bowel as origin could not be excluded. Time from diagnosis and reconstructive surgery was 11/9 years and 2/1 year respectively. Out of the four patients diagnosed with cancer at follow-up three had neither cancer nor dysplasia in the colectomy specimen; one patient with IRA done as a one-stage procedure had cancer in the colectomy specimen as an unexpected finding. Five more patients with cancer or dysplasia in their colectomy specimens were reconstructed with IRA, one of whom developed dysplasia in the rectal mucosa requiring proctectomy. One case of low-grade dysplasia in a locally removed tubular adenoma was found at follow-up among patients with IRA; the patient was reluctant to have surgery and had not developed any further dysplasia at follow-up three years later. Thirteen patients reconstructed with IPAA had cancer or dysplasia in their colectomy specimens but none of those had dysplasia or cancer during follow-up. Estimated cumulative cancer-risk after 10, 20 and 25 year duration of disease was 0.0%, 2.1% and 8.7% for IRA and 0.7%, 1.8% and 1.8% for IPAA (p = 0.59). One patient with rectal carcinoma was alive with no signs of recurrence at follow-up (2.5 years); the others died from their cancer.
3.5. Functional Outcome Patients with an intact IRA at follow-up had no proctitis in 21 (23.1%) of cases, proctitis classified to BG 1 in 38 (41.8%), BG 2 in 23(25.3%), BG 3 in one case (1.1%) and in eight (8.8%) instances rectoscopy had not been performed. Patients with
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
4
P. Andersson et al.
Figure 1
Distribution of reconstructive operations after colectomy for ulcerative colitis 1992–2006
an intact IPAA had a history of episodic pouchitis defined as an episodic need for anti- or probiotic medication due to an increased stool-rate with or without endoscopic changes in 27 (20.5%) of cases. Chronic pouchitis defined as a need for continuous anti-, probiotic or anti-inflammatory medication or having had at least three relapses per year was present in 18 (13.6%) patients and unspecified dysfunction in eight (6.1%). Stool-rate daytime (median, range when recorded (IRA, n = 89 and IPAA, n = 108)) was 4 (1–11) and 5 (2–13) respectively (p = 0.03). There was no association between
IPAA
IRA.
stool-rate or unspecified dysfunction and postoperative complications neither major nor minor.
3.6. Medication At the time of follow-up, after exclusion of failures, 83 (91.2%) patients with IRA and nine (6.9) patients with IPAA were on some kind of inflammatory-moderating medication (p b 0.00001). Medication for patients with IRA was
Table 3 Operative data on 253 patients undergoing reconstructive surgery between 1992 and 2006 after colectomy for ulcerative colitis.
Laparoscopic approach (converted) Type of anastomosis Stapled (n) Hand sewn (n) Anastomotic height (cm) Blood loss (ml) Reconstruction only (n = 84/127) Concurrent colectomy (n = 18/21) a Operating-time (min) Previous colectomy (n = 85/127) Concurrent colectomy (n = 18/21) a Defunctioning ileostomy (n) Time to closure of ileostomy (days)
IRA
IPAA
p
(n = 105)
(n = 148)
29 (9)
1 (–)
102 (97.1) 3 (2.9) 17.2 ± 3.2
143 (96.6) 5 (3.4) –
ns
90.2 ± 86.6 227.8 ± 228.8
367.9 ± 312.4 316.7 ± 227.7
b 0.0001 ns
131.7 ± 44.5 186.7 ± 96.0 5 120.4 ± 19.6
194.1 ± 47.6 211.2 ± 60.8 142 113.2 ± 55.1
b 0.0001 0.04 – ns
–
Data are given as mean values ± SD or as n (%). a Data missing on 3 cases with IRA. Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
Complication
IRA primary reconstruction (all severities of complications) (n = 105)
IPAA primary reconstruction (all severities of complications) (n = 148)
IRA loop-closure (all severities of complications) (n = 5)
IPAA loop-closure (all severities of complications) (n = 142)
IRA reconstruction and loop-closure together (Clavien–Dindo ≥ 3)
IPAA reconstruction and loop-closure together (Clavien–Dindo ≥ 3)
Fluid-balance disturbance/dehydration Small bowel obstruction (complete/partial) Anastomotic leak Abdominal infection (deep/superficial) Hemorrhage/hematoma Prolonged ileus Small bowel perforation Urinary tract infection Bladder retention Compartment syndrome (leg) Enterocutaneous fistula Wound dehiscence Rectovaginal fistula Anal stenosis Thrombo-embolic episodes Pneumonia Dural puncture Pancreatitis Total number of complications p-Value
– 7 3 3 5 2 1 1 1 – – 1 – – – – – – 24
11 5 7 7 4 6 – 2 – 2 – 1 – – 2 1 1 1 50
– – – – – 1 – – – – – – – – – – – – 1
– 4 1 8 – 2 – – 1 – 2 – 1 1 1 – – – 21
– 1 3 2 5 – 1 – – – – 1 – – – – – – 13
1 5 7 1 4 1 – – – 1 – 1 1 1 2 1 – – 26
ns
ns
IRA compared to IPAA in ulcerative colitis
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
Table 4 Any kind of complication within 30 days of operation with ileorectal anastomosis (IRA) or ileal pouch anal anastomosis (IPAA) as well as severe complications (Clavien– Dindo ≥ 3) after primary reconstruction including the closure of an optional temporary diverting loop ileostomy. Data are given as number (n) of patients. One patient with IPAA had complications classified as Clavien–Dindo ≥ 3 on two occasions.
b 0.01
5
6
P. Andersson et al.
Fig. 2 Survival (months) of IRA or IPAA without proctectomy, excision of the pouch, permanent defunctioning or occurrence of cancer.
distributed as follows: 5-ASA as single medication (n = 62) or combined with azathioprine (n = 1) or topical corticosteroids as single medication (n = 5), combined with 5-ASA (n = 12), combined with 5-ASA and azathioprine (n = 1), combined with azathioprine only (n = 1) or combined with anti-TNF only (n = 1). Patients with IPAA were on antibiotics (n = 5),
Table 5 Causes of failure in patients primarily reconstructed with ileorectal anastomosis (IRA) or ileal pouch anal anastomosis (IPAA) after colectomy for ulcerative colitis. Values are numbers (%).
Dysfunction/frequent defecation Dysfunction/evacuation difficulties Dysfunction/short bowel after anastomotic leak Proctitis/intractable Proctitis/fistulas Incontinence/sphincter lesion Chronic pouchitis Anal stenosis Chronic pelvic infection Failed attempt for ileostomy closure Dysplasia Cancer/ileoanal anastomosis Cancer/rectal Cancer abdomen/unknown primary Unclear Total failures
IRA
IPAA
(n = 105)
(n = 148)
– – 1 (1.0)
5 (3.4) 2 (1.4) –
5 (4.8) 4 (3.8) – – – – – 1 (1.0) – 2 (1.9) – 1 (1.0) 14 (13.3)
– – 1 (0.7) 2 (1.4) 1 (0.7) 2 (1.4) 1 (0.7) – 1 (0.7) 1 (0.7) – 16 (10.8)
corticosteroids (n = 1) or azathioprine (n = 3). Furthermore IPAA-patients had a history of episodic antibiotics (n = 21) or probiotics (n = 12).
4. Discussion Here we report the hitherto largest case series comparing the long-term outcomes of IRA and IPAA for reconstructive surgery after colectomy for UC. To the best of our knowledge, no comparative study of IRA and IPAA reconstruction has been published since Parc and coworkers published their study in 1989.16 Parc compared patients operated with IPAA between 1983 and 1988 to a historical cohort operated with IRA at the same single institution between 1961 and 1982. In 2001 we launched a national randomized trial comparing IRA and IPAA but this trial was not possible to carry through due to the unwillingness among surgeons to include patients and also the unwillingness of patients to be enrolled in the study. Therefore we designed and carried out this retrospective study in order to elucidate still important surgical aspects of the two techniques, focusing particularly on IRA. Our data represent the experience from a single center as did the data from Parc's study, but in contrast with Parc, our data are based on operations performed by a limited number of surgeons during a defined time-period comprising both types of reconstruction. Baseline data between the two groups did not differ; there was a preponderance of men in both groups, a gender distribution in colectomy patients previously noted by others.17,18 The most common indications for colectomy in both groups were severe colitis and steroid-dependency. Cancer or dysplasia as indications was evenly distributed between the groups and, surprisingly, cancer prevention was significantly more common in the group later chosen for IRA. It
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
IRA compared to IPAA in ulcerative colitis could be anticipated that patients with dysplasia or cancer who were later reconstructed with IRA would be elderly but this was not always the case, and this might have contributed to one of the deaths. Several authors have claimed that one of the advantages with IRA is that the surgery is less complex than for IPAA.11,12,19 Operating time and bleeding were truly significantly less for IRA in our cohort although, particularly concerning operating time, the difference was not as marked as could have been expected, as the standard IRA procedure still required more than 2 h to complete. As concerns the primary procedure, major postoperative complications were equally frequent after IRA and IPAA but minor complications (Clavien–Dindo b 3) were more common after IPAA. These findings indicate that IRA is not as simple a procedure as is commonly believed. After the inclusion of loop-ileostomy closure the complication rate after IRA, however, was clearly lower than after IPAA but this was mainly due to a notably high rate of minor complications after IPAA. Ten year cumulative failure-rates for IRA and IPAA after surgery for UC are reported to vary between 26–31% and 6– 13% respectively.12,17,20,21 As it was not possible to make a randomized study this retrospective study is not based on the study of fully comparable groups of patients. A majority of patients could not be allocated to IRA because of the state of their rectum. The study reflects the results of a clinical feasibility praxis where patients considered suitable for IRA were selected in many cases for IRA and the rest were offered the standard procedure IPAA. Comparison of failure-rates, bearing these circumstances in mind, therefore has to be made with caution. In this study the estimated cumulative failure-rates after five and ten years for IRA and IPAA did not differ suggesting that with proper selection criteria and topical maintenance treatment of the rectum the failure rate after IRA can be kept to the same level as after IPAA. Our 10-year cumulative failure-rate of 18.6% for IPAA was, however, higher than previously reported.20 It might be speculated that this can be explained by the condition that our patient-population selected for this procedure had more extensive colorectal inflammation than patients in other studies, since most of our patients with no or only mild rectal inflammation were allocated to IRA. Modern reports on cumulative failure-rates after IRA for UC are sparse but the rates reported vary between 16–19% after five years and up to 31% after ten; figures in accordance with ours.12,21,22 Separate failure-analysis of 35 patients receiving an IPAA where the rectum was free of inflammation and where a true choice between IRA and IPAA existed did not deviate from either the rest of the IPAA-patients or from those receiving an IRA. Ever since Baker et al.2 in 1978 reported a cumulative cancer-risk of 6% after 20 years and rising to 18% after 35 years in patients operated with IRA, surgeons have had concerns regarding the method particularly with the advent of IPAA where the risk for malignant transformation seems to be very low.23,24 Recent figures from the Cleveland Clinic confirm a cancer-risk after IRA on an even higher level of 14% after 20 years.12 Two rectal cancers were found in our cohort resulting in a cumulative cancer-risk of 8.6% after 25 year duration of disease. The lower, but still substantial, risk in our cohort compared to the Cleveland Clinic cohort might at least partly be explained by another selection of patients as our patients were operated at a later date and also possibly by a
7 more consequent use of 5-ASA preparations.10 Even though no cancers were found among our patients with IRA before 10 years of disease duration, all in line with previous findings,2,22,25 this reinforces the belief that IRA has to be used with caution. Part of the high cancer-risk in our study is actually the effect of poor selection of patients and in addition insufficient surveillance at the home county hospital, which in one case led to death. The risk for poor or, as in the current case, non-compliance with follow-up protocols from both patients and health-care institutions is consequently a crucial factor that must be taken into account when selecting patients to IRA. However, IPAA seems to carry a risk for developing cancer as well; the cumulative cancer risk after 25 years in our limited cohort was 1.8%. It might be claimed that the patient with the disseminated abdominal adenocarcinoma should not have been included as the origin of the tumor could not definitely be determined, but even after exclusion of this patient a risk of 1.2% persisted. Unfortunately no true epidemiological registry-based studies on the risk for cancer after IPAA exist. Functional outcome regarding the number of loose stools was significantly better among patients with IRA; the median number of four bowel movements during daytime was less than those reported by others but might be due to retrospective data extracted from patient-records.12,22,26 As expected a vast majority of patients with IRA required topical maintenance treatment mainly of 5-ASA but steroids were also commonly used in order to control proctitis whereas for patients with IPAA only a small percentage needed regular medication. Although not part of this study, sexual function and fecundability are important issues regarding surgery for UC as the disease often affects young people. In familial adenomatous polyposis (FAP) female fecundability has been shown to stay close to normal after IRA but to be impaired after IPAA; the latter relationship is even more true for UC.7,27,28 Data are unfortunately lacking as regards fecundability after IRA in UC but there is reason to believe that the results for FAP also are applicable for UC. In our opinion, IRA is justified under certain conditions provided that no or only mild rectal inflammation is present or that this can be controlled with topical 5-ASA. Strict absence of dysplasia or cancer is mandatory if IRA is to be used except possibly in a few elderly patients with a limited life-expectancy. To young patients having a short duration of disease and with concerns regarding fecundability and sexual function IRA might be offered as an interim procedure, but they should be informed that an IPAA must be planned for in the not too distant future due to the established cancer-risk after more than 10 years of disease duration. Patients should tolerate anti-inflammatory topic medication in order to not only control pending proctitis but also reduce the cancer-risk to some extent. Finally and very importantly, patients have to fully understand the need for meticulous surveillance and agree to comply with a plan for at least yearly visits for rectal biopsies. Unless these conditions are fulfilled patients should be offered an IPAA.
Conflict of Interest Statement The authors declare no conflicts of interests. This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014
8
P. Andersson et al.
Acknowledgments Mats Fredriksson, Linköping Academic Research Center, for statistical advice.
Statement of authorship PA ([email protected]) designed the study, acquired and analyzed the data as well as drafted, wrote and revised the manuscript. RN ([email protected]) acquired the data and revised the manuscript. JDS (johan.d.soderholm@ liu.se) designed the study and revised the manuscript. PM ([email protected]) designed the study, drafted and revised the manuscript. All authors approved the final manuscript.
References 1. Aylett SO. Conservative surgery in the treatment of ulcerative colitis. Br Med J 1953;2(4850):1348–51. 2. Baker WN, Glass RE, Ritchie JK, Aylett SO. Cancer of the rectum following colectomy and ileorectal anastomosis for ulcerative colitis. Br J Surg 1978;65(12):862–8. 3. Parks AG, Nicholls RJ. Proctocolectomy without ileostomy for ulcerative colitis. Br Med J 1978;2(6130):85–8. 4. McGuire BB, Brannigan AE, O'Connell PR. Ileal pouch-anal anastomosis. Br J Surg 2007;94(7):812–23. 5. McLaughlin SD, Clark SK, Tekkis PP, Ciclitira PJ, Nicholls RJ. Review article: restorative proctocolectomy, indications, management of complications and follow-up—a guide for gastroenterologists. Aliment Pharmacol Ther 2008;27(10):895–909. 6. Berndtsson I, Oresland T, Hulten L. Sexuality in patients with ulcerative colitis before and after restorative proctocolectomy: a prospective study. Scand J Gastroenterol 2004;39(4):374–9. 7. Ording Olsen K, Juul S, Berndtsson I, Oresland T, Laurberg S. Ulcerative colitis: female fecundity before diagnosis, during disease, and after surgery compared with a population sample. Gastroenterology 2002;122(1):15–9. 8. Landy J, Al-Hassi HO, McLaughlin SD, Knight SC, Ciclitira PJ, Nicholls RJ, et al. Etiology of pouchitis. Inflamm Bowel Dis 2012;18(6):1146–55. 9. Wasmuth HH, Trano G, Midtgard TM, Wibe A, Endreseth BH, Myrvold HE. Long-term function after ileal pouch-anal anastomosis — function does not deteriorate with time. Colorectal Dis 2010;12(10 Online):e283–90. 10. Pinczowski D, Ekbom A, Baron J, Yuen J, Adami HO. Risk factors for colorectal cancer in patients with ulcerative colitis: a case– control study. Gastroenterology 1994;107(1):117–20. 11. Börjesson L, Lundstam U, Öresland T, Brevinge H, Hultén L. The place for colectomy and ileorectal anastomosis: a valid surgical option for ulcerative colitis? Tech Coloproctol 2006;10(3):237–41. 12. da Luz Moreira A, Kiran RP, Lavery I. Clinical outcomes of ileorectal anastomosis for ulcerative colitis. Br J Surg 2010;97(1): 65–9.
13. Lennard-Jones JE. Classification of inflammatory bowel disease. Scand J Gastroenterol Suppl 1989;170:2–6. 14. Baron JH, Connell AM, Lennard-Jones JE. Variation between observers in describing mucosal appearances in proctocolitis. Br Med J 1964;1(5375):89–92. 15. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg 2004;240(2): 205–13. 16. Parc R, Legrand M, Frileux P, Tiret E, Ratelle R. Comparative clinical results of ileal-pouch anal anastomosis and ileorectal anastomosis in ulcerative colitis. Hepatogastroenterology 1989;36(4):235–9. 17. Tulchinsky H, Hawley PR, Nicholls J. Long-term failure after restorative proctocolectomy for ulcerative colitis. Ann Surg 2003;238(2):229–34. 18. Wasmuth HH, Trano G, Endreseth B, Rydning A, Wibe A, Myrvold HE. Long-term surgical load in patients with ileal pouch-anal anastomosis. Colorectal Dis 2009;11(7):711–8. 19. Elton C, Makin G, Hitos K, Cohen CR. Mortality, morbidity and functional outcome after ileorectal anastomosis. Br J Surg 2003;90(1):59–65. 20. Hahnloser D, Pemberton JH, Wolff BG, Larson DR, Crownhart BS, Dozois RR. Results at up to 20 years after ileal pouch-anal anastomosis for chronic ulcerative colitis. Br J Surg 2007;94(3): 333–40. 21. Lepistö A, Järvinen HJ. Fate of the rectum after colectomy with ileorectal anastomosis in ulcerative colitis. Scand J Surg 2005;94(1):40–2. 22. Pastore RL, Wolff BG, Hodge D. Total abdominal colectomy and ileorectal anastomosis for inflammatory bowel disease. Dis Colon Rectum 1997;40(12):1455–64. 23. O'Riordan JM, Kirsch R, Mohseni M, McLeod RS, Cohen Z. Long-term risk of adenocarcinoma post-ileal pouch-anal anastomosis for ulcerative colitis: report of two cases and review of the literature. Int J Colorectal Dis 2012;27(3): 405–10. 24. Egan L, D'Inca R, Jess T, Pellino G, Carbonnel F, Bokemeyer B, et al. Non-colorectal intestinal tract carcinomas in inflammatory bowel disease: results of the 3rd ECCO Pathogenesis Scientific Workshop (II). J Crohns Colitis 2013. http: //dx.doi.org/10.1016/j.crohns.2013.04.009 [published Online First: Epub Date]. 25. Mann CV. Total colectomy and ileorectal anastomosis for ulcerative colitis. World J Surg 1988;12(2):155–9. 26. Saito Y, Sawada T, Tsuno N, Watanabe Y, Higuchi M, Shinozaki M, et al. Total colectomy and ileorectal anastomosis in ulcerative colitis. J Gastroenterol 1995;30(Suppl 8):131–4. 27. Olsen KO, Juul S, Bulow S, Järvinen HJ, Bakka A, Björk J, et al. Female fecundity before and after operation for familial adenomatous polyposis. Br J Surg 2003;90(2):227–31. 28. Nieuwenhuis MH, Douma KF, Bleiker EM, Bemelman WA, Aaronson NK, Vasen HF. Female fertility after colorectal surgery for familial adenomatous polyposis: a nationwide cross-sectional study. Ann Surg 2010;252(2):341–4.
Please cite this article as: Andersson P, et al, Ileorectal anastomosis in comparison with ileal pouch anal anastomosis in reconstructive surgery for ulcerative colitis — a single institution experience, J Crohns Colitis (2013), http://dx.doi.org/10.1016/j.crohns.2013.11.014