|
|
|
Background: Pouchitis is a significant sequel of restorative proctocolectomy. This study was undertaken to document the incidence of pouchitis at the Edinburgh Royal Infirmary and to assess outcome of treatment with metronidazole. Methods: Patients who developed pouchitis following restorative proctocolectomy for ulcerative colitis form the basis of this study. Pouchitis was suspected if patients developed diarrhoea with or without blood, mucus or pus. Diagnosis was confirmed with pouch endoscopy and biopsy. Results: From 1990 to 1999 (10 years) 139 patients underwent restorative proctocolectomy and ileo-anal pouch anastomosis (135 J pouches and 4 W pouches). Their median age was 35 years (range 13-74). There were 68 females and 71 males. The indication for operation was failed medical treatment in 104 patients and toxic megacolon in 35. Forty-seven patients (34%) developed pouchitis (21 females and 26 males). Symptoms were diarrhoea (35), diarrhoea, mucus and pus (5) and diarrhoea and blood (7). Symptoms of pouchitis started at an average of 33.51 + 29.2 months (range 2-102, median 18). All patients were treated with metronidazole for a minimum of one month. Thirty-six patients (77%) resolved on metronidazole alone. Nine patients (19%) went on to develop chronicity and were managed by long-term metronidazole (and/or ciprofloxacin). A further 2 patients (4%) had treatment resistant pouchitis and required pouch excision. Conclusion: Pouchitis is common following restorative proctocolectomy for ulcerative colitis. Treatment with metronidazole is associated with improvement in the majority of cases. In patients with chronic pouchitis maintenance of remission is possible with antibiotics. A high index of suspicion is advocated in patients who develop severe diarrhoea following this procedure.
Keywords: pouchitis, restorative proctocolectomy, ulcerative colitis
J.R.Coll.Surg.Edinb., 46, December 2001, 334-337
Restorative proctocolectomy was introduced in 1978 for the treatment of ulcerative colitis.1 It has the advantage of removing all diseased mucosa while avoiding a conventional ileostomy and it has become the elective procedure of choice for most patients with ulcerative colitis and familial adenomatous polyposis .2 The most significant late sequelae of this procedure is pouchitis. This is a non-specific inflammatory condition of the ileal pouch which can mimic inflammatory bowel disease.1-4 The exact incidence is unknown due to widely varying reports of incidence and therapeutic outcomes and the lack of universally accepted criteria for diagnosis.1- 5
The natural history of acute pouchitis is unknown. It may occur as either a single attack or recur with intermittent acute attacks separated by times of normal pouch function. In a small number of patients a syndrome of chronic pouchitis develops requiring constant maintenance therapy. The clinical behaviour of patients who develop pouchitis does not differ from the rest of the patients with ileo-anal pouches with respect to their original disease extent, severity or pouch configuration.6
This study was undertaken to examine the incidence and cumulative risk of pouchitis following restorative proctocolectomy for ulcerative colitis in Edinburgh and to assess response to treatment with metronidazole.
This is a prospective audit of patients undergoing restorative proctocolectomy with ileal pouch and ileo-anal anastomosis for ulcerative colitis over a ten-year period from 1990 to 1999. The operations were performed in the Colorectal Unit of the Edinburgh Royal Infirmary by one surgeon (DCCB). The indications for restorative proctocolectomy were failure of medical treatment and toxic megacolon. The former underwent restorative proctocolectomy and the latter initially underwent subtotal colectomy with ileostomy, followed by an elective completion restorative proctectomy. The creation of an ileo-anal pouch was accompanied by a covering ileostomy in all cases. Imaging of the pouch (pouchogram) was performed two months after the procedure to exclude sub-clinical leak, prior to performance of ileostomy closure.
Pouchitis was suspected if patients developed diarrhoea with or without blood, mucus or pus. Diagnosis was confirmed with pouch endoscopy and biopsy, performed by one endoscopist (DCCB). We did not use a scoring system in this cohort of patients. Treatment was with metronidazole 200mg three times a day for a minimum of one month. Occasionally, patients required to be commenced on intravenous metronidazole 500 mg three times a day for a few days as inpatients before they were converted to the oral form. Response to treatment was defined as reduction in the frequency of stools and disappearance of mucus, pus and blood. Chronicity was defined as continuous symptoms for more than a month and the continued need for drugs in order to control symptoms. Refractory (treatment-resistant) pouchitis was defined as no or poor response to antibiotic treatment. Patients who were initially diagnosed as having pouchitis following ulcerative colitis, and subsequently found to have Crohn’s Disease, were excluded from the study. Other exclusions were infective diarrhoea and other forms of pouch-related sepsis. Patients were followed up at one month, three months and 6-monthly thereafter.
Figure 1: Cumulative risk of developing pouchitis following restorative proctolectomy
One hundred and thirty nine patients underwent restorative proctocolectomy for ulcerative colitis during this period. Their mean age was 38.16 + 14.84 years (range 13-74, median 35). There were 68 females and 71 males. One hundred and four patients (75%) underwent elective restorative proctocolectomy for failed medical treatment and 35 (25%) underwent emergency subtotal colectomy followed by elective completion proctectomy. There were 135 J pouches and four W pouches. The median pouch capacity was 300 ml (range 180-360).
Forty-seven patients (34%) developed pouchitis (21 females and 26 males). Figure 1 shows the cumulative risk of developing pouchitis following this operation calculated by life table analysis. It was 8% at 1 year, 18% at two years 28% at five years. The indication for initial restorative proctocolectomy in this cohort of patients was failed medical management in 36 (77%) and toxic megacolon in 11(23%). Four patients developed other pouch-related complications before developing pouchitis, including a pouch-vaginal fistula, anal stenosis, pelvic abscess and anastomotic dehiscence. The complications were managed successfully in all cases, with no long-term sequelae. The patients went on to develop pouchitis.
The presenting symptoms were diarrhoea in 35 patients (75%), diarrhoea and blood in 7 patients (15%) and diarrhoea, mucus and pus in 5 (11%). Symptoms of pouchitis started at an average of 33.51 + 29.2 months after creation of the pouch (range 2-102, median 18). All had endoscopy and biopsy. Chronic inflammatory infiltrate and ulceration were seen in all patients. There was villous atrophy in 41 patients and there was no comment on villous atrophy in the other six patients. There was documentation of crypt abscesses only in 19 patients. Six patients required initial inpatient treatment with intravenous metronidazole, which was converted to oral metronidazole on discharge. The rest of the patients were given oral metronidazole from the beginning. The patients were followed up for a median of 60 months (range 2-108). Thirty-six patients (77%) resolved on metronidazole treat-ment alone. Three patients had recurrence of pouchitis, which resolved on metronidazole without progressing to chronic pouchitis. Nine patients (19%) went on to develop chronicity and were maintained on long-term antibiotic treatment (metronidazole, ciprofloxacin or both). One patient with chronic pouchitis died 8 years after restorative proctocolectomy and 7 years after developing pouchitis.
Two male patients required pouch excision because of refractory pouchitis (4%). Both patients were initially given different treatment combinations including metronidazole, ciprofloxacin, topical and systemic corticosteroids, as well as azathioprine, before a decision to excise the pouch was made. One patient underwent a pouch excision four years after restorative proctocolectomy and three years after developing the first attack of pouchitis. He died a few months later from unrelated causes. The other patient underwent excision two years after the procedure and was still alive at the last follow-up.
Although patients undergoing restorative proctocolectomy have a favourable functional outcome, between 7 and 50% will develop pouchitis depending on the quality and duration of follow-up. 1,3, 7-10 It is higher following ulcerative colitis than in patients having restorative proctocolectomy for familial adenomatous polyposis.1,6 The cumulative risk of developing pouchitis in the present series was 18% at two years and 28% at five years. In 1997, Keranen et al studied 137 patients with restorative proctocolectomy for ulcerative colitis and reported a five-year cumulative risk of 25% for developing pouchitis.11 In 1998 Hurst et al studied 291 patients with ulcerative colitis undergoing restorative proctocolectomy and observed a five-year cumulative risk of 44% for developing pouchitis.3 These studies together with the present series confirm pouchitis as the most common sequelae of restorative proctocolectomy following ulcerative colitis.
The aetiology of pouchitis is poorly understood. There has, however, been much speculation about its aetiopathogenesis and a variety of theories have been postulated, including (i) bacterial overgrowth, (ii) chronic ischaemia, (iii) stasis, (iv) deficiency of short chain fatty acids (v) immunological reactivation, and (vi) recrudescence of ulcerative colitis.1, 5-8,10,11 No clear microbiologic factor has been shown to play a part in the aetiology.10
Epidemiological studies following patients up from the indication for surgery to the postoperative phase have identified several independent risk factors for the development of pouchitis, such as high relapse rate, chronic refractory disease, presence of extra-intestinal manifestations and cessation of smoking.1,6,10,12 Studies investigating the association of peri-nuclear anti-neutrophil cytoplasmic antibody (p-ANCA) with pouchitis have yielded conflicting results. In 1995, Sandborn et al noted that 100% of patients with chronic ulcerative colitis and active pouchitis were positive for p-ANCA.13 Yasuda et al, on the other hand, demonstrated no correlation between p-ANCA and the presence of pouchitis in a study of 49 patients in 1998.14
Acute pouchitis presents with sudden onset of diarrhoea, abdominal cramping, rectal urgency, tenesmus and incontinence, often with constitutional symptoms causing anorexia, nausea and flu-like illness.1,4,8,10,15 Bowel frequency may be more than 30 times a day.15 Frequently there is blood, and occasionally pus, with the diarrhoea. If the patient has had previous extra-intestinal manifestations of colitis these extra-intestinal manifestations often relapse with pouchitis, although they may sometimes present for the first time during an attack of pouchitis.1,10,15
The diagnosis of pouchitis has been defined using various criteria. Some authors have favoured a diagnosis based on clinical symptoms, while others recommended the use of endoscopic or histological features.8 This has led to difficulty in interpreting reported data, necessitating the establishment of a gold standard in diagnosis. Recently, therefore, it has been advocated that an unequivocal diagnosis should be based on a diagnostic triad consisting of clinical symptoms, endoscopic features of inflammation and histological evidence of inflammatory infiltrate.1,4,16 Endoscopically, the condition is characterised by oedema, granularity, mucous exudate, contact bleeding and ulceration and histological characteristics are chronic inflammatory infiltrate, crypt abscesses and ulceration.1,7-10,16 By incorporating clinical, endoscopic and histological criteria the Mayo Clinic devised a Pouchitis Disease Activity Index (PDAI).17 Patients with a total score of 7 points or more were considered to have a diagnosis of pouchitis. Whereas the policy at the Edinburgh Royal Infirmary is to obtain both endoscopic and histological evidence of pouchitis before making this diagnosis, we did not use PDAI or any other scoring system at the time of the study.
Patients undergoing routine endoscopy may have endoscopic evidence of inflammation.7 Rubinstein and Fisher (1996), however, caution that these patients should not be labelled as having pouchitis without other clinical, endoscopic and histological evidence of pouchitis, but that they should be investigated for other causes of diarrhoea.7 These conditions include infective diarrhoea, secretory diarrhoea, malabsorption, vitamin B12 deficiency, Crohn’s disease and irritable bowel syndrome and retained rectum.1,7,10 Causes of infective diarrhoea that need to be excluded include cytomegalovirus infection, pseudomembranous enteritis, Enteropathic E coli, Shigella and many others.1,18,19
The treatment of pouchitis is largely empirical because controlled trials are lacking.5 The concept that acute pouchitis may be due to stasis and bacterial overgrowth by anaerobes prompted early clinicians to treat these patients with metronidazole, given at a dose of 200 mg to 400 mg three times a day.1,4,12,16 Treatment with metronidazole has been shown to reduce bacterial count of bacteroides in the pouch1,20, re-establish short chain fatty acids in the lumen of the pouch 21 cause rapid improvement of villous atrophy and crypt cell hyperplasia with reduction in crypt cell production rates and lead to reversal of leukocyte infiltration in the pouch.1,22 Improvement is usually realised within 72 hours of starting treatment.1,7 Indeed, response to metronidazole has been regarded by many as being one of the diagnostic criteria for acute pouchitis.1,10 Treatment with metronidazole in the present study successfully eradicated the disease in 77% of patients. Metronidazole does have side effects, however, the most common of which are nausea, dysgeusea and the development of peripheral neuropathy.7
Most patients have only a single attack of pouchitis that rapidly resolves on treatment.10 Progression to chronic pouchitis has been reported to occur in 4 - 25% of cases of acute pouchitis.3, 6, 9,11,15,23 It occurred in 19% of cases in the present series, and treatment with metronidazole was able to maintain remission on its own or with the addition of ciprofloxacin. The alternative approach in patients resistant to metronidazole therapy is to use systemic or topical corticosteroids, 5-aminosalicylic acid, azathioprine, cyclosporin A and allopurinol.1,5,7,12 Other less uncommonly used agents with documented improvement in the disease process include doxycycline, ampicillin-clavulaenic acid combination, short chain fatty acids and glutamine.1 In a double-blind placebo-controlled trial Gionchetti et al (2000). demonstrated that administration of probiotics is effective in preventing flare-ups of chronic pouchitis.24
Refractory pouchitis requiring ultimate excision of the pouch is reported to occur in 1.4 - 3% in the literature. 6 It occurred in 4% of cases in the present series.11,23 Undiagnosed Crohn’s disease is one of the main causes leading to pouch excision.11,25 Most often it is expressed as anal fistula or abscess but may also appear as refractory pouchitis.11 The present series does not include these cases. Pouch excision may be the only way to improve quality of life in patients with refractory pouchitis, but the decision to remove the pouch should not be undertaken lightly. It certainly should never be considered for acute pouchitis.1
Whether or not there are functional changes in patients with pouchitis has not been clearly defined. Hurst et al (1998) showed that the occurrence of the first episode of pouchitis increases rapidly after restoration of gastrointestinal continuity, but then levels off after approximately 30 months.3 We did not observe this pattern in the present series. Hurst et al also demonstrated a tendency for poor pouch function associated with pouchitis, which commenced even before the first episode of the acute symptomatic pouchitis. They postulated that the chronically poor function was unlikely to result from long term damage to the pouch from an episode of acute pouchitis but rather that pouchitis may represent a chronic condition that displays episodic symptomatic exacerbations. Keranen et al (1997), on the other hand, compared functional parameters concerning defaecation frequency and continence and demonstrated little difference between groups with and without pouchitis.11 They did, however, demonstrate significant functional impairment in patients with chronic pouchitis.
Although the disability caused by pouchitis results in severe inconvenience to the patient, the majority of patients express satisfaction with their present situation despite pouchitis. It is evident, therefore, that avoiding permanent ileostomy is such a valuable factor on quality of life that certain health risks, such as pouchitis are acceptable, and consequently most patients who are offered surgery for ulcerative colitis prefer an anal preserving operation despite this inconvenience.11
Prevention of this condition has received scant attention in the literature. Since the pathogenesis of pouchitis is poorly understood, it is difficult to make recommendations regarding its prevention and that of its recurrence. The use of 5-aminosalicylic acid derivatives for the first 12 months after creation of a pouch, the use of nicotine, delivered transdermally, as well as genetic characterisation have all been suggested.5 Further studies, therefore, are required to assess the role of these management options in the prevention of pouchitis.
This study has demonstrated that pouchitis is common following restorative proctocolectomy for ulcerative colitis. Whereas we advocate a high index of suspicion in patients who develop severe diarrhoea following this procedure, we nevertheless support Rubinstein and Fisher (1996), in cautioning against labelling all patients with diarrhoea as pouchitis.11 The diagnosis must rely, therefore, on a proper history, with subsequent endoscopy and biopsy. Treatment with metronidazole is associated with improvement in the majority of patients, who are then able to maintain pouch function. Patients with chronic pouchitis are more difficult to treat but maintenance of remission is possible with antibiotic treatment.
Presented at the Colorectal Disease in 2001 Pre-Course Symposium in Fort Lauderdale, Florida, USA, 8th February 2001
1. Keighley MRB. Review article: the management of pouch-itis. Aliment Pharmacol Ther 1996; 10: 449-57
2. Cohen Z, Smith D, McLeod R. Reconstructive surgery for
pelvic pouches. World J Surg 1998; 22: 342-6
3. Hurst RD, Chung P, Rubin M, Michelassi F. The implications of acute pouchitis on the long-term functional results
after restorative proctocolectomy. Inflamm Bowel Dis
1998; 4: 280-4
4. Sandborn WJ, McLeod R, Jewell DP. Medical therapy for
induction and maintenance of remission in pouchitis: A
systematic review. Inflamm Bowel Dis 1999; 5: 33-9
5. Merrett MN. Ileal pouches: adaptation and inflammation.
Baillière’s Clinical Gastroenterology 1997; 11:175-93
6. Becker JM, Stucchi AF, Bryant DE. How do you treat
refractory pouchitis and when do you decide to remove the
pouch? Inflamm Bowel Dis 1998; 4:167-9
7. Rubinstein MC, Fisher RL. Pouchitis: Pathogenesis, diagnosis and management. Gastroenterologist 1996; 4:
129-33
8. Schouten WR. Pouchitis. Mediators of Inflammation 1998; 7: 175-81
9. Gionchetti P, Rizzello F, Venturi A, et al. Antibiotic combination therapy in patients with chronic treatment-resistant
pouchitis. Aliment Pharmacol Ther 1999; 13: 713-8
10. Nicholls RJ, Banerjee AK. Pouchitis: Risk factors, etiology
and treatment. World J Surg 1998; 22: 347-51
11. Ker per thousand nen U, Luukkonen P, J per thousand rvinen H. Functional results
after restorative proctocolectomy complicated by pouchitis.
Dis Colon Rectum 1997; 40: 764-9
12. K¸hbacher T, Schreiber S, Runkel N. Pouchitis: pathophysiology and treatment. Int J Colorect Dis 1998; 13: 196-207
13. Sandborn WJ, Landers CJ, Tremaine WJ, Targan SR.Antineutrophil cytoplasmic antibody correlates with
chronic pouchitis after ileal pouch-anal anastomosis. Am J
Gastroenterol 1995; 90: 740-7
14. Yasuda N, Thomas P, Ellis H, et al. Perinuclear antineutrophil cytoplasmic antibodies in ulcerative colitis after
restorative proctocolectomy do not correlate with the presence of pouchitis. Scand J Gastroenterol 1998; 33: 509-13
15. Lohmuller JL, Pemberton JH, Dozois RR, et al. Pouchitis
and extraintestinal manifestations of inflammatory bowel
disease after ileal pouch-anal anastomosis. Ann Surg 1990;
211: 622-9
16. Madden MV, Farthing MJG, Nicholls RJ. Inflammation in
ileal reservoirs: “Pouchitis”. Gut 1990; 31: 247-9
17. Sandborn WJ, Tremaine WJ, Batts K, et al. Pouchitis after
ileal pouch-anal anastomosis: a pouchitis disease activity
index. Mayo Clin Proc 1994; 69: 409-15
18. Mumoz-Juarez M, Pemberton JH, Sandborn WJ, et al. Misdiagnosis of specific cytomegalovirus infection of the
ileoanal pouch after refractory idiopathic chronic pouchitis.
Dis Colon Rectum 1999; 42: 117-20
19. Vesoulis Z, Williams G, Matthews B. Pseudomembranous
enteritis after proctocolectomy. Dis Colon Rectum 2000;
43: 551-4
20. Kmiot WA, Younds D, Tudor R, et al. Mucosal morphology, cell proliferation and faecal bacteriology in acute
pouchitis. Br J Surg 1993; 80: 1445-9
21. Sagar PM, Taylor BA, Godwin P, et al. Acute pouchitis and
deficiencies of fuel. Dis Colon Rectum 1995; 38: 488-93
22. Kmiot WA, Hesslewood SR, Smith N, et al. Evaluation of
the inflammatory infiltrate in pouchitis with indium-labelled granulocytes. Gastroenterology 1993; 104: 981-8
23. Fazio V, Ziv Y, Church JM, et al. Ileal pouch-anal anasto-moses: complications and function in 1005 patients. Ann
Surg 1995; 222: 120-7
24. Gionchetti P, Rizzello F, Venturi A, et al. Oral bacterio-therapy as maintenance treatment in patients with chronic
pouchitis: a double blind placebo-controlled trial.
Gastroenterology 2000; 119: 584-7
25. Wexner SD, Jenesen L, Rothenberger DA, et al. Long-term functional analysis of the
ileo-anal reservoir. Dis
Colon Rectum 1989; 32:275-81
Copyright date: 13th July 2001
Correspondence: Correspondence: T. E. Madiba, Department of Surgery, University of Natal, Private Bag 7, Congella 4013, R.S.A. E-mail: madiba@nu.ac.za