Pouchitis and how to avoid them

There are no validated scoring systems available to define pouchitis. The pouch disease activity index (PDAI)⁵ is used in the research setting but it remains unvalidated and its routine use in clinical practice is rare. 

Importantly, there can be a lack of correlation between symptoms, endoscopic activity and histological findings, leading to potential misdiagnosis.^6,7 Furthermore, there is a tendency to label every patient who has pouch dysfunction as having pouchitis; primary idiopathic pouchitis is often diagnosed without endoscopic and histological confirmation, which can result in other pathologies being missed. In addition, patients often self-diagnose primary idiopathic pouchitis based on their symptoms, without objective markers of inflammation or endoscopic assessment being used. 

It is also important to distinguish between primary and secondary types of pouchitis. Primary idiopathic pouchitis is defined as pouchitis when all secondary causes have been excluded, whereas secondary pouchitis can be due to Crohn’s disease, infection (e.g. with Clostridiodes [formerly Clostridium] difficile, Campylobacter, Salmonella, Candida or cytomegalovirus), pelvic sepsis, faecal stasis, ischaemia or drugs (particularly NSAIDs).⁸ 

When patients present with symptoms related to their pouch it is important to be mindful that other possible causes (e.g. hyperthyroidism, coeliac disease or pouch cancer) are excluded. This is best done by following an algorithm, such as the one presented in figure 1.⁹ 

It is essential to take a robust history, perform a thorough examination and offer timely investigations (including pouchoscopy and radiological examination) before confirming a diagnosis of primary idiopathic pouchitis. An important question to ask is “Has the pouch ever worked well?” If this is not the case, a diagnosis of anastomotic leak/pelvic sepsis needs careful consideration. Importantly, it is essential to establish the baseline pouch function for an individual patient; the median 24-hour stool frequency is four to eight, with roughly half of patients needing to defecate at night.

In one retrospective study, more than a third of patients considered to have antibiotic-dependent pouchitis were actually found to have evidence of pelvic sepsis on MRI.⁷ It is therefore imperative to thoroughly assess a dysfunctional pouch with MRI. The potential consequences of missing this diagnosis are inappropriate use of antibiotics, antibiotic resistance, side effects and worsening clinical state. Furthermore, by missing pelvic sepsis and assuming symptoms are primary idopathic pouchitis, there is the potential to escalate to biologics, which may cause immunosuppression and worsening sepsis. Should pelvic sepsis be found, appropriate antibiotic therapy should be considered, with a multidisciplinary discussion to plan management that may include drainage and diversion.

Thorough endoscopic assessment is vital to help understand the potential reasons for a poorly functional pouch. However, no validated endoscopic reporting systems, key performance indicators, training or certification are available to assess competency in pouchoscopy. Furthermore, there is considerable variation in the quality of endoscopy reports and efforts have only recently been made to develop a reporting template specific to pouchoscopy.¹⁰ The standardised reporting template introduced is a systematic tool that prompts examination of the anal/perineal region, rectal cuff, anastomosis, lower and upper pouch body, pouch inlet and pre-pouch ileum, but it is not widely used. 

Other problems with pouchoscopy include poor intra- and inter-rater reliability between pouchitis scoring systems,¹¹ with some of the PDAI descriptors considered inappropriate for assessing endoscopic disease activity in pouchitis. This suggests that we need better validated scoring systems for assessment of pouch inflammation.¹¹ 

We suggest that endoscopic examination of a pouch is performed by experienced endoscopists and that a systematic template should be used to encourage examination (and biopsy) of all pouch regions and recording of findings. Photographic evidence of the pouchoscopy can also help assessment.

There is a large variation in how different patients perceive their pouch symptoms and how their symptoms impact on their overall quality of life (QoL). There can also be a disconnect between symptom burden and the severity of inflammation noted in the pouch. Indeed, some patients with very minimal objective inflammation noted endoscopically and histologically can have pouch function that severely impacts on their quality of life, whereas other patients can report minimal symptoms but have severe inflammation noted. It is essential to be conscious of this variability and discrepancy, so that investigation and management is tailored to the individual patient. In patients who have a high symptom burden but little inflammation, other causes of their symptoms should be explored along with all supportive options available, including psychological support. 

The clinical scoring systems do not take into account QoL indicators that may be significant to a patient, which means it is vital to consider a patient’s quality of life when assessing symptomatology. We recommend that a broad range of domains (including effect on personal life, work, sexuality and overall quality of life) are taken into account. To aid such assessment, a multidisciplinary team of pouch and stoma nurses, psychologists, clinicians and patient-to-patient support can be valuable when helping patients with pouch-related issues.

Specifically, it is important to assess the mental health of a patient who has poor pouch function and offer them psychological support. There has been evidence to show that adjuncts such as biofeedback may be beneficial.¹² Irritable pouch syndrome (analogous to irritable bowel syndrome) is diagnosed when no other pathology is present to explain symptoms, and may respond to medications such as tricyclic antidepressants.

When forming a pouch, there is an anastomosis between the ileal pouch and the anorectum. In some situations, this area—refered to as the cuff—can become inflamed, resulting in ‘cuffitis’. Essentially this is residual ulcerative proctitis. The symptoms of cuffitis are characterised by the frequent passage of stool with small quantities of blood,¹³ urgency and pain and it can, therefore, be mistaken for pouchitis. 

A digital examination should be performed to manually feel the cuff. A long retained rectal cuff (longer than 2cm) is more prone to inflammation. Endoscopic evaluation is required to directly visualise the cuff, with biopsy samples taken for histological assessment to confirm the diagnosis.^13–15 The incidence of inflammation of the retained anorectum has not been extensively studied; some studies report a 9% incidence.¹⁶ 

The treatment for cuffitis differs from that for pouchitis and so distinguishing the two is key. Although cuffitis is a poorly studied condition, there is evidence that mesalamine suppositories¹⁴ may provide some benefit, with steroid suppositories as a second-line therapy.¹⁷ Essentially the management of cuffitis is the same as treating ulcerative proctitis.

Prior to pouch formation a thorough pre-surgical assessment must be considered to rule out Crohn’s disease, to include small-bowel studies, histological samples reviewed for the presence of granuloma and the absence of a history of perianal disease. A preoperative diagnosis of Crohn’s disease of the pouch is a relative contraindication to pouch surgery¹⁸ because of the high rate of complications and pouch failure.^19–21 Indeed, it has been estimated that pouch excision rates are 45–55% in patients who have a preoperative diagnosis of Crohn’s disease^21,22 and that the pouch retention rate 5 and 10 years after formation is 58% and 50%, respectively.²³ Despite this, some small studies have shown that in the absence of perianal or small-bowel disease, restorative proctocolectomy can be performed with similar outcomes to those who have a pouch for ulcerative colitis.^24,25 Indeed, the European Crohn’s and Colitis Organisation (ECCO) suggest that restorative proctocolectomy can be offered to patients who have Crohn’s disease without perianal disease or small-bowel involvement,²⁶ but we suggest that patients are very carefully counselled about the potential poor outcomes and such surgery is done highly selectively.

There are essentially two scenarios that lead to a diagnosis of ‘Crohn’s disease of the pouch’. The first arises when there has been a preoperative diagnosis of Crohn’s disease, and the second is the development of Crohn’s-like features after the formation of a pouch in a patient who has ulcerative colitis. 

In one long-term study, 2–8% of patients who originally underwent restorative proctocolectomy for presumed ulcerative colitis had their original diagnosis changed to Crohn’s disease.²¹ We believe it is important to diagnose Crohn’s disease of the pouch accurately to help with prognostication and offer appropriate treatment. The criteria utilised to diagnose Crohn’s disease are varied. Some studies have defined Crohn’s disease of the pouch as including: inflammation of the pouch that is resistant to antibiotic treatment, stricturing of the afferent limb, stricturing of the small bowel, or fistulating disease.^27–30 Furthermore, the presence of pre-pouch ileitis is controversial—some studies suggest this may be an endoscopic feature of Crohn’s disease,^31,32 but this has been disputed by others.³³ 

It is likely that Crohn’s disease of the pouch is an overused diagnosis, with one study highlighting that histological confirmation of Crohn's disease was found in only 20% of patients who underwent pouch excision for Crohn's disease of the pouch.³³ It is also important to appreciate that strictures and fistulas that may mimic Crohn’s disease can be caused by other factors such as sepsis, anastomotic complications (e.g. leak and/or stricture) or ischaemia. We therefore suggest that Crohn’s disease of the pouch should only be diagnosed by conclusive histology (i.e. granulomas supporting Crohn’s disease) and/or the presence of characteristic skip lesions in the small bowel. The timing of when the Crohn’s-like problems of the pouch occur can often aid diagnosis—fistulas especially around the anastomosis that occur within a year of pouch formation are likely to be related to the surgery itself, whereas complications beyond this point could represent de novo Crohn’s disease, but this still requires thorough investigation. 

We also believe that it can be unhelpful and confusing for patients to hear that their diagnosis has changed from ulcerative colitis to Crohn’s disease and suggest it is more useful to explain that they have active inflammatory bowel disease. It is more beneficial for the patient to be clear on the strategies of care for managing their problem, whether it be a fistula, stricture or inflammation, using the appropriate combination of surgical/endoscopic and medical therapies, than to become bogged down in semantics regarding terminology.

Chronic primary idiopathic pouchitis develops in approximately 10–15% of patients with acute pouchitis and it can be ‘responsive’ or ‘refractory’ to antibiotic therapy.^3,4 Some of these patients require long-term antibiotics to maintain symptomatic relief.³⁴ The two antibiotics most commonly used to treat primary idiopathic pouchitis are ciprofloxacin and metronidazole. Patients should be counselled on the specific adverse effects of these medications, which include tendinopathy in the case of ciprofloxacin³⁵ and peripheral neuropathy in the case of metronidazole (see figure 1).³⁶ In particular, patients should be made aware of the risks of long-term antibiotic use and other therapeutic options considered. 

The long-term use of antibiotics can be associated with the development of antibiotic-resistant organisms.³⁴ There have been case series suggesting that C. difficile infection occurs in patients who have a pouch and are taking antibiotics and that they should be used cautiously, especially in the elderly.³⁷ We advise regular reassessment of patients on long-term antibiotics, including QoL assessment, endoscopic and radiological investigations and consideration of alternative medications where appropriate. As already described, more than a third of patients considered to have antibiotic-dependent pouchitis were found to have  evidence of pelvic sepsis on MRI7 and hence this should be excluded before a diagnosis of chronic antibiotic-dependent primary idiopathic pouchitis is made. 

If antibiotics fail to control chronic pouchitis, biological therapies can be considered; however, sepsis must be excluded before embarking on this. Limited evidence—based on small case series and not randomised controlled trials—indicate that, overall, anti-TNF therapies are associated with a 45–58% response rate.³⁸ Anti-TNF agents were the first to show some benefit for patients with refractory pouchitis, with emerging data that ustekinumab³⁹ and vedolizumab⁴⁰ may also be effective. Non-biologic alternatives, such as alicaforsen (an antisense oligonucleotide), may be an option but formal results from a phase 3 trial have not yet been published.⁴¹ 

Importantly, if a patient requires biologic therapies, their symptom burden is usually quite severe and hence ongoing objective assessment of drug effectiveness and exploration of quality of life are essential. We recommend that patients are counselled thoroughly prior to starting biologics and, where possible, they have a joint consultation with an experienced pouch surgeon to discuss alternatives (pouch diversion or excision). Chronic pelvic sepsis must be excluded using cross-sectional imaging. Clear timelines must be discussed with the patient when planning a strategy of care, so that if biologic drugs do not achieve the predetermined goals set with the patient, alternative options are sought. 

As for all forms of IBD, validated, robust and consistent data on which to base dietary advice for patients with pouchitis are lacking. There is evidence that fruit consumption can reduce the incidence of primary idiopathic pouchitis⁴² and a reduction of foods rich in antioxidants can predispose to pouchitis.⁴³

In terms of vitamin deficiencies, it has been highlighted that of patients who have undergone restorative proctocolectomy, 10.6% have a vitamin D deficiency and 5% have either a vitamin A or B12 deficiency. Furthermore, patients with pouchitis may be at risk of having lower levels of calcium, total cholesterol, triglycerides, vitamin E and iron.^44,45 It is therefore important that anyone who undergoes proctocolectomy is advised to consider eating foods rich in these vitamins and minerals to keep levels normal. 

Caution must also be exercised with respect to fluid balance in a patient who has a pouch. Normal stool frequency can be four to eight times in a 24-hour period, and a great deal higher when active pouchitis is present. Patients can become dehydrated⁴⁶ and electrolytes depleted, contributing to fatigue. It can be helpful to assess fluid balance by checking the urinary sodium level, and consider the addition of electrolyte mix and antidiarrhoeal agents to reduce fluid loss. Guidance from an experienced dietitian is helpful. 

Restorative proctocolectomy is an operation that is designed to improve overall quality of life and remove the need for an ileostomy; however, poorly functioning pouches may impact on a patient’s life in multiple ways, resulting in nutritional, psychological, physical and emotional problems. It is therefore essential that a patient who has a pouch is given access to a broad range of healthcare professionals who have the skills and experience necessary to support their various needs. If a pouch starts to function poorly, it is vital that early discussions are held with the multidisciplinary team so that all management options, including surgery, are considered. 

Specialist pouch nurses, gastroenterologists, surgeons and dietitians are needed to ensure effective service delivery. Joint medical and surgical clinics can help to achieve a holistic review of a patient’s quality of life and offer the full range of treatment options. As pouch surgery is not frequently performed, we recommend that, whenever possible, pouch care should be centralised to units that perform a high volume of these operations and have the healthcare support staff and infrastructure needed to manage pouch-related problems. Furthermore, patients should have a point of contact, much like an ‘IBD nurse’, so they can discuss problems with their pouch and be directed to the right services early.