Ulcerative Colitis

Ulcerative colitis

Gwo-Tzer Ho

Ray Boyapati

Jack Satsangi

Gwo-Tzer Ho PhD MRCP is a Medical Research Council (MRC) Clinician Scientist and Honorary Consultant Gastroenterologist at the MRC Centre for Inflammation Research, University of Edinburgh and Western General Hospital, UK. Competing interests: none declared.

Ray Boyapati MBBS is a Senior Clinical and Research Fellow at the MRC Centre for Inflammation Research, University of Edinburgh and Western General Hospital, Edinburgh, UK. Competing interests: none declared.

Jack Satsangi DPhil FRCP is Professor of Gastroenterology at the Gastrointestinal Unit, Institute of Genetics and Molecular Medicine, University of Edinburgh, Western General Hospital, Edinburgh, UK.

Competing interests: none declared.

Abstract

Ulcerative colitis (UC) and Crohn’s disease (CD) represent the two major forms of inflammatory bowel disease (IBD). UC is a chronic idiopathic inflammatory condition affecting the colon and rectum. A complex interplay between components of the innate immune system and environmental factors, notably the microflora, regulates colonic mucosal homeostasis which is dysregulated in UC. In this chapter, we review current concepts of the epidemiology, pathogenesis, clinical features and management of UC.

Keywords

Inflammatory bowel disease; management; pathogenesis; ulcerative colitis

Epidemiology

In the United Kingdom (UK), the annual incidence of ulcerative colitis (UC) is approximately 10/100,000 with a point prevalence of 200/100,000. UC can present at any age, but typically does so in the third orfourth decade (later than Crohn’s disease (CD)). Men and women are equally affected. A global systematic review of epidemiologic studies from 1950-2010 showed that the rate of increase in UC incidence in Northern Europe and North America hasslowed whilst it continues to increase in areas with previously low incidence, such as Southern Europe and Asia.1In adults at presentation, about 55% have proctitis, 30% left-sided colitis, and 15% extensive colitis or total colitis.

Pathogenesis

Environment

Both human and animal studies suggest that the intestinal inflammation in UC is likely a consequence of environmental factors triggering a breakdown in the regulatory constraints on mucosal immune responses in genetically susceptible individuals. The relatively weaker concordance in monozygotic twin pairs with UC (6–14%) compared with CD (44–50%) suggests that environmental factors are more important than genetic factors in the pathogenesis of UC.2Migrants and particularly first generation children from low prevalent areas who immigrate to high prevalent areas have a higher risk of UC suggesting an environmental influence in the onset of disease.Cigarette smoking, appendicectomy and diet are well-characterized environmental factors. The protective effect of smoking on UC is in stark contrast to the detrimental effect seen in CD; this is the most consistent environmental factor ininflammatory bowel disease (IBD). The protective effect of appendicectomy may be secondary to the development of appendicitis. The consumption of red and processed meat, alcohol and low dietary fibre has been associated with an increased likelihood of relapse.Epidemiological data (total of 36 studies)investigating the subsets of food including sugar, cereal, fruit, vegetables, fat and carbohydrate associated with onset of incident UC have not yet demonstrated a consistent picture and may reflect difficulties in study design.3

Genetics

UC has lower heritability (λs, sibling recurrence risk = 10–15) compare toCD (λs = 20–35). A meta-analysis of fifteengenome wide association (GWA) studies involving 75, 000 individuals identified 163IBDsusceptibility loci.4Most (110 loci) are associated with both UC and CD; with 23 specific for UC. In UC, these dataprincipally implicate genes that are involved in the maintenance of epithelial integrity (HNF4A, CDH1, LAMB1, ECM1), innate immune function (PLA2G2E, CARD9), immune regulatory function (HLA-region, IL-10, BTNL2, IFNγ-IL25, NKX2-3), and cellular homeostasis in response to endoplasmic reticulum stress (ORMDL3) in UC. Associations within the major histocompatibility complex class II region near HLA-DRA (α-chain) are the most significant and consistent observation.Theoverlap with CD susceptibility genes is notably in the interleukin-23 signalling pathway — IL23R, IL12B, JAK2 andSTAT3. This highlights the complex nature of disease mechanism where UC and CD share some genetic susceptibility factors whilstdisparate environmental/luminal factors drive different clinical presentations.As genetic factors account forless than 10% of disease variance in UC, the role for epigeneticsvia mechanisms including DNA methylation, RNA interference and histone modification may provide further understanding in how the environmentinteracts with genetic susceptibility in the pathogenesis of UC.5

Microbiota, epithelial barrier function and mucosal response

The microbial influences in UC primarily focus on their metabolic effects on epithelial homeostasis and function. The colon harbours 106 predominantly anaerobic bacteria. The colonic epithelium derives 70% of its energy from short-chain fatty acids (SCFAs), principally butyrate, which are products of anaerobic bacterial fermentation. Faecal diversion in UC worsens inflammation in contrast to CD. In UC, a shift in the ‘healthy’ microbiota such as Bifidobacterium and Bacteriodes with a concurrent reduction in SCFAs has been described7 along with an increase insulphate-reducing bacteria (SRB) of the genus Desulfovibrio. SRB convert sulphate to sulphide, which inhibits butyrate oxidation and is directly toxic to colonic epithelial cells. SCFAs can induce the expansion of colonic regulatory T-cells,8thus illustrating how bacterial products can educate or regulatethe mucosal immune system.

Advances in DNA sequencing technology allowing culture-independent approaches have provided an exceptional platform for in-depth analysis on the gut microbiome.The MetaHIT (Metagenomics of the Human Intestinal Tract) project estimated 3.3 million non-redundant bacterial genes within the luminal flora, 150 times larger than the human complement.9 In this study, the relative abundance of 155 bacterial species can differentiate UC, CD and healthy individuals.As gut microbial composition is strongly associated with diet,10further microbiome studies linking diet and IBD are expected to provide mechanistic insights into the triggering factors in UC.

Recent studies have focused on the immunologic factors underlying the aberrant inflammatory response in UC, which is traditionally considered to have a TH2 profile with elevated IL-4 and IL-5 cytokine concentrations. UC is associated with the presence of lamina propria non-invariant (Type II) NKT cells producing IL-13 and mediating epithelial cell cytotoxicity.A ‘self’ glycolipid has been suggested as a trigger for NKT cell activation.11A number of pro-inflammatory damaged associated molecular patterns (DAMPs)such as calprotectin and high-mobility group box 1 (HMGB1) are found in high levels in active UC – uncontrolled release of these immunogenic motifs may have a role in perpetuating the non-resolving nature of inflammation. Overall, de-regulated epithelial function remains the critical element in the pathogenesis of UC.

Clinical features

The major symptoms of UC include diarrhoea with urgency, rectal bleeding and colicky abdominal pain. In patients with distal disease (rectosigmoid involvement), the symptoms of rectal irritation may predominate – namely tenesmus (the sensation of incomplete emptying), small-volume diarrhoea, proximal constipation and rectal bleeding. In contrast, in extensive colitis (beyond the splenic flexure), profuse bloody diarrhoea, abdominal cramping and, in severe cases, systemic features such as weight loss, fever and tachycardia are more prominent. Symptoms tend to present insidiously but may also present acutely, mimicking an infective aetiology. In children, extensive colitis is typical at diagnosis.

Diagnosis
The diagnosis of UC is principally based on clinical, endoscopic and histological grounds. UC invariably affects the rectum and extends proximally to a variable distance. The inflammation is characteristically confluent. The earliest signs tend to be subtle loss of vascular patterns with hyperaemia and oedema of the mucosa (Figure 1). With more active inflammation, the mucosa becomes granular with presence of mucopus and contact bleeding. In advanced cases, deep ulceration may present, which may mimic severe CD. In the diagnosis of UC, the two critical features influential in the subsequent management are disease extent and disease severity. A summary of the differential diagnosis of UC is shown in Table 2.

Assessment of disease extent

Patients with extensive disease have an increased likelihood of medically refractory and severe disease. A review of selected referral centre-based and population-based cohorts hasshown that the risk of colectomy is influenced by disease extent. The risk of colectomy among patients with only limited proctitis ranged from 2 to 9% while patients with extensive colitis had 5-year colectomy rates of 30–44%. Disease extent is not static and proximal extension can occur in patients with previously distal disease.

Assessment of disease severity

The risk of colectomy in UC is also strongly influenced by the severity of disease at initial presentation. The Truelove and Witt’s criteria provide the most widely used definition for severe disease, encompassing descriptions of profuse bloody diarrhoea and the clinical/biochemical features of systemic upset (Table 3).Historically, 15% of patients with UC will develop a severe attack of colitis requiring in-patient medical management and high-dose intravenous corticosteroid therapy; 30–40% of these will fail to respond adequately and require colectomy. The in-patient mortality associated with such attacks has fallen from 31–61% to the present level of 1–2% following the introduction of high-dose corticosteroid therapy, better joint medical–surgical management, and the acceptance of a policy of early surgery in patients not responding to medical therapy.12Despite these advances, severe UC remains an entity with considerable morbidity and mortality. In the UK, reported 3-year mortality rates are approximately 12% following the presentation of severe UC;13 mortality is greatest in the elderly population (>65 years old) and those with significant co-morbidities.14The presence of colonic dilatation, hypoalbuminaemia and elevated C-reactive protein levels are indicative of significantly advanced disease and are strong prognostic markers of medical treatment failure.

Histology

The histological appearances of UC are characterized by the presence of a predominantly acute inflammatory process associated with the destruction of mucosal cells, particularly epithelial cells. Loss of crypts (atrophy) and destruction of crypt architecture are present at all stages of disease. Crypt inflammation (cryptitis) is a feature of UC that can also be found in CD and infective colitis. Histologically, two features are critical in the diagnosis of UC –chronicity and disease distribution. In the latter, the rectum is invariably involved with confluent inflammation limited to the mucosa or superficial submucosa (except in severe fulminant disease). Basal plasma cells and extensive basal lymphoid infiltrate are indicative of chronicity.In active UC, neutrophilic infiltration, goblet cell depletion and crypt abscesses are prominent.

Treatment

Induction of remission

Medical management is based on the induction and maintenance of remission. 5-aminosalicylic acid (5-ASA; sulphasalazine and mesalazine) is considered first-line therapy for active mild–moderate, left-sided or extensive UC. Combinations of oral and rectal (enema) formulations of 5-ASA lead to faster and higher remission rates. Rectal 5-ASA is often sufficient on its own in active proctitis, when it should be administered as a suppository. In left-sided disease, topical 5-ASA or corticosteroid foam or retention enemas should be used. 5-ASA is more effective than corticosteroid, and therefore should be used as first line.

In patients with more extensive or severe disease, or with disease unresponsive to oral 5-ASA, oral or parenteral corticosteroids are required to induce remission. Although effective in the short term (in up to 70% of the cases), the adverse effect profile of corticosteroids needs to be considered becausea substantial proportion of patients will relapse in the medium term.6

Medical therapies in acute severe UC

High-dose intravenous corticosteroids remain the first-line medical therapy in acute severe UC. In patients who respond, this is typically converted to oral prednisolone after 5–7 days. Prognostic models, such as the Edinburgh Colitis Risk Score, Oxford Criteria and Paediatric UC Activity Index can identify subsets of adults and children who are at risk of not responding to corticosteroids and who will benefit from early second-line medical therapies, such as ciclosporin, infliximab or early colectomy.15

Ciclosporin acts mainly by inhibiting T lymphocyte function, which is essential for the propagation of inflammation. Controlled trial data have shown that intravenous ciclosporin induces remission in 60–80% of patients with severe active UC.16 Although most adverse effects are minor and managed with mild adjustments in dose, severe opportunistic infections, most notably Pneumocystis jirovecii pneumonia, and nephrotoxicity have been reported.

Infliximab, a chimeric monoclonal antibody directed against tumour necrosis factor-alpha (TNF-α), is now routinely used for the treatment of CD. The ACT-1/2 study demonstrated efficacy of infliximab in active UC.17Whilst the primary endpoints of disease response at week 8 were achieved, corticosteroid-free remission at 1 year was observed in only 21%. Jarnerot etal. demonstrated that a single infusion of infliximab 5 mg/kg was effective rescue therapy in hospitalized patients with acute severe UC failing to respond to first-line intensive medical therapy (p = 0.017, odds ratio 4.7).18The current strategy and positioning of infliximab, as with ciclosporin, is to use this as a ‘bridge’ to longer-term immunosuppressive maintenance treatment such as azathioprine. As with ciclosporin, infliximab therapy in this clinical setting is associated with major adverse events, mainly serious infections, which may be fatal. Thus, the use of second-line medical therapy as oppose to early colectomy needs careful risk/benefit consideration. A recent randomized controlled trial of 115 patients showed no difference between ciclosporin and infliximab as second line medical therapy for acute severe UC (failed treatment 60% vs. 54%; adverse effects 16% vs. 25% respectively; p=NS).19

Maintenance of remission

The cornerstone of maintenance of remission is based on 5-ASA therapy, whichreduces the likelihood of relapse over oneyear by three-fold (from 70–80% without therapy to 10–30% with5-ASA maintenance therapy). Azathioprine and 6-mercaptopurine (6-MP) are immunomodulators,which conferclinical benefit in up totwo-thirds of patients. Azathioprine or 6-MP is used in patients who are refractory to, or dependent on, corticosteroids or have poor control with frequent relapses. In this group, azathioprine is superior to 5-ASA alone.20Adverse drug reactions to azathioprine/6-MP occur in 15–28% of patients, often necessitatingwithdrawal of therapy. The most serious adverse effect is profound bonemarrow suppression in patients (1 in 300 individuals) who are deficient in thiopurine methyl transferase (TPMT), the enzyme involved in the metabolism of azathioprine/6-MP. The optimal therapeutic effect for azathioprine/6-MPis observed 2–3 months after the initiation of therapy. Patients who take azathioprine/6-MP should avoid allopurinol (a xanthine oxidase inhibitor) as it inhibits the breakdown of azathioprine/6-MP and increases the risk of myelosuppression.

New and Emerging Therapies

New and emerging medical therapiesfor UC include those targeting:TNF (‘biosimiliar’ agents golimumab, dersalazine, HMPL-004); specific cytokines or downstream transcription factors of the inflammatory response (including IL-2, IL-4, IL-5, IL-13, IL-17, NF-κB);leukocyte trafficking (α4β7 integrin (vedolizumab),α4β1 integrin, MAdCAM, β7, ICAM-1); and immunomodulatory pathways (JAK (tofacitinib), CCR9, CXCL-10, TLR-9 agonist).21At a molecular level, transcriptomic studies that predict disease outcome22 and further understanding of sub-disease mechanisms (for example a sub-group of UC patients with high IL-13may benefit from IL-13 inhibition23) promise to allow stratification of patients to optimise treatment in the future.

Surgery

The indications for surgery are usually categorized into emergency and elective operations. In the former, the absolute indications for surgery are perforation and massive haemorrhage, with more commonindications being failed medical therapy and toxic megacolon (Figure 2). The appropriate timing of emergency surgery in the case of failed medical therapy or toxic megacolon is dependent on careful joint medical–surgical decision-making. Delayed surgery in this setting is associated with increased morbidity, operative complications and mortality. There is general agreement that the operation of choice in the emergency situation is sub-total colectomy with ileostomy and the preservation of the rectum with a plan for completion proctectomy at a later date.

The elective indications for surgery are chronic continuous symptoms, corticosteroid dependency, partial response to medical therapy and the presence of malignancy/dysplasia.

Generally, four surgical approaches are considered in the elective setting:

•total colectomy with end ileostomy

•proctocolectomy with the formation of ileo-anal pouch (ileal pouch–anal anastomosis; IPAA)

•proctocolectomy with ileo-rectal anastomosis

•proctocolectomy with the formation of a continent ileostomy or Kock’s pouch.

Total proctocolectomy with construction of an IPAA has become the standard of care for patients with UC who require removal of the colorectum. This involves abdominal colectomy and construction of an ileal pouch that is anastomosed to theanus. In younger patients, the option of IPAA is more attractive as it avoids the need for a stoma. Laparoscopic colectomy formation is being increasingly performed in carefully selected patients with benefits of reduced disability, quicker recovery, smaller incisions and reduced hospital stay.

The most common and widely reported complication is inflammation of the ileal pouch. Such pouchitis presents with watery, frequent diarrhoea and/or rectal bleeding, accompanied by urgency, incontinence, abdominal cramping, malaise, and fever. The cumulative probability of pouchitis, determined on the basis of symptoms, endoscopy, and histopathology in 468 IPAA patients was 20% at 1 year, 32% at 5 years, and 40% at 10years.24Treatment with ciprofloxacin or metronidazole usually provides rapid clinical improvement. The incidence of pouch failure is reported to be 10% at 10 years, with severe pouchitis accounting for 10% of failures.Ileo-rectal anastomosis may be considered in young female patients in order to avoid proctectomy and pelvic dissection, given that the latter is associated with impaired fertility.

Malignancy risk

Patients with UC have a higher risk of colorectal cancer than the general population. Disease extent and duration are the two critical factors underlying this association, although recent data suggest that disease severity may also be influential. Ekbom and colleagues reported standardized incidence ratios for colorectal cancer risk of 1.7 for patients with proctitis, 2.8 for patients with disease extending beyond the rectum but no further than the hepatic flexure, and 14.8 for patients with disease extending beyond the hepatic flexure.25 A family history of colorectal cancer, the presence of primary sclerosing cholangitis and backwash ileitis are also independently associated with increased cancer risk. Continuation of maintenance 5-ASA and sulfasalazine, cigarette smoking, and possibly folic acid and vitamin E supplementation are protective factors. The estimated risk of colorectal cancer is 2%, 8% and 18% at 10, 20 and 30 years respectively post disease onset. The current British Society of Gastroenterology (BSG)guidelines recommend thatall patients should undergo screening colonoscopy ten years post onset of symptoms, irrespective of extent, to define a risk profile which then informs the interval for further surveillance colonoscopies.Routine biopsies and chromoendoscopy using dye spray with targeted biopsies of suspicious lesions is recommended.