Ulcerative colitis

Introduction

Ulcerative colitis (UC) is an idiopathic relapsing inflammatory bowel disease (IBD) involving the mucosa and lamina propria of the rectum and variable extent of the proximal colon. Characterised by remissions and exacerbations, the clinical spectrum of disease can range from inactive to fulminant. Medical management is generally effective in controlling UC, but ultimately 30–40% of patients will require surgical intervention. Criteria for the management of acute and chronic disease are well established, with surgery playing a fundamental role, as removal of the colon and rectum is essentially curative.

Epidemiology

UC is an uncommon disease with varying incidence rates (0.5–24.5/100 000) and discernible differences are seen between different geographic and ethnic regions of the world. Less common in Asia, Africa, South America and South-eastern Europe, UC has a varied incidence of between two and 15 cases per 100 000 persons per year in developed and industrialised Western countries of North America, North-western Europe and the UK. A significant trend of increasing incidence and prevalence rates has been reported in under-developed parts of the world as they become more industrialised, thus supporting the importance of environmental factors in the development of UC.

The onset of symptoms typically plateaus around the fourth decade of life, remaining fairly constant thereafter. A second peak of onset around the sixth to seventh decade has been described, though there is uncertainty as to whether this is truly a subsequent peak or merely difficulty in differentiating it from other colitides.

UC is seen with near equal frequency in males and females. Whites and blacks have a nearly equivalent incidence, while the Jewish populace experiences the highest documented rates. Hispanic, Native American, African and Asian populations have the lowest incidence.

Aetiopathogenesis

The pathogenesis of UC remains enigmatic, though multiple factors have been described as potential causative or protective agents in its occurrence and include: diet, alcohol and tobacco consumption, socioeconomic status, hygiene, urban living conditions, antibiotic usage, gut flora dysbiosis, probiotic use, non-steroidal anti-inflammatory agents, appendicectomy, breastfeeding, oral contraceptive use, stress, and familial and genetic causes.

Though a significant number of dietary factors have been evaluated as potential causative agents for UC, no consensus has emerged. A decreased risk has been associated with alcohol consumption, and the risk declined as daily alcohol consumption increased.

Evidence demonstrates that smoking is protective against disease activity, and it has been shown that those who quit smoking are more likely to have a relapse. Ex-smokers are 70% more likely to develop UC when compared to those who have never smoked, though the causation remains unclear. Supplemental nicotine therapy has not consistently been shown to be more effective than placebo or conventional therapy (steroids/5-aminosalicylic acid [5-ASA]) and has a significant side-effect profile.

The hygiene hypothesis contends that cleaner living environments reduce the amount of organisms one is exposed to early in life, thus reducing the ability of the immune system to become tolerant, and subsequently causing an aberrant response when thus exposed. UC is more common among urban populations, is associated with indoor living, smaller families, and individuals of middle and upper socioeconomic status, who primarily reside in more sanitary surroundings.

Antibiotic usage and the resulting gut flora dysbiosis are commonplace in developed countries, and hypothetically a potential cause of UC when taking into consideration that higher rates of utilisation are seen in industrialised and developed nations, though this is yet to be proven. A correlation has been demonstrated in children with UC, who are more likely to have received antibiotics during their first year of life.

A predisposition for UC has been reported to be as high as 29% in those with a positive family history, and between 10% and 20% of affected individuals have a first-degree relative with IBD. Twin studies have consistently shown a higher concordant disease rate in monozygotic compared to dizygotic pairs (approximately 50% vs. nearly 0%), where the concordance among ordinary siblings was found to be around 5%.

Clinical presentation

Colonic involvement at presentation can vary widely between different geographic regions, though proctosigmoiditis is the most common. In the USA 46% presented with proctosigmoiditis, 37% pancolitis and 17% with left-sided colitis.

Common symptoms associated with UC include urgency, diarrhoea, tenesmus and haematochezia. Constipation, a complaint in 15–20% of patients, is related to incomplete evacuation of the rectum. Symptoms correlate with severity of disease, and increasing severity leads to worsening nausea, emesis, abdominal distension and weight loss. Protein-losing enteropathy may lead to loss of lean body mass and anaemia, and growth retardation in children. Haemodynamically significant haemorrhage is an uncommon complication but is responsible for 10% of emergency colectomies. Severity can also have systemic manifestations, including tachycardia, pyrexia, leucocytosis and increased fluid requirements, indicating toxicity.

Approximately 5–15% of patients with UC develop acute severe colitis, and up to 50% present initially with fulminant disease. Intense medical treatment has a high chance of inducing remission but when unsuccessful, urgent surgery will be necessary in up to 20% of patients. Perforation is a rare but serious occurrence, with a mortality approaching 60%.

Diagnosis and evaluation

With an extensive differential and no one exclusive pathognomonic test, a firm diagnosis of UC is dependent on several factors, including the clinical presentation, radiological workup, endoscopic evaluation and histopathological determination of tissue biopsies. The differential diagnosis can include infectious (viral, bacterial, protozoal) as well as non-infectious causes (Crohn’s disease, indeterminate colitis, collagenous colitis, ischaemic colitis, radiation colitis, diversion colitis, pharmacotherapy-induced colitis), and obtaining a detailed history and physical examination is imperative.

Colorectal cancer and surveillance

Prolonged duration, continuously active disease, severity of inflammation, primary sclerosing cholangitis (PSC) and diffuse involvement (pancolitis) are cumulative risk factors for the development of colorectal cancer in the setting of UC. Incidence rates for the development of cancer correspond to cumulative probabilities of 2% by 10 years, 8% by 20 years and 18% by 30 years. As a general rule, beginning 10 years after the diagnosis of UC, the incidence of colorectal cancer increases by approximately 1% per year as long as the patient has their colon. The relative risk for cancer in relation to ulcerative proctitis has been estimated to be 1.7, left-sided colitis 2.8 and pancolitis 14.8. In relation to the general population, there is an overall eightfold higher risk of colorectal cancer, with a 19-fold higher risk in patients with extensive colitis. It has been shown that roughly 17% of all deaths in UC are a result of colorectal cancer.

Flat low-grade dysplasia (LGD) detected during colonoscopic evaluation confers a reported ninefold increased risk of developing colorectal cancer and a 12-fold risk of developing an advanced lesion (high-grade dysplasia [HGD], or cancer). Progression from colitis without dysplasia to colorectal cancer does not necessarily follow a sequence of LGD, HGD and ultimately carcinoma (inflammation–dysplasia–carcinoma sequence). Rather, LGD can progress directly to colorectal cancer. LGD can present as unifocal or multifocal and treatment is to some extent controversial, with some advocating prophylactic colectomy, while others recommend intensive colonoscopic surveillance. Flat LGD has been shown to be a strong predictor of progression to advanced neoplasia (53% at 5 years) in surveillance colonoscopy, and in patients who underwent a colectomy, an unexpected advanced neoplasia (HGD or cancer) was found in nearly 24%. Other studies have shown the presence of LGD is as likely as HGD (54% vs. 67%) to be associated with an already established cancer. Repeated attempts to show LGD on endoscopic examinations should not be undertaken; rather, proctocolectomy is recommended to prevent progression to HGD or cancer.

Flat HGD has been shown to have a 42–45% rate of associated colorectal cancer at the time of colectomy, thus maintaining the recommendation that colectomy is mandatory in these patients, even with completely resected HGD or with HGD found on random biopsies.

Several studies have suggested that surveillance colonoscopy in patients with UC significantly reduces the risk of developing neoplasia. To date, no randomised controlled trials have documented a reduced risk of colorectal cancer development or death by utilising surveillance colonoscopy. The American Gastroenterological Association and British Society of Gastroenterology share international guidelines recommending surveillance colonoscopy every 1–2 years starting 8–10 years after a diagnosis of pancolitis, or 15 years after left-sided colitis. Recommendations are also advocated for random non-targeted biopsies performed every 10 cm in all four quadrants, equating to 20–40 biopsies per colon. Colitis-associated cancers have been shown frequently to arise from flat mucosa, be multifocal, broadly infiltrating, anaplastic and uniformly distributed throughout the colon. It is estimated that 33 non-targeted biopsies are required to detect dysplasia with 90% confidence, though studies show this is not often achieved.

Patients with PSC and UC have an increased risk of colorectal cancer in comparison to those without PSC. Cumulative colorectal cancer risk has been described as 33% at 20 years and 40% at 30 years after a diagnosis of UC. Surveillance colonoscopy is recommended at the time of diagnosis of PSC and yearly thereafter. In patients without a known diagnosis of UC, diagnostic colonoscopy with random biopsies is recommended to evaluate for subclinical evidence of disease.

Flat and depressed colorectal lesions are often missed with conventional ‘white light’ colonoscopy and the utility of obtaining non-targeted random biopsies has been called into question. Though highly specialised, time-consuming and not universally available, the use of high-magnification chromoscopic colonoscopy (dye spraying of the mucosal surface with indigo carmine or methylene blue) has been shown to have a significantly better correlation between endoscopic and histopathological findings than conventional colonoscopy, and also increased the number of neoplastic lesions identified. Chromoendoscopy (CE) has been repeatedly shown to increase the chance of detecting dysplasia compared to standard colonoscopic surveillance, allowing the endoscopist to take fewer, but rather higher yield, biopsies.

Severity assessment

Disease severity is classified as mild, moderate or severe, and is based on the original descriptions provided by Truelove and Witts. Mild disease is characterised by less than four stools daily, with or without macroscopic blood, no signs of systemic toxicity (fever, tachycardia), mild to no anaemia and a normal ESR. Severe disease results in six or more bloody bowel movements daily, signs of systemic toxicity, anaemia (less than 75% of normal value) and an increased ESR.

Unlike the Crohn’s Disease Activity Index, no gold standard index exists for evaluating the severity of UC. Rather, multiple indices have been developed to measure disease severity and activity in clinical trials, many with overlapping measured variables. Some assess the clinical and biochemical aspects of disease (Truelove and Witts Severity Index, Lichtiger Index, Powel–Tuck Index, Activity Index, Rachmilewitz Index, Physician Global Assessment, Ulcerative Colitis Clinical Score), others focus on endoscopy (Truelove and Witts Sigmoidoscopic Assessment, Baron Score, Powel–Tuck Sigmoidoscopic Assessment, Rachmilewitz Endoscopic Index), while further indices evaluate a combination of clinical and endoscopic criteria (Mayo Clinic Score, Sutherland Index, Ulcerative Colitis Endoscopic Index of Severity). Common to all indices are numerical scoring systems. Scores at the elevated end of the spectrum indicate high disease activity, while lower scores signify milder or quiescent disease.