Review Article

The Practical Management of Acute Severe Ulcerative Colitis

D. Seah; P. De Cruz


Aliment Pharmacol Ther. 2016;43(4):482-513. 

In This Article

Complications Associated With ASUC

The management of ASUC encompasses the appropriate diagnosis and treatment of common complications, encountered in clinical practice. Management of these complications is discussed in further detail below.

Opportunistic Infections

Clostridium Difficile. Infectious precipitants of a flare of colitis must be screened for and excluded in patients presenting with ASUC. Inflammatory bowel disease is associated with an increased risk of C. difficile infection (CDI), particularly among those with colonic disease – even in the absence of recent antibiotic exposure.[124] CDI is formally defined as the presence of diarrhoea with the biochemical detection of C. difficile or C. difficile toxins, or colonoscopic evidence of pseudomembranous colitis.[125] Twenty-five per cent of IBD patients, who develop CDI, have no other risk factors predisposing them to the condition. The incidence of CDI has been found to be 2.9 times higher in IBD patients compared to non-IBD patients, and four times higher in patients with UC.[126] Corticosteroid initiation in IBD has also been shown to triple the risk of developing CDI compared to any other immunosuppressant – including thiopurines and methotrexate (RR: 3.4; 95% CI: 1.9–6.1).[127] IBD admissions complicated by CDI have been associated with a higher mortality rate (adjusted OR = 4.7, 95% CI: 2.9–7.9) than non-CDI IBD admissions, with higher rates of bowel surgery and mortality seen in UC patients compared to their Crohn's disease counterparts.[128]

In particular, CDI must be suspected in patients, who present with signs and symptoms, inconsistent with their regular exacerbations. Clinical suspicion necessitates early diagnosis and appropriate management. Diagnosis is generally made on the basis of stool analysis.[125] Specific investigations include enzyme-linked polymerase chain reaction (PCR) analysis of stool samples or a glutamate dehydrogenase assay; either of these tests can be used in combination with enzyme-linked immunosorbent assays for C. difficile toxins.[129] On meta-analysis, the glutamate dehydrogenase assay exhibited >90% sensitivity and ≥98% specificity when compared with stool culture. Endoscopic and histologic findings of CDI in IBD patients are either less obvious or absent compared with non-IBD patients and therefore cannot be relied upon in isolation for a diagnosis.[124]

Given the morbidity and mortality associated with CDI in the setting of ASUC, first-line therapies should include metronidazole and vancomycin.[130] Fidaxomicin may be an alternative to the latter strategy but has not been formally evaluated in the IBD setting. The initial use of vancomycin, in lieu of metronidazole, has been retrospectively shown to have reduced colectomy rates from 45% to 3% over a 2-year interval.[131] CDI can be classified into severe and nonsevere based on the presence of endoscopic evidence of pseudomembranous colitis, treatment in the ICU or two or more of the following: age >60 years, temperature >38.3 °C, hypoalbuminaemia, or white cell count >15 000 cells/mm3.[132] When compared to metronidazole, vancomycin use in nonsevere CDI was associated with lower 30-day readmission (0% compared to 31%; P = 0.04) and results were suggestive of lower rates of 12-week readmission (15% compared to 31%; P = 0.45) and colectomy during the index admission (15% compared to 24%; P = 0.70), although the latter two associations did not reach statistical significance.[133] Although faecal microbiota transplantation appears to have a role in recurrent CDI, it has not been fully evaluated in patients with ASUC.[134] A concern is the increased risk of bacterial translocation, given the ulcerated and denuded mucosa, which may increase the risk of systemic toxicity.

Management of CDI should occur simultaneously with conventional therapy for ASUC; however, the use of immunomodulatory agents in the context of CDI remains controversial. A small case series in the non-IBD setting, demonstrated steroid efficacy in six patients with pseudomembranous colitis.[135] Based on the latter small study, ECCO guidelines continue to support the use of corticosteroid therapy when treating CDI.[136] A 12% rate of adverse outcome, indicated in a multicentre retrospective study among those who were treated with immunomodulators in addition to antibiotics, has suggested that immunomodulators should be withdrawn in the setting of C. difficile in IBD.[130] Adverse outcomes were defined as death or colectomy within 3 months of admission, in-hospital megacolon, bowel perforation, haemodynamic shock or respiratory failure. Prospective data supporting these findings are required before immunomodulator cessation can be routinely recommended. Recent guidelines suggest that ongoing immunosuppression should be considered on a case-by-case basis and perhaps maintained in severe colitis, while warning against dose escalation in untreated CDI.[137] Infliximab has not been shown to be associated with any increased incidence of serious bacterial infection, including CDI. However, the safety and efficacy of treatment with infliximab in comparison to ciclosporin in patients with ASUC, complicated by CDI requires further investigation.

Cytomegalovirus. In patients with UC, CMV infection (or reactivation) must be carefully distinguished from active CMV colitis – a term used to describe CMV infection leading to colonic inflammation. In most cases, uncomplicated CMV infection in patients with IBD remains subclinical and self-limiting and latent CMV reactivation, while detectable in the blood, does not influence disease activity during a flare.[138,139] A prospective study of 69 patients with moderate to severe UC demonstrated no significant differences in outcome between patients who were seropositive and those who were seronegative for CMV infection.[139] Furthermore, a retrospective case–control study of CMV-infected patients, diagnosed on blood PCR, compared with CMV negative controls, revealed no difference in colectomy rate and admission length between the two groups.[138]

In contrast to this, CMV infection that causes colonic inflammation, termed CMV colitis, has been associated with higher rates of emergency colectomy and poorer patient outcomes.[140] The diagnosis of CMV colitis is generally made on the basis of tissue analysis of colonic biopsy specimens, taken during an acute flare of colitis. The mechanism by which CMV infection precipitates acute colitis is thought to be related to ischaemic, thrombotic and autoimmune changes.[141–144]

Immunosuppressive therapies are a major risk factor for the development of CMV disease in IBD. In particular, prospective data have demonstrated that concomitant treatment with azathioprine and corticosteroids is associated with CMV infection.[145] Diagnostic criteria for CMV infection in this study consisted of a positive result in one or more of the following tests: IgM antibody, CMV PCR of colonic biopsy and inclusion bodies on haematoxylin & eosin stain of a colonic biopsy specimen. Standard induction therapy (week 0, 2, 6) with salvage therapy agent, infliximab, does not appear to increase the risk of developing CMV in patients with IBD.[146] Current ECCO guidelines recommend screening for CMV infection only in the context of acute steroid-resistant colitis.[136] Over a third of patients with steroid-refractory colitis have biochemically active CMV; as such, exclusion of clinically significant infection is of high priority in this patient group. Sigmoidoscopy and colonic biopsy to detect CMV should be performed within 48 h of admission for ASUC and tissue samples analysed on the basis of histopathology (CMV inclusions denoted by an 'owl's eye' appearance), immunohistochemistry and/or PCR results.[136] Timely administration of anti-viral therapy, with 2–3 weeks of ganciclovir, is the current standard of care in patients with confirmed infection.[147] CMV infection should be excluded prior to escalation of immunomodulator therapy to minimise the risk of precipitating opportunistic CMV colitis.[136] Careful consideration should also be given to the option of discontinuing immunomodulators until colonic inflammation subsides. In practice, while delaying immunosuppressant therapy may be appropriate in a subset of patients with ASUC, it may not be feasible in the setting of 3–5 days of steroid nonresponse. In these cases, rapid sequential or concurrent anti-viral therapy may need to be administered alongside a salvage therapy agent. Prompt withdrawal of immunomodulator therapy is required in systemic CMV disease.

Venous Thromboembolism

Venous thromboembolism (VTE) is a potentially fatal complication of IBD that is closely linked with active disease. The risk of VTE in IBD is three times higher than that seen in the general population and the likelihood of developing VTE is even higher during a flare (HR 8.4; 95% CI: 5.5–12.8; P < 0.0001).[148] The mechanism by which IBD increases thrombotic risk is unclear. Low molecular weight heparin and fondaparinux have been shown to significantly reduce VTE risk in hospitalised patients.[149] Despite the latter evidence, surveys indicate that up to 65% of gastroenterologists fail to administer VTE prophylaxis to patients admitted for ASUC.[150–152] Thromboprophylaxis does not appear to precipitate excessive bleeding during flares of IBD, even in patients experiencing bloody diarrhoea, and is recommended in all patients with ASUC.[153,154] The VTE risk in IBD patients appears to be highest in ambulant out-patients following discharge from hospital (HR 15.8; 95% CI: 9.8–25.5, P < 0.0001), suggesting that there may be a role for maintaining prophylaxis in the out-patient setting. However, there are no published data to support this recommendation. Prospective studies to evaluate the efficacy and safety of VTE prophylaxis in IBD are needed.

Toxic Megacolon

Toxic megacolon (TM) is a life-threatening condition, which complicates approximately 5% of presentations with ASUC.[155] It is classically defined according to the Jalan clinical criteria (Table 7) and describes segmental or entire colonic distension accompanied by a systemic inflammatory response.[156–158] A high index of clinical suspicion is required – especially in the setting of severe, refractory or worsening symptoms, as patients with TM are often profoundly unwell and carry a heightened risk of poor outcomes. Complications of TM include colonic perforation, peritonitis, massive haemorrhage and, in rare cases, even death – seen in 0–2% of ulcerative colitis.[157,159]

Clinical signs cannot be relied upon for the diagnosis of TM as corticosteroid therapy is known to attenuate examination findings.[160,161] Daily abdominal plain films should be performed for all patients with ASUC, looking for gas distension of the transverse colon >6 cm in diameter and the disappearance of normal haustral patterns.[158,162] Plain films may also be useful for observing distension of the ascending colon and/or the presence of colonic air fluid levels together with exclusion of a pneumoperitoneum.[157] Although computed tomography imaging is the optimal imaging modality for patients with TM, it carries with it a risk of unnecessary radiation exposure and therefore should not be used as a serial imaging modality.[163–166]

Management of TM involves close collaboration between physicians and surgeons. While some patients may respond to medical therapy with high-dose corticosteroids, broad-spectrum antibiotics and salvage therapies, although not an absolute indication for surgery, there should be a low threshold for surgical intervention.[15,156,157,167–169] Emergent colectomy is associated with a significantly higher mortality rate compared to elective colectomy (30% and 5%).[170] Medications that reduce colonic motility should be withdrawn – anti-diarrhoeals, anti-cholinergics and narcotic drugs. Total parenteral nutrition in order to facilitate bowel rest may be considered as an adjunct to surgery; however, it has not been shown to exert any primary therapeutic effect.[171] Although positional gas redistribution through rolling and adopting the knee-elbow position has had some success, as has endoscopic decompression, these techniques are not a substitute for definitive surgical management in all patients.[172–175]