• Welcome to Crohn's Forum, a support group for people with all forms of IBD. While this community is not a substitute for doctor's advice and we cannot treat or diagnose, we find being able to communicate with others who have IBD is invaluable as we navigate our struggles and celebrate our successes. We invite you to join us.

Management of Acute Severe Ulcerative Colitis

Expert Rev Gastroenterol Hepatol. 2009;3(4):395-405. © 2009

Abstract and Introduction

Ulcerative colitis is a chronic relapsing and remitting inflammatory disorder that can generally be managed successfully with maintenance oral medications. However, approximately 15% of patients with ulcerative colitis will develop a severe exacerbation and require hospitalization. While many patients with acute severe ulcerative colitis will respond to a short course of intravenous corticosteroids, up to a third will fail to improve. In these patients with steroid-refractory colitis, the choice is between rescue medical therapy with ciclosporin or infliximab, or surgery. Well-timed rescue medical therapy is generally safe when administered by experienced physicians, and is effective in the majority of cases. Surgery is unavoidable in some cases, but is the treatment of choice in others. While ileal pouch–anal anastomosis offers the prospect of life without a permanent ileostomy, there are issues with its long-term functional outcome.

Ulcerative colitis (UC) is a chronic idiopathic inflammatory disorder of the GI tract characterized by mucosal inflammation of the rectum that extends proximally through the colon, in a continuous fashion, but to a variable extent. The disorder is characterized by a relapsing and remitting course of variable severity. The majority of patients present with left-sided or distal disease of mild-to-moderate severity. Most remain in remission for long periods with maintenance medical therapy. However, natural history studies suggest that between 10 and 40% will undergo a colectomy at some point during the course of their disease.[1,2]

Although most patients present with mild-to-moderate UC, 10% of patients initially present with severe disease. Additionally, approximately 15% of patients will develop a severe flare during the course of their lifetime. Acute severe UC can be defined according to the original criteria set forth by Truelove and Witts; six or more stools per day, with either a body temperature of more than 37.8°C, a pulse rate of more than 90 bpm, large amounts of blood per stool, a hemoglobin level of less than 10.5 g/dl or an erythrocyte sedimentation rate (ESR) of more than 30 mm/h.[3] Both the American College of Gastroenterology practice guidelines and the European Crohn's and Colitis Organisation position statements define severe colitis similarly as the passage of six or more stools per day with evidence of systemic toxicity (e.g., fever, tachycardia, anemia or elevated ESR).[4,5]

Patients with acute severe UC require hospitalization for optimal management owing to the seriousness of their illness.[5] Although rates of death in severe UC have dropped by up to 25% with the adoption of more aggressive monitoring and treatment,[3] acute severe colitis is still associated with a measurable mortality.[6,7] The use of intravenous steroids and improved surgical techniques probably explain much of the reduction in mortality associated with acute severe UC observed in the decades since their introduction.[8]

Severe UC should be considered a medical emergency, and patients require close monitoring of stool frequency and vitals symptoms. Serial abdominal examinations and plain radiographs should be performed. Close observation allows early identification of patients with complications (notably toxic dilatation and perforation), which require immediate colectomy.[9] Baseline laboratory tests, including full blood count, inflammatory markers (e.g., ESR and C-reactive protein [CRP]), electrolytes, liver tests and serum albumin should be obtained and monitored frequently.[10] Dedicated abdominal imaging by computed tomography (CT) may play an important role in selected cases but is not routinely required. Stool samples should be collected for culture and toxin analysis carried out to rule out enteric infection. In addition, a limited unprepared flexible sigmoidoscopy should be carried out (and biopsies obtained) in order to exclude infections that can trigger relapse in patients with UC, particularly Clostridium difficile and cytomegalovirus.[11,12] Full colonoscopy is not necessary and may be associated with an increased risk of complications.

The cornerstone of management of severe UC remains the use of intravenous corticosteroids, which are effective in the induction of remission in the majority of cases.[3,13,14] A daily intravenous steroid dose of hydrocortisone 300 mg or methylprednisolone 60 mg is adequate. There is no proven benefit to using doses in excess of this, although studies examining a weight-based approach to corticosteroid dosing in adults are lacking.[13,15] Additionally, there is no clear evidence that the use of continuous intravenous infusions has any advantage over bolus dosing.[4]

While a range of other interventions are frequently employed in patients with acute colitis, there are sparse data to support their use. Routine use of intravenous antibiotics (e.g., ciprofloxacin or metronidazole) in patients with uncomplicated acute severe colitis confers no therapeutic advantage,[16,17] although it may be appropriate when incipient colonic perforation, toxic megacolon or infection is suspected. Additionally, there is no evidence that 5-aminosalicylate drugs have any effect on disease of this severity and treatment should not be initiated until the acute episode has clearly resolved. We actually recommend discontinuing 5-aminosalicylate in patients with disease of this severity, owing to the rare risk (3%) of a paradoxical response. There is no proven advantage to the routine use of bowel rest or total parenteral nutrition,[18,19] although an active approach to optimize nutritional status is a valuable component of therapy. Efforts to minimize the risk of venous thromboembolism with compression stockings and prophylactic subcutaneous heparin are now advocated by many.[5,10] Some interventions may, however, prove harmful in the setting of acute severe disease and should be avoided, particularly opioid analgesics, antidiarrheals and anticholinergic agents, all of which decrease colonic motility and may increase the risk of colonic dilatation and toxic megacolon.[4,5,10]

Although most patients with severe UC respond to intravenous steroid therapy, approximately 30% of patients fail to respond after 5–7 days and are termed 'steroid refractory'.[13] This article will focus on this group of patients, since they tend to pose the greatest challenge for successful long-term management. For steroid-refractory patients with severe UC, treatment options are generally limited to either continued rescue-medical therapy or colectomy. Choosing the best option in this situation requires an individualized approach and demands direct collaboration between the patient, their gastrointestinal physician and an experienced colorectal surgeon.

Making an early decision about either proceeding to surgery or stepping up to rescue medical therapy is a key aspect of ensuring favorable outcomes, so discussion of the options must be initiated at an early stage. Early identification of patients who are likely to fail steroid therapy is crucial. It has been recognized for some time that monitoring of basic clinical and laboratory indices can be helpful in predicting which patients with acute severe UC are likely to require colectomy during the same hospital admission.[20] Parameters such as disease extent, stool frequency, temperature, heart rate, CRP, albumin and abnormal radiology are all associated with outcome.[13] One of the simplest algorithms is perhaps the most powerful; at day 3 of intravenous steroid therapy, patients with a stool frequency of greater than eight per day, or three per day plus a CRP greater than 45 mg/dl, have an 85% likelihood of requiring colectomy.[21] These are the patients who may benefit from early escalation of therapy. Once a patient has been deemed at 'high risk' for steroid failure, a colorectal surgeon and a stomatherapist should be involved (if they have not been already) to help make a decision regarding colectomy or rescue medical therapy expeditiously. Ideally, the decision should be made within 5 days of the initiation of intravenous steroid treatment.

Medical treatment of steroid-refractory severe UC has expanded somewhat in recent years with the availability of both ciclosporin and infliximab as rescue agents; however surgery still remains the only 'curative' option. We will review the data supporting the use of both ciclosporin and infliximab, with comparison of the long-term outcomes with use of restorative proctocolectomy and ileal pouch–anal anastomosis (IPAA).


Ciclosporin A (CyA) was the cornerstone of rescue medical therapy for steroid-refractory acute severe colitis until the advent of anti-TNF agents. CyA is a competitive calcinuerin inhibitor with potent immunosuppressive properties. Despite convincing data to support the efficacy of CyA as rescue therapy in severe acute UC, it has failed to become established in a widespread fashion as standard therapy due to concerns regarding its toxicity. In the UK national inflammatory bowel disease audit of 2006, only 7.2% of patients hospitalized with acute severe UC received CyA. However, given the option to make an informed choice, most patients will opt for CyA over surgery more readily than their physicians.[22]
The efficacy of CyA in acute steroid-refractory UC was first demonstrated in the landmark study by Lichtiger et al. in 1994.[23] This randomized, placebo-controlled trial showed an initial response in 82% (nine out of 11) of patients who received CyA 4 mg/kg as a continuous intravenous infusion, compared with none of the nine patients in the placebo group. The study is often criticized due to small patient numbers, but recruitment was stopped early by the safety committee because a clear difference in efficacy had already been demonstrated. Over 30 subsequent open-label trials have confirmed clinical response in at least two-thirds of patients with use of CyA and several systematic reviews support its efficacy as a therapy in acute UC.[24–26]

In addition to its use as rescue therapy, CyA monotherapy (4 mg/kg intravenously) has shown similar efficacy to intravenous methylprednisolone for the induction of remission in acute UC in a randomized trial.[27] Thus, CyA monotherapy may be a good option for those with contraindications to steroid therapy (e.g., severe osteoporosis, avascular necrosis and steroid psychosis). However, it should be noted that a number of patients failing monotherapy with either corticosteroids or CyA responded to combination therapy, suggesting that combination therapy (CyA and methylprednisolone) may be more effective.

Toxicity remains the main concern for physicians, and probably explains why CyA therapy has not become more widely accepted. Significant toxicity has been reported in larger case series of patients with both the standard and the low-dose regimens, with serious infections in up to 5% of patients and mortality rates of up to 1–3%.[28–32] The principal side effects of CyA identified with its use in UC are summarized in Box 1, and with proper precautions, a number of these side effects can be avoided. Few would dispute that CyA, as with all immunomodulators, increases the risk for opportunistic and other infections. However, it is virtually impossible to dissect the relative contribution of prolonged high-dose steroids and CyA when adverse events are observed in such patients. Measures exist that may serve to enhance the safety of CyA as a therapeutic option. At least some of the deaths reported associated with CyA therapy have been attributed to Pneumocystis carinii pneumonia,[28,31] and the risk of this opportunistic infection can be mitigated with the use of prophylactic sulfamethoxazole/trimethoprim. Interventions can be also be put in place to minimize the risk of other forms of CyA toxicity. Seizures can arise as a complication of CyA therapy, but the risk can be lessened by controlling blood pressure, correcting hypomagnesemia, closely monitoring levels and using the drug with caution in patients with hypocholesterolemia.[33] Oral microemulsion CyA (5 mg/kg/day) appears broadly equivalent to intravenous CyA in terms of efficacy, and may be a safer alternative in terms of seizure risk for patients with hypocholesterolemia or hypomagnesemia.[34–36] Nephrotoxicity, as evidenced by a rise in serum creatinine, is relatively common, albeit typically mild, and usually responds to lowering of the dose,[28] although cases of irreversible renal impairment have been reported.[37] Importantly, it has been suggested that CyA therapy has no significant additive negative impacts on subsequent surgical outcomes (over the use of intravenous corticosteroids alone),[38] although others have reported a different experience.[39]

To minimize concerns about CyA toxicity, the use of lower doses of CyA has been investigated. In 2003, a randomized trial demonstrated equivalent response rates between the standard CyA dose of 4 mg/kg intravenously (86%) and a lower dose of 2 mg/kg intravenously (84%), and suggested that CyA levels between 150 and 250 ng/ml were sufficient to achieve the desired clinical response. However, owing to small sample size (n = 73), there remains the possibility of a type II error. Unfortunately, the trial itself was too small to demonstrate any difference in adverse effects, but the hope is that the use of lower doses and the achievement of lower plasma levels may help to minimize side effects. Further investigation into this option is needed.

With use of a low-dose regimen, as outlined previously, along with close therapeutic monitoring, the use of preventative measures and early identification of adverse effects, the risk of serious toxicity from CyA can be minimized. The risk of toxicity may also be reduced by limiting the duration of therapy, and observational studies have suggested that CyA therapy can be limited to just a few weeks,[40] although most centers employ maintenance oral CyA for a period of 3–6 months. Figure 1 contains a useful guide to the safe and effective use of CyA in acute colitis, based on published experience.[25,33] Although these options may enhance the safety of CyA, it is still advisable to avoid CyA use entirely in high-risk patients, such as the elderly and those with significant comorbid disease, particularly pre-existing renal impairment.

Figure 1.

Treatment protocol for acute severe UC at days 0, 3 and 5–7.
BP: Blood pressure; CRP: C-reactive protein; CT: Computed tomography; IPAA: Ileal pouch-anal anastomosis; PPD: Purified protein derivative; TPMT: Thiopurine methyltransferase; UC: Ulcerative colitis.

While use of CyA may prove efficacious in the acute setting, the utility of CyA over the long term is less clear. Several studies have examined this issue and have shown colectomy rates at 7 years of between 58 and 88%,[29,32,41] lending support to the notion that few avoid surgery over the long term. However, several of these follow-up studies have shown better outcomes for those successfully transitioned to immunomodulator therapy with azathioprine/mercaptopurine.[32,41] In one study, 80% of CyA responders who successfully transitioned to azathioprine/mercaptopurine avoided colectomy at 5.5 years,[42] suggesting that use of CyA as a bridge to thiopurine immunomodulator therapy can be a successful strategy.

Tacrolimus (FK506) is an alternative calcineurin inhibitor with a mechanism of action and side-effect profile similar to CyA. It is worthy of mention as an alternative to CyA therapy, with the advantage that it has reliable oral bioavailability and can be administered by an oral route. A single, randomized, controlled trial (n = 63) on the use of tacrolimus in steroid-refractory UC has been performed, with improvement noted in 68% of patients dosed with tacrolimus to achieve a high trough level (10–15 ng/ml), compared with 10% in the placebo group.[43] These observations appeared to confirm the results of previous uncontrolled studies,[44] although disease severity in the population studied was less than in the original CyA studies. Nonetheless, the use of tacrolimus as a rescue agent is increasingly recognized within consensus guidelines.[5]

Well-timed use of CyA (or tacrolimus) in the management of acute severe UC is generally safe and effective when administered by experienced physicians. While safety concerns dictate prudence in the way the drug is administered and monitored, outcomes in experienced hands are impressive. Studies that demonstrate better quality of life in CyA-treated patients than in those who underwent surgery[45] only serve to underscore that that these agents are certainly options that are worth exploring in suitable patients.


More recently, infliximab has become available for the treatment of UC. Infliximab is the prototype monoclonal antibody therapy that specifically targets free and membrane-bound TNF-α. Initially licensed as a therapy for moderate-to-severe and fistulizing Crohn's disease (CD), several studies have since demonstrated the efficacy of infliximab for moderate-to-severe UC. It has also been used as rescue treatment for severe, steroid-refractory UC, despite mixed results. Infliximab was approved for use in moderate-to-severe UC based on two large, randomized, placebo-controlled trials: Active UC Trial (ACT) I and II.[46] However, the patient populations studied in these trials differed from that in the CyA trials. Although 217 out of the 700 patients enrolled were steroid refractory, none were hospitalized. Early clinical response rates of 63–77% were reported in the steroid-refractory group, although subsequent remission rates were not reported. A recent meta-analysis examining all available trials concluded that infliximab improves short- and long-term remission rates in patients with moderate-to-severe UC.[47] Short-term responses rates were similar in steroid-refractory and non-steroid-refractory patients.

Several randomized trials have looked specifically at infliximab in the setting of severe steroid-refractory UC. A small pilot study of patients failing a 5-day course of intravenous steroids demonstrated a clinical response in four out of eight patients (50%) who received infliximab relative to none out of three patients (0%) receiving placebo infusion. The study was stopped early owing to poor recruitment, and the numbers were too small to demonstrate significant differences.[48] A subsequent larger study (n = 43), evaluating the effect of two infliximab infusions (5 mg/kg) administered at 14-day intervals into patients with steroid-refractory UC, also failed to demonstrate significant benefit,[49] although it should be noted that patients with severe disease, deemed at risk of urgent colectomy, were excluded. A further study of similar size, which did enroll those at risk for surgery (n = 45), showed a reduced risk of colectomy at 3 months in patients receiving single-dose infliximab (seven out of 24 [29%] versus 14 out of 21 patients [67%]; p = 0.017).[50] This study has impacted significantly on the care of patients with severe steroid-refractory UC, although it should be noted that when patients were subgrouped into fulminant and severe, infliximab no longer appeared statistically more effective in the sicker population. It is also interesting to note that the placebo response rate in this study was 33%, versus 0% in the original CyA study. Although there is evidence to justify the use of infliximab in acute severe UC, with up to two-thirds of inpatients refractory to intravenous corticosteroids avoiding colectomy in the short term,[51] it appears to work less well in the sickest patients. We believe more evidence is needed to establish the efficacy of this agent in severe steroid-refractory disease before it becomes first-line rescue therapy.

Long-term follow-up of infliximab use in the setting of acute severe UC is limited, but initial observations suggest that medium- to long-term rates of colectomy are unlikely to be radically altered by the use of anti-TNF agents.[52] A 2-year follow-up of patients from the Järnerot study has shown a colectomy rate of 46% in patients who received infliximab versus 76% in patients who received placebo, although the results are difficult to interpret.[53] In total, 13 patients in the infliximab group were transitioned to azathioprine (compared with three patients in the placebo group) and only one patient received maintenance infusions of infliximab. Colectomy rates of between 30 and 53% are reported in other studies with median follow-up of 1–2 years.[54,55] Ideally, if infliximab is used to achieve remission, efficacy should be maximized by administration at 0, 2 and 6 weeks, with maintenance infusions every 8 weeks thereafter.

Infliximab may also impact surgical outcomes and peri-operative complications, although the data are mixed. While some groups have reported no excess of early postoperative complications in CD patients treated with infliximab,[56,57] others have noted infliximab use to be associated with an increased risk of early re-admission, intra-abdominal abscesses and postoperative sepsis following ileocolonic resection.[58] Some studies that have examined UC patients did not detect a significant increase in postoperative complications with infliximab,[59,60] except in patients who received a combination of both CyA and infliximab in the preoperative period.[60] However, others have reported increased rates of infectious complications in patients who received infliximab prior to IPAA surgery.[61]

Similar to CyA, infliximab is not without potentially serious toxicity, including infections, lymphoma, drug-induced liver injury and demyelinating disease. However, data from the Crohn's Therapy Resource Evaluation and Assessment Tool Registry regarding the use of infliximab in CD provide some reassurance.[62] They suggest that the significant factors associated with the increased risk of serious infections and death are disease severity and use of corticosteroids or narcotics (rather than infliximab). It seems likely that this may also be the case in acute severe UC, with serious adverse outcomes as likely related to disease severity and the use of high-dose corticosteroids as to the use of anti-TNF agents. However, the increased risk of serious and opportunistic infections with anti-TNF drugs should not be ignored.

When faced with the choice between infliximab and CyA, the challenge for clinicians is the lack of a head-to-head trial specifically comparing the two agents, although we understand one is currently underway. For the moment, the decision is based largely on physician preference, and many gastroenterologists are more comfortable with infliximab, given its widespread use for the treatment of CD. Overall, however, the data appear to favor CyA, especially in patients with more severe disease and severe steroid-refractory UC, and, particularly for the latter, CyA is our drug of choice. There are specific circumstances where infliximab may be the preferred option, particularly in patients already on azathioprine/mercatopurine who show poor short- and long-term responses to CyA.[32] A number of uncontrolled studies have assessed the response to infliximab in patients with moderate-to-severe UC already receiving azathioprine/mercaptopurine with outcomes similar to those seen in immunomodulator-naive patients.[63–66] Therefore, infliximab appears to be the drug of choice for patients who have failed or are intolerant of azathioprine/mercaptopurine. Sequential administration of rescue medical therapy, either CyA following infliximab, or vice versa, is associated with substantial morbidity and should only be considered in very limited circumstances.
Surgical Management

As opposed to medical management, surgery provides a definitive means of treating acute colitis. Surgery decreases mortality and improves quality of life, but comes with the risk of significant perioperative complications and is associated with functional outcomes that are far from what most patients would consider 'normal'. Surgical treatment for UC is clearly indicated for acute severe UC when complicated by severe hemorrhage, perforation or toxic dilatation with impending colonic perforation. However, the most common reason for choosing surgery for UC is the disease being refractory to medical therapy.[68] For such patients, it is imperative that surgery be considered early in the treatment course. Emergency colectomies performed after protracted medical therapy come with a risk of adverse outcomes, including death.[7]

Before the advent of pouch surgery, total proctocolectomy with permanent (Brooke) ileostomy was the optimal procedure in acute severe UC, and it arguably remains the best choice for elderly patients and those with impaired (anal) sphincter function. For the majority of patients, the procedure of choice is usually abdominal colectomy with end ileostomy and Hartmann closure of the rectum.[9] This procedure presents the lowest risk in the acute setting, avoids pelvic dissection and intestinal anastomosis, can be performed using minimally invasive techniques[69] and allows a more definitive procedure to be performed at a later date. Restorative proctocolectomy with IPPA typically follows in a two- to three-stage procedure and is also now increasingly performed using laparoscopic techniques.[70]

Although surgery 'cures' colitis, improves quality of life – which is generally poor in patients with severe colitis[62] – and eliminates the long-term risk of colitis-associated dysplasia or cancer, there are known morbidities. Restorative proctocolectomy with IPAA restores physiological fecal flow, removing the need for ileostomy, but most patients do not return to what they would consider normal bowel function. In patients who undergo subtotal colectomy, creation of a functional pouch is usually a three-stage procedure, taking approximately 6 months to complete. Postoperative complications occur quite frequently, with pelvis sepsis particularly feared owing to its association with subsequent poor pouch function.[71–73] Functional outcomes following IPAA are somewhat variable and are summarized in Table 1. Many patients need to wear incontinence pads. Mean daytime stool frequency is 5.7 at 1 year and 6.4 at 20 years, by which time the mean nocturnal stool frequency increases from 1.5 to two.[74] Reports of frequent daytime fecal incontinence increase from 5 to 11% with prolonged follow-up, with frequent nocturnal incontinence reported by over 20% at 20 years. Pouchitis is the most common complication of pouch reconstruction. Half of all patients experience an episode of pouchitis within 10 years,[71] and chronic pouchitis develops in up to 15% of patients.[75] A range of other mechanical, inflammatory, neoplastic and functional pouch-related disorders have been described.[76] Overall, pouch failure leading to end ileostomy occurs in up to 10% of cases.[71] There is also increasing awareness about the effects of IPAA on reproductive health and function. In particular, decreased fertility following pouch surgery is increasingly recognized. IPAA may increase the risk of female infertility by as much as threefold.[77] Despite these issues, it is worth stating that most patients are highly satisfied with the results of pouch surgery, report good quality of life and would recommend surgery to others.[78] It is not uncommon to hear patients state, 'I cannot believe I waited so long to have surgery'. The functional outcomes and morbidities of the IPAA must be weighed against the long-term risks of continued medical therapy and the possibility of recurrent flares.

Careful counseling is necessary to fully explain all of these possible outcomes to patients before surgery is undertaken, and patients should be offered this information early on in the course of an episode of acute severe UC. Counseling regarding surgery should be balanced, giving equal weight to the positive aspects of surgery, such as reduced need for ongoing drug therapy (with potential toxicity) and elimination of colon cancer risk. In cases of severe UC, the risks and benefits of all available options should be discussed openly and in detail so that patients can make an informed decision between early surgery and continued medical therapy. Quality-of-life issues must form a key component of the discussion,[45] which should be an open process involving the patient, gastroenterologist and surgeon. Early engagement with the patient about these issues facilitates early decision making, and the importance of avoiding delays in the choice between escalation of medical therapy and surgery cannot be overstated.

Novel/Experimental Therapies

Despite the many advances in recent decades, it is clear that the treatment of severe steroid-refractory UC is not optimal and that new options for therapy are needed. A number of uncontrolled trials or pilot studies have reported the use of novel approaches to the treatment of severe UC. Targeting the key proinflammatory cytokine IL-2, using themonoclonal antibody basiliximab, has been assessed as a possibl e strategy. Early results of a pilot study of the effect of a single infusion for treatment of moderate-to-severe UC were impressive (nine out of ten successfully treated; 90% remission at 8 weeks), but the effect was not sustained.[79] Follow-up of the use of basiliximab, specifically in severe steroid-refractory UC, reported four out of seven patients having colectomy by 24 weeks.[80]

Visilizumab is another novel monoclonal therapy that has been studied in severe steroid-refractory UC. It consists of an IgG2-humanized monoclonal antibody to CD3 expressed on human T lymphocytes. In an open-label study of 32 patients, remission rates of 40% at 30 days were reported, and almost half of the patients had avoided colectomy by 1 year.[81] However, Phase III trials of the efficacy of visilizumab in patients with severe UC who were failing steroid therapy have been stopped owing to concerns about safety and lack of efficacy. Another monoclonal compound, MLN02, a humanized antibody to the α4β7 integrin, has demonstrated efficacy for the induction and maintenance of remission in UC,[82] but it has not been specifically studied in patients with severe colitis refractory to steroid therapy.

The use of a number of other immunosuppressants, such as ciclophosphamide,[83] in severe UC has been reported in pilot or uncontrolled studies. While benefits to such therapies are suggested, they have not received acceptance, owing to a lack of data from placebo-controlled trials. A number of other novel approaches have reported success in uncontrolled studies, including leukocyte apheresis,[84] thalidomide,[85] IFN-β[86,87] and phosphatidylcholine.[88] At present, their use outside of randomized, controlled trials would seem difficult to justify.

Expert Commentary

Although the majority of patients with chronic UC never require admission to the hospital and can be managed successfully with oral medications, approximately 15% of UC patients will require hospitalization for an episode of severe colitis. Fortunately, most patients hospitalized with an acute flare of UC will respond to intravenous corticosteroids within a few days. For the minority who fail to improve, a prompt decision must be made regarding colectomy or continued medical management. This is a choice that must be individualized to the patient's needs and preferences. Patients who present for the first time with an acute UC flare face significant psychological adjustment to their diagnosis and often feel unable to cope with the idea of colectomy. Often, patients who have already lived for a time with symptomatic colitis that has been a challenge to control will feel very differently and may opt for surgery earlier in the course of an acute flare. The physician's responsibility is to ensure that patients understand the risks and benefits of both approaches. This can best be achieved by open dialogue between the patient, their gastrointestinal physician and a dedicated colorectal surgeon at an early stage of the hospital admission. Particular effort should be focused on patients who have markers predicting a high likelihood of colectomy. Such an approach can facilitate early decision making, with the choice between surgery and rescue medical therapy ideally made by day 5 of the hospital admission. Although surgery is curative, one must take into account the morbidity and long-term functional outcomes. Figure 1 provides a useful summary of a practical approach to early decision making and the choice between rescue medical therapy and surgery.
While surgery always remains an option, most patients, given the choice, will opt for a trial of second-line medical therapy prior to colectomy.[89] Clinicians now face the choice between CyA and infliximab as a rescue agent, understanding that there are studies to support the use of either agent in severe UC. At present, published data provide better evidence for the efficacy of CyA in hospitalized patients with steroid-refractory disease. Infliximab is an attractive option in some patients, particularly those patients who have developed a flare while taking a thiopurine immunomodulator or who have proven intolerant to such agents. In these patients, there is no clear transition strategy from acute use of CyA, and these patients are probably best served by induction and maintenance infliximab or surgery. In patients presenting for the first time with an acute steroid-refractory flare or with no prior history of immunomodulator exposure, the optimal approach is less clear. However, in such cases, we typically favor the use of CyA. Infliximab could be considered in those patients with disease of moderate severity or with indeterminate inflammatory bowel disease. When an initial response is achieved with infliximab, we would advocate use of an induction (0, 2 and 6 weeks) and maintenance (every 8 weeks) regimen similar to that used in CD. Owing to the widespread use of infliximab in treatment for CD, many gastroenterologists find the administration of this drug much more within their 'comfort zone' than the use of intravenous CyA. However, this factor alone should not shape practice. In the hands of an experienced physician, and with appropriate precautions and monitoring, CyA use is safe and effective. There are more convincing studies that have examined the use of CyA specifically in hospitalized patients with severe steroid-refractory UC, while specific studies of infliximab have been small and somewhat inconsistent. Use of CyA should be considered for all such patients, and, if local experience is lacking, referral should be made to a specialty center with experience in CyA administration. Clearly, the results of controlled comparisons of the use of infliximab versus CyA in such patients are eagerly awaited.

Five-year View

Much progress has been made in the management of acute severe UC in the last 50 years, and there is no disputing the massive impact of treatment with intensive high-dose intravenous corticosteroids and early colectomy for nonresponders on the natural history of acute severe UC. Further progress in rescue medical therapies has produced less dramatic improvements, and more effective medical treatment options are needed.

The next 5 years will certainly bring us to a better understanding of how to target existing rescue therapy. Results of controlled trials comparing CyA to infliximab, with careful subgroup analysis and appropriate follow-up, will allow us to direct therapy to the groups of patients who will benefit most from one or the other agent. Monitoring of basic clinical and laboratory parameters will help identify high-risk patients, and the coming years may see the addition of more-specific molecular markers (e.g., fecal calprotectin) to further enhance prognostication.

One area where meaningful progress may be possible in the coming years is in our understanding of what triggers flares of UC and, in particular, what specific triggers are associated with steroid-refractory disease activity. Recent advances have demonstrated the importance of infections, such as cytomegalovirus infections. Infections with other specific pathogens, or alterations in the endogenous colonic microflora, may be significant, and an understanding of such factors may also open new therapeutic avenues. Broadly speaking, improvement of the maintenance of remission, with the goal of minimizing the risk of severe flares, is perhaps the area with the greatest potential yield.