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

Salvage Therapies

Historically, failure to induce clinical remission with intravenous corticosteroids invariably led to colectomy. The introduction of medical rescue, or salvage, therapies has provided an alternative option to patients previously facing only surgical management. Salvage therapy appears to have improved clinical outcomes in patients with ASUC and in particular, its use has been associated with a reduction in overall colectomy rate. Kaplan et al. identified patients within the Canadian Calgary database, admitted with ASUC between 1997 and 2009, demonstrating a significant reduction in elective colectomy rate.[48] Applying linear regression models, investigators found that the average annual per cent change (AAPC) in elective colectomy rate was −7.4% (95% CI: −10.8% to +3.9%). Correlating with this, thiopurine use increased across this time period (OR = 1.15, 95% CI: 1.09–1.22) alongside infliximab use, which had increased since 2005 (OR = 1.68, 95% CI: 1.25–2.26). Emergent colectomy rates, which are associated with significant morbidity and mortality in this population, were relatively unchanged (AAPC: −1.4%, 95% CI: −4.8%, +2.0%). Overall, these data suggest that patients with chronic active colitis are more likely to respond to infliximab than patients with an acute severe flare. The current infliximab dosing schedule (weeks 0, 2, 6) does not appear to be adequately effective in preventing progression to emergent colectomy. Clinicians may need to reconsider optimal infliximab dosing and adopt a more intensive dosing strategy in order to minimise colectomy rates in the acute setting.

There are a number of options available when considering salvage therapy. Ciclosporin and infliximab have been the focus of most active research and currently represent the mainstays of salvage therapy; however, a number of emerging therapies are also under investigation.

Salvage therapies are predominantly used in the context of steroid failure; however, their use as a first-line treatment option in selected cases of inflammatory bowel disease has been proposed in order to minimise immunosuppression-associated morbidity.[49] The latter 'top-down' management approach has been suggested for patients at high risk of steroid failure; however, it carries the risks of overtreatment as well as the potential adverse effects associated with the potent salvage therapies themselves.


Short-term Efficacy. Lichtiger et al. first demonstrated the efficacy of the calcineurin inhibitor ciclosporin in ASUC, in a randomised controlled trial of 20 steroid-resistant patients, who had failed to respond to ≥7 days intravenous hydrocortisone together with nightly hydrocortisone enemas.[50] Steroid therapy was maintained in the enrolled patient sample while in addition, 11 patients received 4 mg/kg/day intravenous ciclosporin and 9 received placebo. Of the 11 patients treated with ciclosporin, 9 (82%) showed significant improvement after 7 days – measured as a decrease in clinical activity score from 13 to 6 on a modified Truelove and Witts' severity index (Table S2[7] In contrast, the placebo group demonstrated minimal improvement (P < 0.001) – with a mean score decrease from 14 to 13. The relative risk of failing to induce remission with ciclosporin was 0.18 (95% CI: 0.05–0.64). At 2-week follow-up, 3 of 11 patients (27%) had undergone colectomy in the intervention group, in contrast with four of nine placebo group patients (44%) – with the relative risk of colectomy calculated as 0.6 (95% CI: 0.18–2.06).[51] After failure to achieve the primary endpoint of a reduction in Lichtiger score below 10, five of the placebo treated patients subsequently responded to open-label ciclosporin. Taken together, these findings successfully helped establish the role of ciclosporin in steroid-refractory ASUC.

A randomised double-blinded Belgian study of 30 patients with ASUC directly compared ciclosporin and steroid therapy. Nine of 14 subjects (64%), who received ciclosporin, entered remission, as compared to 8 of 15 (53%) who received methylprednisolone.[52] At 1-year follow-up, seven of nine patients (78%) initially controlled with ciclosporin had maintained remission, compared to three of eight (37%) initially controlled with methylprednisolone. A 2005 Cochrane review suggested that the short-term efficacy of ciclosporin monotherapy is similar to that of conventional steroid treatment; however, this analysis was limited by the inclusion of only two prospective studies.[51]

Randomised controlled trials evaluating ciclosporin use have reported rates of immediate response, at 1–2 weeks, between 64% and 82%.[50,52] Results from the latest round of the UK IBD audit (2014) found that 26% of patients (68/261) treated with ciclosporin following steroid failure progressed to colectomy prior to discharge.[53] In clinical trials, colectomy rates at 3 months have ranged from 26% to 47%.[10,50,52,54,55] Rates of short-term clinical response, 3 months following ciclosporin have been between 40% and 54%.[10,54,55] These data are described in Table 4.

Long-term Efficacy. Despite proven efficacy for induction of remission, ciclosporin has been significantly less successful at maintaining remission at longer term follow-up. Campbell et al. demonstrated a 65% relapse rate at 1-year follow-up post-ciclosporin administration and a 58% colectomy rate at 7-year follow-up, while Actis et al. demonstrated a 1-year relapse rate of 56% and a 100% relapse rate after 7 years.[5,56] Patients in both studies were treated with intravenous ciclosporin infusion or oral ciclosporin, with initial dosing set at 2 or 4 mg/kg/day intravenously or 5 mg/kg/day orally. The heterogeneity of interventions both between the studies and within each study itself may account for the differences in reported treatment effects. Numerous studies advocate the use of purine analogues (azathioprine, mercaptopurine) for maintenance therapy, following induction of remission with intravenous ciclosporin.[5,57,58] In the Actis study, 10% of ciclosporin responders, who were not given azathioprine, had avoided relapse at 3-year follow-up; in contrast, of those maintained on azathioprine, 26% remained relapse-free (P = 0.0635). Differences in route of administration, dosage and maintenance therapy likely contribute to the variability in results. Nevertheless, the literature demonstrates that the long-term relapse rate following initial induction of remission with ciclosporin is high and colectomy is often merely delayed by ciclosporin treatment.[56,59] Overall, ciclosporin response rates in randomised controlled trials, at 1-year follow-up and beyond were in the range of 42–50%, while long-term colectomy rates were between 36% and 58%.[10,52,54,55] Long-term ciclosporin data are described in further detail in Table 4.

There has been recent suggestion that more extended durations of therapy may enhance greater durability of remission and improve rates of colectomy-free survival.[60] While the duration of optimal therapy is yet to be established, in a retrospective study, Molnar et al. found that treatment for 1 year correlated with a 66% probability of avoiding colectomy at 4-year follow-up and beyond.[31,60] Further prospective studies are needed to confirm this association.

Safety Profile. Ciclosporin should be avoided in patients who have relapsed with optimal thiopurine maintenance therapy, those with hypocholesterolaemia and poor renal function, and those affected by active infection or recent immunosuppression secondary to recent infliximab infusion. Treatment with ciclosporin may be associated with potentially serious side effects including renal failure, neurotoxicity (seizures), anaphylaxis and infectious complications, thereby limiting it use as long-term therapy.[60,61] Randomised controlled trials have reported mortality rates between 0% and 5%.[10,50,52,54,55]

Dosing Schedule. In a study by Van Assche et al., no significant difference in short-term colectomy rates, median time to response and side effects was found between patients receiving 2 and 4 mg/kg/day.[62] Current guidelines suggest administration of the lower 2 mg/kg/day dose to offset the risk of long-term ciclosporin toxicity.


Short-term and Long-term Efficacy. Although early randomised controlled trials failed to establish the role of infliximab in the management of ASUC, more recent studies have demonstrated its efficacy and it has become the salvage therapy used most widely in clinical practice, principally due to its ease of administration.[10,55,63]

An initial study by Sands et al. suggested a superior clinical response to infliximab compared to placebo in steroid-refractory ASUC, but was underpowered (n = 11) – preventing a definitive conclusion.[63] In a subsequent randomised controlled trial, Jarnerot et al. established a lower 3-month colectomy rate in patients with moderate to severe attacks of betamethasone-refractory UC, given single infusion 5 mg/kg infliximab (7/24 patients – 29%) compared with those receiving placebo (14/21 patients – 67%). At 3-month follow-up, of the patients who avoided colectomy, clinical and endoscopic remission was confirmed in 6/15 infliximab-treated patients (40%) and 2/6 placebo patients (33%).[64] Three-year follow-up of these patients demonstrated sustained benefits, with a total of 12/24 infliximab-treated patients (50%) undergoing colectomy in comparison to 16/21 patients (76%) given placebo (P = 0.012).[6] Overall colectomy rates remained in favour of infliximab. However, selection of maintenance therapy following induction of remission was not standardised and the higher rate of azathioprine use in the infliximab group (75% vs. 57%) may have biased findings at follow-up.[65,66]

Multiple retrospective analyses support these prospective data.[38,67–70] In general, retrospective studies have suggested that infliximab treatment in ASUC induces higher rates of remission than placebo and reduces progression to colectomy.

The Scottish Society of Gastroenterology Infliximab Group found that 26/39 (67%) ASUC patients given infliximab avoided early colectomy during hospitalisation and remained colectomy-free at 3-month follow-up.[38] A Canadian cohort study similarly found that 16/21 patients (76%) with severe UC avoided early colectomy following infliximab, with 13/21 (62%) remaining colectomy-free at 3-month follow-up.[70] Comparable findings were described in a Danish multicentre analysis, with infliximab administration to glucocorticoid nonresponders with ASUC resulting in 61% (34/56) colectomy-free survival when followed up after a median of 1.5 years.[69]

In randomised controlled trials, clinical response rates following infliximab salvage therapy were similar within the acute admission period, extending to 3 months post-infliximab. Immediate response rates ranged from 50% to 83%, while short-term response rates ranged from 46% to 83%.[10,54,63,64,71] Longer term response was marginally lower and was in the range of 50–65%.[54,64] In the latest round of the UK IBD audit (2014), 15% of patients (79/512) treated with anti-TNF salvage therapy, for steroid-refractory disease, underwent colectomy prior to discharge from hospital.[53] Rates of colectomy in infliximab-treated ASUC were 0–50% and 35–50%, at short-term and long-term follow-up respectively.[10,54,63,64,71] Prospective and retrospective studies evaluating clinical outcomes in ASUC following infliximab therapy are outlined in Table 5.

Multiple Dosing Strategy. Although a single dose of infliximab does confer some benefit in ASUC, outcomes appear better in those who receive more than one dose. In a multicentre Italian trial, a total of 70/83 (84%) patients avoided colectomy within 2 months, with an overall colectomy rate of 30% (25/83) at 2-year follow-up.[68] In the latter study, a combination of retrospective (n = 46) and prospective analyses (n = 37) was undertaken. Results from this trial demonstrated that patients treated with two or more infliximab infusions had lower rates of early colectomy (3/57 – 5%) in comparison to those given a single infusion (9/26 – 35%) (OR = 9.53, 95% CI: 2.31–39.26, P = 0.001). Unlike in previous studies, maintenance therapy used did not influence relapse rates. This was mirrored in a study by Mortensen et al., who noted a lower colectomy rate in patients given three or more infusions (6/24 – 25% colectomy rate after a median of 9 months) when compared to patients given a single infusion only (13/26 – 50% colectomy rate after a median of 7 days).[69]

Intensified Dosing Strategy. The efficacy of an intensified dosing strategy utilising 10 mg/kg of infliximab has not been prospectively evaluated in ASUC, with the exception of a study by Sands et al., which was limited by insufficient statistical power.[63] Nonetheless, doses of 10 mg/kg have been investigated in chronic active colitis. Two double-blinded placebo-controlled studies, the Active Colitis Trials 1 and 2 (ACT1/2), evaluated the use of infliximab for induction and subsequent maintenance of remission in chronic active colitis.[72] Both trials involved 728 patients, who were randomised to receive 5 or 10 mg doses of infliximab or a placebo, administered at weeks 0, 2 and 6, then given every 8 weeks. After 8 weeks, ACT2 reported clinical response rates of 64.5% (78/121) and 69.2% (83/120) with 5 mg and 10 mg infliximab, respectively. There were significantly higher rates of clinical response, remission and mucosal healing at week 8 in those who received 10 mg/kg compared to placebo at both week 8 and 54, suggesting that a higher dose of infliximab is clinically efficacious.[73] Further evaluation of a more intensive dosing regimen in the ASUC setting is warranted.

Accelerated Dosing Strategy. Recent evidence has also indicated that an accelerated infliximab dosing regimen may reduce early colectomy rates in patients with highly active disease. A single-centre retrospective study analysed outcomes in 50 patients with ASUC, admitted for infliximab induction therapy.[74] Thirty-five patients were treated with standard induction therapy, with infliximab infusions at weeks 0, 2, 6. Fifteen patients alternatively received an accelerated dosing regimen, with three induction doses of infliximab administered within 3–4 weeks. Responders were given 8-weekly maintenance infliximab doses. The early colectomy rate during infliximab induction was shown to be significantly lower in the group that had received an accelerated induction regimen compared to those on the standard regimen – 7% (1/15) and 37% (13/35), P = 0.039. Furthermore, the standard infliximab regimen was associated with a reduced time to colectomy. However, 2-year follow-up revealed equivalent colectomy rates in both groups. Further prospective studies are needed to establish any potential role for accelerated induction regimens.

Anti-TNF Pharmacokinetics. The pharmacokinetics of anti-TNF therapy may be utilised to optimise infliximab dosing in patients with ASUC. Therapeutic drug monitoring is becoming increasingly relevant in the clinical setting; however, a significant proportion of patients progress to early colectomy within 1 month of initial salvage therapy – 29% in a study by Jarnerot et al.[64] For these patients, there is a narrow window of opportunity for dose optimisation, during which serum infliximab levels are still unavailable. Introduction of alternative treatment strategies based on known pharmacokinetic behaviours has been suggested. In a recent review, Rosen et al. proposed a three-pronged model, explaining accelerated clearance of anti-TNF drugs in ASUC due to: high TNF burden; anti-TNF neutralisation; and reduced tissue penetration.[75]

Elevated levels of both serum and mucosal TNF have been associated with greater disease severity.[76] Olsen et al. found that higher levels of pre-treatment TNF-alpha gene expression, within the colorectal mucosa, were predictive of clinical nonremission (OR = 2.5, P = 0.01) and endoscopic nonremission (OR = 4.8, P = 0.003) following standard week 0, 2, 6 infliximab induction therapy.[77] High mucosal TNF, in conjunction with low mucosal anti-TNF levels, has been shown to correlate with endoscopically active UC.[78] Higher serum and lower mucosal levels of anti-TNF also appear to predict for mucosal disease activity.[78]

Systemic inflammation associated with severe flares of UC is thought to be responsible for reduced drug efficacy. This is postulated to be due to mononuclear phagocyte upregulation and subsequent ligation and destruction of exogenous anti-TNF monoclonal antibodies.[79] Proteolytic degradation by cells of the reticuloendothelial system is thought to be a major contributor to primary nonresponse to infliximab therapy. This is corroborated by findings that higher CRP levels are associated with faster infliximab clearance.[80]

In addition, highly permeable colonic mucosa in ASUC can result in reduced tissue penetration of therapeutic infliximab. Infliximab clearance is suggested to be faster in IBD patients with higher body weight and hypoalbuminaemia.[81] Faecal drug loss may occur rapidly after infliximab administration, resulting in the delivery of effectively episodic doses. The consequence of this is the promotion of immune sensitisation and the formation of antibodies against infliximab, detectable within the blood. The detection of these antibodies may have implications for future management. In their retrospective study, Afif et al. demonstrated that in the presence of antibodies against infliximab, rates of clinical response following dose escalation were significantly lower than response rates after changing to another anti-TNF agent [1/6 (17%) and 11/12 (92%), P < 0.004].[82] Studies exploring the potential clinical effects of antibodies against infliximab have been performed in the broader IBD population and there are currently no studies that specifically investigate the impact of antibody formation in ASUC. Although, as in Crohn's disease, it is likely that combination therapy with a thiopurine may reduce immunogenicity compared with infliximab monotherapy this is yet to be proven in ASUC. Moreover, it remains to be demonstrated as to which dosing strategy is most likely to reduce the development of antibodies in the acute setting.

In patients who have been previously nonresponsive to infliximab therapy, the adoption of a high-dose induction regimen may enable greater treatment success. It has been postulated that intensive dosing (10 mg/kg) may overcome faecal infliximab losses and successfully establish a therapeutic serum infliximab level. Bridging to infliximab with nonprotein-based therapy may represent an alternative strategy in these patients. These management approaches are yet to be evaluated in the clinical setting. It is increasingly clear that patients with ASUC do not respond to infliximab equally and methods to assess pharmacokinetic suitability for anti-TNF therapy are needed to guide selection of medical salvage.

Overall, taking into account pharmacokinetic considerations, there is growing evidence that the standard week 0, 2, 6 infliximab induction regimen may be inadequate in the context of ASUC. Accelerated or intensified dosing regimens may offer better clinical and endoscopic outcomes. Moreover, integration of pharmacokinetic models and dose optimisation strategies into the current management paradigm will enable clinicians to offer individualised administration protocols to ASUC patients. Further research into this area is required to facilitate improvements in anti-TNF therapy dosing schedules.

Safety Profile. Infliximab is considered safe overall in the setting of ASUC; however, contraindications to infliximab therapy include untreated latent tuberculosis infection, congestive cardiac failure (New York Heart Association Class III or IV), demyelinating disease and active infection.[35] The significant immunosuppression caused by infliximab necessitates a number of screening tests before initiation of therapy, including tuberculosis and hepatitis screening. Infliximab carries risks of reactivation of latent tuberculosis, opportunistic upper respiratory tract and urinary tract infections, as well as cellulitis and sepsis.[63,64,83,84] Mortality following infliximab use is comparable to the mortality risk associated with ciclosporin and randomised controlled trials report rates to be between 0% and 2%.[10,54,55,63,64,71]

Comparing Infliximab and Ciclosporin

Current international guidelines recommend both ciclosporin and infliximab as rescue therapy.[31] In the last 5 years, head-to-head research comparing infliximab and ciclosporin has been undertaken and suggest that they are equally effective.

An Italian randomised trial of 21 patients with steroid-refractory ASUC demonstrated no statistically significant difference in 1-month outcomes between patients treated with infliximab (infusions at weeks 0, 2, 6) and those given oral ciclosporin (5 mg/kg, adjusted on plasma levels).[85] Remission rates (infliximab 57%; ciclosporin 43%) and colectomy rates (43% in both groups) were very similar. Infliximab was associated with infective complications in two patients [systemic cytomegalovirus (CMV) infection and one fatal case of Pneumocystis jiroveci pneumonitis], whereas patients given ciclosporin avoided any adverse effects.

In the seminal randomised controlled trial CySIF (Cyclosporin With Infliximab in Steroid-refractory Severe Attacks of Ulcerative Colitis), 116 patients were enrolled into either the ciclosporin group (2 mg/kg daily IV ciclosporin for 1 week followed by 4 mg/kg daily oral ciclosporin until day 98) or infliximab group (5 mg/kg infusions at weeks 0, 2, 6).[10] The two groups demonstrated no differences in immediate outcomes (day 7) or short-term outcomes (day 98). Rates of clinical response at day 7 were comparable – 86% (50/58) of patients treated with ciclosporin demonstrated improvement in Lichtiger score, compared to 84% (48/57) of the infliximab group (P = 0.76). At 3-month follow-up, treatment failure was observed in 35/58 patients (60%) receiving ciclosporin compared to 31/57 patients (54%) receiving infliximab. Similar rates of mucosal healing at 3-month endoscopic assessment were achieved in the two treatment streams – 47% (26/55) of ciclosporin-treated patients and 45% (25/55) of infliximab-treated patients (P = 0.85). Colectomy rates did not differ either – colectomy was performed in 17% (10/58) of the ciclosporin group and 21% (12/57) of the infliximab group. Safety profiles were also similar (adverse events – ciclosporin 16%; infliximab 25%).

Preliminary results from the recently completed CONSTRUCT study (Comparison Of INfliximab And Ciclosporin In STeroid-Resistant Ulcerative Colitis Trial) demonstrated no significant differences between infliximab and ciclosporin in steroid-refractory ASUC.[55] CONSTRUCT was a UK-based, multicentre randomised controlled trial with 3.5 year follow-up, involving 270 participants – evenly randomised into infliximab and ciclosporin treatment groups. Quality of life questionnaires were completed by participants on enrolment, at 3 and 6 months and then at 6-monthly intervals. No significant difference in quality-adjusted survival was found between infliximab and ciclosporin groups (infliximab: mean 614.6 days, standard deviation 229.8 days; ciclosporin: mean 626.0 days, standard deviation 226.8 days). Investigators also reported no significant differences in rates of colectomy (infliximab: 41%, vs ciclosporin: 48%), mortality or adverse effects.

A number of retrospective analyses have similarly found no differences in rates of clinical remission and short-term colectomy between patients treated with either agent.[86,87] Nevertheless, some studies have advocated for the use of infliximab over ciclosporin. In their open-label study, Croft et al. evaluated 83 cases of steroid-refractory ASUC.[54] Primary outcomes were defined as colectomy-free survival at different time points – on discharge from hospital, and 3 and 12 months following admission. Twenty-four of forty-three patients (56%) avoided colectomy at discharge after receiving ≥72 h ciclosporin 2–4 mg/kg, whereas 32/38 (84%) patients were discharged without colectomy after receiving a single infliximab 5 mg/kg infusion (P = 0.006). Infliximab remained more efficacious at maintaining colectomy-free survival at 3 months (28/37 – 76% and 23/43 – 53%, P = 0.04) and 12 months (24/37 – 65% and 18/43 – 42%, P = 0.04). Two serious adverse events were reported – both in the ciclosporin group. The authors concluded that infliximab produces a lower rate of colectomy and adverse effects in the short and medium term. However, it is important to note that patients who were on immunosuppressant therapy at the time of salvage therapy initiation had lower colectomy rates, with the authors acknowledging that further investigation into the effect of prior immunosuppression is required.[88] Results from a Dutch respective review, presented as an abstract at ECCO 2015, showed that rates of colectomy-free survival in a single IBD centre were significantly higher following infliximab use compared to ciclosporin, at 1 month [98% (136/139) and 74% (32/43), P < 0.0005], 6 months [84% (117/139) and 56% (24/43), P < 0.0005], 12 months [80% (108/135) and 49% (21/43), P < 0.0005] and 36 months [67% (67/100) and 45% (19/42), P = 0.015].[89] Although, these findings were among patients with moderate to severe UC, they suggest that infliximab may be associated with better long-term treatment response.

Existing literature supporting the efficacy of ciclosporin over infliximab has been largely derived from retrospective studies. A cohort study of 49 patients found that rates of colectomy-free survival at 2 weeks, 3 months and 12 months were higher following intravenous ciclosporin, compared to infliximab.[90] However, patients included in this study were only given a single infliximab infusion instead of three doses, as is now commonly used for standard induction therapy.

When choosing the optimal rescue therapy for a patient, clinicians must also consider factors beyond clinical efficacy and safety profile. Lowenberg et al. established that when comparing ciclosporin with infliximab, use of the anti-TNF agent is associated with a shorter length of hospital stay – median length of stay was 4.0 days (IQR 4.0–5.75) in the infliximab group and 11.0 (IQR 7.75–13.25) in the ciclosporin group (P < 0.01). However, total costs of infliximab administration and treatment at 3 months post-initiation of salvage therapy were significantly higher than costs associated with ciclosporin use.[91] In comparison to this, recent findings from the multicentre CONSTRUCT study demonstrated no significant difference in length of stay between the two treatments and while infliximab was found to be more expensive than ciclosporin, other associated costs were similar. Patients enrolled in the CONSTRUCT trial reported greater treatment satisfaction with infliximab, with the treating teams also preferring infliximab to ciclosporin. Finalised data may provide added insights into this ongoing debate.[55]

Differences in drug half-life must also be taken into consideration. Ciclosporin has a relatively short half-life of a few hours, whereas infliximab can persist in the blood for weeks. While the head-to-head effects of these agents on post-operative recovery are still being researched, if emergent colectomy is required following initiation of salvage therapy, any potential risks associated with immunosuppression, such as post-operative sepsis, is likely to be more pronounced with infliximab treatment.

The CySIF and Croft trials prospectively compared infliximab and ciclosporin but have yielded conflicting results. From the outset, Croft et al. aimed to illustrate salvage therapy use in a real-life clinical cohort typically encountered in clinical practice. In contrast, the CySIF study was a multicentre randomised controlled trial. When comparing these two studies – both have reasonably large sample sizes (CySIF: n = 116; Croft: n = 83); however, due to uncertainty about the relative efficacy of the two agents, power calculations were not performed in the Croft study. Patient inclusion criteria in CySIF were based on the Lichtiger CAI score (>10), while the Croft trial enrolled patients who satisfied the stricter Truelove and Witts' criteria for ASUC. Within the literature, European IBD centres adopt more inclusive criteria for hospitalisation for ASUC, when compared to centres in other regions. Relevant raw clinical data are required to properly compare baseline disease severity in these two studies, in order to interpret their findings with respect to each other and existing literature. Both trials were open-label and in addition, interventions were not randomised in the Croft cohort. Perhaps most importantly, treatment failure was defined according to composite criteria in the CySIF trial – based on Lichtiger CAI, Mayo score, colectomy and death. Colectomy at discharge, 3 and 12 months constituted treatment failure in the Croft study – providing a significantly narrower scope for failure.

In summary, while multiple earlier studies favoured either infliximab use in ASUC over ciclosporin or vice-versa, emerging prospective data have demonstrated no difference in efficacy between the two salvage therapy agents. More recent studies have featured larger sample sizes and utilised infliximab induction dosing at weeks 0, 2, 6, which yields comparable clinical outcomes to those achieved with intravenous ciclosporin. However, it remains to be proven as to whether a more intensive infliximab induction schedule will make any difference to outcome when compared to ciclosporin.