Y-ECCO Literature Review: Joshua McGuire

Joshua McGuire

Infliximab induction regimens in steroid‐refractory acute severe colitis: A multicentre retrospective cohort study with propensity score analysis

Sebastian S, Myers S, Argyriou K, Martin G, Los L, Fiske J, Ranjan R, Cooper B, Goodoory V, Ching HL, Jayasooriya N, Brooks J, Dhar A, Shenoy AH, Limdi JK, Butterworth J, Allen PB, Samuel S, Moran GW, Shenderey R, Parkes G, Lobo A, Kennedy NA, Subramanian S, Raine T

Aliment Pharmacol Ther. 2019;50:675–683. doi: 10.1111/apt.15456..


Joshua McGuire picture
Joshua McGuire
© Joshua McGuire

Acute Severe Ulcerative Colitis (ASUC) is a medical emergency which necessitates a colectomy in up to 30% of cases on index presentation [1]. The first-line treatment is with intravenous corticosteroids but up to 40% of patients will fail to respond [2]. Ciclosporin and infliximab are then well-recognised options for rescue therapy to avert the need for a colectomy and, whilst there appear to be no difference in response rates between these two choices [3], many experts favour infliximab owing to convenience and familiarity [4]. Up to 55% of patients do not respond to the standard dosing regimen of infliximab extrapolated from the outpatient setting [5]. The exaggerated clearance of infliximab in ASUC is increasingly better characterised [6]; this has led to the concept of accelerated dosing regimens although the efficacy of such regimens has yet to be evaluated by randomised controlled trials. A recent meta-analysis [7] of the available cohort studies showed no benefit of accelerated induction in reducing colectomy rates in steroid‐refractory disease; however, provider bias represents a significant barrier to answering this question. Propensity score matching seeks to address this provider bias.


This multi-centre, retrospective cohort study included consecutive hospitalised patients with ASUC, defined by modified Truelove and Witts criteria, who received at least three doses of IV steroids in 11 centres over 2 years. Patients with a diagnosis of IBD unclassified, Crohn's colitis, infective colitis, co‐existent cytomegalovirus, admission for elective surgery and prior therapy with anti‐TNF were excluded. Those who received infliximab were stratified into two groups: those who received 5 mg/kg at week 0 and then no further doses until week 2 and those who received either 10 mg/kg at week 0 or at least two doses of 5 mg/kg with a second dose on or before day 7. The primary outcome measure was colectomy rates at 30 days. Secondary outcomes included index admission colectomy rates, colectomy rates at 90 days, 6 months and 12 months, the length of hospital admission and adverse events, including postoperative complications and mortality.

With regards to propensity score-adjusted matching, patients were matched in a 1:1 fashion based on bivariate propensity scores generated using the variables of C‐reactive protein (CRP), serum albumin, CRP/albumin ratio, haemoglobin and the presence of pancolitis. Outcomes were analysed in both the entire and the propensity-score matched cohorts.

Key Findings

The study included data on 131 patients: 102 on a standard regimen and 29 on an ‘accelerated’ regimen. The overall colectomy rate in the unmatched cohort was 29%, mirroring previous studies. There was no difference in colectomy rates between the two regimens (28.4% vs 31%, p=0.99). However, there was a clear separation in the biochemistry of the two groups: on both day 1 and day 3, compared with the standard regimen group those on the accelerated regimen had a higher CRP (day 1: median IQR 56 mg/L ± 78 standard vs 101 mg/L ± 36 accelerated, p=0.001), a lower mean albumin (day 1: 33 g/L standard vs 30 g/L accelerated, p=0.006) and more often had a CRP/albumin ratio >2. These differences speak to the need for propensity score matching; the authors were able to match 52 patients with no significant difference in baseline characteristics and demographics. The majority of those in the matched group had pancolitis (23/26 standard, 21/26 accelerated), endoscopically graded at Mayo 3 (24/26 standard, 26/26 accelerated). In this matched group with severe disease, there was no difference in overall colectomy rates (57% vs 31%, p=0.09) but both the index admission colectomy rates (53% vs 23%, p=0.045) and the 30-day colectomy rates (57% vs 27%, p=0.048) were lower. Of note, the model used does not account for variability in management, including dose optimisation, during the maintenance period. This variability could account for the fact that the short-term reduction in colectomy rates does not translate into the overall rates. Of the 26 in the accelerated group, only four received the front-loaded 10 mg/kg dose, which precluded any separate analysis. The authors acknowledge that their study is limited by not having data on therapeutic drug monitoring to support the rationale for accelerated dosing. Also, significantly, there was one death from postoperative sepsis in the accelerated group. There was no difference in complication rates between the two groups (18.6% vs 20.7%, p=0.8).


This multicentre cohort study of patients with ASUC has demonstrated, in a real-world setting, that in a subgroup analysis controlling for bias of treatment choice, accelerated dosing of infliximab reduces both in-hospital and 30-day colectomy rates, without increasing complication rates. It emphasises the utility of propensity score matching in evaluating treatments when head-to-head trials are challenging. It raises the question of what the optimal accelerated dosing strategy would be and advocates prospective studies. Moreover, it points to another unanswered question: Is a ‘hot’ or a delayed colectomy preferable?


1. Ananthakrishnan AN, McGinley EL, Binion DG, Saeian K. A nationwide analysis of changes in severity and outcomes of inflammatory bowel disease hospitalizations. J Gastrointest Surg. 2011;15:267–76.

2. Turner D, Walsh CM, Steinhart AH, Griffiths AM. Response to corticosteroids in severe ulcerative colitis: a systematic review of the literature and a meta-regression. Clin Gastroenterol Hepatol. 2007;5:103–10.

3. Williams JG, Alam MF, Alrubaiy L, Arnott I, Clement C, Cohen D, et al. Infliximab versus ciclosporin for steroid-resistant acute severe ulcerative colitis (CONSTRUCT): a mixed methods, open-label, pragmatic randomised trial. Lancet Gastroenterol Hepatol. 2016;1:15–24.

4. Clement C, Rapport F, Seagrove A, Alrubaiy L, Williams J. Healthcare professionals' views of the use and administration of two salvage therapy drugs for acute ulcerative colitis: a nested qualitative study within the CONSTRUCT trial. BMJ Open. 2017;7:e014512.

5. Aratari A, Papi C, Clemente V, Moretti A, Luchetti R, Koch M, et al. Colectomy rate in acute severe ulcerative colitis in the infliximab era. Dig Liver Dis. 2008;40:821–6.

6. Rosen MJ, Minar P, Vinks AA. Review article: applying pharmacokinetics to optimise dosing of anti-TNF biologics in acute severe ulcerative colitis. Aliment Pharmacol Ther. 2015;41:1094–103.

7. Sebastian S, Myers S, Nadir S, Subramanian S. Systematic Review: Efficacy and Safety of Accelerated Induction Regimes in Infliximab Rescue Therapy for Hospitalized Patients with Acute Severe Colitis. Dig Dis Sci. 2019;64:1119–28.


Joshua is a core medical trainee with aspirations towards gastroenterology specialty training in IBD. He gained IBD experience at the John Radcliffe Hospital in Oxford and is currently part of the IBD team at UCLH in London. He is part of the Gastro London Investigative Network for Trainees (GLINT) and is currently working on a London-wide project to optimise endoscopic reporting in UC.

Posted in ECCO News, Y-ECCO Literature Reviews, Committee News, Y-ECCO, Volume 14, Issue 4