Outcomes Following Acute Severe Colitis at Initial Presentation: A Multi-centre, Prospective, Paediatric Cohort Study

Abstract

Aim: To assess contemporary outcomes in children with acute severe ulcerative colitis [ASUC] at initial presentation.

Methods: Between April 2014 and January 2019, children aged <17 years, with new onset ASUC (Paediatric Ulcerative Colitis Activity Index [PUCAI ≥65) were prospectively followed in a Canadian inception cohort study. 16S rRNA amplicon sequencing captured microbial composition of baseline faecal samples. Primary endpoint was corticosteroid-free clinical remission with intact colon at 1 year [PUCAI <10, no steroids ≥4 weeks].

Results: Of 379 children with new onset UC/IBD-unclassified, 105 [28%] presented with ASUC (42% male; median [interquartile range; [IQR]) age 14 [11-16] years; extensive colitis in all). Compared with mild UC, gut microbiome of ASUC patients had lower α-diversity, decreased beneficial anaerobes, and increased aerobes; 54 [51%] children were steroid-refractory and given infliximab [87% intensified regimen]. Corticosteroid-free remission at 1 year was achieved by 62 [61%] ASUC cohort (by 34 [63%] steroid-refractory patients, all on biologics; by 28 [55%] steroid responders,13 [25%] on 5- aminosalicylic acid [5-ASA], 5 [10%] on thiopurines, 10 [20%] on biologics). By 1 year, 78 [74%] escalated to infliximab including 24 [47%] steroid-responders failed by 5-ASA and/or thiopurines. In multivariable analysis, clinical predictors for commencing infliximab included hypoalbuminaemia, greater PUCAI, higher age, and male sex. Over 18 months, repeat corticosteroid course[s] and repeat hospitalisation were less likely among steroid-refractory versus -responsive but -dependent patients (adjusted odds ratio [aOR] 0.71 [95% CI 0.57-0.89] and 0.54 [95% CI 0.45-0.66], respectively).

Conclusion: The majority of children presenting with ASUC escalate therapy to biologics. Predictors of need for advanced therapy may guide selection of optimal maintenance therapy.

Keywords: Ulcerative colitis; paediatric acute severe colitis.

Graphical Abstract

Figure 1

Patient disposition from first presentation until 1 year.

Figure 2

Comparison of the baseline intestinal microbiome of paediatric UC patients with mild or acute severe UC at first presentation using 16S rRNA amplicon sequencing. [A] UPGMA tree constructed based on the Aitchison distances. The bars below the tree indicate UC severity at first presentation. [B] Relative abundance of taxa at the family level. [C] RDA analysis of Aitchison distances. Permutational multivariate analysis of variance [PERMANOVA: Adonis2, p < 0.05] of Aitchison distances suggests the overall gut microbiome composition varies between patients with mild and acute severe UC. [D] Boxplots of Shannon diversity index show the mean within-sample diversity differs between patients with mild and acute severe UC [ANOVA, *p <0.05]. UC, ulcerative colitis.

Figure 3

Specific OTUs are associated with ASUC at first presentation. Differentially abundant OTUs [99% clustering] present more than 10 times in at least 20% of samples were determined using ANCOM-BC. Negative log fold change values indicate OTUs enriched in ASUC patients and positive log fold change values indicate OTUs enriched in mild UC. OTU, operational taxonomic units; ASUC, acute, severe ulcerative colitis.

Figure 4

Time to infliximab initiation among children and adolescents with ASUC at first presentation. [A] All patients. [B] Patients stratified by lowest serum albumin measured during first 5 days of hospitalisation. ASUC, acute, severe ulcerative colitis.

Figure 5

Nomogram to predict 2-week and 3-month probability of remaining biologic-naïve.