Impact of thiopurines and anti-tumour necrosis factor therapy on hospitalisation and long-term surgical outcomes in ulcerative colitis

Table 1 Summary of key research investigating impact of thiopurines and tumour necrosis factor inhibitors therapy on long-term surgical outcomes in ulcerative colitis

	Ref.	Study design	Population	n	Key findings
Thiopurines	Ardizzone et al[11]	RCT comparing AZA vs 5-ASA	Steroid dependent UC	72	No difference in colectomy rates at 6 mo between AZA and 5-ASA groups
	Kaplan et al[13]	Population based time trends analysis of colectomy rates	Unselected UC	N/A	Reduction in elective colectomy rates of 7.4% per year
					Doubling of TP use over the study period
					Emergency colectomy rates remain static
	Targownik et al[14]	Population based analysis of colectomy rates	Unselected UC	3752	10.4% colectomy rate at 10 yr post diagnosis
					> 16 wk TP therapy associated with reduced colectomy requirement
	Chhaya et al[15]	Population based time trends analysis of colectomy rates	Unselected UC	8673	TP use > 12 mo associated with a 71% reduction in risk of colectomy
					Early TP use not associated with added benefit
					No significant change in colectomy rates over study period
	Cañas-Ventura et al[16]	Retrospective descriptive cohort study of UC patients receiving AZA	Unselected UC	1334	5 yr colectomy rate at 8.8%
					TP use within 33 mo of diagnosis associated with increased risk of colectomy
aTNF	Sjöberg et al[24]	Multi-centre retrospective analysis of IFX rescue therapy	Acute severe UC	211	64%, 59% and 53% colectomy-free survival at years 1, 3, 5
					Majority of colectomies within first 2 wk of IFX therapy
	Gustavsson et al[26]	RCT comparing IFX rescue therapy vs placebo	Acute severe UC	45	3 yr colectomy free survival 50%
	Laharie et al[29]	Head to head RCT comparing IFX vs CSA as rescue therapy	Acute severe UC	115	No significant differences in colectomy rates between two therapies at 3 mo
	Sandborn et al[19]	ACT 1 and 2 RCT of IFX vs placebo	Moderate to severe UC	728	Colectomy rate significantly lower in IFX group (10% vs 17%) at 54 wk
	Feagan et al[41]	ULTRA 1 and 2 RCT of ADA vs placebo	Moderate to severe UC	963	Very low colectomy rates reported at 52 wk (approximately 4%)
					No difference in colectomy rates between ADA and placebo
	Reich et al[45]	Time trends analysis of colectomy rates following introduction of IFX	Unselected UC	481	19% annual decrease in elective colectomy in biologic era
					15% annual decrease in emergency colectomy in biologic era
	Costa et al[50]	Meta-analysis of aTNF use in UC	Moderate to severe UC	836	Reduced risk of surgery at 1 yr in patient treated with IFX compared to placebo (OR = 0.55)
					NNT was 11

UC: Ulcerative colitis; aTNF: Tumour necrosis factor inhibitors; RCT: Randomised controlled trial; AZA: Azathioprine; TP: Thiopurine; 5-ASA: 5-aminosalicylic acid; IFX: Infliximab; CSA: Ciclosporin; ADA: Adalimumab; NNT: Number needed to treat; N/A: Not applicable; ACT: Active ulcerative colitis trials; ULTRA: Ulcerative colitis long-term remission and maintenance with adalimumab.

Table 2 Summary of key research investigating impact of thiopurines and tumour necrosis factor inhibitors therapy on hospitalisation in ulcerative colitis

	Ref.	Study design	Population	n	Key findings
Thiopurines	Actis et al[61]	Retrospective study comparing hospitalisation before and after AZA induction	Severe UC	17	Significant decrease in hospitalisation for patients with UC up to 5.8 yr following AZA induction
					Most of patients were also treated with ciclosporin at AZA induction
	Herrinton et al[62]	Population based cohort study of prescribing trends in UC	Unselected UC	5895	150% increase in immuno-modulator use in UC between 1998-2005
					Concurrent reduction in UC hospitalisations in the same period by a third
	Vester-Andersen et al[63]	Prospective descriptive study of IBD inception cohort	Unselected UC	300	26% exposure to immuno-modulator during follow up
					Hospitalisation rates decreased from 4.7 d/person-years in year 1 after diagnosis to 0.4 d in year 5
					Immuno-modulator therapy found not to be significant in predicting need for hospitalisation
aTNF	Carter et al[65]	Medical insurance cost analysis study	Unselected UC	420	UC patients with a prescription for infliximab for > 80% of the study period had less hospitalisation requirement, lower admission costs and shorter inpatient stays
	Oussalah et al[37]	Multicentre retrospective study on outcomes in UC patients post aTNF	Unselected UC	191	Estimated hospitalisation-free survival at 1, 2, 3 and 6 yr were 66.7%, 60.2%, 57.1% and 44.6% respectively
					Earlier use of aTNF predictive of need for hospitalisation
	Sandborn et al[19]	ACT 1 and 2 RCT comparing IFX with placebo	Moderate to severe UC	728	Of patients treated with IFX, 84% remained free of hospitalisation at 54 wk, compared to 75% in the placebo group
	Feagan et al[41]	ULTRA 1 and 2 RCT comparing ADA with placebo	Moderate to severe UC	963	Significantly reduced all-cause and UC-related admissions at both 8 wk and 52 wk in patients treated with ADA compared to placebo
	Lopez et al[47]	Meta-analysis of aTNF in UC outcomes	Moderate to severe UC	964	aTNF therapy was superior to placebo in reducing UC-related hospitalisations, with a relative risk of 0.71 (95%CI: 0.56-0.90)

UC: Ulcerative colitis; aTNF: Tumour necrosis factor inhibitors; RCT: Randomised controlled tria; AZA: Azathioprine; IFX: Infliximab; ADA: Adalimumab; IBD: Inflammatory bowel disease; ACT: Active ulcerative colitis trials; ULTRA: Ulcerative colitis long-term remission and maintenance with adalimumab.