Advanced IBD therapies and cardiovascular events: trials and registries disagree

Design

systematic review and meta-analysis of randomised and observational studies, 2002-2024

Population

43 studies (36 randomised trials, 7 observational) of advanced therapies in inflammatory bowel disease

Primary outcome

major adverse cardiovascular events versus placebo

Effect

pooled odds ratio 0.60 (95% CI 0.24-1.51) in randomised trials; no class estimate significant

Inflammatory bowel disease carries raised cardiovascular risk, and in other immune-mediated conditions advanced therapies appear cardioprotective. Whether that holds in IBD has been unclear. This meta-analysis pooled 43 studies from 2002 to 2024 - 36 randomised trials with nine long-term follow-up reports, and seven observational studies - estimating odds of major adverse cardiovascular events against placebo, with sensitivity analyses for sparse and zero-event data.

Across placebo-controlled trials the pooled estimate was 0.60 (95% CI 0.24 to 1.51): a trend towards fewer events, nowhere near significance. By class, IL-12/IL-23 inhibitors came out at 0.35 (95% CI 0.05 to 2.21), JAK inhibitors at 0.57 (0.16 to 2.06), and anti-TNF agents at 3.04 (0.31 to 29.47). Every one of those intervals spans a range from large benefit to large harm, because cardiovascular events are rare in IBD trial populations and many studies contributed no events at all.

The observational data then point the other way for the class that mattered most: anti-TNF therapy associated with lower risk (0.29, 95% CI 0.21 to 0.40), while IL-12/IL-23 inhibitors looked worse (4.41, 0.49 to 39.28) and JAK inhibitors neutral. A reversal that complete between designs usually means confounding by indication rather than a real effect - patients started on anti-TNF agents in practice differ systematically from those who are not. What can be said is that no clear cardiovascular signal has emerged in either direction, which is reassurance of a limited and honest kind.

• Do not use this to prefer or avoid any drug class on cardiovascular grounds; no estimate reached significance.
• Treat the observational anti-TNF benefit as likely confounding by indication, not a protective effect.
• Continue to manage conventional cardiovascular risk factors actively - that evidence is not in doubt.
• For JAK inhibitors, the existing class label warnings derived from rheumatoid arthritis populations are unchanged by this.
• Cardiovascular event counts in IBD trials are too low to answer this; the evidence has to come from well-designed long-term cohorts.

Why it matters

It removes cardiovascular risk as a usable basis for choosing between advanced therapies in IBD, which is how it has sometimes been argued.

Don't overread it

Non-significant class differences are not evidence of equivalence; the trials were too small in cardiovascular events to detect a moderate difference.

The statistics, in plain English

An odds ratio of 3.04 with a confidence interval from 0.31 to 29.47 carries essentially no information: the data are compatible with anti-TNF agents tripling events or cutting them by two-thirds. Intervals this wide come from very few events, and the authors' sensitivity analyses for zero-event studies show they knew it. When a randomised estimate and an observational one point in opposite directions, the randomised one is usually the less biased but here it is also the less precise - which is why the answer is 'unresolved' rather than either result.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.