Bariatric surgery versus GLP-1 receptor agonists on hard cardiovascular endpoints

Design

PRISMA systematic review and network meta-analysis with node-splitting consistency checks, databases searched January 2000 to September 2024

Population

43 studies, 932,380 patients with obesity receiving bariatric surgery, GLP-1 receptor agonists or control

Primary outcome

major adverse cardiovascular events, heart failure and myocardial infarction

Effect

surgery vs control: MACE HR 0.66 (95% CI 0.60-0.74), HF 0.45 (0.38-0.53), MI 0.53 (0.45-0.63); GLP-1RA vs control: MACE 0.85 (0.77-0.94), HF and MI non-significant; I-squared 66-93%

This network meta-analysis pooled 43 studies and 932,380 patients with obesity to place bariatric surgery and GLP-1 receptor agonists in the same comparison, using direct and indirect evidence. Against control, surgery was associated with reductions in major adverse cardiovascular events (HR 0.66, 95% CI 0.60 to 0.74), heart failure (HR 0.45, 95% CI 0.38 to 0.53) and myocardial infarction (HR 0.53, 95% CI 0.45 to 0.63). GLP-1 receptor agonists reduced MACE (HR 0.85, 95% CI 0.77 to 0.94) but their effects on heart failure and myocardial infarction did not reach significance.

Before reading that as a ranking, note what generated it. Almost none of the surgical evidence can be randomised, so patients who had bariatric surgery were selected for fitness for an operation and for the motivation to pursue one, and no statistical adjustment removes that. Heterogeneity ran from 66% to 93%, which is very high, and the search closed in September 2024 — meaning the newer agents and the recent cardiovascular outcome trials that most changed prescribing are underrepresented. The authors say as much.

What survives all that is a reasonable clinical shape rather than a league table: for a patient with obesity at high cardiovascular risk who is a surgical candidate, surgery has the broader and more consistent evidence across endpoints; for a patient who is not, a GLP-1 receptor agonist has real evidence for MACE and weaker evidence for the others. In Indian practice the constraint is usually access and cost rather than a genuine choice between the two, and where bariatric surgery is realistically available, this is an argument for referring earlier rather than treating it as a last resort after drugs fail.

• Discuss bariatric surgery earlier in high cardiovascular risk obesity rather than only after pharmacotherapy fails
• Do not tell a patient GLP-1 receptor agonists do not help the heart — they reduced MACE; the heart failure and infarction results were inconclusive, not negative
• Weigh surgical candidacy honestly; the comparison is confounded precisely by who can have an operation
• Note the evidence cut-off of September 2024 — newer agents and their outcome trials are not fully represented
• Where surgery is not accessible or affordable, a GLP-1 receptor agonist remains an evidence-based choice, not a consolation

The statistics, in plain English

A hazard ratio of 0.66 means events occurred at about two-thirds the rate over the follow-up period, but these are pooled observational hazard ratios, so they measure the difference between people who had surgery and people who did not, which includes every reason one group had surgery. Heterogeneity of 66% to 93% means the contributing studies disagreed substantially, so the pooled figures are averages across quite different populations and the intervals are narrower than the real uncertainty. The non-significant GLP-1 results for heart failure and myocardial infarction mean those analyses could not distinguish an effect from no effect — a different statement from showing there is none.

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.