Who comes off diabetes drugs after metabolic surgery, and who relapses

Design

Long-term follow-up of pooled randomised trials with predictor modelling

Population

262 adults with type 2 diabetes and obesity (mean BMI 36.4, HbA1c 8.5%)

Primary outcome

Diabetes remission (HbA1c below 6.5% off medication) and relapse

Effect

Remission 58% after surgery vs 3% with medical-lifestyle care over median 11 years

ARMMS-T2D brought together US randomised trials comparing metabolic-bariatric surgery with intensive medical and lifestyle treatment in 262 people with type 2 diabetes and a mean BMI of about 36. This analysis followed them for a median of 11 years, up to 14, to find which baseline features predicted remission and relapse.

Remission — HbA1c below 6.5% with no glucose-lowering drugs — was reached by 58% of those randomised to surgery and 3% of those on medical-lifestyle care. In multivariable analysis, shorter diabetes duration, not using insulin and greater weight loss at one year predicted remission. Among those who reached remission after surgery, longer diabetes duration and use of more drug classes at baseline predicted later relapse.

The practical message is about timing. The patients who gain most are those referred early, before insulin and before several drug classes have been added. A patient who has had diabetes for many years and is on insulin may still benefit from surgery for weight and glucose control, but should not be promised a drug-free outcome. The analysis is exploratory and from US centres; it predicts likelihood, not a guaranteed result for one person.

• Discuss metabolic surgery early in eligible patients, before insulin is needed, when remission is most likely.
• Record diabetes duration and number of drug classes when referring; both shape the likely outcome.
• Set expectations: long duration and insulin use lower the chance of coming off all drugs.
• Watch one-year weight loss after surgery; smaller losses predicted lower remission.
• Keep monitoring after remission, especially in those with long-standing disease, where relapse was likelier.

Why it matters

Surgery is often offered as a last resort, which is the point at which its chance of remission is lowest.

Don't overread it

The predictor analysis is exploratory; it shows who is more likely to remit, not that delaying surgery causes failure.

The statistics, in plain English

58% versus 3% is the randomised comparison and is robust. The predictors come from modelling within a few hundred patients, so they describe groups and carry uncertainty for any single person.

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.