GLP-1s: muscle, rebound, eyes and pancreas. What the data say

Do you lose muscle?

It is the most repeated criticism on social media. The body-composition data say something more nuanced. In the SURMOUNT-1 DXA substudy (160 adults), tirzepatide was associated at 72 weeks with −21.3% body weight, −33.9% fat mass and −10.9% lean mass: about 75% of the weight lost was fat and 25% lean mass, the same split as placebo¹.

A network meta-analysis of 22 trials (2,258 people) found an average lean-mass loss of −0.86 kg, also about a quarter of total weight lost, with no change in lean mass relative to body weight². Two caveats: in that analysis, tirzepatide (15 mg) and semaglutide (2.4 mg) were the most effective for weight and fat reduction but among the least effective at preserving lean mass²; and DXA lean mass is not the same as muscle function.

What happens when you stop?

In STEP 1, semaglutide was associated with −14.9% body weight at 68 weeks vs −2.4% with placebo⁴. In the trial extension (327 people), one year after stopping, participants regained about two-thirds of the weight they had lost, and cardiometabolic improvements drifted back toward baseline³.

SURMOUNT-4 tested this with randomization: after 36 weeks on tirzepatide (−20.9% weight), 670 people either stayed on the drug or switched to placebo. From week 36 to 88 the placebo group regained 14.0% and the continued group lost a further 5.5%⁵. For many researchers, this signals that obesity behaves as a chronic condition³⁵.

What about vision? The NAION case

Non-arteritic anterior ischemic optic neuropathy (NAION) is a kind of "stroke" of the optic nerve. The alarm came in 2024 from a single-center study that found more cases among semaglutide users; the authors stressed it did not establish causality⁶. In people with type 2 diabetes, a Danish-Norwegian population study found a hazard ratio of 2.81 vs SGLT-2 inhibitors, but about 1.4 extra cases per 10,000 person-years: "the absolute risk remains low"⁷. A meta-analysis of 78 trials found no increase in eye disorders overall or in diabetic retinopathy; for NAION it found an association (OR 3.92; 95% CI 1.02-15.02), but the authors judged the evidence insufficient for definitive conclusions⁸.

In June 2025 the EMA's pharmacovigilance committee concluded its review: NAION is a very rare side effect of semaglutide that may affect up to 1 in 10,000 people, and the EMA recommended updating the product information⁹. In the US, by contrast, the Wegovy label (June 2026) does not mention NAION¹⁴.

Pancreas, stomach and gut

A US insurance-claims study of people without diabetes using GLP-1s for weight loss associated these drugs with more pancreatitis, bowel obstruction and gastroparesis than bupropion-naltrexone, with very few events and wide confidence intervals¹⁰. By contrast, a meta-analysis of 21 randomized trials (99,599 patients) found no difference in pancreatitis; it did find more gastrointestinal (+63%) and gallbladder (+26%) disorders, and lower all-cause mortality¹¹.

And suicidal ideation?

The EMA reviewed the question and in April 2024 concluded that "the available evidence does not support a causal association" between GLP-1s and suicidal thoughts or behaviors¹³. A study of more than 240,000 people with overweight or obesity even associated semaglutide with lower risk of suicidal ideation compared with other obesity drugs (HR 0.27), although the evidence is observational¹².