Skip to content
This GLP
← Research
Evidence

What Else Comes Back When You Stop a GLP-1? Almost Everything

Across eighteen trials, stopping reversed HbA1c, waist, blood pressure and fasting glucose as well as weight — and the rebound was roughly twice as large after semaglutide as after liraglutide.

Dana Sullivan6 min read
What returned after stopping, by drugweight (kg)systolic (mmHg)semaglutide8.217.09liraglutide4.291.56Eighteen trials, 3,771 people, and they disagreed enormously.

More than the weight. Pooled across eighteen randomized trials and 3,771 participants, stopping reversed HbA1c, waist circumference, BMI, systolic blood pressure and fasting glucose together. In people with obesity HbA1c rose 0.25 percentage points.

The rebound differed by drug. It ran 8.21 kg after semaglutide against 4.29 kg after liraglutide, and systolic pressure rose 7.09 mmHg against 1.56. It also grew with time: 7.31 kg beyond 26 weeks against 2.51 kg in shorter follow-ups.

⛔ Heterogeneity reached 99.57%. The eighteen trials disagreed with each other almost entirely, so the pooled figures describe a direction rather than an amount. The randomized withdrawal studies behind the weight half of this are STEP 1 [2] and STEP 4 [3].

That weight returns after stopping is well established, and we cover it in stopping and weight regain. What is less discussed is that the scale is not the only thing that reverses, and that the size of the reversal depends on which drug was stopped.

What was pooled

A 2025 systematic review searched five databases without language restriction and included eighteen randomized trials with at least twelve weeks of follow-up after treatment ended, covering 3,771 participants [1]. Every included study was rated low risk of bias.

What came back

Among people with obesity, body weight rose 5.63 kg after stopping (95% CI 3.52 to 7.73) and HbA1c by 0.25% (95% CI 0.18 to 0.32). Waist circumference, BMI, systolic blood pressure and fasting plasma glucose all deteriorated significantly.

In the type 2 diabetes studies the pattern differed. Weight rose less, at 2.03 kg (95% CI 1.63 to 2.42), while HbA1c rose more, at 0.65% (95% CI 0.22 to 1.08). Fasting glucose remained stable there. The two populations lose different things and get different things back, which is the same lesson as the dose-response gradient in tirzepatide against semaglutide.

The differences worth noticing

Two subgroup findings stand out. The rebound was larger after semaglutide than after liraglutide — 8.21 kg against 4.29 — with waist circumference up 3.80 cm against 2.69 and systolic blood pressure up 7.09 mmHg against 1.56. And it grew with time: studies following people beyond 26 weeks reported 7.31 kg against 2.51 kg in shorter ones.

The blood pressure figure is the one least likely to be mentioned by anyone selling the drug. Seven millimeters of mercury is a clinically meaningful move. It is the kind of thing a person who stopped six months ago would have no reason to connect to a decision they made then.

What it means before buying

It means the purchase is longer than it looks. The authors conclude that guidelines need to address discontinuation and maintenance, not only initiation and titration. That is a polite way of saying that the exit is unplanned for. On the seller side that gap is total. Nothing in this market describes what happens after, and most of it does not describe what happens during, which is the silence counted in what sellers will not tell you.

It also means a year of purchases is the realistic unit of decision, not a month. and the prices that decide whether a year is affordable are the ones this site measures in when the headline is not the bill.

Frequently asked

Does anything besides weight come back?
Yes. In people with obesity, HbA1c rose 0.25% and waist circumference, BMI, systolic blood pressure and fasting glucose all deteriorated significantly after stopping.
Does the drug make a difference?
In the subgroup analysis it did: 8.21 kg regained after semaglutide against 4.29 after liraglutide, with systolic pressure up 7.09 mmHg against 1.56.
How reliable are these numbers?
As a direction, well supported. As a prediction for an individual, not at all — heterogeneity reached 99.57% on the weight outcome, meaning the pooled trials disagreed almost entirely.
Does the rebound stop?
It grew with follow-up length rather than settling: 7.31 kg beyond 26 weeks against 2.51 kg in shorter studies.

Sources

  1. [1] Tzang CC, et al. (2025). Metabolic rebound after GLP-1 receptor agonist discontinuation: a systematic review and meta-analysis EClinicalMedicine. PMID 41399474
  2. [2] Wilding JPH, Batterham RL, Davies M, Van Gaal LF, Kandler K, Konakli K (2022). Weight regain and cardiometabolic effects after withdrawal of semaglutide: The STEP 1 trial extension Diabetes, Obesity and Metabolism. PMID 35441470
  3. [3] Rubino D, Abrahamsson N, Davies M, et al. (2021). Effect of Continued Weekly Subcutaneous Semaglutide vs Placebo on Weight Loss Maintenance in Adults With Overweight or Obesity: The STEP 4 Randomized Clinical Trial JAMA. PMID 33755728

Where to get it

Best GLP-1 injections

Every injectable seller we can verify, with the price each one publishes and an honest read of what the trials measured.

Compare providers →

More in Evidence