GLP-1 Drugs and the Gut Microbiome: What 38 Studies Could Not Agree On

GLP‑1 drugs change the bacteria in your gut. That much is agreed. What is not agreed is how — a systematic review of 38 studies found the direction of the change differs by drug, differs between animals and people, and in the case of semaglutide points two ways at once: a bacterium associated with better metabolic function increases while overall microbial diversity falls.^[1] There is no clinical action to take from any of this yet, and the reason to publish it is that plenty of people will try to sell you one.

What the systematic review found

The review followed PRISMA and gathered every study it could find on how GLP‑1 analogs affect the composition, richness and diversity of gut microbiota, in animal and human models. Thirty-eight studies met the bar. The conclusions split by drug rather than converging.^[1]

The semaglutide row is the one to sit with. Akkermansia muciniphila is the genus most often cited as a marker of metabolic health, and it went up. Diversity is the other measure most often cited as a marker of gut health, and it went down. A story that uses only the first half is available to anyone selling something.

The randomized human data is very small

The MIGHTY study is a three-month parallel clinical trial in African American youth with type 2 diabetes, randomized to metformin alone (14 people) or metformin plus liraglutide (11 people). Twenty-five participants in total, eleven of them on the GLP‑1. Any sentence drawn from it carries those numbers.^[2]

• Metformin alone was associated with more Eubacterium and Eubacterium rectale, and less Bacteroides ovatus.
• Metformin plus liraglutide was associated with more Bacteroides fragilis and less Streptococcus thermophilus.
• Secondary bile acid metabolites rose in the metformin group, and one of them correlated with lower fasting glucose.
• ⛔ Shifts in microbiota taxa were not associated with plasma short-chain fatty acids, HbA1c or glucose.

That last line is the one that matters, because it is the opposite of the popular claim. The bacteria moved. The glucose control that the bacteria are supposed to explain did not move with them. The authors’ own suggestion is that bile acid pathways, rather than the taxa themselves, may be doing the work.^[2]

What you can actually do with this

Very little, and that is the honest answer. No microbiome test sold to consumers can tell you whether your response to a GLP‑1 is good or bad, because nobody has established what a good response looks like. No probiotic has been shown to improve outcomes on these drugs. The gastrointestinal side effects people actually experience — nausea, constipation, reflux — have explanations in gastric emptying that do not require the microbiome at all.

If your interest here is symptoms rather than science, our pages on constipation and burping and reflux deal with what is treatable. If it is nutrition, eating too little on a GLP-1 covers what genuinely runs low.