Here's the short version. Metformin is a cheap, old diabetes pill, and lab studies show it calms inflammation [1]. In randomized human trials, though, the drop in inflammation markers has been small or absent [2] [3]. In the DPP population with impaired glucose tolerance, the intensive lifestyle intervention reduced CRP more than metformin.
Which kind of inflammation this is about
The research here is about low-grade, body-wide inflammation, the kind tracked by blood markers like CRP. Higher CRP has been linked to obesity and insulin resistance [2].
That's different from a swollen knee or a sore muscle after a workout. None of the trials below tested pain, swelling or injury recovery, so metformin isn't a stand-in for a pain reliever.
What the lab studies show
In cells and animals, metformin suppressed the IKK/NF-κB inflammatory pathway and the release of inflammatory messengers such as interleukin-6 and interleukin-1β [1]. It did this in liver cells and in immune cells called macrophages.
The same researchers showed this effect could be separated from metformin's glucose-lowering actions [1]. That's the main reason scientists think it might help people without diabetes too.
What happened in people: CRP in randomized trials
Randomized trials show small or no changes in CRP. Two trials carry most of the weight.
The Diabetes Prevention Program enrolled 3,234 adults with impaired glucose tolerance [2]. At one year, metformin reduced CRP in women compared with placebo. In men the median changes were smaller: −7% on metformin vs +5% on placebo.
The same trial's lifestyle program did far better: CRP fell 33% in men and 29% in women [2]. The authors credit weight loss, more than extra activity, for most of that drop.
Keep in mind what the DPP was built for. Its main goal was preventing type 2 diabetes, which both metformin and the lifestyle program did [5]. CRP was a secondary look, measured once at one year.
CAMERA tested metformin in 173 adults without diabetes who had heart disease and were already on statins [3]. Over 18 months, hsCRP didn't differ significantly from placebo. Neither did artery-wall thickness, the main outcome.
Metformin did move other numbers in CAMERA: HbA1c, insulin, insulin resistance and a clotting-related protein all came in lower than on placebo [3]. Stomach side effects were more common too, 28 reports vs 5. So the drug was working on metabolism. It just didn't shift CRP.
| Study | Type | Population | Finding | Strength of evidence |
|---|---|---|---|---|
| Cameron 2016 [1] | Lab, cohort and small trial | Cells and mice; people with diabetes; heart failure | NF-κB and cytokines suppressed; log-transformed neutrophil-to-lymphocyte ratio 0.09 lower (about 9% lower geometric mean) than on a sulfonylurea | Mechanistic and observational |
| Haffner 2005, DPP [2] | Randomized trial, 1 year | 3,234 adults with impaired glucose tolerance | CRP: women −14% vs 0% on placebo; men −7% vs +5%; lifestyle −29% (women) and −33% (men) | Moderate |
| Preiss 2014, CAMERA [3] | Randomized trial, 18 months | 173 adults without diabetes, with heart disease, on statins | No significant hsCRP difference | Moderate (small) |
| Barzilai 2016 [4] | Trial proposal | Older adults (planned) | TAME aging trial proposed; proposal does not establish benefit | None yet |
Observational and smaller signals
In a large group of people with newly treated diabetes, those starting metformin had a slightly lower neutrophil-to-lymphocyte ratio after 8 to 16 months than those starting a sulfonylurea [1]. That ratio is a rough blood marker of inflammation.
A small placebo-controlled trial in people with heart failure and no diabetes also saw lower blood levels of several cytokines, including CCL11 [1]. These are hints, not proof. Observational comparisons can reflect who gets prescribed which drug, and the cytokine finding came from a sub-analysis.
Inflammation, aging and the TAME trial
The idea that metformin slows aging leans partly on its anti-inflammatory effects. The cited clinical evidence does not establish that metformin slows aging in people without diabetes.
TAME (Targeting Aging with Metformin) was proposed in 2016 as a first step toward testing aging itself as a drug target [4]. The proposal is not a trial result. The studies cited here do not establish an anti-aging treatment benefit.
The trade-off for active people
In two trials of healthy older adults, metformin blunted the gains from exercise training. That matters if you take it for longevity and also train for it.
- Aerobic training (53 adults, 12 weeks). Metformin weakened the rise in insulin sensitivity and VO2 max, and wiped out the rise in muscle mitochondrial respiration [6].
- Strength training (94 adults aged 65 and up, 14 weeks). People on placebo gained more lean mass and thigh muscle than people on metformin [7].
The resistance-training trial is telling. Its authors expected metformin's anti-inflammatory effect to help muscle grow, and it did the opposite [7]. If training is part of your plan, raise both trials with your prescriber.
Side effects and the boxed warning
The label's most common side effects are diarrhea, nausea or vomiting, gas, weakness, indigestion, stomach discomfort and headache [8]. In the label's immediate-release tablet trial, diarrhea affected 53% of people on metformin vs 12% on placebo; that figure does not describe every formulation.
Its boxed warning is lactic acidosis, a rare but serious buildup of acid in the blood [8]. The label also warns about vitamin B12 deficiency, and about low blood sugar when metformin is combined with insulin or drugs that push insulin release. Your prescriber reviews your kidney function and other medicines before starting.
If inflammation or metabolic health is your goal
Weight loss and exercise lowered CRP more than metformin did in the DPP [2]. That's the place to start. Ask your clinician whether measuring hsCRP would help your particular care plan.
If you want a medicine in the mix, the metabolic shortlist covers metformin and the GLP-1 options side by side. For those drugs and inflammation, see semaglutide and inflammation and tirzepatide and inflammation. If recovery between workouts is the real goal, try the sleep and recovery shortlist. Not sure where you fit? Start the quiz.