GLP-1 drugs cut major heart events by about 17 percent, and that cut was the same whether a patient's kidneys were working normally or already failing. A meta-analysis published this week in Diabetes, Obesity and Metabolism ↗ pooled eight cardiovascular outcome trials and found the relative benefit did not budge across kidney function. The twist is what happens when you turn that relative number into actual prevented events.
The drugs are the GLP-1 receptor agonists, the injectable and now oral medicines built to copy a gut hormone that controls blood sugar and appetite. Beyond weight and glucose, the big trials have shown they lower the rate of heart attack, stroke, and cardiovascular death, a bundle clinicians call MACE for major adverse cardiovascular events. This analysis asked a narrower question: does that heart protection weaken in people whose kidneys are damaged, the patients who are sickest and often treated most cautiously?
What the analysis found
The authors searched the literature through late April 2026 and pulled nine publications from eight randomized placebo-controlled trials, including the recent SOUL trial of oral semaglutide ↗. Together the trials covered 70,822 people with type 2 diabetes or excess weight, and 70,534 of them had heart-event data split by kidney function, measured as estimated glomerular filtration rate, or eGFR, the standard gauge of how well the kidneys filter blood.
They split patients at an eGFR of 60, the line below which kidney function is considered reduced. Above 60, GLP-1 drugs lowered MACE risk with a hazard ratio of 0.83 (95 percent confidence interval 0.77 to 0.90), a 17 percent relative reduction. Below 60, the hazard ratio was also 0.83 (interval 0.74 to 0.93). The formal test for whether kidney function changed the effect came back flat: a ratio of hazard ratios of 1.02 (interval 0.85 to 1.21), with a p value of 0.84. In plain terms, the drug worked just as well, proportionally, in damaged kidneys as in healthy ones.
Why the absolute number flips the story
A constant relative benefit does not mean a constant payoff. People with reduced kidney function start with more heart events to prevent, so the same percentage cut removes a bigger slice in raw terms. The analysis estimated an absolute risk reduction of 2.6 percent in the reduced-kidney group versus 1.6 percent in the better-kidney group. Translated into the number of patients you have to treat to prevent one heart event, that is 39 people with reduced kidney function against 62 with preserved function.
That difference matters clinically because it points the other way from instinct. Patients with chronic kidney disease are the ones doctors often hesitate to start on a new drug, worried about a fragile patient. Here the math says they are exactly the group that gets the most heart events prevented per person treated.
What it does and does not settle
This is a meta-analysis of trial-level summary data, not patient-level records, and the absolute-risk numbers are described by the authors as exploratory, derived from the placebo-group event rates and the pooled hazard ratios. It does not speak to kidney failure as an outcome, only to heart events sorted by baseline kidney function. The drugs studied span the class, from daily liraglutide ↗ to weekly dulaglutide ↗ to oral semaglutide, all hitting the same receptor, GLP-1R ↗.
The headline most readers will take from a paper like this is "the benefit holds in kidney disease." The more useful reading is that holding steady on the relative scale, in a sicker population, quietly means a larger absolute return. The relative number is reassuring. The absolute number is the argument for treating.