Two pills lowered blood pressure by almost exactly the same amount. Over the next year and a half, the patients on one of them were about two-thirds less likely to die. That is the striking finding, and also the reason to handle it carefully, from a retrospective study published August 29 in eClinicalMedicine ↗.

The setting is resistant hypertension. That is blood pressure that stays high even after a patient is taking three different drugs, including a diuretic, at proper doses. Treatment guidelines say the fourth drug to add should be a mineralocorticoid receptor antagonist, or MRA. The two common ones are spironolactone and eplerenone, cheap generics that block the hormone aldosterone and help the kidneys shed salt and water.

The researchers, led by Zhejia Tian and Bernhard M W Schmidt at Hannover Medical School in Germany, asked a different question. In people who have resistant hypertension and are also overweight or obese, could a GLP-1 drug work as an alternative fourth agent instead? GLP-1 drugs like semaglutide ↗ and tirzepatide ↗ are known for weight loss, and they lower blood pressure a little, but nobody had compared them head to head against the guideline-recommended MRA in this group.

What the database showed

The team used TriNetX, a network of electronic health records from 67 US health systems. Out of 213,309 eligible adults, 22,694 started a GLP-1 drug as their fourth blood-pressure agent and 5,673 started an MRA. The two groups differ in almost every way that matters, so the researchers used propensity score matching to build two balanced arms of 4,153 patients each, then followed them for a median of 1.4 years.

The GLP-1 arm did better on nearly everything. Major adverse cardiovascular events, meaning heart attacks, strokes, and cardiovascular death, were about 37 percent lower (hazard ratio 0.63). Death from any cause was about two-thirds lower (hazard ratio 0.34). Kidney outcomes moved the same direction: major adverse kidney events down about 36 percent, acute kidney injury down about 38 percent.

Here is the part that should stop you. Blood pressure fell by essentially the same amount in both groups. At twelve weeks the GLP-1 arm was down 5.7 mmHg and the MRA arm down 6.3 mmHg, a difference small enough to be noise, and if anything the MRA lowered pressure slightly more. So whatever produced that survival gap, it was not the blood pressure.

Two ways to read it

One reading is that GLP-1 drugs genuinely protect the heart and kidneys through routes that have nothing to do with blood pressure. That is not a stretch. Semaglutide's large cardiovascular outcome trial (SELECT) and its kidney outcome trial (FLOW) both found benefit that held up after accounting for blood pressure and weight. A real edge over an MRA in this population is directionally plausible.

The other reading is the one the study cannot rule out, and it is about the comparator. Spironolactone and eplerenone are not just hypertension drugs. They are workhorse treatments for heart failure and advanced chronic kidney disease. So the patients who get started on an MRA are, on average, sicker in exactly the ways that lead to death and kidney failure, and no amount of matching on baseline codes fully erases that channeling. Add the healthy-user pattern on the other side, where the expensive weekly injectable tends to reach more engaged, better-resourced patients, and you have a setup that manufactures a survival gap on its own.

A two-thirds drop in deaths over 18 months from swapping one add-on pill for another, at identical blood pressure, is larger than any randomized GLP-1 trial has delivered. That magnitude is the tell. The authors, whose funding was listed as none, say plainly that prospective studies are needed before anyone rewrites the treatment ladder.

This is a familiar shape. When a records database shows a drug beating its comparator on every single outcome at once, the clean explanation is often not the drug but who was selected to receive it, the same caution that applied when GLP-1 users turned out to have less painful diabetic nerve damage in a single snapshot ↗.

Both semaglutide and tirzepatide act on the GLP-1 receptor ↗, with tirzepatide adding a second action on the GIP receptor. MRAs are small molecules, not peptides, and target a different receptor entirely. The result here is a reason to run a randomized trial of GLP-1 drugs against an MRA as fourth-line therapy in overweight patients with resistant hypertension. It is not evidence that the trial has already passed.