A widely prescribed class of blood-pressure drugs has been linked to a higher risk of serious kidney problems in people with type 2 diabetes — but the researchers are clear it cannot prove the drugs are to blame.
The analysis examined health data from 31,031 adults with type 2 diabetes between 2016 and 2021. Those taking dihydropyridine calcium-channel blockers (DCCBs) had a 33% greater risk of a major adverse kidney event — hazard ratio 1.33 — compared with patients on other blood-pressure drugs. Findings were presented at the European Renal Association Congress.
What was measured
“Major adverse kidney events” here meant a 40%+ drop in eGFR — a large loss of kidney filtering function — or progression to end-stage kidney disease requiring dialysis or transplant.
These are hard, unambiguous outcomes rather than laboratory fluctuations. For people with diabetes, who already face elevated kidney risk, they represent the trajectory the entire treatment approach is designed to prevent.
All participants were also on RAS inhibitors and SGLT2 inhibitors, and were followed for a median of 3.5 years.
Why the background therapy matters
That detail strengthens the study considerably and is easy to skip past.
RAS inhibitors and SGLT2 inhibitors are the two drug classes with demonstrated kidney-protective effects in diabetes, and both are standard care. Requiring all participants to be on them means the comparison is between patients receiving equivalent kidney protection who differ in their additional blood-pressure drug.
Without that constraint, an apparent difference could simply reflect some patients receiving better baseline kidney care than others.
What DCCBs do
Dihydropyridine calcium-channel blockers — amlodipine being the most widely used — lower blood pressure by relaxing arterial smooth muscle. They are effective, inexpensive, well tolerated and prescribed to enormous numbers of people.
There is a plausible mechanistic reason they might be less kidney-protective in this setting. The kidney’s filtering units are supplied by an incoming and an outgoing arteriole, and pressure across the filter depends on the balance between them. DCCBs preferentially dilate the incoming vessel, which can raise pressure inside the filter even while systemic blood pressure falls.
RAS inhibitors act on the outgoing vessel, reducing that internal pressure — which is part of why they protect kidneys beyond their blood-pressure effect. The concern about DCCBs is not new; this study is a large observational test of it.
The confounding problem, stated by the authors
The study is observational, so it can reveal a link but not causation. The authors state plainly that it “cannot show that DCCBs directly caused the poorer kidney outcomes” and call for prospective studies and randomised trials.
Confounding by indication is the central difficulty. Patients receive a DCCB in addition to RAS and SGLT2 inhibitors because their blood pressure is not controlled on those alone — meaning they have more resistant hypertension, which is itself a marker of more advanced vascular and kidney disease.
So the comparison may be between patients with harder-to-control blood pressure and patients with easier-to-control blood pressure, with the drug marking that difference rather than causing the outcome.
Why this is hard to adjust away
Statistical adjustment can account for measured differences — baseline eGFR, protein in urine, blood pressure readings, diabetes duration. It cannot account for the clinical judgement that led a physician to add a particular drug to a particular patient, which encodes information no dataset captures.
That is why this class of finding recurs across observational pharmacoepidemiology and why randomised evidence remains the only resolution.
What would settle it
A trial randomising patients with type 2 diabetes already on RAS and SGLT2 inhibitors to a DCCB versus an alternative third agent, with kidney outcomes measured over years.
Such a trial is feasible and unlikely to be funded, since all the drugs involved are generic and no commercial sponsor has an incentive. Publicly funded outcome trials comparing established generics are exactly the studies that most need doing and least often get done.
How to read this
The signal is worth studying and is not a reason to stop any medication. Uncontrolled blood pressure damages kidneys reliably and substantially, and a patient stopping an effective antihypertensive on the strength of an observational signal would be trading a possible risk for a certain one.
Why conference presentations warrant extra caution
One procedural detail deserves flagging: this was presented at a congress rather than published in a peer-reviewed journal, and that difference matters more than coverage usually conveys.
Conference abstracts are selected on interest and relevance, reviewed briefly, and presented before the full analysis has been scrutinised by independent reviewers. Methods are compressed into a few hundred words, sensitivity analyses are frequently absent, and the numbers can change between presentation and publication.
A meaningful proportion of findings presented at major medical conferences are never published in full at all, and among those that are, effect sizes commonly shrink.
None of that makes the finding wrong. It means the appropriate posture is to note the signal, wait for the full paper, and be particularly wary of acting on a conference report describing an association that contradicts current prescribing practice for widely used generic drugs.
The reasonable response for clinicians is awareness when choosing a third agent, not a change in existing prescriptions. This summarises research presented at a conference and is not medical advice; do not change medications without consulting your doctor.