SGLT2 inhibitors linked to lower death and kidney injury after cardiac surgery
In a study of over 15,000 cardiac surgery patients, those taking SGLT2 inhibitors before surgery had lower death rates at 30 days, 90 days, and one year, plus lower acute kidney injury, compared to non-users.
Why it matters: These findings could influence perioperative management decisions for millions undergoing cardiac procedures, though current evidence is observational, not from clinical trials.
Keep in mind: This was an observational study matching patient groups by statistical methods; it cannot prove SGLT2 inhibitors caused the benefits, only that use was associated with them.
AI-assisted explanation based on the supplied abstract. Read the linked source for full context.
What this changes
The finding has not established an approved change to our evidence assessment.
Source date: Sep 2, 2026 · Discovered: Sep 15, 2026 · Assessed: — · Research update
What it does not prove
- Limitations have not yet been verified. Read the primary source.
Extracted from supplied source material; model confidence is not a measured probability of correctness.
Study details
- Design
- Cohort
- Sample size
- 15,052
- Duration
- 12 months
- Population
- Adults (≥18 years) undergoing cardiac procedures, identified from the TriNetX U.S. Collaborative Network. Propensity score-matched cohorts with mean age 65.8 ± 10.6 years, 70.5% male, 29.5% female.
- Primary outcome
- All-cause mortality at 30 days, 90 days, and 1 year.
- Effect size
- All-cause mortality at 30 days: RR 0.566 [95% CI 0.443-0.722]; at 90 days: RR 0.643 [95% CI not fully specified]; at 1 year: RR 0.657 [95% CI not fully specified]. Acute kidney injury at 30 days: RR 0.870 [95% CI not fully specified]. Hospital admission rates: RR 1.111 [95% CI 1.089-1.133].
- Significance
- Significant
- Journal
- Journal of clinical medicine
- Published
- Sep 2, 2026
Funding: Not reported.
Conflicts: Not reported.
Sources
Primary source
- Journal of clinical medicine ↗Peer reviewed
Assessment history
9/23/2026 · SUPERSEDED: Source updated; assessment withdrawn pending review.
Discussion (0)
One thread per scientific event, no matter how many outlets covered it.
No comments yet
Be the first to weigh in on this evidence.