Too sweet to be true? Prophylactic SGLT2 inhibitors before cardiac surgery

Welcome to AudioScholar. Today we're summarizing "Too sweet to be true? Prophylactic SGLT2 inhibitors before cardiac surgery", from NephJC.

Acute kidney injury after cardiac surgery is common — occurring in as many as half of cases — and between two and five percent of those patients need renal replacement therapy. It drives length of stay, morbidity, mortality and cost, and yet prevention remains stubbornly supportive: fluid balance, hemodynamics, avoiding nephrotoxins. Diuretics, calcium channel blockers, bicarbonate and statins have all failed. So there is real appeal in the idea that SGLT2 inhibitors — sodium-glucose cotransporter 2 inhibitors, the flozins — might protect the kidney through the bypass ischemia-reperfusion window by improving renal oxygenation and reducing tubular workload. Countering that, most current guidance says hold these drugs three to four days before surgery because of euglycemic ketoacidosis and volume depletion.

Into that gap comes MERCURI-2, reported by Oosterom-Eijmael and colleagues in JAMA. This was a multicentre, double-blind, placebo-controlled randomised trial in the Netherlands enrolling adults scheduled for elective cardiac surgery, both coronary bypass and non-bypass procedures. Patients already on an SGLT2 inhibitor or insulin, those with prior ketoacidosis, systolic pressure under 100, or estimated glomerular filtration rate below 20 were excluded. Seven hundred and eighty-four patients were randomised to dapagliflozin or placebo, with the first dose the day before surgery and daily dosing through the second postoperative day — four doses in total. Median age was 68, median estimated filtration rate was 80, and adherence was high in both arms. The primary outcome was KDIGO-defined acute kidney injury within seven days, with daily creatinine and a urinary catheter in place throughout for urine output measurement.

The headline number is large. Acute kidney injury occurred in 28 percent of the dapagliflozin group versus 52 percent of the placebo group — a risk difference of about 24 percentage points, relative risk 0.54. But the NephJC commentary argues persuasively that this result does not survive scrutiny. When acute kidney injury was defined by creatinine alone, the incidence was 14 percent versus 15 percent — a relative risk of 0.93, entirely null. When defined by urine output alone, it was 21 percent versus 48 percent, relative risk 0.44. In other words, the whole treatment effect lives in the urine output criterion, and most events occurred within the first 24 to 48 hours. Dapagliflozin is a glucosuric osmotic diuretic; a drug that increases urine output will mechanistically make patients less likely to trip a low-urine-output threshold. Post-cardiac-surgery patients are frequently hyperglycaemic from surgical stress, meaning more filtered glucose and more osmotic diuresis precisely during the adjudication window. Crucially, perioperative glucose and haemodynamic measurements were prespecified but explicitly not analysed for this report.

Secondary outcomes reinforce the concern. Stage 3 acute kidney injury was numerically higher with dapagliflozin, three cases versus one, not statistically significant. The maximum fall in estimated filtration rate was 7.4 millilitres per minute with dapagliflozin versus 1.75 with placebo — larger than the classic initiation dip seen in CREDENCE or EMPA-REG. There were no differences in thirty-day major adverse cardiac or kidney events, new-onset ECG-confirmed atrial fibrillation at roughly a third of patients in each arm, length of stay, days at home, or patient-reported quality of life. On safety, reassuringly, there was one episode of ketoacidosis and one mycotic infection, both in the dapagliflozin arm.

For practice, the takeaway is that this trial does not establish that perioperative dapagliflozin prevents clinically meaningful cardiac surgery-associated kidney injury. It may, however, offer modest reassurance that short perioperative exposure is not dangerous. Signals from a smaller open-label empagliflozin trial and from meta-analyses of chronic dosing suggest a more modest creatinine-based benefit may exist — but MERCURI-2 did not reproduce it. The population was also highly homogeneous, predominantly male and nearly all white.

This is a single-trial report with substantial commentary layered on top, so review the primary paper and its accompanying editorial before changing your perioperative approach.

That was a summary of Too sweet to be true? Prophylactic SGLT2 inhibitors before cardiac surgery, from NephJC. For the full piece, visit the original source. Until next time.

This is an automated summary generated by artificial intelligence, which can make mistakes. Always review the original source materials.