Dapagliflozin after cardiac surgery: reducing AKI or reducing the diagnosis of AKI?

Cardiac surgery-associated acute kidney injury (CSA-AKI) sets in motion a cascade of adversities, including prolonged hospitalisation and recovery, cardiovascular complications such as arrhythmias and myocardial infarction, and reduced long-term survival.^1,2 Having poorer baseline renal function further elevates the risks, and such patients are more likely to experience major adverse kidney events (MAKE) encompassing persistent renal dysfunction, progression to end-stage renal failure and in-hospital mortality.³ In 1% to 5% of cases, renal replacement therapy is required.⁴

Depending on the definition used and the type of surgery undertaken, the incidence of CSA-AKI can be as high as 42%.^1,2,5 Thus, it is far from a benign, transient post-operative insult that merely requires observation until renal function recovery.

Given the emerging evidence for the pleiotropic effects of sodium–glucose co-transporter 2 (SGLT2) inhibitors, their unequivocal cardiovascular and renal benefits and the current lack of a drug to prevent CSA-AKI, it is only natural to question whether these agents may have a role in mitigating its risk. Thus, Oosterom-Eijmael and colleagues⁶ conducted the multi-centre, randomised, double-blind, placebo controlled MERCURI-2 (Promoting Effective Renoprotection in Cardiac Surgery Patients by Inhibition of SGLT-2) superiority trial to answer this very question.

The study was conducted across seven sites in the Netherlands and recruited 784 patients undergoing elective cardiac surgery, randomised in a 1:1 ratio (n=392 in each arm) to receive either 10 mg of dapagliflozin or placebo. A total of 4 doses were administered, starting on the day before surgery and continuing until day 2 post-surgery. This contradicts current European Association for Cardiothoracic Surgery guidelines that recommend pausing SGLT2 inhibitors at least 72 hours prior to open-heart surgery to reduce risk of euglycaemic ketoacidosis (Class IIa recommendation) with a view to restarting it once the patient is haemodynamically stable (Class IIb).⁷

The trial excluded several clinically relevant groups, including patients already on SGLT2 inhibitors, individuals with insulin-dependent type 2 diabetes and a body mass index <25 kg/m², people with type 1 diabetics, those with a baseline estimated glomerular filtration rate (eGFR) of <20 mL/min/1.73 m², a history of diabetic ketoacidosis (DKA), systolic blood pressure <100 mmHg at enrolment, and women who were pregnant, breastfeeding or of child-bearing potential who were not using regular contraception. 

The primary outcome was onset of AKI within 7 days of cardiac surgery, defined according to the Kidney Disease: Improving Global Outcomes (KDIGO) criteria as an increase in serum creatinine of ≥0.3 mg/dL (26.5 μmol/L) within 48 hours, a ≥1.5-fold increase in creatinine within 7 days of surgery, or urine output of <0.5 mL/kg/h for 6 to 12 hours (oliguria) after surgery. A number of secondary outcomes, including the onset of stage 3 AKI, major adverse cardiac events (MACE) and MAKE at 30 days and safety outcomes, were also evaluated. 

Study participants were predominantly male (76%) and White (97%), with a median age of 68 (IQR, 61–74) years, a median BMI of 27 (IQR, 25–30) kg/m² and a median eGFR of 80 (67–89) mL/min/1.73 m². Notably, only 12% of participants had type 2 diabetes and 6% had heart failure. Regarding intra-operative characteristics, about 40% of patients underwent two or more procedures, while approximately 30% underwent isolated coronary artery bypass graft (CABG) surgery and non-CABG surgery each, across both study groups. Proportions of patients that underwent surgery on and off cardiopulmonary bypass (CPB) and durations of surgery, CPB and cross-clamping, which are independent predictors of CSA-AKI, were also similar across both arms.⁸ 

The key finding from the study was a nearly 50% reduction in both the incidence (28% vs 52%) and risk of AKI (relative risk [RR], 0.54 [95% confidence interval (CI), 0.45–0.65]) with dapagliflozin compared to placebo. Closer evaluation, however, revealed an important caveat: the primary endpoint was driven predominantly by the urine output component of the AKI definition. In fact, when AKI was defined exclusively by creatinine criteria, there was an underwhelming lack of difference between the dapagliflozin and placebo groups (RR, 0.93 [95% CI, 0.66–1.32]. This suggests that the headline result relied heavily on the definition of AKI used, raising a question about whether there was a true reduction in kidney injury. Additionally, a greater reduction in eGFR from baseline was seen in the dapagliflozin group compared to placebo, consistent with previous studies, with no differences in adverse events between the groups.^9–11 

Whilst urine output is an important marker of perfusion and haemodynamic status, post-operative oliguria, defined as a reduced urine output <0.5 mL/kg/h for ≥ 6 hours, is commonly observed and has been described as an epiphenomenon and adaptive response to anaesthetic and surgical trauma, often unrelated to kidney dysfunction measured by serum creatinine. In a recent study examining this, oliguria was noted in 34% of patients that underwent intermediate- and high-risk surgeries; however, a creatinine rise consistent with AKI was not observed in 88% of these patients.¹² 

Nonetheless, the substantial difference in urine output between the study group warrants attention. Although, as noted by the authors, SGLT2 inhibitors are not considered diuretics, several studies have demonstrated natriuresis and associated diuresis in the early stages of treatment. It remains uncertain whether this effect is what was captured by the study or if other mechanisms, such as differences in fluid input that have not been accounted for, contributed to the findings. Further, whether this observation translates into a meaningful outcome remains unclear. It is also debatable whether the definition of AKI in the post-operative setting requires revision.

Thus, while MERCURI-2 is a well-designed trial that addresses an important unmet need, its headline finding raises as many questions as it answers. The results are insufficient to justify routine perioperative use of dapagliflozin for the prevention of CSA-AKI, particularly given the limited generalisability of the findings. The study was based on an elective surgical cohort that was predominantly White and male, with relatively preserved renal function and little representation of individuals with diabetes or other comorbidities. 

A digest of the study can be read here.

References

1. Hobson CE, Yavas S, Segal MS et al (2009) Acute kidney injury is associated with increased long-term mortality after cardiothoracic surgery. Circulation 119: 2444–53
2. Oosterom-Eijmael MJP, Hermanns H, Lankadeva YR, Hulst AH (2026) Cardiac surgery-associated acute kidney injury. BJA Educ 26: 92–100
3. Palevsky PM, Molitoris BA, Okusa MD et al (2012) Design of clinical trials in acute kidney injury: report from an NIDDK workshop on trial methodology. Clin J Am Soc Nephrol 7: 844–50
4. Mehta RH, Grab JD, O’Brien SM et al (2006) Bedside tool for predicting the risk of postoperative dialysis in patients undergoing cardiac surgery. Circulation 114: 2208–16
5. Wang XD, Bao R, Lan Y et al (2024) The incidence, risk factors, and prognosis of acute kidney injury in patients after cardiac surgery. Front Cardiovasc Med 11: 1396889
6. Oosterom-Eijmael MJP, Hulst AH, de Oliveira NPM et al; MERCURI-2 Study Group (2026) Dapagliflozin and acute kidney injury following cardiac surgery: A randomized clinical trial. JAMA 30 Jul [Epub ahead of print]
7. Jeppsson A, Rocca B, Hansson EC, et al; EACTS Scientific Document Group (2024) 2024 EACTS Guidelines on perioperative medication in adult cardiac surgery. Eur J Cardiothorac Surg 67: 355
8. Nadim MK, Forni LG, Bihorac A et al (2018) Cardiac and vascular surgery-associated acute kidney injury: The 20th International Consensus Conference of the ADQI (Acute Disease Quality Initiative) Group. J Am Heart Assoc 7: e008834
9. Lambers Heerspink HJ, De Zeeuw D, Wie L et al (2013) Dapagliflozin a glucose-regulating drug with diuretic properties in subjects with type 2 diabetes. Diabetes Obes Metab15: 853–62
10. Shibata R, Taguchi K, Kaida Y, Fukami K (2022) Effect of dapagliflozin on the initial estimated glomerular filtration rate dip in chronic kidney disease patients without diabetes mellitus. Clin Exp Nephrol 27: 44
11. Heerspink HJL, Stefánsson BV, Correa-Rotter R et al; DAPA-CKD Trial Committees and Investigators (2020) Dapagliflozin in patients with chronic kidney disease. N Engl J Med 383: 1436–46
12. de Moura Pedro RA, Cunha GB, Pietrobom I et al (2025) Postoperative oliguria after intermediate and high-risk surgeries in critical care. A cohort analysis. J Crit Care 85: 154976