A SINDONE1, S D ANKER2, P PONIKOWSKI3, C WANNER4, E PFARR5,S HAUSKE6, B PEIL5, A SALSALI7, I RITTER8, M NORDABY6, A KOITKA-WEBER6, M BRUECKMANN6, W JAMAL6, J LINDENFELD9, W T ABRAHAM10
1Concord Clinical School, The University of Sydney School of Medicine, Sydney, Australia, 2Department of Cardiology, Charité University, Berlin, Germany, 3Department of Heart Diseases, Wrocław Medical University, , Poland, 4Department of Medicine, Würzburg University Clinic, Würzburg, Germany, 5Biostatistics + Data Sciences, Boehringer Ingelheim Pharma GmbH & Co KG, Ingelheim, Germany, 6TA CardioMetabolism Respiratory Med, Boehringer Ingelheim International GmbH, Ingelheim, Germany, 7Clinical Development, Boehringer Ingelheim Pharmaceuticals, Inc., Ridgefield, USA, 8Pharmacovigilance, Boehringer Ingelheim International GmbH, Ingelheim, Germany, 9Department of Medicine, Vanderbilt University Medical Center, Nashville, USA, 10Division of Cardiovascular Medicine, The Ohio State University, Columbus, USA
Aim: To assess estimated glomerular filtration rate (eGFR) after starting/stopping empagliflozin in heart failure (HF) patients with and without type 2 diabetes (T2D) from the EMPERIAL (Effect of EMPagliflozin on ExeRcise ability and heart failure symptoms In patients with chronic heArt faiLure)-Reduced and -Preserved trials (NCT03448419/NCT03448406).
Background: Empagliflozin induces an initial decline in eGFR in T2D patients. Although considered haemodynamic and reversible, it may raise concerns in clinical practice and needs to be better understood in HF patients, beyond T2D.
Methods: Patients with HF, baseline eGFR ≥20 mL/min/1.73 m², elevated NT-proBNP, and left ventricular ejection fraction ≤40% in EMPERIAL-Reduced and >40% in -Preserved, received placebo or empagliflozin 10 mg. eGFR changes were analysed for empagliflozin versus placebo (pooled data from both trials) using a mixed-model-repeated-measures analysis.
Results: Of 626 patients, 55.6% had T2D and mean baseline eGFR±SD was 57.6±19.3 mL/min/1.73 m². Placebo-adjusted mean eGFR±SE change from baseline with empagliflozin was –2.3±0.7 (95% confidence interval [CI] – 3.6, –1.0) and –3.3±0.8 (95% CI –4.9, –1.8) mL/min/1.73 m² at weeks 6 and 12, respectively. Change in eGFR±SE from last value on treatment to follow-up was –0.3±0.5 with placebo and +3.9±0.5 mL/min/1.73 m² with empagliflozin. These findings were consistent across subgroups with/without T2D. The incidence of acute renal failure (ARF) was similar with placebo and empagliflozin (11.8 and 10.7 per 100 patient-years, respectively), and also consistent across subgroups by baseline T2D status. Serious ARF occurred in 4 placebo and 1 empagliflozin patient.
Conclusions: In HF patients with baseline eGFR ≥20 mL/min/1.73 m², eGFR dynamics after starting/stopping empagliflozin, and kidney adverse events, were similar to empagliflozin’s known eGFR change on/off treatment, and kidney safety profile, irrespective of T2D status.
Professor Andrew Sindone B. Med (hons), MD, FRACP, FCSANZ, FNHFA is Director of the Heart Failure Unit and Department of Cardiac Rehabilitation at Concord Hospital and Head of Department of Cardiology at Ryde Hospital. He has received multiple awards, has been Principal Investigator in over 40 multicentre clinical trials and has presented over 85 research papers. He is a Co-chairman of the NSW Cardiovascular Expert Reference Group, Fellow of the Heart Foundation of Australia was on the Scientific Committee for the World Congress of Cardiology in 2002 and is co-author of the Australian Guidelines for the Management of Heart Failure.