Levosimendan in acute heart failure with severely reduced kidney function, a propensity score matched registry study

BACKGROUND: Patients with heart failure frequently present with kidney dysfunction. Kidney function is relevant, as prognosis declines with reduced kidney function and potentially beneficial drugs like levosimendan are contraindicated for missing safety data. MATERIALS AND METHODS: A single-center retrospective registry study was conducted including all patients receiving levosimendan on a medical intensive care unit between January 2010 and December 2019. Exclusion criteria were a follow-up less than 24 h or missing glomerular filtration rate (eGFR) before administration of levosimendan. The first course of treatment was evaluated. Patients were stratified by eGFR before drug administration and the primary endpoint was a composite of supraventricular-, ventricular tachycardia and death within 7 days after administration of levosimendan. An internal control group was created by propensity score matching. RESULTS: A total of 794 patients receiving levosimendan were screened and 368 unique patients were included. Patients were predominantly male (73.6%) and median age was 63 years. Patients were divided by eGFR into three groups: >60 ml/min/1.73 m(2) (n = 110), 60–30 ml/min/1.73 m(2) (n = 130), and <30 ml/min/1.73 m(2) (n = 128). ICU survival was significantly lower in patients with lower eGFR (69.1, 57.7, and 50.8%, respectively, p = 0.016) and patients with lower eGFR were significantly older and had significantly more comorbidities. The primary combined endpoint was reached in 61.8, 63.1, and 69.5% of subjects, respectively (p = 0.396). A multivariate logistic regression model suggested only age (p < 0.020), extracorporeal membrane oxygenation (p < 0.001) or renal replacement therapy (p = 0.028) during day 1–7 independently predict the primary endpoint while kidney function did not (p = 0.835). A propensity score matching of patients with eGFR < 30 and >30 ml/min/1.73 m(2) based on these predictors of outcome confirmed the primary endpoint (p = 0.886). CONCLUSION: The combined endpoint of supraventricular-, ventricular tachycardia and death within 7 days was reached at a similar rate in patients independently of kidney function. Prospective randomized trials are warranted to clarify if levosimendan can be used safely in severely reduced kidney function.