Predictors of diuresis response to levosimendan administration in patients with acute heart failure

Funding Acknowledgements Type of funding sources: None. Levosimendan, a calcium sensitizer and potassium channel-opener, is appreciated for its effects on systemic and pulmonary hemodynamic and for the relief of symptoms in acute heart failure (AHF). Positive effects of levosimendan on renal function have been also described. The aim of the present analysis was to assess the predictors of the diuresis response to levosimendan administration in high risk acute heart failure patients. Methods. We analysed 34 consecutive patients admitted with high risk AHF to one centre and treated in intensive cardiac care unit. Levosimendan was administered on top of other treatment as a 24-hour infusion of 12.5 mg total dose except for 7 patients (1 patient - terminated earlier due to intolerance, 5 patients – 48h infusion, 1 patient - 72h infusion). Decision of levosimendan administration was based on clinical status and left to attending physician. Diuresis and diuretic dosage before (24 hours) and after levosimendan infusion (48 hours) were taken into account for the present study. Results. The AHF was primary of cardiac origin in all patients. In 6 (18%) it was due to recent acute myocardial infarction. In-hospital mortality was 24%. Median length of hospitalization was 26 days (range 6 to 107 days). Mean age of the patients was 66 ± 12 years, 25 (74%) were men. Mean INTERMACS score was 3.4 ± 1.4 with wet-cold clinical profile present in 13 (38%) of patients. Mean left ventricle ejection fraction (LVEF) was 27 ± 13%, mean NTproBNP was 17176 ± 12464 pg/ml, and mean eGFR 48 ± 22 ml/min/1.73m2. At the time of levosimendan administration patients had background treatment with catecholamines (mean number per patient 1.4 ± 1.1, range 0-3) and with diuretics (mean dosage of furosemide 167 ± 102 mg/24h, range 20-500). 48-hours diuresis after levosimendan administration varies from 950 to 11300 ml (mean 4307 ± 2418 ml). It was significantly lower in patients with cold-wet profile (2646 ± 1335 vs. 5335 ± 2381 ml in other clinical profiles, p = 0.0002). Additionally, 48-hour diuresis was negatively correlated with age (r=-0.46, p = 0.0062) and the number of background catecholamines (r=-0.47, p = 0.0047), and not significantly with the furosemide dosage (r=-0.28, p = 0.10) – figure. No association with diuresis was found for LVEF, NTproBNP, and eGFR. In multiple regression analysis (model R2 = 0.63, p = 0.0085) both older age (p = 0.026) and cold-wet profile (p = 0.0074) were significant predictors of poor diuresis after levosimendan administration. Conclusion. Older age and cold-wet profile were significant predictors of poor diuresis response to levosimendan administration in high risk acute heart failure patients. Although concomitant catecholamines and high diuretic dosage use cloud also be markers of non-responders to levosimendan in terms of diuresis. Abstract Figure