Multicenter, Double-blind, Placebo-controlled Randomized Trial to Evaluate the Efficacy and Safety of Intravenous Administration of Intermittent Doses of Levosimendan in Ambulatory Patients With Advanced CHF: the LION-HEART Study

The LION-HEART study (Levosimendan® Intermittent administration in Outpatients: effects on Natriuretic peptides in advanced chronic HEART failure) was a multicenter, double-blind, randomized, parallel group, placebo-controlled trial evaluating the efficacy and safety of intravenous administration of intermittent doses of levosimendan in outpatients with advanced chronic heart failure.

Study Design and Oversight Between November 2010 and December 2012,69 patients fulfilling inclusion criteria were enrolled from 12 recruiting centers in Spain (Figure 1). The study protocol was approved by institutional review board of each participating center and conducted in accordance with the principles of the Declaration of Helsinki (1996), International Conference on Harmonization Good Clinical Practice, and local and national regulations. All enrolled patients provided written informed consentbefore any study-related procedure was undertaken. The study was registered on the website www.ClinicalTrials.gov (unique identifier: NCT01536132) and the EudraCT database (2009-014242-28). The trial was designed, implemented and overseen by the Steering Committee. On-site monitoring of the study, data collection and data management was performed by a Clinical Research Organization (CRO, 3DHealth). The manuscript was written and submitted by the Steering Committee members. All contributing authors had full access to study data and analyses.

Study population, eligibility and recruitment The study was divided in three different parts: 1) screening (1 week), 2) treatment (12 weeks) and 3) follow-up (12 weeks). An additional vital status assessment was planned after 12 months of enrolment (Figure 2, design of the study).

Eligibility was assessed at the screening phase, once the written informed consent was obtained. Inclusion criteria for this study were: age over 18 years, left ventricular ejection fraction (LVEF) of less than 35% measured in the previous 6 months and clinical diagnosis of advanced chronic HF (Metra M et al. Eur J Heart Fail 2007; 9(6-7):684-694)according to the following criteria: a) presence for >3 months of typical signs and symptoms of HF b) persistent ambulatory NYHA functional class III or IV for the last 4 weeks, c) no signs of congestion or low cardiac output at the time of enrolment, d) episodes of pulmonary and/or systemic congestion requiring intravenous administration of diuretics (either hospitalized or in an ambulatory basis) in the previous 12 months, e) all the previous criteria were present despite optimal medical management (including use of diuretics, antagonists of the renin-angiotensin-aldosterone system and beta-blockers) and device therapy (including implantable cardioverter defibrillator-ICD and/or cardiac resynchronization therapy-CRT) or attempts to optimize it. Major exclusion criteria were: concurrent inclusion in another study, the presence of left ventricular tract obstruction, uncorrected significant primary valve disease, recent acute coronary syndrome or stroke, hypertrophic or restrictive cardiomyopathy, administration of amrinone, milrinone, enoximone, dopamine or dobutamine in the previous 3 days, administration of levosimendan in the previous 31 days, serum potassium below 3.5mmol/L, estimated glomerular filtration rate <30ml/min/m2 (MDRD-4 formula), systolic blood pressure < 90 mmHg orheart rate > 110 bpm at screening, planned or ongoing evaluation for any of the following procedures: CRT, ICD, coronary revascularization, heart transplant or left ventricular assist device (LVAD) implant, other acute or chronic conditions that would made the patient unsuitable for this study according to the investigator's judgement, anticipated poor compliance and inability or unwillingness to give informed consent.

Randomization and Blinding Eligible patients that signed the informed consent were randomized in a 2:1 ratio to receive either Levosimendan or placebo. Randomization was centrally conducted and supervised by the CRO. The computer-generated randomization scheme used random permuted blocks stratified per center. Once the patient was allocated to Levosimendan of placebo, the CRO communicated the exact treatment number to the local hospital pharmacy which handed out the treatment to the local investigator at each infusion cycle. Levosimendan and placebo had the same appearance thus the treatment was concealed to both investigators and study patients (double blind).