Is levosimendan effective in paediatric heart failure and post-cardiac surgeries?

Abstract

A best evidence topic in cardiothoracic surgery was written according to a structured protocol. The question addressed was ‘do children with heart failure post-cardiac surgery undergoing treatment with levosimendan have an acceptable haemodynamic improvement?’ The use of levosimendan as a vasoactive drug is an accepted intervention for patients with altered haemodynamics post-cardiac surgeries. However, the role of levosimendan and its efficacy have been debated. Eleven relevant papers were identified, which represented the best evidence to answer the question. The author, journal, date, country of publication and relevant outcomes are tabulated. The 11 studies comprised 3 randomized trials, 2 of which compared levosimendan and milrinone. A single-centre randomized study that included 40 infants showed that cardiac output (CO) and cardiac index (CI) increased overtime in the levosimendan group compared with the milrinone group. The significant interaction for CO (P = 0.005) and CI (P = 0.007) indicated different time courses in the two groups. A similar, European randomized study undertaken on neonates (n = 63) showed better lactate levels [P = 0.015 (intensive care admission); P = 0.048 (after 6 h) with low inotropic scores in the levosimendan group. Although the length of mechanical ventilation and mortality were less, this was statistically insignificant. A retrospective cohort analysis (n = 13) in children reported a reduced use of dobutamine and improvement in the ejection fraction from 29.8 to 40.5% (P = 0.015) with the use of levosimendan. In a questionnaire-based study from Finland, 61.1% of respondents felt that it had saved the lives of some children when the other treatments had failed. No study reported any adverse effect attributable to use of levosimendan. In conclusion, the above studies were in favour of levosimendan as a safe and feasible drug providing potential clinical benefit in low cardiac output syndrome (LCOS) and post-cardiac surgeries when other vasoactive drugs were insufficient to maintain stable haemodynamics. A small sample size was indeed a limitation in all the above studies. Furthermore, it is best used as a rescue drug on a named-patient basis. A small sample size was indeed a limitation in all the above studies. Larger, well-designed trials are required to further evaluate the efficacy and feasibility of levosimendan in paediatric heart failure and post-cardiac surgeries.

INTRODUCTION

A best evidence topic was constructed according to a structured protocol. This is fully described in the ICVTS [1].

THREE-PART QUESTION

Do [children with heart failure post cardiac surgery] undergoing [treatment with levosimendan] have an acceptable [hemodynamic profile]?

CLINICAL SCENARIO

You are on the paediatric intensive care unit and a 36-month old baby is admitted post-congenital cardiac surgery. Two hours postoperatively, the child is tachycardic with a low blood pressure and a trailing urine output. Mixed venous saturation (SvO₂) is 56%, serum lactate is 12 mmol/l, left atrial pressure is 25 mmHg, and echocardiography showed reduced left ventricular function with a fractional shortening (FS) of 10%. The intensivists suggest the use of levosimendan in this situation. You are unsure of its role in the critically ill paediatric population and hence resolve to check the literature.

SEARCH STRATEGY

The MEDLINE database was searched from the date of inception to December 2012 using Medical Subject Headings (MeSH) search terms ‘levosimendan’, ‘paediatric cardiac surgery’, ‘paediatric heart failure’ and ‘congenital heart disease’. The Cochrane database of systemic reviews, EMBASE was also searched. In addition, related articles and references were screened for suitable articles.

SEARCH OUTCOME

Seventeen studies were found using the reported search. From these, eleven papers were identified that provided the best evidence to answer the question. These are presented in Table 1.

Table 1:

Best evidence papers

Author, date, journal and country Study type (level of evidence)	Patient group	Outcomes	Key results	Comments
Lechner et al. (2012), Pediatr Crit Care Med, [2] Prospective double-blind randomized study (level I)	The study enrolled 40 infants comparing use of levosimendan with milrinone	CI (as a primary endpoint) Haemodynamic parameters Inotropic score, NIRS, length of ICU stay	CO and CI increased overtime in the levosimendan group The other parameters were statistically insignificant heart rate (P = 0.172), systolic blood pressure (P = 0.62), lactate (P = 0.80) Inotropic score (P = 0.52), NIRS (P = 0.42) ICU stay (P = 0.72)	The study group did not include pre-existing congestive cardiac failure and complex congenital heart lesions (single ventricle lesions)
Momeni et al. (2010), J Cardiothorac Vasc Anesth, Belgium [3] Prospective randomized double-blind study (level I)	41 children in the age group 0–5 years were randomized in a double-blind fashion to a continuous infusion of either levosimendan or milrinone started at the onset of cardiopulmonary bypass 0.36 patients completed the study	Serum lactate at 4 h (as a primary endpoint) Haemodynamic parameters	Change in serum lactate was insignificant Lower HR (<0.05) and a lower rate pressure index (P < 0.001) were noted in the levosimendan group	Authors noted levosimendan is as efficacious as milrinone Limitations: Heterogenous study group No quantitative measurement of cardiac function done No loading dose used No risk stratification of patient groups has been stated
Ricci et al. (2012), Intensive Care Med, Italy [4] Case control study (level of evidence I)	Study of levosimendan in comparison to a standard inotrope with risk stratification (RACHS 3–4) The study involved 63 neonates (32 cases and 31 controls)	Lactate levels Inotropic score Ventilation days and ICU stay Survival Side effects	Better lactate levels (P = 0.015 at ICU admission; P = 0.048 at 6 h) Better ionotropic scores in the levosimendan group (P < 0.0001) Length of mechanical ventilation (5.9 ± 5 vs 6.9 ± 8 days, P = 0.54) and ICU stay (11 ± 8 vs 14 ± 14 days, P = 0.26) though better was not statistically significant Mortality (1 vs 3 patients, P = 0.35) None reported	Levosimendan was well tolerated with benefit on haemodynamic and metabolic parameters Limitations: The study was a pilot open label uncommitted trial Patients with only biventricular anatomy included; RACHS 5–6 not included Inadequate sample size (type II error) Conventional Inotropic score used which did not take levosimendan into consideration when calculating MAP not in the definition of LCOS was used to guide ionotrope use
Magliola et al. (2009), Intensive Care Med, Spain [5] Prospective cohort study (level II-2)	Open, quasiexperimental cohort involving 14 children with refractory LCOS (18 opportunities)	CO was the primary endpoint inotropic score Mixed venous saturation	Successful in 9/18 (P = 0.004) Ionotropic score (P = 0.01) and A-VDO2 (P = 0.029) showed reduction, SvO2 showed improvement (P = 0.03)	Improved CO in 50% of the interventions and no adverse effect reported Limitations: Heterogenous sample with no risk stratification
Namachivayam et al. (2006), Pediatr Crit Care Med, Australia [6] Retrospective cohort analysis (level II-2)	15 children aged 7 days to 18 years (median age 38 months) with severe myocardial dysfunction secondary to end-stage heart failure, or acute heart failure, who were inotrope-dependent (requiring at least one catecholamine)	Effect on catecholamine use Haemodynamic parameters Ejection fraction	Reduced dose of dobutamine (from 6.4 μg/kg/min pre-levosimendan to 1.8 μg/kg/min on day 5 (P < 0.01) Hear rate, blood pressure and central venous pressure were unchanged Ejection fraction improved from 29.8 to 40.5% with levosimendan (P = .015)	Showed substantial reduction in catecolamine use Limitations: Retrospective data; Heterogenous group Inotropic score not commented Diastolic myocardial performance assessment may have been useful
Osthaus et al. (2009), Eur J Pediatr, Germany [7] Case series (level III)	Case series involving seven infants (body weight range 2.6–6.3 kg) with severe myocardial dysfunction after complex congenital heart surgery	Haemodynamic parameters	Lactate levels decreased Significant increase in central venous oxygen saturation	Small number of patients No risk stratification
Pertti et al. (2011), BMC Anesth, Finland [8] Retrospective study (level III)	Analysis of data involving 293 patients and 484 infusions (4 h 21 years age group) over 10 years Web-based questionnaire survey concerning their clinical perceptions of levosimendan were also evaluated	The indication of using levosimendan Need for mechanical assist device Survival	The most common indication for the use of levosimendan (94%) was when the other inotropic agents were insufficient to maintain stable haemodynamics 70.1% successfully weaned during study period 44% of respondents thought that the mechanical support had been totally avoided in some patients 61.1% of respondents felt that it had saved the lives of some children when the other treatments had failed 11% compared with overall mortality of 2.3% in cardiac surgery patients	Authors attributed higher mortality for its use in very sick children Limitations: Heterogenous patient data Retrospective study No risk stratification No data on haemodynamic parameters evaluated
Bravo et al. (2011), Neonatology, Spain [9] Prospective case series (level III)	Case series study involving neonates seven neonates	Cerebral and peripheral intravascular oxygenation	NIRS showed increased intravascular oxygenation but no change in tissue oxygenation index Reduced need of cardiovascular support and decreased serum lactate levels Mixed linear model analysis identified blood pressure changes and levosimendan as factors independently associated with cerebral oxygenation	Levosimendan was used as a rescue therapy Limitations: Small group Echocardiographic performance lacking
Lobacheva et al. (2010), Anesteziol Reanimatol, Russia [10] Cohort study (level II)	Study of 75 cases with postoperative LCOS within the age group 3 days to 2 years 10 months	Haemodynamic parameters LVEF by echocardiography Side effects	Increase in mean blood pressure [from 42 to 53 mmHg; (P < 0.05)]. Reduction in left atrial pressure [from 25 to 17 mmHg; (P < 0.05)] CVP unchanged LVEF increased from 21 to 27% Systemic hypotension in first hour requiring fluid boluses and epinephrine	Authors recommend levosimendan as an alternative to PDE III inhibitors in LCOS as well as a basic drug during extracorporeal circulation and after its cessation
Lechner et al. (2007), Pediatr Crit Care Med, Austria [11] Case report (level III)	Single case report of premature 32 weeks (1.5 kg)	Haemodynamic parameters	Left atrial pressure decreased to 7 from 24 mmHg Systolic arterial pressure increased to 60 mmHg from 40 mmHg Serum lactate level normalized to 1.7 mmol/l from 14.8 mmol/l Mixed venous saturation increased to 81 from 56% Improvement in left ventricular function (FS increased from 10–25%)	It is difficult to extrapolate the results in a premature neonate to the haemodynamics of older infants
Braun et al. (2004), Eur J Cardiothorac Surg, Berlin [12] Case report (level III)	Case report of a 2 month baby	Survival	Improved left ventricular function and decreased vascular resistance	Single case

A-VDO₂: arterio-venous difference in O2-content; CI: cardiac index; CO: cardiac output; CVP: central venous pressure; FS: fractional shortening; HR: heart rate; ICU: intensive care unit; LCOS: low cardiac output syndrome; LVEF: left ventricle ejection fraction; MAP: mean arterial pressure; NIRS: near infrared spectroscopy; PDE: phosphodiesterase; RACHS: risk adjustment for congenital heart surgery.

Lechner et al. [2] reported a lower cardiac output (CO) and cardiac index (CI) values initially in the levosimendan group compared with milrinone, and this increased at the end of 48 h. The significant interaction for CO (P = 0.005) and CI (P = 0.007) indicated different time courses of CO and CI in the two groups. However, P-values for heart rate, serum lactate, blood pressure, mixed venous saturation and near infrared spectroscopy (NIRS) were insignificant. The use of additional catecholamines, as reflected in the inotropic score did not differ between the groups (74 vs 70%).

Momeni et al. [3] reported a lower rate pressure index, an indicator of myocardial oxygen demand, at 24 and 48 h postoperatively in the levosimendan group (P < 0.001) in comparison with the milrinone group. Although not significantly different, the troponin values in the levosimendan group were less at 1 h {median [p (25)-p (75): 20.7 (15.3–48.3) vs 34.6 (23.8–64.5) ng/ml]} and 4 h postoperatively [30.4 (17.3–59.9) vs 33.3 (25.5–76.7) ng/ml]. Lactate levels were non-significant. They concluded that levosimendan is at least as efficacious as milrinone after corrective congenital cardiac surgery in neonates and infants.

Ricci et al. [4] showed no significant differences in mortality (1 vs 3 patients, P = 0.35), length of mechanical ventilation (5.9 ± 5 vs 6.9 ± 8 days, P = 0.54), and paediatric cardiac intensive care unit stay (11 ± 8 vs 14 ± 14 days, P = 0.26) against a standard post-cardiopulmonary bypass inotropic infusion, with better controlled postoperative heart rate and lactate levels at admission, 6, 12 and 24 h.

Magliola et al. [5] concluded that levosimendan improved CO in 50% of the interventions with post-surgical LCOS, and no adverse effect was observed. Both inotropic score (12.1 vs 6.1, P = 0.01) and arterio-venous difference in O2-content (26.78 ± 11.5 vs 20.81 ± 7.72%, P = 0.029) showed reduction, while SvO₂ improved (69.5 ± 11.4 vs 76 ± 9.29%, P = 0.03).

Namachivayam et al. [6] showed that levosimendan allowed for substantial reduction in catecholamine infusions in children with end-stage or acute heart failure and also produced an objective improvement in myocardial performance in children with acute heart failure.

Osthaus et al. [7] reported that the administration of levosimendan in seven infants with severe myocardial dysfunction was well tolerated intraoperatively. The mean arterial lactate declined. Central venous oxygen saturation increased significantly 24 and 48 h after the onset of levosimendan infusion.

Pertti et al. [8] studied 484 levosimendan infusions delivered to 293 patients over 10 years, as administered to children with cardiac surgery (72%), cardiomyopathy (14%) and with cardiac failure (14%). The most common indication for the use of levosimendan (94%) was when the other inotropic agents were insufficient to maintain stable haemodynamics. The results of a questionnaire concerning the perceptions of clinicians were also evaluated. Eighty-nine percent of the respondents believed levosimendan administration postponed the need for mechanical assist device in some children with cardiomyopathy. Furthermore, 61.1% of respondents felt that it had saved the lives of some children when the other treatments had failed, and 44% thought mechanical support was totally avoided in a few patients post-cardiac surgery after receiving levosimendan.

Bravo et al. [9] reported that levosimendan as a rescue therapy produced an increase in cerebral (P < 0.05) and peripheral (non-significant) intravascular oxygenation, a decrease in heart rate (P < 0.001) and serum lactate (P < 0.05) along with reduction in cardiovascular support. The study concluded that levosimendan improves cerebral and systemic perfusion and oxygenation in critically ill neonates suffering from LCOS.

Labacheva et al. [10] observed that during levosimendan infusion, there was a significant increment in mean blood pressure and a reduction in left atrial pressure. The left ventricular ejection fraction significantly rose by 6%. The major adverse reaction noted was a tendency towards systemic hypotension within the first hour of levosimedan infusion.

Lechner et al. [11] reported a premature infant postatrial switch operation in heart failure. Administration of levosimendan when conventional inotropes failed resulted in the increase in systemic pressure, decrease in left atrial pressure with improvement of left ventricular function and fractional shortening.

Braun et al. [12] published that a breakthrough in treatment was achieved in a two-month old baby by using levosimendan to improve left ventricular function and to decrease vascular resistance.

CLINICAL BOTTOM LINE

The current best evidence suggests that levosimendan is beneficial in improving cardiac performance and reducing the left ventricular afterload. In addition, it may be effective in reducing the need for catecholamine and the duration of critical care. Furthermore, it is safe and well tolerated. It is promising as a rescue drug on the named-patient basis for a potential clinical benefit in low-CO syndrome and post-cardiac surgeries. No data are available to validate its role with regard to its cost effectiveness in comparison with milrinone. These encouraging results need to be evaluated by larger, well-designed clinical trials and its indications further elucidated. However, the present evidence may not be enough to recommend it to change the current practice in paediatrics.

Conflict of interest: none declared.