Inotropes for the prevention of low cardiac output syndrome and mortality for paediatric patients undergoing surgery for congenital heart disease: a network meta-analysis

Background: Paediatric patients undergoing surgery for congenital heart disease (CHD) are at risk for postoperative low cardiac output syndrome (LCOS) and mortality. LCOS affects up to 25% of children after heart surgery. It consists of reduced myocardial function and increases postoperative morbidity, prolongs mechanical ventilation, and lengthens the duration of intensive care unit (ICU) stay. Pharmacological prophylaxis involves inotropes, including catecholamines, phosphodiesterase III inhibitors, or calcium sensitisers, to enhance myocardial contractility. It is unclear whether they are effective in preventing LCOS or death in this vulnerable population.

Objectives: 1. To evaluate the relative benefits and harms of inotropes for the prevention of LCOS and mortality in paediatric patients undergoing surgery for CHD. 2. To generate a clinically useful ranking of prophylactic inotropes for the prevention of LCOS and mortality in paediatric patients undergoing surgery for CHD according to benefits and harms.

Search methods: We searched CENTRAL, MEDLINE, Embase, Web of Science, and clinical trial registries, most recently in December 2023 and April 2024. We also checked reference lists from identified studies and review articles. We did not apply any language restrictions.

Selection criteria: We included randomised controlled trials comparing inotropes from one drug class (catecholamines, phosphodiesterase type III inhibitors, calcium sensitisers) to another (either alone or in combination) or placebo, in paediatric patients (birth to 18 years of age) undergoing surgery for CHD.

Data collection and analysis: Two review authors independently selected studies, extracted data, assessed risk of bias, and rated the certainty of evidence using the CINeMA framework. We performed random-effects network and pairwise meta-analyses comparing the relative effects of each possible pair of medications with each other or placebo. Where meta-analysis was not possible, we provided a narrative description of the results. We ranked the prophylactic medications according to their effects relative to each other. The primary outcomes were all-cause mortality within 30 days, time to death, and LCOS incidence; secondary outcomes were length of ICU stay, length of hospital stay, duration of mechanical ventilation, inotrope score, mechanical circulatory support, and adverse events.

Main results: We included 13 studies with 937 participants. All except two multicentre studies were conducted at single tertiary care hospitals. Participants comprised children from birth to 14 years of age undergoing surgery for different types of CHD on cardiopulmonary bypass. Five studies compared levosimendan versus milrinone; two compared levosimendan versus placebo; two compared milrinone versus placebo (one comparing two different doses); one compared levosimendan versus dobutamine, another milrinone versus dobutamine. Two studies used combinations of inotropes. Study duration was between less than one year and 5.3 years, with follow-up mostly during ICU or hospital stay. Funding sources included governmental bodies and hospital departments, but also drug manufacturers. We downgraded the certainty of evidence for high risk of bias at study level, or imprecision at comparison level. Primary outcomes Compared to placebo, levosimendan likely results in a large reduction in mortality (risk ratio (RR) 0.57, 95% confidence interval (CI) 0.15 to 2.13) and milrinone likely results in no difference (RR 0.97, 95% CI 0.11 to 8.49), whereas for dobutamine, no effect was estimable; all moderate-certainty evidence (9 studies, 557 participants, 14 events). LCOS was largely reduced with levosimendan (RR 0.45, 95% CI 0.24 to 0.83; high-certainty evidence), likely largely reduced with milrinone (RR 0.46, 95% CI 0.24 to 0.89; moderate-certainty evidence), and may be reduced with low-dose milrinone (RR 0.7, 95% CI 0.39 to 1.28; low-certainty evidence), compared with placebo (5 studies, 513 participants, 85 events). Secondary outcomes The length of ICU stay may be no different with levosimendan (ratio of means (ROM) 1.12, 95% CI 0.77 to 1.63; low-certainty evidence), and is likely no different with milrinone (ROM 1.13, 95% CI 0.75 to 1.69) or with dobutamine (ROM 1.11, 95% CI 0.66 to 1.86), compared with placebo (9 studies, 577 participants); both moderate-certainty evidence. The length of hospital stay, compared with placebo, is likely no different with levosimendan (ROM 1.03, 95% CI 0.84 to 1.27) or with milrinone (ROM 1, 95% CI 0.78 to 1.3), but is likely reduced with dobutamine (ROM 0.68, 95% CI 0.37 to 1.26); all moderate-certainty evidence (7 studies, 297 participants). The duration of mechanical ventilation, compared with placebo, is likely increased with levosimendan (ROM 1.17, 95% CI 0.65 to 2.12) or with milrinone (ROM 1.25, 5% CI 0.67 to 2.36) and is likely no different with dobutamine (ROM 1.04, 95% CI 0.45 to 2.38); all moderate-certainty evidence (9 studies, 577 participants). There is moderate-certainty evidence that adverse events are likely increased with levosimendan (incidence rate ratio (IRR) 1.23, 95% CI 0.78 to 1.96) or dobutamine (IRR 1.24, 95% CI 0.75 to 2.03) and low-certainty evidence that they may be increased with milrinone (IRR 1.31, 95% CI 0.96 to 1.79) and decreased with low-dose milrinone (IRR 0.84, 95% CI 0.47 to 1.5), compared with placebo (8 studies, 706 participants, 380 events).

Authors' conclusions: Levosimendan likely results in a large reduction in mortality compared to placebo in paediatric patients undergoing surgery for congenital heart disease, whereas milrinone likely results in no difference, and the effect of dobutamine is unknown. Low cardiac output syndrome (LCOS) is largely reduced with levosimendan, likely largely reduced with milrinone, and may be reduced with low-dose milrinone, compared to placebo. The length of ICU stay may be no different with levosimendan and is likely no different with milrinone or with dobutamine, compared to placebo. The length of hospital stay is likely no different with levosimendan or with milrinone, but is likely reduced with dobutamine, compared to placebo. The duration of mechanical ventilation is likely increased with levosimendan or with milrinone and is likely no different with dobutamine, compared to placebo. Adverse events are likely increased with levosimendan or dobutamine, and may be increased with milrinone and decreased with low-dose milrinone, compared to placebo. The evidence is based on few, heterogeneous studies, with small numbers of patients and short follow-up periods. Future research should include large numbers of patients, consistently report all co-interventions, and ensure the longest possible follow-up.