Perioperative Corticosteroid Therapy in Children Undergoing Cardiac Surgery: A Systematic Review and Meta-Analysis.

Background: The benefit-risk profile of perioperative corticosteroids in pediatric patients undergoing cardiac surgery remains controversial. Objective: To investigate the influence of perioperative corticosteroids on the postoperative mortality and clinical outcomes in pediatric patients undergoing cardiac surgery with cardiopulmonary bypass. Methods: We conducted a systematic search using MEDLINE, EMBASE, and Cochrane Database through August 31, 2019. We included randomized controlled trials comparing perioperative corticosteroids with other clinical interventions, placebo, or no treatment in children between 0 and 18 years of age undergoing cardiac surgery. The primary outcome of interest was all-cause in-hospital mortality. The secondary outcomes were length of intensive care unit stay (LOIS), duration of mechanical ventilation (DMV), postoperative insulin therapy, postoperative low cardiac output syndrome (LCOS), postoperative infection, maximal temperature ( T max ) in the first 24 h postoperatively, urine output (UO) in the first 24 h postoperatively, serum lactate at postoperative day (POD) 1, blood glucose at POD 1, vasoactive inotrope score (VIS) at POD 1, and postoperative acute kidney injury (AKI). Study quality was assessed using the Cochrane Risk of Bias Assessment Tool. Results: Our analysis included 17 studies and 848 pediatric patients. The data demonstrated that children receiving corticosteroids showed no significant difference on the all-cause in-hospital mortality with a fixed-effect model (RR = 0.59, 95% CI = 0.28-1.25, P = 0.55) compared with controls. For the secondary outcomes, corticosteroids had a statistically significant reduction on the VIS at POD1 (MD = -2.04, 95% CI = -3.96 -0.12, P = 0.04), while it might be significantly associated with an increased blood glucose at POD1 (MD = 1.38, 95% CI = 0.68-2.09, P = 0.0001) and a 2.69-fold higher risk of postoperative insulin therapy (RR = 2.69, 95% CI = 1.37-5.27, P = 0.004). No statistical significance was shown in other secondary outcomes. Conclusion: Perioperative corticosteroids might not significantly improve clinical outcomes identified as mortality, LOIS, DMV, AKI, and LCOS other than VIS at POD1. However, it might increase the blood glucose and episodes of insulin therapy. Perioperative corticosteroids to attenuate the inflammatory response are not supported by available evidence from our study. Further results from ongoing randomized controlled trials with a larger sample size are required.