Why is this question important?
Children who are born with a heart defect, or who develop heart disease after birth, may need heart surgery. To operate, surgeons often need to stop the heart and lungs temporarily. To keep the child alive, they use a heart-lung machine that takes over the work of the heart and lungs. The machine adds oxygen to the blood, removes carbon dioxide from it, and pumps the blood back into the child’s body.
We know that heart surgery involving a heart-lung machine causes inflammation across the body. This can cause complications ranging from low blood pressure to major organ dysfunction. In some cases, patients may die.
Corticosteroids (a type of anti-inflammation medicine) have been widely used to prevent inflammation in children who undergo heart surgery that requires a heart-lung machine, but their benefits and risks are unclear. To find out whether they prevent inflammation, and whether they are associated with any unwanted effects (such as poor wound healing, increased risk of infection or increased risk of death), we reviewed the evidence from research studies.
How did we identify and evaluate the evidence?
First, we searched the medical literature for randomized controlled studies (studies in which people are randomly divided into different treatment groups), because these studies provide the most robust evidence about the effects of a treatment. We then compared the results and summarised the evidence from all the studies. Finally, we assessed how certain the evidence was. To do this, we considered factors such as the way studies were conducted, study sizes, and consistency of findings across studies. Based on our assessments, we categorised the evidence as being of very low, low, moderate or high certainty.
What did we find?
We found 13 studies that involved a total of 1087 children. The studies lasted for between 14 months and 30 months (duration was not reported for seven studies). Three corticosteroids were investigated: methylprednisolone (five studies), hydrocortisone (two studies) and dexamethasone (six studies). The studies compared these corticosteroids against a placebo (medicine that is exactly the same apart from it does not have the active medicine in it).
The evidence shows that:
- corticosteroids probably make little or no difference to the number of children who die in hospital after surgery (five studies, 313 children (participating in the studies), moderate-certainty evidence);
- corticosteroids probably make little or no difference to the number of children who die from any cause (five studies, 313 children, moderate-certainty evidence) or from heart and circulation problems specifically (three studies, 109 children, moderate-certainty evidence) at the longest follow-up time after surgery;
- corticosteroids may make little or no difference to whether children are taken off the heart-lung machine after surgery (one study, 40 children, low-certainty evidence).
- corticosteroids reduce the number of hours for which children need a breathing machine (six studies, 421 children, high-certainty evidence);
- corticosteroids make little or no difference to the length of time children spend in the intensive care unit (six studies, 421 children, high-certainty evidence);
- corticosteroids probably make little or no difference to the total length of time children spend in hospital after surgery (one study, 176 children, moderate-certainty evidence).
It is unclear whether corticosteroids are associated with non-fatal unwanted effects because the studies did not report on unwanted effects consistently.
What does this mean?
Giving corticosteroids to children who have heart surgery that requires a heart lung-machine:
- probably makes little or no difference to the number who die after surgery at any point or from any cause;
- may make little to no difference to whether children are taken off the heart-lung machine after surgery;
- probably reduces the length of time spent on the breathing machine after surgery, but this does not lead to a shorter stay in the intensive care unit or hospital.
Future studies need to collect information on non-fatal unwanted effects in a standardised way, so that we can evaluate the risks of corticosteroids.
How-up-to date is this review?
The evidence in this Cochrane Review is current to June 2020.
Corticosteroids probably do not change the risk of mortality for children having heart surgery using CPB at any time point. They probably reduce the duration of postoperative ventilation in this context, but have little or no effect on the total length of postoperative ICU stay or total postoperative hospital stay. There was inconsistency in the adverse event outcomes reported which, consequently, could not be pooled. It is therefore impossible to provide any implications and policy-makers will be unable to make any recommendations for practice without evidence about adverse effects. The review highlighted the need for well-conducted RCTs powered for clinical outcomes to confirm or refute the effect of corticosteroids versus placebo in children having cardiac surgery with CPB. A core outcome set for adverse event reporting in the paediatric major surgery and intensive care setting is required.
Corticosteroids are routinely given to children undergoing cardiac surgery with cardiopulmonary bypass (CPB) in an attempt to ameliorate the inflammatory response. Their use is still controversial and the decision to administer the intervention can vary by centre and/or by individual doctors within that centre.
This review is designed to assess the benefits and harms of prophylactic corticosteroids in children between birth and 18 years of age undergoing cardiac surgery with CPB.
We searched CENTRAL, MEDLINE, Embase and Conference Proceedings Citation Index-Science in June 2020. We also searched four clinical trials registers and conducted backward and forward citation searching of relevant articles.
We included studies of prophylactic administration of corticosteroids, including single and multiple doses, and all types of corticosteroids administered via any route and at any time-point in the perioperative period. We excluded studies if steroids were administered therapeutically. We included individually randomised controlled trials (RCTs), with two or more groups (e.g. multi-drug or dose comparisons with a control group) but not ‘head-to-head' trials without a placebo or a group that did not receive corticosteroids. We included studies in children, from birth up to 18 years of age, including preterm infants, undergoing cardiac surgery with the use of CPB. We also excluded studies in patients undergoing heart or lung transplantation, or both; studies in patients already receiving corticosteroids; in patients with abnormalities of the hypothalamic-pituitary-adrenal axis; and in patients given steroids at the time of cardiac surgery for indications other than cardiac surgery.
We used the Covidence systematic review manager to extract and manage data for the review. Two review authors independently assessed studies for inclusion, extracted data, and assessed risks of bias. We resolved disagreements by consensus or by consultation with a third review author. We assessed the certainty of evidence with GRADE.
We found 3748 studies, of which 888 were duplicate records. Two studies had the same clinical trial registration number, but reported different populations and interventions. We therefore included them as separate studies. We screened titles and abstracts of 2868 records and reviewed full text reports for 84 studies to determine eligibility. We extracted data for 13 studies. Pooled analyses are based on eight studies. We reported the remaining five studies narratively due to zero events for both intervention and placebo in the outcomes of interest. Therefore, the final meta-analysis included eight studies with a combined population of 478 participants.
There was a low or unclear risk of bias across the domains. There was moderate certainty of evidence that corticosteroids do not change the risk of in-hospital mortality (five RCTs; 313 participants; risk ratio (RR) 0.83, 95% confidence interval (CI) 0.33 to 2.07) for children undergoing cardiac surgery with CPB. There was high certainty of evidence that corticosteroids reduce the duration of mechanical ventilation (six RCTs; 421 participants; mean difference (MD) 11.37 hours lower, 95% CI -20.29 to -2.45) after the surgery. There was high-certainty evidence that the intervention probably made little to no difference to the length of postoperative intensive care unit (ICU) stay (six RCTs; 421 participants; MD 0.28 days lower, 95% CI -0.79 to 0.24) and moderate-certainty evidence that the intervention probably made little to no difference to the length of the postoperative hospital stay (one RCT; 176 participants; mean length of stay 22 days; MD -0.70 days, 95% CI -2.62 to 1.22). There was moderate certainty of evidence for no effect of the intervention on all-cause mortality at the longest follow-up (five RCTs; 313 participants; RR 0.83, 95% CI 0.33 to 2.07) or cardiovascular mortality at the longest follow-up (three RCTs; 109 participants; RR 0.40, 95% CI 0.07 to 2.46). There was low certainty of evidence that corticosteroids probably make little to no difference to children separating from CPB (one RCT; 40 participants; RR 0.20, 95% CI 0.01 to 3.92). We were unable to report information regarding adverse events of the intervention due to the heterogeneity of reporting of outcomes.
We downgraded the certainty of evidence for several reasons, including imprecision due to small sample sizes, a single study providing data for an individual outcome, the inclusion of both appreciable benefit and harm in the confidence interval, and publication bias.