A Bayesian re-analysis of the STRESS trial

Area of Science:

• Cardiovascular Surgery
• Pediatric Cardiology
• Pharmacology

Background:

• Prophylactic steroids are commonly used to mitigate systemic inflammatory responses during infant heart surgery involving cardiopulmonary bypass.
• The original STRESS trial found no significant difference in outcomes between methylprednisolone and placebo, though secondary analyses suggested potential benefits.
• Bayesian analytics were employed to re-evaluate the STRESS trial data for a more nuanced understanding of methylprednisolone's effects.

Purpose of the Study:

• To re-analyze the STRESS trial data using Bayesian methods to assess the probability of benefit from methylprednisolone.
• To investigate potential benefits of methylprednisolone beyond the primary endpoint of the original STRESS trial.
• To explore the impact of varying prior beliefs on the probability of methylprednisolone efficacy.

Main Methods:

• A covariate-adjusted proportional odds model was used with the STRESS trial's composite endpoint (death, transplant, major complication, length of stay).
• Markov Chain Monte Carlo simulations were performed to determine the probability of benefit (OR < 1) versus harm (OR > 1).
• Sensitivity analyses explored various prior beliefs (pessimistic, neutral, optimistic) and prior strengths, alongside a secondary analysis using priors from previous steroid trials.

Main Results:

• The posterior probability of benefit from methylprednisolone was 92%, with an 8% probability of harm.
• Methylprednisolone was associated with an absolute risk difference of -2% for composite death or major complication.
• Nine sensitivity analyses consistently showed a higher probability of benefit than harm, with 8 demonstrating >80% probability of benefit and ≥1% absolute risk reduction.

Conclusions:

• Bayesian re-analysis indicates a high probability that perioperative methylprednisolone reduces the risk of death or major complications in infants undergoing cardiopulmonary bypass.
• This in-depth analysis provides stronger evidence for methylprednisolone's efficacy compared to the initial STRESS trial findings.
• The findings support the use of methylprednisolone as a prophylactic measure in this patient population.