Comparative safety profile of chloral hydrate versus other sedatives for procedural sedation in hospitalized infants

Le résumé fourni par la source

BACKGROUND: Given the limited available evidence on chloral hydrate safety in neonatal populations and the discrepancy in chloral hydrate acceptance between the US and other countries, we sought to clarify the safety profile of chloral hydrate compared to other sedatives in hospitalized infants. METHODS: We included all infants <120 days of life who underwent a minor procedure and were administered chloral hydrate, clonidine, clonazepam, dexmedetomidine, diazepam, ketamine, lorazepam, midazolam, propofol, or pentobarbital on the day of the procedure. We characterized the distribution of infant characteristics and evaluated the relationship between drug administration and any adverse event. We performed propensity score matching, regression adjustment (RA), and inverse probability weighting (IPW) to ensure comparison of similar infants and to account for confounding by indication and residual bias. Results were assessed for robustness to analytical technique by reanalyzing the main outcomes with multivariate logistic regression, a doubly robust IPW with RA model, and a doubly robust augmented IPW model with bias-correction. RESULTS: Of 650 infants, 497 (76%) received chloral hydrate, 79 (12%) received midazolam, 54 (8%) received lorazepam, and 15 (2%) received pentobarbital. Adverse events occurred in 41 (6%) infants. Using propensity score matching, chloral hydrate was associated with a decreased risk of an adverse event compared to other sedatives, risk difference (95% confidence interval) of -12.79 (-18.61, -6.98), p < 0.001. All other statistical methods resulted in similar findings. CONCLUSION: Administration of chloral hydrate to hospitalized infants undergoing minor procedures is associated with a lower risk for adverse events compared to other sedatives.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.