Midazolam for sedation of infants in the neonatal intensive care unit.

Rationale: Proper sedation for neonates undergoing uncomfortable procedures may reduce stress and prevent complications. Midazolam is a short-acting benzodiazepine that is used in neonatal intensive care units. This is an update of a Cochrane review first published in 2003.

Objectives: To assess the benefits and harms of midazolam - compared with placebo, other pharmacological interventions, and non-pharmacological interventions - for sedation of newborn infants in neonatal intensive care units.

Search methods: We searched MEDLINE, Embase, CINAHL, CENTRAL, trial registries, and conference abstracts up to January 2025. We also screened reference lists of relevant systematic reviews and included studies.

Eligibility criteria: We selected randomised controlled trials (RCTs) and quasi-RCTs of midazolam for sedation in newborn infants compared with placebo, other pharmacological interventions, and non-pharmacological interventions. We excluded studies that used midazolam in combination with an analgesic for painful procedures, as an anaesthetic or anticonvulsant, or in infants undergoing endotracheal suctioning.

Outcomes: Critical outcomes included sedation and analgesia assessed at 30 minutes after administration, neurodevelopmental outcomes at 18 to 24 months, hypotension, and apnoea. Important outcomes included all-cause mortality until hospital discharge and days on mechanical ventilation.

Risk of bias: We used the Cochrane risk of bias tool RoB 1.

Synthesis methods: We conducted meta-analyses using fixed-effect models to calculate risk ratios (RRs) for categorical variables and mean differences (MDs) for continuous variables, each with its 95% confidence interval (CI). To assess statistical heterogeneity, we calculated the I² statistic. We evaluated the certainty of evidence according to GRADE methods.

Included studies: We included six trials (three new) in 388 neonates. The indications for sedation with midazolam were ventilatory support (4 studies) and magnetic resonance imaging (MRI; 2 studies). The gestational age and bodyweight of infants in the trials varied substantially. The administration route of midazolam was intravenous in five studies and intranasal in one. In two studies, the control arm received placebo, and in one study each, the control arm received opioids, both placebo and opioids, an oral sweet solution, and phenobarbital. We downgraded the overall certainty of evidence for all outcomes because of limitations in study design (e.g. selection bias due to lack of allocation concealment) and serious imprecision of results (wide CIs and few events). We identified three ongoing trials.

Synthesis of results: No studies reported sedation, analgesia, or neurodevelopmental outcomes at 18 to 24 months for any comparison. The evidence is very uncertain about the effects of midazolam compared with placebo on hypotension (RR 1.22, 95% CI 0.50 to 2.97; 1 study, 46 participants; very low-certainty evidence; ), apnoea (RR 1.57, 95% CI 0.45 to 5.52; 1 study, 33 participants; very low-certainty evidence; ), all-cause mortality until hospital discharge (RR 0.79, 95% CI 0.40 to 1.56; I² = 0%; 3 studies, 122 participants; very low-certainty evidence), and days on mechanical ventilation (MD 3.60, 95% CI -0.25 to 7.44; I² = 0%; 2 studies, 89 participants; very low-certainty evidence). [Figure: see text] [Figure: see text] The evidence is very uncertain about the effects of midazolam compared with opioids on hypotension (RR 1.00, 95% CI 0.07 to 15.26; 1 study, 60 participants; very low-certainty evidence; ), apnoea (RR 21.00, 95% CI 1.29 to 342.93; 1 study, 60 participants; very low-certainty evidence; ), all-cause mortality until hospital discharge (RR 3.26, 95% CI 0.14 to 76.10; 1 study, 46 participants; very low-certainty evidence), and days on mechanical ventilation (MD 0.74, 95% CI -0.30 to 1.79; I² = 77%; 2 studies, 106 participants; very low-certainty evidence). [Figure: see text] [Figure: see text] The evidence is very uncertain on the effects of midazolam compared with oral sweet solutions on apnoea (RR 17.00, 95% CI 1.00 to 287.62; 1 study, 112 participants; very low-certainty evidence; ). No studies reported the other critical outcomes for this comparison. [Figure: see text] AUTHORS' CONCLUSIONS: The evidence is very uncertain on the benefits and harms of midazolam - compared with placebo, other pharmacological interventions, and non-pharmacological interventions - for sedation of newborn infants in neonatal intensive care units. No studies reported sedation, analgesia, or long-term neurodevelopmental outcomes. This review raises concerns about the use of midazolam in neonates. There is a need for well-conducted multicentre trials with blinded outcome assessment, standardised outcome reporting, and long-term follow-up to guide safe and effective sedation practices in this vulnerable population.

Funding: This Cochrane review had no dedicated funding.

Registration: Original review (2003): doi.org/10.1002/14651858.CD002052 Review update (2017): doi.org/10.1002/14651858.CD002052.pub3.