JAMA Netw Open. 2026 Mar 2;9(3):e263852. doi: 10.1001/jamanetworkopen.2026.3852.
ABSTRACT
IMPORTANCE: In preterm neonates supported with continuous positive airway pressure (CPAP), early caffeine administration and less invasive surfactant administration (LISA) results in a lower frequency of endotracheal intubation. It is unknown whether this regimen improves outcomes at a corrected age of 2 years in this population.
OBJECTIVE: To determine whether LISA improves neurodevelopmental impairment (NDI) and pulmonary outcomes at a corrected age of 2 years.
DESIGN, SETTING, AND PARTICIPANTS: This prespecified secondary analysis and follow-up of a randomized clinical trial was performed at 3 academic medical centers in California. Participants included infants born at gestational ages ranging from 24 weeks 0 days to 29 weeks 6 days with follow-up assessments available. Follow-up visits occurred from May 2, 2022, to April 3, 2025.
INTERVENTIONS: Infants received intravenous caffeine by 2 hours of life followed by either less invasive surfactant administration (intervention group) or CPAP without initial surfactant administration (control group).
MAIN OUTCOMES AND MEASURES: The primary outcome was the composite of death or moderate to severe NDI at a corrected age of 2 years, as measured by the Bayley Scales of Infant Development (BSID). Secondary outcomes included components of the BSID composite; the third edition of the Ages and Stages Questionnaire (ASQ-3); developmental screening; pulmonary outcomes of bronchodilator use, oral and/or inhaled corticosteroid use, or hospitalizations with respiratory diagnoses after discharge; and results of an autism screen.
RESULTS: Of 180 randomized infants, 147 (81.7%) had follow-up assessments available (74 in the LISA group and 73 in the CPAP group); 75 infants (51.0%) were male. The mean (SD) gestational age at the time of the Bayley assessment was 24.6 (1.5) months corrected age for the LISA group and 24.7 (2.2) months corrected age for the CPAP group. The mean (SD) chronological age for the ASQ-3 was 28.1 (2.4) months for both groups. Death or moderate to severe NDI occurred in 17 of 74 children (23.0%) in the LISA group and 23 of 70 (32.9%) in the control group (odds ratio [OR], 1.56 [95% CI, 0.77-3.17]; P = .22). On the screening ASQ-3, infants in the LISA group had no statistically significant differences in rates of typical development (22 of 69 [31.9%] vs 12 of 67 [17.9%]; OR, 0.47 [95% CI, 0.21-1.04]; P = .06) and scores indicating possible delay (43 of 69 [62.3%] vs 47 of 67 [70.1%]; OR, 1.42 [95% CI, 0.70-2.90]; P = .34) compared with the control group. Children in the LISA group were more likely to have a mean (SD) fine motor z score in the reference range (-0.59 [1.10] vs -1.00 [1.01]; P = .03). There were no differences between the LISA and CPAP groups in bronchodilator use, oral and/or inhaled corticosteroid use, or postdischarge hospitalizations with respiratory diagnoses.
CONCLUSIONS AND RELEVANCE: In this secondary analysis of a randomized clinical trial of preterm infants supported with CPAP, early caffeine administration plus LISA did not reduce the incidence of death or moderate to severe NDI noted on the BSID. Infants who received LISA were more likely to have a fine motor domain score in the reference range on ASQ-3 developmental screen.
TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04209946.
PMID:41915392 | DOI:10.1001/jamanetworkopen.2026.3852
