Purpose: Cognitive outcomes in preterm infants may be adversely affected by use of sedation and anesthetic agents. We investigated the associations between anesthetics/sedatives and full-scale intelligence quotient (FSIQ) measured at 36 months corrected age (CA) in very preterm infants (born < 29 weeks gestational age).

Methods: This retrospective cohort study included preterm infants born at < 29 weeks of gestation between 1 January 2006 and 31 December 2012, whose cognitive outcomes were assessed at 36 months CA. Imputed and complete case univariable and adjusted multivariable linear regressions were used to investigate the associations between FSIQ [standardized to mean (standard deviation) 100 (15)] and exposure to volatile anesthetics, propofol, benzodiazepines, barbiturates, and ketamine. These agents were the subject of a 2016 warning from regulatory authorities in the USA recommending caution for administration to children and pregnant women.

Results: A total of 731 infants met the inclusion criteria. Unadjusted associations were -7 (95% confidence interval [CI], -10 to -4; P < 0.001) and -6 (95% CI, -10 to -3; P < 0.001) FSIQ points with exposure to warned medications using imputed and complete case analyses, respectively. Imputed and complete case adjusted associations between FSIQ and warned medications were -3 (95% CI, -7 to 0; P = 0.045) and -4 (95% CI, -8 to 0; P = 0.071) FSIQ points, respectively. Adjusted associations between volatile anesthetic exposure only and FSIQ were -3 (95% CI, -6 to 0; P = 0.072) and -5 (95% CI, -9 to -2; P = 0.004) FSIQ points using imputed and complete case data sets, respectively. FSIQ was not associated with opioid exposure.

Conclusion: Exposure of very preterm infants to anesthetics/sedatives on the United States Food and Drug Administration warning list was associated with a decrease in FSIQ points at 36 months CA. There was no association between opioid exposure and FSIQ.