Objective: To assess site variability and concomitant respiratory support related to the timing of caffeine discontinuation, and compare clinical characteristics of infants who discontinued caffeine before vs. within the last week of hospitalization.
Study design: Cohort study of 81,110 infants <35 weeks gestational age and <1500 g birth weight discharged from 304 neonatal intensive care units from 2001-2016.
Results: The mean postmenstrual age at caffeine discontinuation ranged from 32 to 37 weeks among sites. Respiratory support at the time of discontinuation was common, but variable, with 0-57% of infants receiving positive airway pressure at caffeine discontinuation by site. Infants who discontinued caffeine within the last week of hospitalization had longer total duration of caffeine, but were discharged from the hospital at an earlier postmenstrual age.
Conclusion: There was substantial variability among sites in the timing of caffeine discontinuation before discharge and respiratory support at the time of caffeine discontinuation.