279 - Little Stimulus: Unpacking Caffeine Use in Late Preterm Infants

Publication Number: 279.5722

Kristin Weimer, Duke University School of Medicine, Durham, NC, United States; Lakshmi Katakam, Duke, Durham, NC, United States; Kevin G. Williams, Duke University School of Medicine, Durham, NC, United States; Keyaria Gray, Duke University School of Medicine, Durham, NC, United States; Kamlesh Athavale, Duke University, Durham, NC, United States; Simone Schneider, Duke University School of Medicine, Durham, NC, United States; Kwai Tei Chan Poon, Duke University School of Medicine, Durham, NC, United States; Veeral Tolia, Pediatrix, Dallas, TX, United States; Rachel G. Greenberg, Duke Clinical Research Institute, Durham, NC, United States; Samia Aleem, Duke University School of Medicine, Durham, NC, United States

Background: Caffeine citrate is a methylxanthine used to treat and prevent apnea of prematurity (AOP) and is the most commonly prescribed non-antimicrobial drug in neonatal intensive care units (NICUs) in the United States (U.S.). Late preterm infants (LPI), born at 340/7 - 366/7 weeks gestation, are an increasing population in the NICU and are at increased risk for morbidity and mortality compared to term infants. Although caffeine is approved by the U.S. Food and Drug Administration for treatment of AOP in infants < 33 weeks, clinically its use has extended to LPI. Little is known about the prevalence, efficacy, and dosing of caffeine in this population.

Objective: Characterize the use of caffeine in LPIs in U.S. NICUs over time

Design/Methods: We performed a cohort study of infants born at 34 0/7 - 36 6/7 weeks gestation at Pediatrix Medical Group NICUs between 2011-2022. We excluded infants with major congenital anomalies, those exposed to prostaglandins or transferred prior to discharge. We divided infants into 2 epochs (2011-2016 and 2017-2022) based on year of hospital discharge. We used chi-square tests to compare demographics and outcomes of LPIs by caffeine exposure and by epoch, including respiratory support, length of stay (LOS) and death. We examined timing of caffeine administration over time and by site.

Results: 212,348 LPIs from 395 NICUs met inclusion criteria, 4% (n=8632) were exposed to caffeine. LPIs exposed to caffeine had a lower gestational age, were more likely to receive surfactant, had higher and longer respiratory support needs and longer LOS (p < 0.001, Table 1). There was no difference in mortality between the two groups (Table 1). Caffeine use in LPIs decreased from epoch 1 to 2 (Table 2) , and there was variation in use by site (0-34%) (Figure 1). Compared to epoch 1, LPIs in epoch 2 were exposed to caffeine at an earlier postnatal age (2 vs. 3 days), for a shorter duration (1 vs 6 days) and had caffeine discontinued at an earlier postmenstrual age (35 vs 35.4 weeks) (all p< 0.001, Table 2).

Conclusion(s): LPIs exposed to caffeine had higher and longer respiratory support needs than unexposed LPI, likely reflecting the indications for use, rather than treatment effect. The longer LOS seen in caffeine exposed LPIs may be related to clinical illness or a desire to monitor infants off caffeine prior to discharge. Although caffeine use was relatively stable over time, prescribing practices varied by site and changed significantly over time, including earlier PNA at initiation and increased use of single dosing. These changes highlight areas for future study.