Time of Initiation of Pumping vs Breastfeeding at Discharge of Preterm Infants

Background: The AAP policy statement on breastfeeding recommends that all preterm infants receive breast milk (BM) feeds. However, mothers delivering prematurely often have difficulty establishing full milk supply. Early BM expression within 1-6 hours of delivery is associated with increased BM production. Baseline data at our large, urban, county hospital revealed that, among all mothers with infants admitted to the NICU, median time to first BM expression was 18 hours in September 2019. In some NICUs, introduction of donor BM led to decreased maternal BM production. Donor BM (DBM) was introduced in our NICU in January 2020. We aimed to (1) assess what characteristics are associated with longer time to first BM expression and (2) implement QI interventions to decrease time to first BM expression, with the goal of helping mothers establish supply to provide BM for their preterm baby beyond NICU discharge.

Methods: Retrospective cohort study of deliveries at <33 weeks gestational age (GA) or birthweight <1500 grams between March 2019 and February 2021. A QI project aiming at encouraging NICU mothers to express BM as soon as possible after delivery was initiated in September 2019. Interventions included education of personnel and mothers, increased number of breast pumps, and earlier in-person contact between mothers and lactation consultants.

Results: Among 512 neonates, infants 137 were born in Epoch-1 and 375 in Epoch-2; time to first BM expression was available in 448 mothers. Demographics were similar among both Epochs (Table 1a). Average time to first BM expression did not change significantly over time (Figure 1a); however, the range and standard deviation decreased transiently in Epoch-2 and increased again when DBM was introduced and during the COVID-19 pandemic (Figure 1b). In multivariate analysis (n=443), time to first BM expression increased with increasing GA, severe preeclampsia and cesarean delivery but did not decrease in Epoch-2 (Table 1b) and was not affected by DBM. Among women who delivered by cesarean section, time to first BM expression (n=291) was not affected by general anesthesia. The odds of receiving BM feeds at NICU discharge (n=424) increased with increasing GA and was lower with multiple gestations and lower in African-American and non-Hispanic White than in Hispanic White neonates (Table 1c); it was not affected by time of first BM expression, administration of DBM, or epoch.

Conclusion: Time to first BM expression increased with increasing GA, cesarean section and preeclampsia. The odds of BM feeding at time of NICU discharge increased with increasing GA, was lower with multiple gestation and highest in Hispanics but was not associated with earlier BM expression or use of DBM. The data suggests that limited access of lactation consultants to mothers during the COVID-19 pandemic negated the initial benefits observed during QI initiation.