Summary Preterm infants in the neonatal intensive care unit were given a soother or were breastfed during blood collection. No overall between-group differences on pain indices were found.

ABSTRACT Managing acute procedural pain effectively in preterm infants in the neonatal intensive care unit remains a significant problem. The objectives of this study were to evaluate the efficacy of breastfeeding for reducing pain and to determine if breastfeeding skills were altered after this treatment. Fifty-seven infants born at 30–36 weeks gestational age were randomized to be breastfed (BF) or to be given a soother during blood collection. Changes in the Behavioral Indicators of Infant Pain (BIIP) and in mean heart rate (HR) across 3 phases of blood collection were measured. In the BF group, the Premature Infant Breastfeeding Behaviors (PIBBS) scale was scored before and 24 hours after blood collection. Longitudinal regression analysis was used to compare changes in Lance/squeeze and Recovery phases of blood collection between groups, with gestational age at birth, baseline BIIP scores, and mean HR included as covariates. Differences in PIBBS scores were assessed using a paired t-test. Relationships between PIBBS scores, BIIP scores, and HR were evaluated with Pearson correlations. No differences between treatment groups were found: BIIP (P = 0.44, confidence interval [CI] −1.60–0.69); HR (P = 0.73, CI −7.0–10.0). Infants in the BF group showed improved PIBBS scores after the treatment (P < 0.01, CI −2.7 to −0.2). Lower BIIP scores during the Lance/squeeze were associated significantly with more mature sucking patterns (r = −0.39, P < 0.05). Breastfeeding during blood collection did not reduce pain indices or interfere with the acquisition of breastfeeding skills. Exploratory analyses indicate there may be benefit for infants with mature breastfeeding abilities.