Ultrasound Assessment of Gastric Emptying in Premature Infants Treated With Non-Invasive Ventilatory Support

Background: Nasal continuous positive airway pressure (CPAP) introduces positive pressure of air into both the trachea and stomach, which may affect gastric emptying. The rate of gastric emptying can be estimated by ultrasound (US) in neonates by two validated techniques: "antral cross-sectional area" (ACSA, two-dimensional estimate of the surface area at the gastric antrum), and "spheroid gastric volume" (spheroid, three-dimensional estimate of the stomach volume). Objective: To compare gastric emptying rates in neonates on machine-derived nasal CPAP (MD-nCPAP, Avea and RAM cannula) with those on bubble CPAP (bCPAP, Fisher Paykel and Babi.Plus nasal prongs). Methods: Ultrasound measurements of the amount of the milk in the stomach were performed before feeding and at 1, 2, and 3 hours after the start of feeding, using both the ACSA and spheroid methods. Rates of gastric emptying were calculated during the "early" (1-2 hours) and "late" (23 hours) phases after feeding. Results: We recruited 32 infants (25-34 weeks gestational age, full enteral tube feedings, on nasal CPAP). Seventeen infants were treated with MD-nCPAP (median birth weight 1015 g [interquartile range (IQR): 870-1300], gestational age 28 weeks [IQR: 27-29], postnatal age 20 days [IQR: 1428]), whereas 15 infants were treated with bCPAP (median birth weight 960g [IQR: 855-1070], gestational age 27 weeks [IQR: 26-28], postnatal age 17 days [IQR: 15-25]). Gastric emptying rates (% emptied/min) were significantly faster in the "early" compared to the "late" phase for all infants. There were no significant differences in the rates of gastric emptying (either "early" or "late") or volumes of gastric residuals between infants receiving MD-nCPAP or bCPAP, measured by either method. Although no feeding intolerance was seen in either group, the volumes of residual gastric contents measured by both methods were higher than the volumes traditionally considered abnormal when obtained by gastric tube aspiration. Conclusions: Gastric emptying is faster during the "early" compared to the "late" phase. Gastric emptying rates are not different in infants receiving MD-nCPAP versus bCPAP. The presence of large residual gastric contents in infants who are tolerating feedings challenges the value of traditional gastric aspiration for the assessment of feeding tolerance in infants.