Standardized Perioperative Feeding Protocol Improves Outcomes in Patients With d-Transposition of the Great Arteries Undergoing Arterial Switch Operation

Objective: To determine whether the presence of a standardized feeding protocol improves outcomes in a subset of neonates undergoing cardiac surgery. Design: Retrospective cohort study. Setting: Cardiovascular ICU at a freestanding academic children's hospital. Patients: Neonates with a diagnosis of d-transposition of the great arteries undergoing arterial switch operation from January 2007 to June 2017. Interventions: Initiation of perioperative feeding protocols. Measurements and Main Results: Patients were evaluated before and after implementation of standardized perioperative feeding protocols in neonates with d-transposition of the great arteries undergoing arterial switch operation. Low-risk patients born after initiation of nurse-driven protocols were compared with a similar historical group. Data obtained included time to achievement of feeding goals, with primary outcome being weight gain at hospital discharge. Other measures analyzed included duration of mechanical ventilation and postoperative hospital length of stay. Overall, 33 patients in the protocol group were compared with 44 patients in the historical group. No significant baseline differences existed between the two cohorts. The protocol group achieved improved feeding outcomes in nearly all measured categories, including introduction to enteral feeds preoperatively (91% vs 59%; p < 0.01) and earlier attainment of postoperative full enteral feeds of 120mL/kg/d (2 vs 5 d; p < 0.01). Protocol patients had significantly improved weight gain at the time of discharge (60 vs 1g; p < 0.01), while achieving shorter postoperative length of stay (10.1 vs 12.6 d; p = 0.04). Conclusions: An aggressive, but safe, perioperative feeding protocol implemented in a homogenous low-risk neonatal cardiac surgical population improves feeding outcomes, including increased weight gain, as well as decreased postoperative length of stay. Consideration for perioperative feeding protocol implementation and further study should be given.