Relationships Between Early Nutrition and Blood Glucose Concentrations in Very Preterm Infants

Objectives: The aim of the study was to determine whether changes to early nutrition are associated with levels of glycemia in very preterm infants. Methods: A retrospective, observational study of infants <1500 g or <30 weeks' gestation admitted to Neonatal Intensive Care, National Women's Hospital, New Zealand, before (Old Protocol) and after (New Protocol) a change in nutritional protocol. Nutritional intakes were calculated and averaged by day for postnatal days 1 to 7 (week 1) and 1 to 28 (month 1). Relationships between glycemia measures, macronutrient intakes, and achievement of 10% enteral feeds (>= 10% total intake) were explored using logistic regression. Results: Old Protocol (n = 190) andNew Protocol (n = 267) groups had similar baseline characteristics. In week 1, New Protocol infants receivedmore protein, less fat, and carbohydrate, had lower mean blood glucose concentrations (BGCs) (mean +/- SD 4.9 +/- 1.2 vs 5.6 +/- 1.4mmoll/L, P < 0.0001), less hyperglycemia (BGC> 8.5mmol/L, 71 [27%] vs 80 [42%], P = 0.0005), but similar hypoglycemia (BGC<2.6). In month 1, New Protocol infants also had less hyperglycemia (105 [39%] vs 96 [51%], P = 0.02) and lower mean BGC (5.0 +/- 1.1 vs 5.5 +/- 1.1 mmol/L, P < 0.0001), but insulin usage was similar. After adjustment for birth weight z score and gestational age, hyperglycemia was significantly associated with week 1 intakes (g.kg(-1) . day(-1)) of protein (odds ratio [95% confidence intervals] 0.47 [0.23-0.79], P = 0.004), fat (0.54 [0.40-0.74], P < 0.0001), and carbohydrate (1.25 [1.09-1.44], P < 0.0001). These relationships were similar for month 1. Each additional day to achieve 10% enteral feeds was associated with increased odds of hypoglycemia (1.09 [1.00-1.18], P = 0.04) and hyperglycemia (1.16 [1.06-1.28], P = 0.002). Conclusions: In very preterm infants, macronutrient balance and small, early enteral feeds may assist glycemic control.