Hyperglycemia and retinopathy of prematurity in very low birth weight infants

Objective: Retinopathy of prematurity (ROP) remains a leading cause of morbidity in the very low-birth-weight (VLBW) infant. This study investigates a possible association between serum/blood glucose and the development of ROP. Methods: A retrospective case - control study of all infants born between 1992 and 1997 at the Johns Hopkins Hospital with birth weights less than 1000 g who developed Stage 3 or 4 ROP was conducted. Controls either had Stage 1 ROP or no eye disease and were matched 2:1 with ROP patients for gestational age, birth weight and year of birth. Odds ratios (ORs) of ROP were calculated for multiple exposures over the first month after birth, including oxygen concentration (FiO2), blood glucose levels, vitamin E, mean airway pressure and mean blood pressure. Results: In a simple logistic regression analysis, we found an increased ROP risk for: (1) each 10 mg/dl increase of mean glucose (OR 1.96; 95% CI 1.13 to 3.42) (2) each 1% increase of mean FiO2 (OR 1.06; 95% CI 1.004 to 1.13), (3) history of dopamine infusion (OR 5.4; 95% CI 1.16 to 25.2) and (4) intraventricular hemorrhage Grade 3 or 4 (OR 7.3; 95% CI 1.53 to 34.7). Using a multiple regression model, we found an increased ROP risk for each 10 mg/dl increase of mean glucose (OR 2.7; 95% CI 1.003 to 7.27). Each IU/kg/day of vitamin E supplementation reduced ROP risk (OR 0.37; 95% CI 0.16-0.86). Conclusion: In this study, we could demonstrate that glucose levels in the first month of life are associated with the development of ROP. Further studies have to determine if this association is causal or if hyperglycemia is just an expression of severity of illness.