Objective: To examine whether hypercapnia in very low birth weight (VLBW) infants during the first 3 days of life is associated with severe intraventricular hemorrhage (IVH).
Study design: Retrospective cohort study of inborn VLBW infants between January 1999 and May 2004 with arterial access during the first 3 days of life. A multiple logistic regression analysis was used where IVH was dichotomized ((grades 0/1/2) = non-severe; (grades 3/4) = severe). Measures of hypercapnia were entered into the model to ascertain their association with severe IVH.
Results: In total, 574 VLBW infants met entry criteria. Worst IVH grade was 0 in 400; 1: 54; 2: 42; 3: 47; and 4: 31 infants. The logistic regression model consisted of the following predictors of severe IVH: gestational age, gender, 1 min Apgar score (dichotomized into two groups: >3 vs < or =3), multifetal gestation, vasopressor use, and maximum PaCO(2).
Conclusion: In addition to traditional risk factors, it appears maximum PaCO(2) is a dose-dependent predictor of severe IVH during the permissive hypercapnia era.