Risk factors for cerebral palsy in preterm infants.

Objective: To identify crucial factors that precipitate cerebral palsy by controlling confounding factors in logistic regression analyses.

Design And Patients: We retrospectively investigated a cohort of all 922 infants with gestational ages of less than 34 weeks (22-33 weeks), who were admitted to our neonatal intensive care unit between 1990 and 1998. Thirty (3.7%) were diagnosed to have cerebral palsy. We analyzed the prenatal and postnatal clinical variables of the cerebral palsy cases and compared them with 150 randomly selected controls.

Results: Risk factors for cerebral palsy identified in univariate analysis were: twin pregnancy, long-term ritodrine tocolysis, respiratory distress syndrome, air leak, surfactant administration, intermittent mandatory ventilation, high frequency oscillation, lowest PaCO2 levels, prolonged hypocarbia during the first 72 h of life, and postnatal steroid therapy. In a conditional multiple logistic model, long-term ritodrine tocolysis, prolonged hypocarbia and postnatal steroid therapy remained associated with an increased risk of cerebral palsy after adjustment for other antenatal and postnatal variables (OR [Odds Ratio] = 8.62, 95% CI [Confidence Interval], 2.18-33.97; OR = 7.81, 95% CI, 1.42-42.92; OR = 21.37, 95% CI, 2.01-227.29, respectively).

Conclusions: Our results suggest that long-term ritodrine tocolysis underlines the development of cerebral palsy. Further assessments of the effect of ritodrine on fetal circulation and nervous system are required. Moreover, possible alternatives to systemic postnatal steroids are needed, and carbon dioxide levels should be more strictly controlled.