Prevalence and Neonatal Factors Associated with ASD in Preterm Infants

Background: Preterm infants have a higher rate of positive results on screening tests for autistic features compared to term infants.  However, the prevalence of confirmed diagnoses later in childhood and the neonatal factors that may affect prevalence are unclear.

Objectives: To determine the association between gestational age and confirmed cases of Autism Spectrum Disorders (ASD) and assess risk factors in the neonatal period.

Methods: The study population included all live births occurring at Northern California Kaiser Permanente Medical Care Program hospitals between 2000-2007 who remained in the healthcare system at two years of age (n=177,549).  Infants with major congenital anomalies or hypoxic ischemic encephalopathy were excluded.  Definite ASD cases were defined as children evaluated and diagnosed with an ASD at a Kaiser Permanente Autism Center through August 2011. We assessed the association between gestational age and other risk factors and a diagnosis of ASD using Cox proportional hazards regressions to account for differential time of follow-up, adjusting for sex, maternal and paternal age, maternal race/ethnicity, maternal education, Cesarean section, and multiple gestation as well as clustering by mother.

Results: The point prevalence of definite ASD cases (N=1286) was 0.72% in the entire study population, 1.08% (147/13,639) in preterm infants (<37 weeks gestation) and 0.69% (1,139/163,910) in term infants (≥ 37 weeks gestation).  Among infants with an ASD, the percentage with a diagnosis of Autistic Disorder (AD) was similar in both groups (65% in preterm, 67% in term).  Compared to term infants, infants 34-36 weeks (Hazard Ratio (HR) 1.32, 95% CI 1.06-1.63, P=0.01), 27-33 weeks (HR=1.34, 95% CI 0.91-1.98, P=0.14), and 24-26 weeks (HR=2.90, 95% CI 1.43-5.86, P=0.003) were at an increased risk of ASD, although statistical significance was not reached in the 27-33 week group.  A 5-minute Apgar score of <6, being small or large for gestational age, bacteremia, inotropic support, necrotizing enterocolitis, surfactant administration, blood transfusion, intraventricular hemorrhage grade >2, cystic periventricular leukomalacia, and a discharge diagnosis of maternal chorioamnionitis were not independent risk factors for ASD in preterm infants. There was a trend toward those who received mechanical ventilation being at higher risk for ASD (HR=1.72 95% CI 0.94-3.17, P=0.08).

Conclusions: Preterm delivery is a risk factor for the diagnosis of ASD later in childhood; however, the magnitude of the increased risk was substantially lower than the 4- to 5-fold increase in positive screening prevalence reported previously.  This suggests that a significant percentage of the positive screens in preterm infants may be due to developmental impairments related to their prematurity rather than to ASD.  No specific diagnoses or interventions in the neonatal period were associated with an increased risk of a diagnosis of ASD.  While infants 24-26 weeks were at a significantly increased risk, this finding should be interpreted with caution given the small number of infants (n=222), 6 with an ASD diagnosis.