Linda Dodds1, Deshayne B. Fell2, Sarah Shea3, Alexander Allen2, B. Anthony Armson4, and Susan E. Bryson3. (1) Obstetrics & Gynecology and Pediatrics, Dalhousie University, Halifax, NS B3K 6R8, Canada, (2) Perinatal Epidemiology Research Unit, IWK Health Centre, Halifax, NS B3K 6R8, Canada, (3) Pediatrics, Dalhousie University, Halifax, NS B3K 6R8, Canada, (4) Obstetrics and Gynecology, Dalhousie University, Halifax, NS B3K 6R8, Canada
Background:
It remains unresolved whether obstetric or neonatal complications are independent factors in autism etiology, contribute via gene-environment interactions or reflect an
epiphenomenon, (i.e., familial factors predispose to obstetric complications and to autism).
Objectives: Our objective was to identify obstetric/neonatal factors associated with the subsequent development of autism and assess whether these contribute to its etiology.
Methods: Women who gave birth between 1988 and 2002 were identified from a population-based perinatal database in Nova Scotia, Canada. Diagnoses of autism among children of women in this cohort were identified from anonymous linkages to administrative databases with relevant diagnostic information from 1992 to 2005. Information on maternal conditions, prenatal factors, obstetrical and neonatal conditions was evaluated. Cox proportional hazards regression models were used to estimate adjusted relative risks and 95% confidence intervals. Analyses controlled for factors suggestive of a strong genetic etiology (e.g., having an affected sibling or maternal psychiatric/neurologic conditions).
Results: Among 129733 children born between 1988 and 2002, there were 924 children with an autism diagnosis between 1992 and 2005. After controlling for factors suggestive of a strong genetic etiology, the following obstetric/neonatal factors were significant: pre-pregnancy weight ³ 90 kg (RR=1.6, 95% CI 1.2-2.0), ³ 18-kilogram pregnancy weight gain (RR=1.2, 95% CI 1.0-1.4), <18-month inter-pregnancy interval (RR=1.5, 95% CI, 1.1-2.0), maternal prescription drug use during pregnancy (RR=1.7, 95%CI 1.0-2.8), induced labour (RR=1.2, 95% CI 1.0-1.5) or no labour (RR=1.3, 95% CI 1.0-1.7) (versus spontaneous labour), male infant sex (RR=4.3, 95% CI 3.5-5.2), 5-minute Apgar score ≤ 3 (RR=3.6, 95% CI 1.2-11.3), and presence of a major congenital anomaly (RR=7.2, 95% CI 3.6-14.5).
Conclusions: Some obstetrical and neonatal factors appear to have an independent role in autism etiology. The association between pre-pregnancy obesity and excessive weight gain during pregnancy and increased autism risk are novel findings that require further investigation.