Background: Frequency of exposure to antenatal ultrasound (U/S) has increased since it’s introduction in the 1970’s, and U/S protocols and techniques have also changed, leading to concerns that antenatal U/S exposure may contribute to autism spectrum disorders (ASDs). Although prior studies have not evaluated an association with ASDs, some studies have suggested an association with dyslexia or left-handedness.

Objectives: To evaluate antenatal B-mode U/S exposure and risk of ASDs in a case-control study within a large integrated health care delivery system.

Methods: Singleton children enrolled in Kaiser Permanente (KP) and born 1995-1999 were included.  Cases were identified from diagnostic codes and controls were randomly sampled within gender and birth month strata. Primary analyses were limited to simplex cases (N=361) and controls without siblings with ASD (N=393).  Information on U/S scans and other clinical data were obtained from multiple electronic files and medical records and demographic data were obtained from birth certificates. Chi-square tests and multivariable logistic regression models were used to evaluate case-control differences.

Results: Demographic characteristics of cases were similar to those of controls, but cases had slightly lower gestational age at birth (p=0.054) and higher maternal education (p<0.001). Thirteen percent of both cases and controls had no U/S scans, 29% of cases and 28% of controls were exposed in the 1^st trimester, 78% of cases and 79% of controls in the 2^nd trimester, and 28% of both cases and controls in the 3^rd trimester. Cases were statistically similar to controls in adjusted analyses with regard to the number and type (complete OB, limited OB, transvaginal) of U/S scans performed throughout pregnancy and for each trimester. 

Conclusions: Children with autism and control children born 1995-1999 were similar with regard to antennal U/S exposure, indicating that fetal exposure to B-mode ultrasound is unlikely to increase the risk of ASD.