19355
An Altered Olfactory Profile in Children Diagnosed with Autism Spectrum Disorder

Saturday, May 16, 2015: 1:45 PM
Grand Salon (Grand America Hotel)
L. Rozenkrantz1, I. Heller1, A. Plotkin1, A. Weissbrod1, D. Zachor2 and N. Sobel1, (1)Department of Neurobiology, Weizmann Institute of Science, Rehovot, Israel, (2)Autism Center, Asaf Harofeh Medical Center, Zerifin, Israel
Background:  Many anecdotal observations report altered responses to odors in children and adults with ASD. Indeed, when considering the neural substrates most notably implicated in autism, ranging from ventral temporal structures to the cerebellum, they are highly overlapping with the neural substrates of olfaction. The sniff-response is a phenomenon whereby nasal airflow parameters are rapidly modulated in accordance with odorant content. For example, pleasant odors are automatically sampled with strong sniffs, yet unpleasant odors are sampled with weak sniffs. Thus, by measuring nasal airflow (sniffs) alone we obtain a non-verbal measure of olfactory perception and processing. 

Objectives:  test the hypothesis that olfactory responses are altered in children with ASD, and that these altered responses may serve early diagnosis. 

Methods:  We developed a pediatric olfactometer that delivers pleasant and unpleasant odors and measures the concurrent sniff-response. The 10-minute procedure consisted of 20 trials (10 for each valence), each 1-2 seconds in duration, separated by a 30 second inter-trial-interval. During the task, participants watched a cartoon. 

Results:  Eighteen children with ASD (17 boys, mean age 7±2.3 y) and 18 typically developing (TD) age and gender-matched children (17 boys, mean age 6.7±2.1 y) completed the non-verbal sniff response test. A repeated-measures ANOVA revealed a significant interaction between sniff parameters (volume, mean, peak and duration), odorant valence (pleasant versus unpleasant) and group (TD versus ASD), reflecting significantly larger sniffs for pleasant versus unpleasant odors in the TD group but not the ASD group (F3,34=6.16, p<0.001). In addition, we found a strong correlation between the sniff-response and autism severity (R=-0.75, p<0.0005) in which longer sniff durations for unpleasant rather than pleasant odors predicted more severe autism.

Conclusions:  Taken together, these results imply a pronounced alteration of olfactory perception that is evident at the early stages of ASD, and is more pronounced with increased autism severity.