Abnormalities in pain sensitivity among individuals with autism spectrum disorder: Evidence from meta-analysis

Pain is defined as an unpleasant sensory and emotional experience associated with, or resembling that associated with, actual or potential tissue damage. Pain is of vital functional significance, as it signals threat and initiates behavioral adaptations to avoid harm so as to protect the body. Previous studies have shown abnormalities in pain sensitivity among individuals with autism spectrum disorder (ASD), which have been associated with their clinical core symptoms, including restricted and repetitive behaviors as well as deficits in social behaviors. Evidence from case studies and surveys suggested the hyposensitivity to pain for individuals with ASD. Nevertheless, results from experimental studies that involved the application of noxious stimulations and psychophysical measurements were heterogeneous, e.g., some studies reported hypersensitivity to pain in ASD individuals, others reported their hyposensitivity or even normal sensitivity to pain. In this study, we utilized a meta-analysis approach to systematically review experimental studies that investigated pain sensitivity among individuals with ASD and were published before August 10, 2020. The meta-analysis was performed according to the rigorous PRISMA Protocol. Studies were included in the analysis if they included both clearly diagnosed ASD individuals and healthy controls, reported data relevant to pain sensitivity, including pain threshold, pain tolerance, pain ratings, and pain-evoked physiological responses. Relevant studies were obtained from databases including China National Knowledge Infrastructure, Web of Science, PsycInfo, and PubMed by searching for keywords including pain, nociception, autis*, and Asperger. The meta-analysis was conducted in STATA12, and the effect size Hedge's g with +/- 95% confidence intervals (CIs) was calculated using a random effect statistical model. Further, we assessed possible moderating effects from variables of pain modality, pain site, the age of involved participants, the sample size of the ASD group and sample locations. Sixteen experimental studies were included in the meta-analysis, with a total sample size N = 822. Pain threshold was not significantly different between ASD individuals and healthy controls (g = 0.34, 95% CI = [-0.14, 0.82]), and this estimate was moderated by variables of pain modality, the age of involved participants, and the sample size of the ASD group. Specifically, individuals with ASD exhibited lower pain thresholds than those of healthy controls selectively for pressure pain (g = 1.62, 95% CI = [0.46, 2.77]). For the outcome variable of pain evoked physiological response, individuals with ASD showed significantly greater physiological responses to medical procedures than those of healthy controls (g = 2.87, 95% CI = [1.07, 4.67]). Nevertheless, ASD and control groups had comparable pain ratings (g = -0.26, 95% CI = [-0.64, 0.11]). These results suggest that the abnormalities of pain sensitivity among individuals with ASD were modality-dependent, with the abnormality selectively applicable to pressure pain or medical pain. Future studies should combine behavioral, physiological, and neuroimaging measures to comprehensively investigate the pain sensitivity profiles of individuals with ASD. The potential link between pain sensitivity and clinical core symptoms among individuals with ASD should be characterized. Relevant results would potentially expand our understanding of ASD neurobiological mechanisms and provide the theoretical basis for pain assessment among individuals with ASD.