Anaesthesia is neurotoxic in developing primates. Retrospective clinical studies show a correlation between exposure to anaesthesia during infancy and the occurrence of learning disorders (LD). Prospective studies failed to detect any influence of a single exposure to anaesthesia on neurodevelopment. We hypothesised that some specific populations of children were electively sensitive to anaesthesia-related neurotoxicity.

Using a case-control design, we analysed the medical histories of children with LD, compared to those of their normally reading siblings. Interviews were conducted and medical records were reviewed. The numbers of hospitalisations and anaesthesia exposures before the age of five years were determined.

Four hundred fourteen dyslexic children were screened over a one-year period. Two hundred and seventy patients were excluded due to confounding variables (single child, all siblings showing LD or any condition placing the neurological prognosis at risk (N = 107/414 for the latter)) or inability to accurately collect evaluation criteria. In the 144 case-control pairs studied, the mean number of hospitalisations was significantly different (N = 1.097 ± 0 .135/case versus 0.667 ± 0.097/control, p = 0.0052), as was the proportion of hospitalised patients (54.2% versus 38.9%, p = 0.0031). The mean number of anaesthesia exposures per individual was not statistically different (N = 0.958 ± 0.183/case versus 0.569 ± 0.107/control, p =  0.0732), but the proportion of children anaesthetised at least once was (43.8% (cases) versus 33.3% (controls), p =  0.0301).

One or more hospitalisation(s) may reflect a health status and/or have an iatrogenic effect disrupting the normal setting up of learning abilities. Anaesthesia may play a role, but a correlation between LD and anaesthesia is of a lower magnitude than between LD and hospitalisation.