Sleep in subjects with autistic disorder: a neurophysiological and psychological study

Abstract

Polysomnography (EOG, EEG, EMG) was carried out in 17 male children and adolescents with autistic disorder, in seven patients with mental retardation and fragile X syndrome, and in five age- and sex-matched normal male subjects. Density of rapid eye movements was not significantly different in the three groups of subjects; however, some sleep parameters such as time in bed, sleep period time, and total sleep time were significantly lower in subjects with autistic disorder than in normal controls; moreover, patients with autistic disorder showed values of sleep period time, first REM latency and percent (%) sleep stage 1 lower than those of patients with fragile X syndrome with mental retardation. Density of muscle twitches was significantly higher in patients with autistic disorder than in normal controls. In contrast only minor differences were observed between patients with autistic disorder and those with fragile X syndrome with mental retardation. Furthermore, some psychoeducational profile-revised items such as perception and eye-hand coordination, showed significant correlation with some sleep parameters (time in bed, sleep latency, stage shifts, first REM latency and wakefulness after sleep onset). Childhood Autism Rating Scale (CARS) scores to visual response and non-verbal communication showed significant correlation with some tonic sleep parameters, such as sleep period time, wakefulness after sleep onset, and total sleep time. Relating to people and activity level items were found to be significantly correlated with rapid eye movement density. Our results suggest the existence of a sleep pattern in autistic patients different from that observed in subjects with mental retardation and from that of normal controls. In addition, these findings indicate that sleep parameters in these patients are correlated with some psychological indices generally used for the diagnosis of autistic disorder; for this reason, polysomnographies might be useful in the comprehension of the neurophysiological mechanisms underlying this condition.

Introduction

Polysomnographic recording is a reliable non-invasive neurophysiological tool for the study of basic pathophysiological mechanisms of mental retardation (MR) and other developmental psychiatric disorders.

Jackson's sleep-cognition hypothesis [1] postulates that cognitive deficits might be associated with alterations in sleep mechanisms and structure. On this basis, several studies were carried out in the past on sleep patterns of subjects with MR of different etiology [2], [3], [4], [5], [6]. The main findings of such studies were: a reduction in percentage of rapid eye movement (REM) sleep and a prolonged latency of the first REM period in MR subjects.

Taking into account the fact that, in animals, REM sleep percentage increases after intensive learning sessions [7], [8] the hypothesis that this sleep stage is involved in cognitive processes was postulated and REM sleep itself was indicated as an index of brain ‘plasticity’, i.e. the ability of the brain to retain information.

As for the modulation of REMs, it was observed that MR patients also show a significant reduction in REMs occurring in shorter intervals (<1 s), as opposed to those separated by longer time intervals (>2 s) [6]. High-frequency REMs show a marked increase with age [9] and also increase after training [10], in normal subjects. For these reasons, they were considered as an index of the brain ‘organizational’ abilities, i.e. the ability to organize information from a random pool of elements into long-term memory. Also, some pharmacological attempts to modify sleep patterns were carried out in subjects with MR in order to ameliorate sleep-related psycho-behavioral functions [11], [12].

For these reasons, we decided to test for the existence of an eventual correlation between tonic and phasic parameters of sleep and psychological characteristics also in a different group composed of children with autistic disorder (AD). Thus, our aims were: (1) to quantify the organization of REMs; (2) to analyze the density of another phasic phenomenon of sleep (muscle twitches) during the different stages; and (3) to evaluate the possible correlation between neurophysiological and psychological data in these patients.