Do high or low doses of anxiolytics and hypnotics affect mismatch negativity in schizophrenic subjects? An EEG and MEG study

Abstract

Objective: Many studies have demonstrated mismatch negativity (MMN) attenuation in schizophrenia. Recently, investigators have shown that GABAergic inhibitory neurons may regulate MMN generation. Considering that a substantial proportion of schizophrenic patients receive anxiolytics and hypnotics that have affinity to GABA_A receptors to reduce their comorbid symptoms of anxiety and sleep disturbances, we need to assess whether anxiolytics/hypnotics might affect their MMN generation. The aim of this study is to assess the possibility that high or low doses of anxiolytics/hypnotics received by schizophrenic subjects affect their mismatch negativity (MMN), using event-related potentials (ERPs) and magnetoencephalography (MEG).

Methods: Twenty-three and 16 patients with schizophrenia participated in the ERP and MEG studies, respectively. Three types of MMN (MMN in response to a duration change of pure-tone stimuli, within-category vowel change (Japanese vowel /a/ with short versus long duration), and across-category vowel change (vowel /a/ versus /o/)) were recorded.

Results: High or low doses of benzodiazepine had no significant effects on MMN amplitude/magnetic MMN power, topography/laterality, or latency under any conditions of the ERP or MEG study.

Conclusions: These results suggest that chronic administration of anxiolytics/hypnotics does not significantly affect MMN in schizophrenia.

Introduction

Since the early clinical description by Bleuler (1911), it has been recognized that attentional deficits are one of the primary dysfunctions in schizophrenic psychopathology. The advent of event-related brain potential (ERP) studies has enabled an electrophysiological assessment of attentional dysfunction in schizophrenia. Earlier investigations focused on the evaluation of conscious, controlled aspects of attention in schizophrenia, and thus abnormalities of P300 (Roth et al., 1981, Kutcher et al., 1987, McCarley et al., 1989), N2b (Brecher et al., 1987), and the processing negativity (Michie et al., 1990) component of ERPs have been repeatedly reported. However, in 1991 Shelley et al. (1991) reported on impaired mismatch negativity (MMN) in response to a duration change of tones, an ERP index of auditory preattentive processing (Näätänen et al., 1978) in schizophrenia. Subsequent studies have replicated a reduction in MMN amplitude in response to a duration (Catts et al., 1995, Kasai et al., 1999c, Javitt et al., 2000b, Michie et al., 2000, Todd et al., 2000) and frequency (Javitt et al., 1993, Javitt et al., 1995, Javitt et al., 1998, Javitt et al., 2000a, Javitt et al., 2000b, Shutara et al., 1996, Alain et al., 1998, Hirayasu et al., 1998, Umbricht et al., 1998, Shelley et al., 1999) changes of tones in schizophrenia, although others did not find a significant reduction in the MMN in response to a frequency change of tones (O'Donnell et al., 1994, Kathmann et al., 1995, Oades et al., 1996, Kirino and Inoue, 1999). Recently, magnetoencephalography (MEG) (Kreitschmann-Andermahr et al., 1999) and functional magnetic resonance imaging (Wible et al., 2001) investigations have also demonstrated abnormalities of auditory mismatch detection in response to frequency changes of tones in schizophrenia.

Regarding the effects of psychotropic drugs on the MMN in schizophrenia, several reports have described no significant correlation between reported MMN amplitudes and the dose of neuroleptics administered to patients with chronic schizophrenia (Shelley et al., 1991, Shutara et al., 1996, Kasai et al., 1999c, Michie et al., 2000, Todd et al., 2000). Catts et al. (1995) and Javitt et al. (1995) independently found no significant difference in MMN amplitude between medicated and neuroleptic-free patients. Umbricht et al., 1998, Umbricht et al., 1999 showed that MMN reduction was not ameliorated by either typical (haloperidol) or atypical (clozapine or risperidone) medication. These reports provide a consensus that neuroleptics have no effect on MMN amplitude in schizophrenia. On the other hand, there is evidence from primate and human studies indicating that the interplay between excitatory glutamatergic and inhibitory GABAergic neurons regulates MMN generation (Javitt et al., 1996, Umbricht et al., 2000). Supporting this hypothesis, Nakagome et al. (1998) reported MMN amplitude reduction in the morning following administration of triazolam in normal subjects. Furthermore, Smolnik et al. (1998) observed a tendency towards an increase in MMN amplitude after the administration of the benzodiazepine antagonist, flumazenil, in normal subjects although the effect was not statistically significant. Considering that a substantial proportion of patients with schizophrenia receive anxiolytics and hypnotics to reduce the comorbid symptoms of anxiety and sleep disturbances, respectively (Carpenter et al., 1999, Hardy et al., 1999, Ito et al., 1999, Voirol et al., 1999), and that these drugs have affinity to GABA_A receptors (Schroder et al., 1997, Fujita et al., 1999), we need to assess the possibility that anxiolytics/hypnotics might have some effect on MMN amplitude in schizophrenic patients. However, to our knowledge, only one preliminary study from our group (Murakami et al., 2001) has evaluated the possible effects of anxiolytics/hypnotics on the MMN in schizophrenic patients. Using a 16-channel montage, Murakami et al. compared the MMN and N2b in response to pure-tone stimuli between two subgroups of schizophrenic patients, one comprised of patients who did not receive benzodiazepines (benzodiazepine-off group; n=7) and the other of those who were administered benzodiazepines during daytime (benzodiazepine-on group; n=7). They found no significant differences in MMN and N2b amplitudes between the two subgroups, whereas the N2b latency was significantly prolonged in the benzodiazepine-on group relative to the benzodiazepine-off group. These results suggested that benzodiazepine had no marked effects on the MMN amplitude/latency in schizophrenia. The present study is an extension of our preliminary study; here, we determine whether the dose of anxiolytics/hypnotics affects MMN (or its magnetic counterpart, MMNm) amplitude (or MMNm power)/latency/topography (or laterality) in response to pure-tone and speech sounds in two new groups of patients with schizophrenia, using ERP and MEG recordings, respectively.