Longitudinal associations between mismatch negativity and disability in early schizophrenia- and affective-spectrum disorders

Background: Impaired mismatch negativity (MMN) is a robust finding in schizophrenia and, more recently, similar impairments have been reported in other psychotic- and affective-disorders (including at early stages of illness). Although cross-sectional studies have been numerous, there are few longitudinal studies that have explored the predictive value of this event-related potential in relation to clinical/functional outcomes. This study assessed changes in MMN (and the concomitant P3a) amplitude over time and aimed to determine the longitudinal relationship between MMN/P3a and functional outcomes in patients recruited during the early stage of a schizophrenia- or affective-spectrum disorder. Methods: Sixty young patients with schizophrenia- and affective-spectrum disorders and 30 healthy controls underwent clinical, neuropsychological and neurophysiological assessment at baseline. Thirty-one patients returned for clinical and neuropsychological follow-up 12-30. months later, with 28 of these patients also repeating neurophysiological assessment. On both occasions, MMN/P3a was elicited using a two-tone passive auditory paradigm with duration deviants. Results: Compared with controls, patients showed significantly impaired temporal MMN amplitudes and trend-level deficits in central MMN/P3a amplitudes at baseline. There were no significant differences for MMN measures between the diagnostic groups, whilst the schizophrenia-spectrum group showed reduced P3a amplitudes compared to those with affective-spectrum disorders. For those patients who returned for follow-up, reduced temporal MMN amplitude at baseline was significantly associated with greater levels of occupational disability, and showed trend-level associations with general and social disability at follow-up. Paired t-tests revealed that MMN amplitudes recorded at the central-midline site were significantly reduced in patients over time. Interestingly, those patients who did not return for follow-up showed reduced frontal MMN and fronto-central P3a amplitudes compared to their peers who did return for repeat assessment. Conclusions: This study provides some evidence of the predictive utility of MMN at the early stages of schizophrenia- and affective-spectrum disorders and demonstrated that MMN impairments in such patients may worsen over time. Specifically, we found that young patients with the most impaired MMN amplitudes at baseline showed the most severe levels of disability at follow-up. Furthermore, in the subset of patients with repeat neurophysiological testing, central MMN was further impaired suggestive of neurodegenerative effects. MMN may serve as a neurophysiological biomarker to more accurately predict functional outcomes and prognosis, particularly at the early stages of illness.