Reduced mismatch negativity predates the onset of psychosis Madiha Shaikh ⁎, Lucia Valmaggia, Matthew R. Broome, Anirban Dutt, Julia Lappin, Fern Day, James Woolley, Paul Tabraham, Muriel Walshe, Louise Johns, Paolo Fusar-Poli, Oliver Howes, Robin M. Murray, Philip McGuire, Elvira Bramon PO63, NIHR Biomedical Research Centre for Mental Health at the Institute of Psychiatry, King's College London and The South London and Maudsley NHS Foundation Trust, London, SE5 8AF, UK abstract article info Article history: Received 21 June 2011 Received in revised form 7 September 2011 Accepted 18 September 2011 Available online 24 October 2011 Keywords: MMN Prodromal Biomarker Psychosis ARMS EEG Background: Individuals with an “At Risk Mental State” have a 20–30% chance of developing a psychotic dis- order within two years; however it is difficult to predict which individuals will become ill on the basis of their clinical symptoms alone. We examined whether mismatch negativity (MMN) could help to identify those who are particularly likely to make a transition to psychosis. Method: 41 cases meeting PACE criteria for the At Risk Mental State (ARMS) and 50 controls performed a duration-deviant passive auditory oddball task whilst their electroencephalogram was recorded. The ampli- tude of the MMN wave was compared between groups using linear regression. The ARMS subjects were then followed for 2 years to determine their clinical outcome. Results: The MMN amplitude was significantly reduced in the ARMS group compared to controls. Of the at- risk subjects who completed followed up (n = 41), ten (24% of baseline sample) subsequently developed psychosis. The MMN amplitude in this subgroup was significantly smaller across all three recording sites (FZ, F3 and F4) than in the ARMS individuals who did not become psychotic. Conclusion: Among those with the ARMS, MMN amplitude reduction is associated with an increased likelihood of developing frank psychosis. © 2011 Elsevier B.V. All rights reserved. 1. Introduction The prodrome is a critical stage in the development of psychosis in which initial symptoms first arise. The basis of the vulnerability to psychosis in this group is not clearly known, and whilst evidence shows that psychological and pharmacological interventions may im- prove outcomes (Bechdolf et al., 2005; McGorry et al., 2002; Morrison et al., 2007; Politi et al., n.d.; Yung et al., n.d.), treating high-risk subjects remains controversial since most of them are not destined to develop psychosis (Simon et al., n.d.). Indicated prevention is currently regarded as the most promising strategy to attenuate, delay, or even avert psychosis. Studies have found significant specificity regarding prediction of conversion using clinical and other variables (Cannon et al., 2008; Velthorst et al., 2009), but only at the cost of sensitivity and vice versa. However a recent study by Ruhrmann et al (2010) introduced a risk modelling procedure with the use of prognos- tic indexes for a multivariate clinical staging approach to improve spec- ificity and individual risk assessment to allow for better targeted and earlier interventions. The multivariate clinical staging approach of at- risk symptoms further demands neurobiological markers for early detection, and researchers have tried to find trait markers using several imaging techniques (Bramon et al., 2008; Fusar-Poli et al., 2011a; Howes et al., 2007; Pantelis et al., 2003). Indeed, the better characterisa- tion of the prodrome of psychosis by means of biological investigations is a prerequisite for both prevention and intervention (Broome et al., 2005b; Ruhrmann et al.; Ruhrmann et al., 2010; Yung et al., 2003; Yung et al., 2004). Direct in-vivo measures of cortical activity observed in the human EEG are useful markers of brain dysfunction in psychosis (Schulze et al., 2007, 2008; Shaikh et al., 2010; Turetsky et al., 2007) and provide high temporal resolution (Bramon et al., 2005). The MMN, in particu- lar, is a change in the EEG (an event related potential) that emerges when a novel stimulus infringes the regularity of the preceding ones. Auditory MMN can be elicited by different qualitative and quan- titative types of deviance, including intensity, location, frequency, duration or pattern of auditory stimuli. The deviant type as well as the attentional condition may have substantial effects on the stability and replicability of MMN potentials. Duration deviants, similar to the ones used in this study, generally produce MMN amplitudes of better intraindividual stability and a task demanding focused attention minimizes measurement error (Kathmann et al., 1999). The MMN waveform is an early phenomenon, thought to reflect cortical activation in basic, pre-attentive stages of auditory information pro- cessing and perception (Näätänen et al., 1989). It is thought that the Schizophrenia Research 134 (2012) 42–48 ⁎ Corresponding author at: Tel.:+44 207 848 0541; fax: +44 207 848 0287. E-mail address: madiha.1.shaikh@kcl.ac.uk (M. Shaikh). 0920-9964/$ – see front matter © 2011 Elsevier B.V. All rights reserved. doi:10.1016/j.schres.2011.09.022 Contents lists available at SciVerse ScienceDirect Schizophrenia Research journal homepage: www.elsevier.com/locate/schres