Sensory Gating in Schizophrenia: P50 and N100 Gating in Antipsychotic-Free Subjects at Risk, First-Episode, and Chronic Patients Anke Brockhaus-Dumke, Frauke Schultze-Lutter, Ralf Mueller, Indira Tendolkar, Andreas Bechdolf, Ralf Pukrop, Joachim Klosterkoetter, and Stephan Ruhrmann Background: Abnormal sensory gating in schizophrenia has frequently been reported; however, only limited data on unmedicated patients and patients at risk to develop a psychosis have, as yet, been available. Methods: P50 and N100 suppression were assessed with an auditory double-click paradigm in five groups: 18 at-risk subjects who did not develop a full psychosis within the follow-up period of 2 years, 21 truly prodromal subjects who developed frank psychosis within the follow-up period, 46 antipsychotic-naïve subjects with first-episode schizophrenia, 20 antipsychotic-free subjects with chronic schizophre- nia, and 46 healthy control subjects. Results: P50 and N100 suppression indices differed significantly between groups and were lowest in chronic schizophrenia patients. Compared with healthy control subjects, P50 suppression was significantly impaired in at-risk subjects, truly prodromal and first-episode patients (stimulus 2 [S2]/stimulus 1 [S1] P50 amplitude ratio), and chronic schizophrenia patients (difference and ratio), and N100 suppres- sion was significantly reduced in truly prodromal and first-episode patients (S1–S2 difference) and in chronic schizophrenia patients (difference and ratio) but not at-risk subjects. At-risk subjects with and without conversion to psychosis did not significantly differ on any test parameter. Conclusions: Sensory gating is already impaired in early stages of schizophrenia, though this is most prominent in chronic stages. Future studies will have to clarify the type and impact of variables modifying sensory gating disturbances, such as illness progression and genetic load. Furthermore, the meaning and nature of differences between P50 and N100 suppression need further elucidation. Key Words: Early recognition, EEG, event-related potentials, first episode, prodrome, schizophrenia, sensory gating M idlatency auditory evoked potentials (e.g., the P50 component that appears about 50 msec after stimulus onset) have been used in studies of auditory informa- tion processing in schizophrenia. Their core finding is that, in a conditioning-testing paradigm, decreased attenuation of the P50 component to the second of two clicks occurs in schizophrenia patients, which can be related to a deficit in sensory gating (1–5). This was also supported by a recent meta-analysis (6). Nonethe- less, overall results are less consistent than for other neurophys- iological parameters, such as the P300 amplitude, since some studies failed to show differences in P50 suppression between schizophrenia patients and control subjects (7–10). Yet, there is indication that P50 gating is associated with a disorganized subtype of schizophrenia and prominent negative symptoms, respectively (10 –15). Recent research has focused on the evaluation of P50 sup- pression as an endophenotype of schizophrenia. As such, it should be heritable, i.e., show a familial co-segregation with illness and be present in some unaffected relatives, and state- independent (16). So far, a reduced P50 suppression has been linked to several gene loci (17) and was found in healthy relatives of schizophrenia patients (18 –25) and in schizotypal personality disorder (26). Myles-Worsley et al. (27) compared a genetically defined high-risk group and a clinically defined sample of at-risk adolescents and showed that P50 suppression was impaired in both groups. Yet, in the genetically high-risk group, P50 suppression abnormalities were found only in those with clinically defined prodromal symptoms (27). Recently, Cadenhead et al. (28) showed that subjects at risk of developing a psychosis had modestly lower levels of P50 suppression relative to control subjects, in particular, if they had a first-degree relative with schizophrenia (28). This difference was nonsignifi- cant, probably due to 1) the fact that a mixed sample was investigated (less than half of the at-risk group was predicted to develop a psychotic episode at least in the next future), and 2) the fact that one third of the group was receiving an atypical antipsychotic that might have “normalized” P50 suppression. The effects of antipsychotic medication on the modulation of the P50 are still not elucidated. Under treatment with first- generation antipsychotics (FGAs), a reduction of the P50 sup- pression was described that was not present under treatment with second-generation antipsychotics (SGAs) (2,4,5). However, a review of the effects of antipsychotics on sensory gating concluded that there was little evidence to suggest that FGAs adversely affect sensory gating (15). And although clozapine might normalize P50 gating, there is scarce evidence of a restorative effect on sensory gating for other SGAs (15). How- ever, to preclude possible confounding effects of antipsychotic medication, only subjects free of antipsychotic medication at the time of recordings were included in the current study. Recent results have extended the findings of reduced P50 suppression in schizophrenia to later event-related potential (ERP) components. Accordingly, sensory gating appears as a per- vasive abnormality in schizophrenia patients that is not limited to From the Department of Psychiatry and Psychotherapy (AB-D, FS-L, RM, AB, RP, JK, SR), University of Cologne, Cologne, Germany; and Department of Psychiatry (IT), University Medical Center Nijmegen, University of Nijme- gen, Nijmegen, Netherlands. Address reprint requests to Anke Brockhaus-Dumke, M.D., Department of Psy- chiatry and Psychotherapy, University of Cologne, Kerpener Str. 62, D-50924 Cologne, Germany; E-mail: anke.brockhaus-dumke@uk-koeln.de. Received June 7, 2007; revised February 7, 2008; accepted February 7, 2008. BIOL PSYCHIATRY 2008;64:376 –384 0006-3223/08/$34.00 doi:10.1016/j.biopsych.2008.02.006 © 2008 Society of Biological Psychiatry