||N100, N200 and P300 in male first-episode psychosis: relationship between brain function and psychopathology
Schizophrenia and Allied Psychoses
||Event-related potential (ERP) techniques offer insight into the time course of abnormal cognition in schizophrenia. Reduction in amplitude of early (N100) and late (P300) cognitive components are reliably reported in patients performing auditory selective attention tasks (OÃ¢â‚¬â„¢Donnell et al, 2004; Gallinat et al, 2002; Mathalon et al, 2000; Laurent et al, 1999; Turetsky et al, 1998; Frodl et al, 1998). However, the nature of these reductions including topography and associated subcomponents may depend on chronicity, symptom profile and sex (Sumich et al, 2005; Turetsky et al, 1998; Gallinat et al, 2002). Study of patients with recent-onset illness minimises the effects of chronicity such as neuronal degeneration, hospialisation and long-term medication use. Demiralp et al (2002) found a specific frontal reduction in P300 amplitude in recent-onset patients, suggesting that anterior areas are primarily affected in the early phases of psychosis. However, their patient group consisted predominantly of women who according to studies of chronic schizophrenia are more likely to demonstrate a reduction in frontal and left temporal P300 (Turetsky et al, 1998). This is concordant with symptom profiles more common in women. Reduced P300 amplitude at frontal sites is associated with hallucinations and depression (Mathalon et al; Turetsky et al, 1998), whilst mechanisms measured by P300 deficits at left temporal sites may underlie thought disorder and active symptoms. Men with chronic schizophrenia, on-the-other hand, show greater reduction at right parietal sites, which is more evident in patients with negative symptoms (Turetsky et al, 1998). Concordantly, studies of recent-onset patients that include a male majority show reduced P300 amplitude at parietal as well as left temporal and frontal sites. However, it is unclear whether the latter deficits found in these studies are due to the inclusion of female patients.
Reduced N100 amplitude was also reported in a study of 26 men and 14 women with recent-onset psychosis (Brown et al 2002). In contrast, a study of first-episode patients by Valkonen-Korhonen et al (2003) (15 females, 10 males) showed no effect on N100 of diagnosis in global field power or mean ERP measures. However, visual inspection suggested that the current density in the N100 time-frame was reduced in patients over bitemporal and frontal areas. This suggests diagnosis and sex may interact to effect N100 amplitude.
Valkonen-Korhonen et al (2003) also show a negative correlation between N100 and the severity of hallucinations. In chronic patients, N100 amplitude has been shown to correlate negatively with both positive and negative symptoms (ref). Gallinat et al (2002) used source localisation techniques to demonstrate that positive symptoms are associated with the N100 subcomponent generated by the superior temporal gyrus, an area implicated in reality distortion by neuroimaging studies (Sumich et al, 2005; Gaser et al, 2004). On-the-other hand, negative symptoms correlated negatively with the subcomponent generated by the anterior cingulate (Gallinat et al 2002). Again this is concordant with neuroimaging studies that associate AC dysfunction with negative symptoms (Ashton et al, 2003).
The AC has a well established role in cognition and affect, and is implicated in depression (Ashton et al, 2003). Thus it is unclear whether the association between AC and negative symptoms is due to its involvement in attention or mood (Ashton et al, 2003). Another component that relies on AC function is the N200, subcomponents of which reflect mechanisms involved in affect (Van Veen and Carter, 2000). N200 amplitude reduction is reported in chronic and FE schizophrenia (OÃ¢â‚¬â„¢Donnell et al, 2004; Laurent et al, 1999), but with much less consistency than P300 or N100 deficits (Guillem et al, 2003; Oades et al ). This may be due to differential effects of subdiagnosis and inter-study differences in the measured N200 subcomponents as defined by topography. Enhanced negativity in the N200 range is seen at fronto-central sites in patients with severe reality distortion (Guillem et al, 2003). At centro-parietal sites, enhanced N200 is associated with negative mood in major depression and with high trait suspiciousness in healthy men (Sumich et al, submitted; Bruder et al, 1998). The relationship between reduced N200 and psychopathology is not fully understood. However, we recently report right frontal N200 amplitude reduction in subclinical depression as measured using the Depression Anxiety and Stress Scale (DASS) (Sumich et al, 2005). This was more prominent in men and is compatible with reduced volume and function of right anterior cingulate in depression found in neuroimaging studies.
||We are exploring sex differences in FEP. Given that sample size for women is not yet high enough would have begun this project with an investigation into right-handed men with FEP. Comparison with women will follow with increased numbers.
The current study explores the relationship between clinical symptoms, ERPs and neuropsychology performance. Reduced amplitude of N100, parietal P300 and left temporal P300 was predicted in patients compared to healthy comparisons. Reduced N200 was predicted in depressed patients. Frontal N100 is expected to be associated with negative symptoms and impaired attention. Left temporal N100 will be associated with hallucinations. Left temporal P300 will be associated with thought disorder. Right N200 amplitude will correlate negatively with severity of depression.
We will also use source localisation to confirm involvement of the AC in N100 and N200 effects.
||Brain Resource Company (BRC) standardised procedures was used to assess patients and comparisons at different clinic sites. The BRC standardised protocol has high interlab reliability (for details see www.brainresource.com and www.brainnet.org.au ) (Williams et al, in press) and previous studies show no significant effect of clinic site (Sumich et al, in press). Assessment procedures were approved by the local ethics committee of each clinic.
Twenty two men who had recently experienced their first psychotic episode were recruited from Westmead Hospital. Diagnosis of schizophrenia (paranoid n=, disorganized n=, undifferentiated n=), schizophreniform (n=) psychosis or schizoaffective (n=) disorder was confirmed by structured clinical interview for DSM-IV (Spitzer et al, 1990) administered by qualified research psychiatrists. Patients were also assessed on the Positive and Negative Syndrome Scale (PANSS) for which interrater reliability was high (kappa=???). Reality distortion was scored as the sum of subscales P1 (delusions), P3 (hallucinations), P5 (grandiosity) and G9 (unusual thought content).
Depression, anxiety and stress were assessed using an internet based demographics questionnaire that included the Depression Anxiety and Stress Scale (DASS) (Crawford et al, 2003; Lovibond and Lovibond, 1995).
Patients were matched as close as possible with 44 healthy individuals recruited from the general public through various means (eg. newspaper advertisements, universities, job centres, pamphleting) from seven world-wide brain-resource company (BRC) clinics. These clinics are located in London, Nijmegen, Sydney, Adelaide, Melbourne, New York and Rhode Island. Information on exclusion criteria were obtained via a structured telephone interview. These criteria included a personal or family history of psychiatric disorders (including depression), personal history of neurological disorder, history of past or current substance abuse or other major health problems related to the heart, thyroid gland or cancer. Data were not collected on applicants excluded under these criteria. After a full description of the assessment protocol all enrolled participants gave written informed consent.
Event-related potential methods are detailed elsewhere (Sumich et al revised submission). Briefly, continuous electroencephalographic data was collected from 21 electrode sites according to the 10-20 system relative to linked mastoids with a forehead ground (Low pass filter =100Hz, 40dB; impedence< 6 kOhms). Both horizontal and vertical eye-movements were used in offline correction of artefact (Gratton et al 1983). Participants completed an auditory-oddball task identical to that reported by Sumich et al (revised submission). N200 was scored as the most negative peak between 180-320ms; and P300 as the most positive peak between 270-550ms.
SPSS 11.0 was used for all statistical analysis. Alpha level for significance testing in all analyses was kept at p < .05 unless stated otherwise.
Relationship between symptoms
Pearson correlations were performed to test the relationship between depression, anxiety, stress, positive symptoms, negative symptoms, general symptoms and total symptoms.
Between group variables used in ERP analysis were Diagnosis (patient, comparison).. For midline and medial electrodes, the within group variable Site had 5 levels (anterior (Fz, F3, F4), anterior-central (FCz, FC3, FC4), central (Cz, C3, C4), central-posterior (CPz, CP3, CP4), posterior (Pz, P3, P4)). For lateral electrodes, Site had 3 levels (anterior (F7, F8), anterior-temporal (T3, T4), posterior-temporal (T5, T6). Analysis of medial and lateral electrodes also included the within group variable Hemisphere (left, right). The main ANOVA was followed by lower order ANOVA as appropriate.
Patients were divided into two groups based on a mean split of the depression score. ANOVA were performed similar to that described above with Depression (depressed, not depressed), rather than Diagnosis as between groups factor.
LORETA source localisation will be used to identify brain areas involved in significant effects. (ie N100 and P300 between patients and controls; and N200 between depressed and nondepressed patients)