Fasting insulin serum levels and psychopathology profiles in male schizophrenic inpatients treated with olanzapine or risperidone.

OBJECTIVE Recent studies have suggested that higher insulin levels are associated with better psychopathology profiles in cross-sectional samples of patients with schizophrenia. This study examines whether drug-induced fasting insulin changes between third and eight week of treatment are related to clinical improvement in non-diabetic patients receiving the atypical neuroleptics: risperidone or olanzapine. METHODS non-diabetic men with a diagnosis of schizophrenia according to the DSM-IV diagnostic classification were recruited from psychiatric inpatient units. Following a drug-free period, neuroleptic treatment was initiated (risperidone n=36, olanzapine n=35) and doses were adjusted to achieve maximal clinical efficacy. All patients were hospitalized throughout the study. Initial and final evaluations of serum insulin levels and psychopathology (assessed with the Positive and Negative Syndrome Scale, PANSS), were carried out at weeks 3 and 8 after the onset of treatment, respectively. RESULTS There were no differences between and within the risperidone and olanzapine groups in changes of serum insulin level between the third and eighth week of treatment. In the olanzapine group, Pearson correlation analysis revealed a significant negative correlation between changes in fasting serum insulin levels and the PANSS-Total, Positive and General Psychopathology subscale scores. Only improvement in the PANSS-Negative Symptom subscale score was not correlated with insulin level change between the third and eighth week of treatment. In the risperidone group, correlations between PANSS subscales scores and the corresponding serum insulin levels change were positive, albeit statistically non-significant. In both groups the improvement in PANSS-Total scores was not correlated with changes in BMI. CONCLUSION Olanzapine-related changes in endogenous fasting insulin levels were correlated with clinical improvement in acutely ill non-diabetic schizophrenic patients. Because the interesting linkage between insulin and positive and negative symptoms could be an epiphenomenon, randomized studies are needed to further explore the role of insulin in therapeutic responses in patients with schizophrenia.