Effects of pioglitazone on metabolic abnormalities, psychopathology, and cognitive function in schizophrenic patients treated with antipsychotic medication: A randomized double-blind study

Robert C. Smith, Hua Jin, Chunbo Li, Nigel Bark, Anantha Shekhar, Sauburah Dwivedi, Catherine Mortiere, James Lohr, Qiaoyan Hu, John M. Davis

Research output: Contribution to journalArticle

21 Scopus citations


Background: Schizophrenic patients treated with antipsychotic drugs (AP) have an increased frequency of glucose-lipid metabolic abnormalities and diabetes. Pioglitazone has been shown to be effective in the treatment of glucose and lipid abnormalities in diabetes and decreasing longer-term conversion of impaired glucose tolerance to frank diabetes. Some studies also suggest possible pro-cognitive and antidepressant effects of pioglitazone. We studied the effects of pioglitazone on potential metabolic, symptomatic and cognitive benefits in schizophrenic patients treated with AP. Methods: 54 schizophrenic patients with at least both a)impaired glucose and b) triglycerides ≥ 120. mg/dL and/or low HDL levels, participated in a double-blind placebo controlled study of 3. month treatment with Pioglitazone (30-45. mg/day) or matched placebo, at 5 sites (4 U.S., 1 China). Fasting glucose and lipid parameters, and psychopathology (PANSS scale) were assessed monthly, and patients had a glucose tolerance test and cognitive testing (RBANS and CPT) at baseline and at the end of study. Statistical analysis used mixed model repeated measures analysis, supplemented by completer and LOCF analysis. Results: In the total sample there was an overall effect (P's < .05 to < .01) of pioglitazone on preventing deterioration in fasting glucose and improving HDL and PANSS depression scores; in the pioglitazone group comparison of baseline vs 3. month values also showed significant (P < .05) decreases in fasting insulin, 2. h glucose in GTT and insulin resistance (HOMA-IR). However there were marked differences between the responses of patients in the U.S. sites vs the China site. In the U.S. sample, patients treated with pioglitazone, when compared to placebo treated patents, had significantly lower fasting glucose (F = 3.99, P = 0.02), improved insulin sensitivity (lower H0MA-IR, F = 6.24, P = .002), lower triglycerides (F = 2.68, P = .06) and increased HDL (F = 6.50, P = .001). By the end of the study 52% of the pioglitazone treated patients compared to 15% of the placebo patients had fasting glucose in the normal range (Fisher's exact test P = .02). Pioglitazone also significantly improved PANSS depression factor scores (F = 2.82, P = 0.05). It did not improve cognitive performance on the RBANS or CPT tasks. Pioglitazone did not increase weight or produce any other significant side-effects. In the small mainland China site sample, pioglitazone treatment, as compared to placebo, did not show greater improvement in metabolic parameters or psychopathology ratings. Conclusions: In the sample of patients from the U.S., pioglitazone was an efficacious and safe treatment for glucose and lipid metabolic abnormalities in schizophrenic patients treated with AP, and it may also have beneficial effects on depressive symptoms. It may be particularly useful in patients whose weight gain effects from antipsychotics have plateaued and where weight loss is not the primary goal. The risk vs. benefits of longer term treatment with pioglitazone has to be carefully evaluated for individual patients.

Original languageEnglish (US)
Pages (from-to)18-24
Number of pages7
JournalSchizophrenia Research
Issue number1
StatePublished - Jan 1 2013
Externally publishedYes


  • Cognition
  • Depression
  • Diabetes
  • Glucose metabolism
  • Pioglitazone
  • Schizophrenia

ASJC Scopus subject areas

  • Psychiatry and Mental health
  • Biological Psychiatry

Fingerprint Dive into the research topics of 'Effects of pioglitazone on metabolic abnormalities, psychopathology, and cognitive function in schizophrenic patients treated with antipsychotic medication: A randomized double-blind study'. Together they form a unique fingerprint.

Cite this