¹College of Medicine, Taipei Medical University, Department of Psychiatry, Taipei, Taiwan, ²Wan Fang Hospital, Psychiatry, Taipei, Taiwan

Background: Patients with schizophrenia are at elevated risk for metabolic disturbances, particularly when treated with second-generation antipsychotics such as olanzapine. Prolactin elevation is a common adverse effect of antipsychotic therapy, yet its role in glucose homeostasis remains unclear.

Aims & Objectives: This present study aimed to investigate the association between serum prolactin levels and insulin resistance among patients with schizophrenia receiving long-term olanzapine monotherapy. By restricting the sample to a single antipsychotic, we aimed to minimize pharmacological heterogeneity and allowed us to isolate the potential role of prolactin in metabolic outcomes. We hypothesized that the relationship between prolactin and insulin resistance would be non-linear, characterized by an increased risk at both low and high prolactin concentrations.

Method: A cross-sectional study was conducted among 310 patients with schizophrenia maintained on olanzapine monotherapy for ≥ 6 months. Insulin resistance was calculated using the Homeostatic Model Assessment for Insulin Resistance (HOMA-IR), with a cut-off value of 2.0 indicating insulin resistance. Curve estimation, quartile-based analyses, and multivariate regression models were employed to investigate both linear and nonlinear associations between prolactin and metabolic indices.

Results: The prevalence of metabolic syndrome and hyperprolactinemia was 36.8% and 51.0%, respectively. Prolactin levels were significantly associated with insulin, BMI, triglycerides, and HOMA-IR. A quadratic regression model provided the best fit for HOMA-IR (R² = 0.384, p < 0.001), revealing a U-shaped relationship: both low and high prolactin levels were associated with higher insulin resistance. Multivariate analysis confirmed prolactin level and body mass index as independent predictors of insulin resistance. Participants were grouped by quartiles of serum prolactin. Compared with Q2, both Q1 and Q4 showed elevated insulin levels and HOMA-IR, indicating nonlinear prolactin–metabolic associations in schizophrenia.

Discussion & Conclusions: This study reveals a U-shaped relationship between prolactin and insulin resistance in olanzapine-treated schizophrenia. Both low and high prolactin concentrations were associated with increased HOMA-IR and insulin levels, suggesting that moderate prolactin levels may confer metabolic protection. These findings highlight the complex interplay between endocrine and metabolic systems in psychiatric populations and underscore the need for the clinical value of integrated monitoring and personalized care. By advancing our understanding of prolactin’s role in metabolic regulation, this research contributes to the broader goal of optimizing treatment outcomes and reducing cardiometabolic burden in schizophrenia.