Risk factors of hyperprolactinemia induced by risperidone and olanzapine and their correlations with plasma glucose and lipids

BACKGROUND: Hyperprolactinemia is a common adverse reaction in patients with schizophrenia who take antipsychotic drugs; it often leads to treatment non-compliance in patients and has an adverse effect on their prognosis. AIMS: This study aimed to explore the risk factors of elevated prolactin (PRL) caused by risperidone (RIS) and olanzapine (OLZ) and the relationship between PRL and fasting plasma glucose and lipids. METHODS: Patients with schizophrenia were divided into two groups: 264 patients who were taking RIS and 175 patients who were taking OLZ. These two groups were further divided according to serum PRL levels: an elevated PRL group (>30 ng/mL) and a normal PRL group (PRL ≤30 ng/mL). The demographics, medication dosage, fasting plasma glucose, total cholesterol and triglycerides were compared in the two groups. Logistic regression analysis was performed to explore the risk factors of elevated PRL levels. RESULTS: Compared with the OLZ group, the RIS group had a greater number of patients with elevated PRL (155/264 vs 58/175). Either the RIS or the OLZ group, the proportion of elevated PRL was greater in female patients (RIS: χ(2)=6.76, p=0.009; OLZ: χ(2)=12.98, p<0.001) and with higher doses of the related drugs (RIS: U=−3.73, p<0.001; OLZ: U=−2.31, p=0.021). In patients taking RIS, the elevated PRL subgroup took the drug for a longer period (U=−2.76, p=0.006) and had lower triglyceride levels (U=2.76, p=0.006). In patients taking OLZ, the elevated PRL subgroup had lower fasting plasma glucose levels (U=2.29, p=0.022). Logistic regression analysis showed that gender, dose and fasting glucose levels were significantly associated with elevated PRL levels (RIS: p=0.001, OLZ: p<0.001; RIS: p<0.001; OLZ: p=0.003; RIS: p=0.020, OLZ: p=0.001, respectively). CONCLUSION: Compared with OLZ, RIS had a greater effect on PRL in patients with schizophrenia, and in patients with schizophrenia taking RIS or OLZ, gender and dose were significantly correlated with the PRL value. Moreover, the plasma glucose level of the group with elevated PRL was lower than that of the group with normal PRL. The results also showed that high serum PRL may be associated with a favourable glucose metabolic profile in patients with schizophrenia taking RIS or OLZ. Further studies are warranted to confirm this association. TRIAL REGISTRATION NUMBER: NCT02640911