BACKGROUND To explore the influence of CYP2D6 genetic polymorphism on risperidone metabolism, thereby affecting risperidone's effects and safeties in patients with chronic schizophrenia. METHODS Sixty-nine subjects with chronic schizophrenia treated with risperidone were recruited. CYP2D6 genotypes was determined using targeted sequencing and translated into phenotype using activity system. Risperidone plasma concentrations were measured using HPLC. Positive and Negative Symptom Scale (PANSS) and Brief Psychiatric Rating Scale (BPRS) were used to evaluate the existence and severity of psychiatric symptoms, Barnes Akathisia Scale (BAS) and Extrapyramidal Symptom Rating Scale (ESRS) for neurological side effects. Metabolic and endocrine status assess were also included. RESULTS The plasma drug concentrations varied hugely among individuals. Intermediate metabolizer (IM) group had higher plasma levels of RIP and dose corrected RIP concentration, RIP/9-OH-RIP ratio and C/D ratio than normal metabolizer (NM) group (p < 0.01). There was no statistic difference between responders and non-responders in dose-adjusted plasma concentrations and ratios of RIP/9-OH-RIP and C/D. The occurrence of EPS was related to active moiety levels in 4th week (p < 0.05). The prolactin (PRL) levels in two follow-ups were both significantly higher than baseline (p < 0.01). PRL change from baseline to week 4 and week 8 were both positively associated with active moiety concentration detected in week 4 (p < 0.05). CONCLUSIONS The risperidone plasma levels have great inter- and intraindividual variations, and are associated with the CYP2D6 phenotypes, as well as the changes in serum prolactin in patients diagnosed with chronic schizophrenia.