Background: Treatment of arterial hypertension in patients with severe mental illnesses often results in polypharmacy, potentially leading to drug-drug interactions. The objective of the study was to analyse the in vivo inhibitory potential of two antihypertensive drugs, amlodipine and metoprolol on CYP2D6 catalysed 9-hydroxylation of risperidone (RIS). Methods: A therapeutic drug monitoring database with plasma concentrations of RIS and 9-hydroxyrisperidone (9-OH-RIS) of 1584 patients was analysed. Three groups were considered; a group of patients receiving RIS without a potentially cytochrome influencing co-medication (control group, R-0, n=852), a group co-medicated with amlodipine (R-A, n=27) and a group, co-medicated with metoprolol (R-M, n=41). Plasma concentrations, concentration-to-dose ratios (C/Ds) of RIS, 9-OH-RIS and the active moiety (AM), as well as the metabolic ratios were computed and compared using the Kruskal-Wallis test, the Mann-Whitney U test and the Jonckheere-Terpstra test to determine the means and different patterns of distribution of plasma concentrations as well as the concentration-to-dose ratios. Results: The median daily dosage of RIS did not differ between the groups (p=0.708). No differences were found in median plasma concentrations of RIS, 9-OH-RIS and AM. However, concentration-to-dose ratios for RIS, 9-OH-RIS and AM were significantly higher in the amlodipine group (p=0.025, p=0.048 and p=0.005). In the metoprolol group, the concentration-to-dose ratio for RIS was significantly higher than in the control group (p=0.017), while the C/D for 9-OH-RIS and AM was not. Conclusions and limitations: Our data show a potential pharmacokinetic interaction, most likely via CYP3A4 between amlodipine and RIS, reflected in significantly different C/Ds for RIS, 9-OH-RIS and AM. Although the interaction did not result in significantly higher plasma levels, changes in C/Ds and their distribution with regard to the median concentrations were observed.