The aim of the study was to validate a test based on analyses of urine to identify the propafenone metaboliser phenotype during routine chronic therapy. Twenty seven patients chronically treated with propafenone were studied. A debrisoquine test was performed in 10. Propafenone and its metabolites in plasma and urine were measured by HPLC. Propafenone, 5-hydroxypropafenone and N-depropylpropafenone concentrations in plasma were 1.09, 0.182 and 0.101 ng.ml-1, respectively. Total recovery of the administered dose in urine was 30.7%. Two patients were identified as PM, based on the result of the debrisoquine test (log D/4OHD of 1.26 and 1.36). This finding was confirmed by the propafenone metabolic ratio in urine, but the plasma data did not permit clearcut separation of the phenotypes. Propafenone/5-hydroxypropafenone in plasma was not a good predictor of metabolizer phenotype. Although the number of patients who completed all three tests was limited, it is concluded that analysis of propafenone/5-hydroxypropafenone in urine collected between two consecutive doses at steady-state is more practical than the debrisoquine test and more specific than determining the propafenone/5-hydroxypropafenone ratio in plasma, for identification of the propafenone metaboliser phenotypes.