Previous reports of an association between the polymorphic cytochrome P450 CYP2D6 and systemic lupus erythematosus are conflicting. Following the elucidation of the molecular basis of the CYP2D6 genetic polymorphism, we re-examined the hypothesis of an association of this gene with a susceptibility to system lupus erythematosus by analysing the complete CYP2D6 coding sequence. For this purpose, we studies the occurrence of 16 mutations in genomic DNA from 69 systemic lupus erythematosus patients and a large control group using a previously described polymerase chain reaction-single strand confirmation polymorphism analysis. In addition, we studied the occurrence of 11 alleles and 21 genotypes in the same individuals by the combined use of restriction fragment length polymorphism and allele-specific polymerase chain reaction followed by polymerase chain reaction-single strand confirmation polymorphism analysis. No significant differences in the distribution of overall genotypes and predicted phenotypes were observed between system lupus erythematosus patients and controls. The only new finding of our study is the higher frequency of one non functional allele, namely the CYP2D6*4A, in systemic lupus erythematosus versus control individuals (P = 0.007). This increased frequency was not statistically significant in multiple comparison analysis and was not related to any specific clinical features of systemic lupus erythematosus. These results suggest that CYP2D6 genotype as well as CYP2D6 phenotype are not determinant of susceptibility to systemic lupus erythematosus but the presence of the inactive CYP2D6*4A allele may be a contributory factor.