In the August 1994 issue of theArchives, Popli et al1make several points relevant to the interactions between selective serotonin reuptake inhibitors (SSRIs) and tricyclic antidepressants (TCAs) with which we must take issue. First, we believe it possible to make reasonable predictions of in vivo effects from in vitro data. Second, we believe more care must be taken in drawing conclusions about drug interactions from case reports. Four research groups, not just one, have made in vitro assessments of the extent to which the SSRIs (and some of their metabolites) inhibit cytochrome P-450 2D6 (CYP2D6).2-7We have derived estimates of the corresponding in vivo relevance using the following strategy. All of the published in vitro assessments include a determination of the inhibition constant (Ki) of fluoxetine. If the Kiof each SSRI is normalized to the Kiof fluoxetine in the same study, averaging those