To the Editor: I thank Dr. Lund and colleagues for their thoughtful comments. Pharmacokinetic as well as pharmacodynamic interactions are always an appropriate concern when more than one medication is administered. I agree that assessing plasma levels of antipsychotic medications is a valuable way of monitoring pharmacokinetic effects when antidepressants are added to antipsychotic agents, although that measurement is not so simple when concentrations and proportions of active metabolites are considered or such issues as competitive protein binding come into play. Certainly, appropriate cautions are to be used to avoid toxicity, but the proof is still not in laboratory test results but in the patient’s clinical and functional outcome.

I definitely also share the concern of Dr. Lund et al. that a new onset of depression may be a potential early sign of psychotic relapse in schizophrenia. I therefore advocate responding to a new onset of depression with greater attention and psychosocial support and would consider adding an adjunctive antidepressant only if the depression syndrome becomes a stable feature in the absence of psychotic relapse.

I think it unlikely, however, that antidepressants are helpful in the treatment of certain schizophrenic patients with depression simply because they might raise plasma antipsychotic concentrations by means of metabolic influences. Indeed, Kramer et al. (1) found slower resolution of active psychotic symptoms when an antidepressant was added to patients’ antipsychotic treatment under randomized, double-blind conditions. Overall, the best results involving adjunctive antidepressant therapy have occurred among schizophrenic patients who were persistently and syndromally depressed but not flagrantly psychotic at the time of their antidepressant trial (2, 3, my article). In this light, the lack of an observed antidepressant effect in the Buchanan et al. (1996) study of fluoxetine in addition to clozapine therapy is not surprising, because this was a study of positive and negative symptoms specifically, involving patients who were not selected for having depression at baseline.

I certainly agree with Dr. Lund et al. that care should be taken when an antidepressant is added to the medication regimen of a schizophrenic patient and that the patient should be observed closely. This is not only because of potential medication interactions or the possibility that depressive symptoms could be a potential harbinger of psychotic relapse but also because of the possibility that the antidepressant could work effectively. Such an action could result in new energy, confidence, and/or enthusiasm for the patient, ushering in a transition period for the patient and his or her environment, during which help and support might be particularly useful (4, 5).