Clinical and biological parameters as predictors for antidepressant drug responses in depressed patients.

Clinical and biological variables were investigated for their predictive value with respect to an antidepressant drug treatment. Thirty patients received clomipramine and thirty patients maprotiline. Characteristic features of the biography, the family anamnesis and the previous course of illness (apart from intermittent course) have no predictive value. The psychopathological symptoms before the start of treatment are also not suitable for prediction (except vegetative syndrome). The activity of the enzymes MAO, COMT and DBH before the start of treatment have no predictive value. The serum level of maprotiline on the seventh day of treatment does not correlate with the outcome of treatment. It is possible that patients, who have relatives with suicidal tendencies, are more likely to be clomipramine non-responders; patients with relatives who have a psychiatric history but without suicidal tendencies, are more likely to be maprotiline responders; i.a., relatives of the first degree manifesting psychiatric problems speak against a response to clomipramine and indicate a response to maprotiline. Patients with diurnal variations before the start of treatment are possibly more likely to respond to maprotiline than to clomipramine. There are statistically established findings for only the following variables: Diurnal variations during treatment speak in favour of an antidepressant response. A positive SD reaction indicates clomipramine response. A serum level of more than 75 ng clomipramine and more than 30 ng desmethyl-clomipramine/ml serum on the 7th day of treatment clearly predict a response to clomipramine.