Clinical Prediction of Antidepressant Response

Abstract

Depression is not a unitary disease and not all cases of depression can be treated alike. Therefore, predictors of antidepressant response to different drugs need to be identified. Premorbid personality, age and sex and family history are useful in prediction. As an example, only females respond to the addition of triiodothyronine to the antidepressant therapy. Endogenous depressions respond favourably to antidepressants, so also do psychotic depressions. On the other hand, delusional depressions do not respond to antidepressants. While atypical depressions are best treated with MAOI, motor retardation improves with imipramine and obsessive symptoms with chlorimipramine. In addition to clinical variables, biochemical and statistical approaches have also yielded significant results in prediction.

Predictive Value of Two Endogenicity Measures

“…Only about 65-75% of depressed patients respond to heterocyclic antidepressants [1], An enormous effort has been invested in the search for variables having a potential pre dictive value as to the therapeutic response (mostly short-term) to antidepressants: clini cal (psychopathological and sociodemo graphic) and biological variables have been investigated [for recent reviews, see [2][3][4], At the clinical level endogenicity has been con sidered to have a major predictive value [5][6][7][8], although this has not always been con firmed [9], A source of variation may be found in the operationalization of the con cept of endogenicity: the best-known scale for measuring endogenicity is probably the Newcastle Scale [10]. Recently however two other systems have been proposed to iden tify endogenous and nonendogenous depres sive patients: the Michigan Discriminant In dex (MDI: table 1) [11] and the Hamilton Endogenomorphy Subscale (HES: table 2) [12], We took advantage of a study in which we examined the differential effects of two antidepressants (maprotiline and clomi pramine) on hostility [unpubl.…”