Although many attempts have been made, to date no convincing evidence exists of a relationship between plasma concentrations of amitriptyline (AT), its active metabolite nortriptyline (NT) and clinical response. Fifteen patients with primary depression (according to DSM-IV) were divided in two groups according to given doses: (I) 6 patients received 3 x 50 mg of AT daily; and (II) 9 patients received 3 x 25 mg of AT daily, for 6 weeks. The clinical status was determined with Hamilton Depression Rating Scale. Both investigated doses were therapeutically effective. AT and NT plasma concentrations were assayed by high performance liquid chromatography. Following administration of 3 x 50 mg of AT daily, the correlation of concentrations of AT, NT, total AT + NT and clinical response were rAT = -0.702 (P < 0.1), rNT = -0.761 (P < 0.1), rAT + NT = -0.741 (P < 0.1). The linear and very high correlation were also present with concentrations of AT, NT, total AT + NT and clinical response in depressive patients on 3 x 25 mg AT daily: rAT = -0.785 (P < 0.02), rNT = -0.811 (P < 0.01), rAT + NT = -0.848 (P < 0.01). Our results support a high correlation between AT/NT plasma concentrations and clinical response indicating that therapeutic monitoring of AT and its metabolite, NT, can provide eventual clinical response.