Major depressive disorder and bipolar disorder are the most frequent mood disorders. Potential biological markers are supposed to be helpful in the investigation the etiology of these disorders, as well as in the establiishing the diagnosis and pharmacological treatment. Among others, serum lipid levels, platelet serotonin levels and platelet MAO activity are candidates for such a marker. The main goal of the study was to determine peripheral biochemical markers (platelet serotonin concentration, platelet MAO activity, total serum cholesterol, HDL, LDL and trigliceryde concentration) in males and females with: bipolar disorder depressed phase, bipolar disorder manic phase, reccurent depressive disorder and healthy males and females. We hypothesized that there were differences between aforementioned biochemical markers in patients with affective disorders. These findings could have contributed to better insight into etiology as well as to improve pharmacological treatment of affective disorders. The study sample consisted of the following four groups of males and females, older than 18 years: a) 67 patients with bipolar disorder, currently in manic phase (45 males, 22 females), b) 33 patients with bipolar disorder, currently in depressed patients (14 males and 22 females), c) 68 patients with reccurent depressive disorder (34 males and 34 females), d) 87 patients with reccurent depressive disorder who were randomised to antidepressant treatment (50 patients were treated with paroxetine, 20 mg once daily, and 37 patients were treated with tianeptine, 12.5 mg three times dialy), d) 69 healthy subjects (32 males and 37 females). All four groups have signed an informed consent document, which was approved by local Ethics Committee. All three patient groups were diagnosed according to both ICD-X and DSM-IV classifications. They were required to be free of psychotropic medication for at least previous 2 weeks, except for low-dose benzodiazepine treatment, which was allowed throught the study. They had no other major psychiatric disorder, and no history of dyslipidemia. All participans have never been diagnosed with any lipid disorder. The patients were recruited from both outpatients and inpatients in University Hospital Center Zagreb, Department of Psychiatry. Platelet serotonin level and platelet MAO activity were determined at Research Institute „Ruđer Bošković“, and serum lipid levels were determined at Biochemical laboratory of University hospital Center Zagreb. Serum cholesterol and LDL values were significantly lower in male manic patients than in healthy controls. We have observed significantly higher platelet 5-HT concentration and significantly lower serum cholesterol and LDL values in male patients with psychotic mania than in both male nonpsychotic manic patients and healthy controls. Significantly lower serum HDL-C values were observed in female patients with bipolar disorder in both manic and depressive episode and with major depressive disorder, compared to the healthy controls. Female patients with bipolar disorders in manic or depressive episodes had significantly higher TG values than healthy controls. There were no differences in cholesterol and LDL levels across all three groups of female patients. Platelet serotonin level was significantly lower in males with major depressive disorder than in healthy males. We have found no differences in serum lipid levels between males in the depressed phase of bipolar disorder, males with major depressive disorder and healthy males. Platelet MAO activity was similar in all four groups of participants. Bolesnici s lošim terapijskim odgovorom na paroksetin imali su značajno višu bazalnu koncentraciju trombocitnog serotonina te serumskog kolesterola i TG u odnosu na bazalnu koncentraciju trombocitnog serotonina, kolesterola i TG u bolesnika s dobarim terapijskim odgovoom i u zdravoj kontroli. Nismo pronašli razliku u koncentraciji trombocitnog serotonina između bolesnika s dobrim ili lošim terapijskim odgovorom na terapiju tianeptinom, i zdrave kontrole. Bolesnici sa slabim terapijskim odgovorom nakon 4 tjedna liječenja tianeptinom su imali značajno višu bazalnu koncentraciju serumskih TG u odnosu na koncentraciju TG u serumu zdravih osoba. Nonresponders to paroxetine had significantly higher baseline platelet serotonin levels, cholesterol and triglyceride levels compared to both paroxetine responders and healthy controls. We observed no difference in baseline platelet serotonin levels between tianeptine responders and nonresponders and healthy control group. Tianeptine nonresponders had significantly increased baseline triglyceride levels in comparison to healthy controls. Several conclusions may be drawn from our study. First of all, given the increased platelet serotonin concentration in our sample of psychotic manic patients, and the literature reports of increased platelet serotonin levels in other psychotic dysorders (psychotic depression, schizophrenia, dementia with psychotic features) we conclude that platelet serotonin levels might be increased in psychotic patients beyong the diagnosis We have confirmed hypothesis that decreased cholesterol levels is state marker of manic phase in male patients with bipolar disorder. We hypothesize that decreased HDL levels and increased TG levels are state markers in female patients with bipolar affective disorder, at least in manic and depressed phase, while decreased HDL levels are also found in female patients with major depressive disorder. Given the well-determined connection between dyslipidemia and cardiovascular disorders, both decreased HDL and increased TG levels could contirbute to consistently reported increased cardiovascular mortality and morbidity in these patients The findings of increased platelet serotonin and serum cholesterol levels in nonresponders to SSRI have led us to speculate that there might be an interaction between baseline levels of easy accessible biochemical markers and response to antidepresants. Furter investigation in this area seems important.