Lithium augmentation is one of the best studied strategies for resistant depression. The lithium dosage usually given is around 900 mg/day and plasma level is maintained in the range of 0.5-0.8 mEq/L. However, the administration of lithium in this dosage necessitates monitoring of plasma concentration and increases the risk of toxicity and side effects. Since it has been shown that low lithium levels increase serotonin turnover and enhance serotonin neurotransmission, we thought it of interest to assess the efficacy of low dosage lithium augmentation for patients with resistant depression. Fifty one patients suffering from severe unipolar or bipolar depression who had failed to respond to treatment with venlafaxine 300-375 mg/day were included in the study and treated as outpatients. Patients had previously been exposed to unsuccessful treatment with various antidepressants, mostly SSRIs. After a washout period for previously administered antidepressants of one week, the dosage of venlafaxine was rapidly titrated to 300 or 375 mg/day, corresponding to about 5 mg/kg. The dose remained stable during the next six weeks. Additional antipsychotic medication was allowed to treat psychotic symptoms. Forty seven severely depressed patients who failed to respond to 300-375 mg/day venlafaxine were, in addition, given lithium carbonate in low dosage (300-450 mg/day). The Clinical Global Impression Improvement scale was used as the treatment outcome. A score of 1 or 2 was considered as non-response. All patients gave informed consent to participate in the study. Ratings were performed at baseline and after 1,2 and 5 weeks. Lithium plasma concentration measurements were performed after 1 and 4 weeks. After 5 weeks of augmentation, 51% of the patients were rated as "much" or "very much" improved. Bipolar patients showed a better response than unipolar (64.3% vs 45.5%, p<0.038). Most patients (76%) showed a rapid response (up tp 7 days), and only 2 patients (4.6%) responded after more than 2 weeks The mean lithium plasma level was 0.33±0.09 mEq/L. No significant differences were found in treatment response with regard to sex, family history, psychotic symptomatology and suicidal ideation. No troublesome side effects were reported. Our results show that treatment augmentation with low lithium dosage may be as effective as augmentation with higher dosage, is well tolerated and does not necessitate monitoring of plasma level. Hence, an initial trial of ugmentation at low dosage lithium may be the preferred first choice in non-emergent situations. The low dosage also minimizes the risk of side effects and drug-drug interactions. Prospective controlled studies to confirm our findings are needed as are larger scale comparisons with therapeutic dose lithium augmentation.