To the Editor: Refractory depression is common in clinical practice. At least 30% of patients fail to respond to initial antidepressant therapy.1 In patients who have failed 1 course of antidepressants, there is evidence from 9 randomized controlled trials suggesting that approximately 50% of treatment-refractory patients do respond to lithium augmentation compared with placebo (p < .001).2 The British Association for Psychopharmacology guidelines recommend that patients be treated with 2 antidepressants before an augmentation strategy is used.3 Our aim was to review the evidence for use of lithium in those patients who are resistant to treatment with 2 or more antidepressants at recognized therapeutic doses. Method. A systematic review was undertaken of placebo-controlled, double-blind, randomized trials of lithium that assess its effectiveness in treatment-resistant depression. English-language studies of adults (aged 18 years or older; both sexes), suffering from major depressive disorder diagnosed by the International Classification of Diseases, Tenth Revision (ICD-10), the Diagnostic Statistical Manual of Mental Disorders, Fourth Edition (DSM-IV), or Research Diagnostic Criteria (RDC) were included. The following keywords were used: lithium OR Camcolit OR Carbolith OR Duralith OR Eskalith OR Licarbium OR Liskonum OR Litarex OR Lithane OR Lithocarb OR Lithizine OR Lithonate OR Lithotabs OR Manialith OR Phasal OR Priadel OR Quilonorm OR Quilonum OR Li-Liquid OR resistant depression OR refractory depression OR treatment resistant depression OR treatment non respon(*) depression OR lithium augment(*). Exclusion criteria were bipolar depression, previous prescription of lithium, co-morbid schizophrenia, or drug and alcohol dependence. All participants should have received 2 antidepressants at recognized therapeutic doses (eg, 150 mg of imipramine daily) for a minimum of 4 weeks. Participants in the active arm should have received at least 4 weeks of lithium treatment at a minimum plasma level of 0.4 mmol/L or a dose of 600 mg daily. The search included the computerized databases of the Cochrane Library, MEDLINE/PubMed (1966+), EMBASE (1980+), CINAHL (1982+), and PsycINFO (1974+). Reference lists were hand searched along with major books. Pharmaceutical companies were contacted. Leading researchers and experts in the field were also contacted. Methodological quality was categorized as (A) adequate, (B) uncertain, or (C) inadequate. Results. A total of 12 randomized controlled trials involving lithium augmentation for resistant depression were identified, of which only 14 had ensured a trial of 2 antidepressants. There was little information given in the published reports regarding the methods used to achieve random allocation. Thus, on the basis of the published reports, the studies initially received a B rating according to the Cochrane criteria. In the 1 study that included 2 antidepressants,4 there were 18 patients in the lithium augmentation arm, of whom 2 dropped out. The proportion who responded to lithium augmentation in the completer and intent-to-treat (ITT) analyses were 12.5% and 11.1%, respectively. There was no statistically significant difference in scores on the 17-item Hamilton Rating Scale for Depression. The lithium group achieved a mean reduction of 6.3/5.5 points (completer/ITT). The mean reduction in the placebo group was 6.5/5.8 points (completer/ITT) (p = .91). The mean blood lithium level at week 2 was 0.63 mmol/L (range, 0.3–1.4 mmol/L) and at week 6 was 0.61 mmol/L (range, 0.6–0.9 mmol/L). The duration of the trial was 6 weeks. In the recent Sequenced Treatment Alternatives to Relieve Depression (STAR*D) study,5 there was no placebo group, and, as there was no significant difference to T3 augmentation (which has not yet been demonstrated as being effective for treatment-resistant depression), we cannot conclude that either lithium or T3 is an effective augmentation strategy for patients who are resistant to at least 2 antidepressants. There is limited evidence to support the use of lithium augmentation after 2 or more failed antidepressant therapies. More research needs to be undertaken to compare lithium augmentation with other augmentation strategies. Those patients who have failed to respond to 2 or more antidepressants are a difficult group to treat.