Institutional members access full text with Ovid®

Share this article on:

Prediction of Treatment Response in Geriatric Depression From Baseline Folate Level: Interaction With an SSRI or a Tricyclic Antidepressant

Alpert, Murray PhD; Silva, Raul R. MD; Pouget, Enrique R. BA

Journal of Clinical Psychopharmacology: June 2003 - Volume 23 - Issue 3 - pp 309-313
Brief Report

Depressed geriatric patients have lower levels of folate (FOL) than controls. Also, FOL supplement can reduce depressive morbidity. One hypothesis consistent with this is that FOL deficiency causes a lowering of CNS serotonin that contributes to depression. The present report is from one site of a multicenter study that compared an SSRI (sertraline) with a nonspecific tricyclic antidepressant (nortriptyline) in geriatric depressed patients. We added measures of FOL at baseline and outcome for 22 depressed patients older than 60 years. Both treatments were effective. At baseline, FOL levels were within the normal range. Higher FOL levels at baseline predicted greater improvement. Further study of FOL interaction with SSRI is warranted. For the group treated with the SSRI, baseline FOL level was a more efficient predictor of improvement, especially for results on a self-rating depression scale (POMS).

Folate (fol), an essential nutrient, participates in a number of metabolic processes. Reactions that intensively involve DNA replication—for example, hematopoiesis—require FOL, and an insufficiency is reflected in the RBC changes of megaloblastic anemia. 1 The recommended FOL intake is 400 μg/day, 2 but the body has extensive stores, and an inadequate intake would have to continue for weeks or months before a deficiency emerged. 3 A number of drugs and alcohol can interfere with one or another stage of FOL absorption or metabolism, and thus an insufficiency could develop despite an adequate diet.

FOL and B12 deficiencies often occur together. Both can be associated with megaloblastic changes. Shorvon and associates 4 attempted to dissociate the effects of one deficiency from the other in a study of 84 patients with megaloblastosis identified by the hematology service. It was established that 50 subjects had a deficiency of B12 but not FOL and 34 had a deficiency of FOL but not B12. The FOL-deficient group was younger than the B12-deficient group (40 years vs. 60 years) and was less likely to present with peripheral neuropathies (18% vs. 40%) but more likely to have depression (56% vs. 20%). The two groups were equally likely (27% vs. 26%) to show “organic mental changes.”

However, since B12 is required for the reduction of FOL to an active form, insufficient B12 can trap FOL and prevent its participation in a number of reactions. 1 Thus, subjects with a B12 deficiency and normal FOL assay findings may in fact have an occult FOL deficiency. On the whole, it appears that B12 deficiencies are more related to peripheral neuropathies, for instance, and that FOL is more implicated in central disorders of cognition, motivation, and mood. 4,5 These mental disturbances probably are not related to the processes that mediate the contribution of FOL to DNA replication but may reflect the participation of FOL in serotonin metabolism 6,7 or the role of FOL in transmethylation or in a range of other CNS metabolic activities. 8

In geriatric populations, both depression and dementia have been found to be associated with lower FOL levels, 9,10 but it remains unclear whether psychiatric disturbance leads to poor diet or the reverse. It is easy to imagine that mildly demented or depressed elderly subjects might fail to obtain adequate nutrition. Conversely, chronic malnourishment could cause mental changes. Possibly both processes feed on each other in a reciprocal cycle. However, FOL supplements can reduce mood disorders, suggesting that at least one path runs from vitamin deficiency to depression. 11,12

Different case-finding strategies (e.g., in a hematology clinic, a psychiatric service, or a psychogeriatric service) can affect estimates of the association between FOL and the incidence and type of psychiatric disorders, and results should be interpreted with consideration of the study design. Research done in a psychiatric facility is more likely to reveal an association with mood, whereas studies in a psychogeriatric facility are more likely to find an association with memory.

A number of investigators have measured FOL levels in psychiatric patients with mixed diagnoses, while exploring a possible association between FOL level and diagnosis. Some studies have shown relations between the severity of disorder and the FOL level, even within the normal range of FOL. For example, Sommer and Wolkowitz 13 examined a small group (N = 13) of elderly patients presenting to a memory clinic and found an association between the Mini Mental State Exam score and RBC FOL level within the range of 200 and 600 ng/mL.

Similarly, Bell and associates, 14 in a retrospective chart review, found that only 1% of a sample of geropsychiatric patients had FOL values below the normal range (<200 ng/mL), and division of subjects into groups with levels above or below 400 ng/mL efficiently predicted the age at onset of “organic psychosis” as well as the length of hospital stay. Also, Ghadirian and associates 15 found a negative correlation between serum FOL levels and Hamilton Depression Scale 16 scores across the normal range of FOL levels.

Evidence implicating FOL deficiency in mood disorders has come from: (1) biochemical surveys in which lower FOL levels were found in depressed patients 17; (2) diminishment of depression when FOL supplements were administered to depressed patients 12,18; (3) reports of lower FOL levels in patients with more severe depression ratings 19; and (4) greater treatment resistance in depressed patients with lower FOL levels. 14,20 Thus, there is converging evidence, especially in geriatric patients, that low FOL levels can contribute to aspects of depression.

Since folate is involved in serotonin metabolism, we hypothesized that baseline folate levels would predict treatment response among individuals treated with an SSRI, whereas the relationship with folate would be more remote in those patients treated with agents that primarily affect other neurotransmitter systems.

Department of Psychiatry, New York University Medical Center, New York, New York

Received March 22, 2002; accepted September 24, 2002.

This work was supported by a grant from the U.S. Pharmaceuticals Group of Pfizer, Inc.

Address requests for reprints to: Raul R. Silva, MD, Department of Psychiatry, NYU Medical Center, NB21S6, 550 First Avenue, New York, NY 10016. Address e-mail to:

© 2003 Lippincott Williams & Wilkins, Inc.