Antidepressants Ease Depression in Patients With a Physical Illness

Megan Brooks

March 24, 2010

March 24, 2010 — Antidepressants are more effective than placebo in treating depression in people with a physical illness, according to an updated Cochrane review published online March 17 in the Cochrane Database of Systematic Reviews.

The new review supports a Cochrane review published in 2000. Several large trials assessing the efficacy of antidepressants in the setting of physical illness published since 2000 prompted the updated review, lead author Lauren Rayner of King's College London, United Kingdom, and colleagues note.

The review included 51 randomized controlled trials comparing antidepressants — mostly tricyclic antidepressants (TCAs) and selective serotonin reuptake inhibitors (SSRIs) — vs placebos in a total of 3603 adults with major depression; adjustment disorder or dysthymia; and a physical illnesses such as stroke, HIV/AIDS, Parkinson's disease, cancer, and chronic obstructive pulmonary disease. Only trials in which depression was a primary outcome were included.

In 25 studies (1674 patients) contributing data on short-term response (6 - 8 weeks), the odds of response were significantly greater with antidepressants vs placebo (odds ratio [OR], 2.33; 95% confidence interval [CI], 1.80 - 3.00; P < .00001). At 6 to 8 weeks, the number needed to treat with an antidepressant for one to benefit was 6.

The superiority of antidepressants to placebo was also apparent in the 6 studies (713 patients) contributing long-term data (> 18 weeks; OR, 2.13; 95% CI, 1.31 - 3.47; P = .002), the reviewers report.

They caution, however, against drawing firm conclusions from the long-term efficacy data because of the "small numbers of patients included in the > 18 week analysis." They also say "it is likely that publication and reporting biases exaggerated the effect sizes obtained."

In subgroup analysis, both TCAs and SSRIs were superior to placebo, suggesting, the researchers say, that TCAs, "which are often considered inappropriate in physically ill patients," are as effective and acceptable to patients as SSRIs. They emphasize, however, that "there are no grounds to recommend one antidepressant over another on the basis of this review."

As expected, patients receiving antidepressants were more likely to experience adverse effects including dry mouth (particularly with TCAs) and sexual dysfunction (particularly with SSRIs).

At 6 to 8 weeks of treatment, fewer patients receiving placebo withdrew from treatment vs patients receiving an antidepressant (OR, 1.32; 95% CI, 1.00 - 1.75; P = .05; 22 studies, 1555 patients). No difference in dropout rates was evident at other time points, and there was no evidence of a difference in dropout rates between TCAs and SSRIs.

Based on their review, the authors recommend that antidepressants "be considered for the treatment of depression in physical illness. In the absence of data on comparative efficacy and acceptability in particular physical illnesses, choices of antidepressant should take account of patients' preferences, symptoms, and possible interactions with other medications."

The authors urge caution in generalizing the findings to all physically ill depressed persons, as the most unwell patients — the very frail or disabled patients with a severe physical illness — were not adequately represented in the trials reviewed.

Cochrane Database Syst Rev. Published online March 17, 2010. Abstract