Antidepressants Linked to Increased Risk for Death, Stroke in Postmenopausal Women

Pam Harrison

December 22, 2009

December 22, 2009 — Postmenopausal women taking either a tricyclic antidepressant (TCA) or a selective serotonin-reuptake inhibitor (SSRI) appear to be at increased risk for all-cause mortality, and SSRI users seem to be at increased risk for hemorrhagic and fatal stroke, although the absolute event risks are low, according to an analysis from the Women's Health Initiative (WHI) study.

In an article published in the December issue of the Archives of Internal Medicine, Jordan Smoller, MD, from Massachusetts General Hospital, Boston, and colleagues report that new use of either a TCA or an SSRI during a mean follow-up of 5.9 years was not significantly associated with an increased risk for coronary heart disease (CHD) in this large prospective cohort of postmenopausal women.

In contrast, compared with women who did not use antidepressants, "those using SSRIs had a 45% increased relative risk of incidence stroke and a 32% increased risk of death in models stratified on propensity and adjusted for multiple covariates," the investigators report. TCA use in turn was associated with a 67% higher relative risk for all-cause death (hazard ratio [HR], 1.67; 95% confidence interval [CI], 1.33 - 2.09). The TCAs also increase stroke risk, but not significantly so.

"Depression is still a much more established risk factor for cardiovascular disease (CVD) than antidepressants, so it's not as though not taking an antidepressant removes the risk because then you have untreated depression, which itself is risky," Dr. Smoller told Medscape Psychiatry. "But if a woman is concerned about taking medication, there are alternative treatments for depression, such as cognitive behavioral therapy, which can be effective."

Association Explored

For the analysis, investigators examined the association between new antidepressant use and first occurrence of CHD, fatal or nonfatal stroke, and all-cause mortality occurring after the first follow-up visit.

The first follow-up visit for women enrolled in the longitudinal cohort observational study was at 3 years; for women enrolled in 1 of the 3 overlapping clinical trials, it took place at 1 year after baseline.

A total of 136,293 postmenopausal women were involved in the WHI, and women taking no antidepressants at study entry and who had at least 1 follow-up visit were included in the analysis.

During the follow-up interval, 5496 women initiated antidepressant treatment. Antidepressant users had higher levels of several CVD risk factors than nonusers. To address potential confounders, investigators obtained a propensity score from a logistic regression model.

Annualized rates per 1000 person-years of stroke with no antidepressant use were 2.99 vs 4.16 for SSRI users. Death rates for non–antidepressant users were 7.79 per 1000 person-years and 12.77 for SSRI users. Annualized mortality rates for TCA users were 14.14 deaths per 1000 person-years.

As investigators point out, the excess risk for stroke seen in association with SSRIs was largely hemorrhagic (HR, 2.12; 95% CI, 1.10 - 4.07), whereas the risk for ischemic stroke with SSRI use did not reach statistical significance (HR, 1.21; 95% CI, 0.80 - 1.83).

In contrast, both the SSRIs and the TCAs were associated with an increased risk for all fatal strokes (HR, 2.10; 95% CI, 1.15 - 3.81; and HR, 2.56; 95% CI, 1.26 - 5.26, respectively).

Table. CV Events and Annualized Rates in All Cohorts (Observational Study and Clinical Trials)

AD Status Participants (n) CHD Events (n) CHD Annualized Rate/1000 Person-Years Stroke Events (n) Annualized Rate/1000 Person-Years
No AD at Follow-up 130,797 2843 3.81 2232 2.99 5881 7.79
Incident SSRI 3040 73 4.73 64 4.16 201 12.77
Incident TCA 1490 41 5.18 39 4.92 114 14.14
Incident other or multiple ADs 966 26 5.38 22 4.55 66 13.42

Key Question

A key question is whether the association between antidepressant use and CV morbidity and mortality is truly related to drug exposure or to underlying differences in other CV risk factors, including depression.

As Dr. Smoller explained to Medscape Psychiatry, the propensity score used in this study did help control for risk factors that could have explained differences in CV outcomes between antidepressant users and nonusers, "but you can't fully tease out the effect that depression itself and some of these risk factors may have had vs the contribution of the medication."

He continued, "If you look at the annualized death rates in all cohorts, the absolute differences [between AD users and nonusers] is very small, so our findings have to be weighed against the risk of CV disease and mortality associated with untreated depression vs antidepressant use because we know it can have a dramatic effect on the ability of patients to enjoy life."

In an invited commentary, Christopher O'Connor, MD, and Mona Fiuzat, PharmD, from Duke University Medical Center, Durham, North Carolina, agreed that the most difficult aspect of these trials is that patients who are likely to be treated with antidepressants have additional risk factors for mortality and CV risk that are difficult to control for.

They also point out that the study cannot fully address situations in which patients begin treatment with antidepressants but do not respond to treatment, and that these patients may represent the highest-risk mortality group.

Nevertheless they indicate that the findings — in the largest cohort of women yet studied — provide "additional warning" that antidepressant use does have certain negative consequences in this particular demographic and that its risks should be considered against the potential benefits of treating depression.

The WHI program is funded by the National Heart, Lung, and Blood Institute. Dr. Smoller has served as a consultant for Eli Lilly and received honoraria from Hoffman-La Roche Inc, Enterprise Analysis Corporation and MPM Capital. He has also served on an advisory board for Roche Diagnostic Corporation. Dr. O'Connor has disclosed no relevant financial relationships.

Arch Intern Med. 2009;169:2128-2139; 2140-2141. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: