Can Depression Raise the Risk of Sudden Cardiac Death in Women? Interview with William Whang, MD, MS

Dr. Whang is located at Columbia University Medical Center in New York. In this interview, he describes the results of the Nurses’ Health Study. Tell us about the Nurses’ Health Study. When and why was it started? How many patients does it involve? What is the age range? The main Nurses’ Health Study began in 1976 with approximately 121,701 female registered nurses aged 30-55 years. These women have been followed up every 2 to 4 years with extensive questionnaires about their medical history, coronary heart disease risk factors, and lifestyle factors. A large amount of research has been generated from the members of this cohort by investigators at Harvard Medical School, where the study was conceived with funding from the National Institutes of Health. Is the Nurses’ Health Study ongoing? It certainly is. There are multiple investigators who perform research in the Nurses’ Health Study. There are genetic substudies, an asthma substudy, and a diabetes study, to name a few, so people from a wide range of disciplines perform research in this cohort. What were your findings? How did the use of antidepressants affect the risk of sudden cardiac death (SCD)? Our main finding was that in this group of women without heart disease, worse depressive symptoms were associated with a higher risk of fatal cardiac outcome. This risk was largely mediated through risk factors for coronary heart disease such as hypertension, diabetes, high cholesterol and smoking. The biggest clinical implication was that the management of coronary heart disease risk factors might be especially important among women with depression. However, our study also raised a question about the mechanism by which depression is associated with worse cardiac prognosis. A prior study that we did in patients with defibrillators showed a higher risk of appropriate ICD shock among patients who had worse symptoms of depression (Journal of the American College of Cardiology, 2005). From the Nurses’ Health Study, we found that when we looked at variables for clinical depression in the category of women who had worse symptoms or who used antidepressant medications, there was a strong association not only with coronary heart disease death but also with SCD. Our findings are consistent with the hypothesis that one mechanism by which depression worsens cardiac prognosis is through sudden cardiac death. Delving a little deeper into our findings, when we looked separately at the worse depressive symptoms and antidepressant use, it seemed there was a specific association between antidepressant use and sudden cardiac death. I think most likely this is due to antidepressant use identifying those women with particularly severe depression. However, I also think it is an issue worth studying further in terms of any possible cardiovascular effects of antidepressant medications. What percentage of patients in this study taking antidepressants experienced sudden cardiac death? The risk was extremely low. Among the women who were taking antidepressants, there were 46 sudden cardiac deaths per 100,000 person-years of observation. Therefore, the risk was still low, even though the relative risk was higher in that group of women who were taking antidepressant medication. Has depression and its risk of SCD been studied in men as well? In 2006 there was a case-control study from Seattle that found an elevated risk of cardiac arrest associated with clinical depression; the results were published in the Archives of Internal Medicine in 2006, and showed that the risk was higher in both men and women. What should people know about the use of antidepressants in regards to this study? Who is most at risk? We would not recommend, based on this study, that anyone stop their antidepressant medication. Large randomized trials, such as the SADHART Trial, showed that the selective serotonin reuptake inhibitor (SSRI) sertraline was well tolerated from a cardiac standpoint. However, one limitation of the randomized trials that have been done is that they are not large enough to detect a rare event such as sudden cardiac death. It is an issue still worth studying, perhaps with a larger randomized trial, but the only randomized data that we know about so far indicates that antidepressant medications (SSRIs) are well tolerated from a cardiac standpoint. Is there a certain kind of antidepressant that most often triggers sudden cardiac death? We did not find a difference between selective serotonin reuptake inhibitors and other types of antidepressant medications. Most of the women in the study were taking selective serotonin reuptake inhibitors and not other kinds such as tricyclic antidepressants, which are already known to be associated with arrhythmia. What are the possible limitations of this study? Because our study was an observational one, it is difficult to account for reverse causality (i.e., people who are already sick may be depressed because they are sick, and they are also at risk of fatal events because they are sick; rather than they experience depression first and then become at risk for events because they are depressed). We tried to minimize the chance of reverse causality by excluding women who had cardiovascular disease at the start of the analysis. Another limitation is the fact that we did not have a measure of adherence to medications in our study, since depression might worsen prognosis through behavioral factors such as reduced medication adherence. What makes these findings unique? Our study is relatively large compared to any other study done on depression and sudden cardiac death. In addition, this prospective study has had a long follow-up period, there was standard documentation and analysis of endpoints, and an updated measurement of coronary heart disease risk factors and depressive symptoms. Therefore, I believe the biggest strength is the size of the study and length of the follow up. Will you be involved in further research in this area? I would certainly like to be; this is a particular research interest of mine, and I think that there is much more to learn about the interaction between psychosocial factors and the risk of arrhythmia.