The investigators drew on 10-year data from the UK Biobank on over 220,000 adults and compared the risk of developing adverse health outcomes among those taking antidepressants with the risk among those who were not taking antidepressants.
After adjusting for preexisting risk factors, they found that 10-year antidepressant use was associated with a twofold higher risk of CHD, an almost-twofold higher risk of CVD as well as CVD mortality, a higher risk of cerebrovascular disease, and more than double the risk of all-cause mortality.
On the other hand, at 10 years, antidepressant use was associated with a 23% lower risk of developing hypertension and a 32% lower risk of diabetes.
The main culprits were mirtazapine, venlafaxine, duloxetine, and trazodone, although SSRIs were also tied to increased risk.
“Our message for clinicians is that prescribing of antidepressants in the long term may not be harm free [and] we hope that this study will help doctors and patients have more informed conversations when they weigh up the potential risks and benefits of treatments for depression,” study investigator Narinder Bansal, MD, honorary research fellow, Centre for Academic Health and Centre for Academic Primary Care, University of Bristol (England), said in a news release.
“Regardless of whether the drugs are the underlying cause of these problems, our findings emphasize the importance of proactive cardiovascular monitoring and prevention in patients who have depression and are on antidepressants, given that both have been associated with higher risks,” she added.
The study was published online in the British Journal of Psychiatry Open.
Monitoring of CVD risk ‘critical’
Antidepressants are among the most widely prescribed drugs; 70 million prescriptions were dispensed in 2018 alone, representing a doubling of prescriptions for these agents in a decade, the investigators noted. “This striking rise in prescribing is attributed to long-term treatment rather than an increased incidence of depression.”
Most trials that have assessed antidepressant efficacy have been “poorly suited to examining adverse outcomes.” One reason for this is that many of the trials are short-term studies. Since depression is “strongly associated” with CVD risk factors, “careful assessment of the long-term cardiometabolic effects of antidepressant treatment is critical.”
Moreover, information about “a wide range of prospectively measured confounders ... is needed to provide robust estimates of the risks associated with long-term antidepressant use,” the authors noted.
The researchers examined the association between antidepressant use and four cardiometabolic morbidity outcomes – diabetes, hypertension, cerebrovascular disease, and CHD. In addition, they assessed two mortality outcomes – CVD mortality and all-cause mortality. Participants were divided into cohorts on the basis of outcome of interest.
The dataset contains detailed information on socioeconomic status, demographics, anthropometric, behavioral, and biochemical risk factors, disability, and health status and is linked to datasets of primary care records and deaths.
The study included 222,121 participants whose data had been linked to primary care records during 2018 (median age of participants, 56-57 years). About half were women, and 96% were of White ethnicity.
Participants were excluded if they had been prescribed antidepressants 12 months or less before baseline, if they had previously been diagnosed for the outcome of interest, if they had been previously prescribed psychotropic drugs, if they used cardiometabolic drugs at baseline, or if they had undergone treatment with antidepressant polytherapy.
Potential confounders included age, gender, body mass index, waist/hip ratio, smoking and alcohol intake status, physical activity, parental history of outcome, biochemical and hematologic biomarkers, socioeconomic status, and long-term illness, disability, or infirmity.
Mechanism unclear
By the end of the 5- and 10-year follow-up periods, an average of 8% and 6% of participants in each cohort, respectively, had been prescribed an antidepressant. SSRIs constituted the most commonly prescribed class (80%-82%), and citalopram was the most commonly prescribed SSRI (46%-47%). Mirtazapine was the most frequently prescribed non-SSRI antidepressant (44%-46%).
At 5 years, any antidepressant use was associated with an increased risk for diabetes, CHD, and all-cause mortality, but the findings were attenuated after further adjustment for confounders. In fact, SSRIs were associated with a reduced risk of diabetes at 5 years (hazard ratio, 0.64; 95% confidence interval, 0.49-0.83).
At 10 years, SSRIs were associated with an increased risk of cerebrovascular disease, CVD mortality, and all-cause mortality; non-SSRIs were associated with an increased risk of CHD, CVD, and all-cause mortality.
On the other hand, SSRIs were associated with a decrease in risk of diabetes and hypertension at 10 years (HR, 0.68; 95% CI, 0.53-0.87; and HR, 0.77; 95% CI, 0.66-0.89, respectively).
“While we have taken into account a wide range of pre-existing risk factors for cardiovascular disease, including those that are linked to depression such as excess weight, smoking, and low physical activity, it is difficult to fully control for the effects of depression in this kind of study, partly because there is considerable variability in the recording of depression severity in primary care,” said Dr. Bansal.
“This is important because many people taking antidepressants such as mirtazapine, venlafaxine, duloxetine and trazodone may have a more severe depression. This makes it difficult to fully separate the effects of the depression from the effects of medication,” she said.
Further research “is needed to assess whether the associations we have seen are genuinely due to the drugs; and, if so, why this might be,” she added.
Strengths, limitations
Commenting on the study, Roger McIntyre, MD, professor of psychiatry and pharmacology and head of the mood disorders psychopharmacology unit at the University of Toronto,, discussed the strengths and weaknesses of the study.
Dr. Roger S. McIntyre
The UK Biobank is a “well-described, well-phenotyped dataset of good quality,” said Dr. McIntyre, chairperson and executive director of the Brain and Cognitive Discover Foundation, Toronto, who was not involved with the study. Another strength is the “impressive number of variables the database contains, which enabled the authors to go much deeper into the topics.”
A “significant limitation” is the confounding that is inherent to the disorder itself – “people with depression have a much higher intrinsic risk of CVD, [cerebrovascular disease], and cardiovascular mortality,” Dr. McIntyre noted.
The researchers did not adjust for trauma or childhood maltreatment, “which are the biggest risk factors for both depression and CVD; and drug and alcohol misuse were also not accounted for.”
Additionally, “to determine whether something is an association or potentially causative, it must satisfy the Bradford-Hill criteria,” said Dr. McIntyre. “Since we’re moving more toward using these big databases and because we depend on them to give us long-term perspectives, we would want to see coherent, compelling Bradford-Hill criteria regarding causation. If you don’t have any, that’s fine too, but then it’s important to make clear that there is no clear causative line, just an association.”
The research was funded by the National Institute of Health Research School for Primary Care Research and was supported by the NI Biomedical Research Centre at University Hospitals Bristol and Weston NHS Foundation Trust and the University of Bristol. Dr. McIntyre has received research grant support from CI/GACD/National Natural Science Foundation of China and the Milken Institute and speaker/consultation fees from numerous companies. Dr. McIntyre is a CEO of Braxia Scientific.
A version of this article first appeared on Medscape.com.