ARB Use Linked to Suicide in New Study, but Is the Finding Real?

A new study is raising concerns about the risk of suicide among patients treated with angiotensin receptor blockers (ARBs).  

In an analysis of patients from several Canadian databases, the study showed that those exposed to ARBs had a significantly increased risk of suicide compared with those treated with ACE inhibitors (OR 1.63; 95% CI 1.33-2.00). This association remained after excluding patients with a history of self-harm (OR 1.60; 95% CI 1.29-1.98).

Led by Muhammad Mamdani, PharmD (St. Michael’s Hospital, Toronto, Canada), the researchers caution that the results are preliminary and need to be confirmed in future studies. Moreover, the mechanism in which ARBs or ACE inhibitors might lead to differential risks of suicide are not known. Nonetheless, given the prevalence of ARB use and severity of the adverse effect, the researchers state in their paper published October 16, 2019, in JAMA Open Network that “clinicians may opt for preferential use of ACE inhibitors over ARBs where possible,” since both drug classes have similar efficacy in treating the same conditions.

The analysis of administrative claims data on patients treated with ARBs and ACE inhibitors from 1995 to 2015 perturbs Eugene Yang, MD (University of Washington School of Medicine, Seattle), a member of the American College of Cardiology’s Prevention of Cardiovascular Disease Council. He said his worry, however, is not the potential risk of suicide with ARB use but rather that the study may alarm patients and physicians, and possibly lead to the discontinuation of medical therapy based on unfounded claims.

“I think there is a lot of concern for patients potentially taking these medications,” he told TCTMD. “To me, this creates hysteria in the public to have studies such as this that are not strongly scientifically validated. It could lead to significant downstream effects, particularly for providers who are now going to have to field calls from patients concerned about this particular study.”

To me, this creates hysteria in the public to have studies such as this that are not strongly scientifically validated. Eugene Yang

Yang said that any decision to change prescribing based on this observational study is premature. Right now, even recommendations for those with a history of mental illness to switch to an ACE inhibitor is unwarranted given the many unknowns. “This is not going to compel me to change my use of the medication,” he said.

In an editorial, Ira Katz, MD, PhD (Philadelphia VA Medical Center, PA), calls the findings “important,” but also stresses that they require replication from other data sets. One problem is that within the current system, which prizes innovation, there are limited incentives for replicating research. Still, given the importance of preventing suicide and the number of patients exposed to ARBs, there is a “need to encourage additional studies and to translate the combined findings into guidance about prescribing,” writes Katz. 

Observational Study of Canadian Patients

For the population-based nested case-control study, the researchers defined cases as patients who died by suicide within 100 days of receiving a prescription for either an ACE inhibitor or ARB. For each case, four controls were matched by age, sex, previous diagnosis of hypertension, and diabetes. Over the 18-year study, 964 cases were matched to 3,856 controls.

The majority of cases and controls were men, and the median age was 76 years. Cases were more likely than controls to have histories of alcohol abuse, anxiety or sleep disorders, affective disorder, and other mental health conditions. Cases were also more likely than controls to be taking antidepressants, benzodiazepines, and mood stabilizers. The most commonly used ACE inhibitors were ramipril and enalapril, while the most popular ARBs were valsartan, telmisartan, and candesartan.

Among the suicide cases, 260 of people were taking ARBs and 704 were taking an ACE inhibitor. Among the controls, 741 were exposed to ARBs and 3,115 were exposed to ACE inhibitors.

For Yang, the study lacks sufficient detail to make any strong conclusions about the risk of suicide among those taking ARBs. For example, there were less than 50 suicides per year with ARB use over the study period, but the researchers provide no details about the annualized rates or absolute risk of suicide with ARB use.

“This odds ratio looks extremely worrisome, but in the context of the limited number of suicide events, it’s probably a very, very small incremental risk, if there is any,” he said. “It does a disservice to not provide the absolute rate of events relative to time. It looks much more worrisome when it’s expressed in this format with only the odds ratio instead of an absolute percentage difference in risk. It doesn’t give us the frame of reference to understand how significant a problem this is.”

The question is whether ARB use is just a bystander. It doesn’t prove causality. Sripal Bangalore

Similarly, Sripal Bangalore, MD (NYU Langone Medical Center, New York, NY), said he doesn’t exactly know what to make of the new findings. Nearly 10 years ago, a meta-analysis published in Lancet Oncology raised concerns that ARB use was associated with an increased risk of new cancers, particularly lung cancer. “That created a big media splash, and everybody was looking at their own data,” Bangalore reminded TCTMD.

Those concerns were eventually laid to rest after several subsequent analyses, including a large-scale network meta-analysis conducted by Bangalore, found no evidence of cancer in patients treated with the drugs. “It’s very difficult to control for unmeasured confounders,” said Bangalore, referring to analyses derived from administrative claims. “The question is whether ARB use is just a bystander. It doesn’t prove causality.”

Like Yang, Bangalore said the new study won’t compel him to alter how he treats patients.

In their paper, Mamdani and colleagues suggest there is a possible pathophysiological mechanism linking ARB use with mental health issues, noting that higher levels of angiotensin II could lead to increases in the peptide substance P and heightened hypothalamic-pituitary-adrenal axis activity, which could provoke stress and anxiety. To TCTMD, Yang said that while there may be a conceptual framework for the association, without further details it’s difficult to tease out the extent of the problem.

“We have been burned many times, and we need to be careful,” said Yang, referring to side effects that turn up over time. “There are potential risks and we need to think about them, but the problem is that this [study] may to lead to behaviors that will really impact patient practice and clinical care. That’s what I’m most concerned about.”