ARBs Linked to Increased Suicide Risk, Experts Skeptical

October 18, 2019

UPDATED with comments October 21, 2019 // Angiotensin receptor blockers (ARBs), which are extensively used to treat hypertension, heart failure, chronic kidney disease, and diabetes, have been linked to an increased risk for suicide in new research. However, leading cardiovascular experts are skeptical.

"In our study, we found that patients taking ARBs were more likely to die from suicide than similar patients taking ACE inhibitors (ACEIs), with an odds ratio of 1.63. This means that for every 100 suicides in any population taking ACEIs, there would be 163 suicides in a similar population taking ARBs," lead investigator Muhammad Mamdani, PharmD, St. Michael's Hospital, Toronto, Canada, told Medscape Medical News.

The study, which was supported by the Ontario Ministry of Health and by the Institute for Clinical Evaluative Sciences, was published online October 16 in JAMA Network Open.

Err on the Side of Caution?

The investigators conducted their study after a previous study found an increase in the risk for suicide with ARBs.

"That study was not intended to specifically examine the association between ACEIs, ARBs, and suicide and was limited by a small number of patients being exposed to ARBs, but we were curious about the results," said Mamdani, who receives honoraria from Allergan, Novo Nordisk, and Celgene.

"We looked into possible mechanisms for this observation and found that ARBs may increase angiotensin II levels in the brain more than ACEIs, and patients with a gene polymorphism producing higher levels of angiotensin II in the brain were more likely to have mood disorders. So we wanted to investigate this further," he added.

However, leading cardiovascular experts told Medscape Medical News that they did not consider the study findings convincing or relevant to clinical practice. Pointing out the many limitations of epidemiologic studies ― in particular, the inability to control for all confounding factors ― they said they would continue prescribing ARBs as usual.

Mamdani acknowledged that although the results need to be validated, "most doctors would say ACEIs and ARBs are interchangeable ― they have very similar effects on cardiovascular endpoints ― so why not err on the side of caution and use an ACE inhibitor instead of an ARB if possible, especially if a patient has a history of mental health issues?"

Higher Suicide Risk

For the study, Mamdani and colleagues examined the association between suicide and exposure to ARBs and ACEIs in a nested case-control study among residents of Ontario, Canada, who were aged 66 years or older from 1995 to 2015.

They used population-based databases that document health service utilization and outcomes, including prescription drugs dispensed, inpatient hospitalization data, physician visits, basic demographic information, data on the prevalence of hypertension and diabetes, and information on deaths, including suicides.

Case patients were defined as persons who died by suicide within 100 days of having received a prescription for an ACEI or an ARB, excluding individuals who received drugs from both of these classes during the same period.

The date of suicide served as the index date. For each case patient, up to four control persons were matched for age, sex, and previous diagnoses of hypertension and diabetes.

Control persons were assigned the same index date as their matched case patient and were required to have been exposed to an ACEI or an ARB within 100 days of the index date.

During the 18-year study period, 964 case patients were matched to 3856 control persons. Most case patients and control persons were men (80%), and the median age was 76 years.

As expected, comorbidities and psychotropic drug use were more common among case patients. Case patients were more likely than control persons to have histories of alcohol abuse, anxiety or sleep disorders, psychoses, agitation and related disorders, affective disorder, and other mental health conditions.

Case patients were also more likely than control persons to use antidepressants, antipsychotics, benzodiazepines, and mood stabilizers.

The most commonly used ACEIs were ramipril (38.8%) and enalapril (15.0%). The most commonly used ARBs were valsartan (16.7%), telmisartan (16.7%), and candesartan (16.7%).

Results showed that 26% of case patients were exposed to ARBs and that 18.4% were exposed to ACEIs. Among control persons, 74.0% were exposed to ARBs and 81.6% were exposed to ACEIs.

In the primary analysis, compared with exposure to ACEIs, exposure to ARBs was associated with a higher risk for suicide (adjusted odds ratio, 1.63; 95% confidence interval [CI], 1.33 – 2.00)

A sensitivity analysis that excluded individuals with a history of deliberate self-harm yielded findings that were consistent with those of the primary analysis (odds ratio, 1.60; 95% CI, 1.29 – 1.98).

"Our findings suggest a possible increased risk of suicide associated with ARBs compared with ACE-inhibitors among adults aged 66 years and older. Given their high prevalence of use, the severity of the outcome, and the similar efficacy of these drug classes in treating the same conditions, clinicians may opt for preferential use of ACE-inhibitors over ARBs where possible," the investigators write.

CVD Experts Sceptical

Three leading cardiovascular experts said they were skeptical. "It is unfortunate that papers like this get published, and, like many others – such as coffee linked to cancer – the associations are normally untrue," said George Bakris, MD, in commenting on the findings for Medscape Medical News.

"There are many problems with such epidemiologic studies," noted Bakris, who is professor of medicine and director of the Comprehensive Hypertension Center, University of Chicago Medicine.

He pointed out that although the investigators admit that case patients were more likely to have received a psychiatric diagnosis, "they were not formally or prospectively assessed to elevate the magnitude of these conditions."

In addition, said Bakris, ARBS are used to treat conditions associated with depression, such as declining kidney function, yet "no data are provided on this association.

"Unless there is evidence that ARBs interfere with the mechanism affecting serotonin in the brain and somehow ACEIs do not also speaks against this association. I have raised only a few problematic issues with the findings, and I suspect, like the coffee-and-cancer scare, this too will be questioned and disproven.

"ARBs have been used extensively in billions of people since the 1980s without such issues, and I plan to continue to prescribe them," Bakris said.

Business as Usual

"These data would not affect my prescribing habits at this time," Wllliam White, MD, told Medscape Medical News.

"In case control studies, there may be clinical differences in those prescribed ARBs vs ACE inhibitors that may be difficult to ascertain, such as differences in mental health conditions, other medications that could affect mood, and other comorbidities that are difficult to deal with, such as heart failure, COPD, neurologic disorders," said White, from the University of Connecticut School of Medicine. He is a past president of the American Society of Hypertension.

"The authors are well known in pharmacoepidemiology research, and their database certainly has its strengths, as it's been developed for quite some time now," he added.

Also commenting, Franz H. Messerli, MD, professor of medicine cardiology, University of Bern, Switzerland, said he was also skeptical about the finding and pointed out that ACEIs have also been associated with angioedema, which can be fatal. "How does this compare with the risk of possible ARB-associated suicide?" he asked. He noted that both of these outcomes would be exceedingly rare.

In an accompanying editorial, Ira R. Katz, MD, writes that the identification of a serious but rare adverse drug effect "could only come from real-world data" and emphasized that the study needs replication. Katz is with the Office of Mental Health and Suicide Prevention, Department of Veterans Affairs, in Philadelphia.

"The strength of the methods, the importance of preventing suicide, and the number of people exposed to ARBs all support the need to encourage additional studies and to translate the combined findings into guidance about prescribing. To ensure that this occurs in a timely manner, there is a need for incentives and structures that reward replication," he writes.

Investigators Respond

In response to concerns about his study, Mamdani said the investigators were careful to minimize confounding.

"We put in a lot of checks and balances to reduce confounding as much as possible. If we had compared ARBs to no medication, then yes, I could see there would be a high potential for bias. But everyone in our study was on either an ACE inhibitor or an ARB ― very similar drugs for very similar conditions. This really minimizes confounding," he said.

"What I think really strengthens our results is that when we just include patients who had a history of self-harm, we still saw an increased risk with ARBs over ACE inhibitors, with a similar odds ratio to the whole study," he added.

"Another sign that we have minimal confounding is that our unadjusted and adjusted odds ratios are very similar (1.64 vs 1.63). This shows that even after we adjusted for many confounders, the results were the same, indicating that the patient populations were quite homogeneous to start with," said Mamdani.

He questioned how cardiovascular experts can be sure there is no increase in suicide risk with the ARBs, "as this is the first study to have really looked at this in a meaningful way.

"I am not suggesting everyone is switched from an ARB to an ACE inhibitor on the basis of these results. But I would say these data should give us something to think about, and if my patient had a history of mental health issues, then this would make me lean towards prescribing an ACE inhibitor rather than an ARB," he said.

JAMA Netw Open. Published online October 16, 2019. Full text, Editorial

For more Medscape Neurology news, join us on Facebook and Twitter


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: