Older patients with dementia face a time-limited increased risk of heart attack if they are placed on antipsychotic medications, researchers reported.

In a retrospective cohort study, community-dwelling seniors placed on antipsychotic drugs had a significantly increased risk of heart attack in the first month after they began taking the medications, with a hazard ratio of 2.19 (95% CI 1.11 to 4.32) compared with dementia patients not taking antipsychotic medications, according to Yola Moride, PhD, of the Université de Montréal in Montreal, and colleagues.

But the risk gradually trailed off over the next year, Moride and colleagues reported online in Archives of Internal Medicine.

While antipsychotic medications are not indicated in dementia, they are commonly prescribed to help control some of the side effects, such as aggression and hallucinations, the researchers noted.

One clinical implication of the current study is that doctors should be cautious about using the medications, according to Sudeep Gill, MD, and Dallas Seitz, MD, both of Queen’s University in Kingston, Ontario.

In an accompanying comment article, Gill and Seitz argued that physicians should “use other techniques when available, such as environmental and behavioral strategies, to keep these patients safe and engaged.”

Although the underlying mechanisms remain unclear, several studies have found an increased risk of cardiovascular events and death when demented patients are treated with antipsychotics. Regulatory agencies have issued warnings about the issue.

But the effect of antipsychotic drugs on the risk of acute myocardial infarction (MI) hasn’t been studied in depth, Moride and colleagues noted.

To help fill the gap, they turned to data from the public prescription drug and medical services programs in the Canadian province of Quebec, which cover more than 98% of the province’s citizens who are 65 and older.

For this analysis, the researchers studied 37,138 people 66 and older who lived in the community and had become new users of cholinesterase inhibitors between Jan. 1, 2000, and Dec. 31, 2009, but were not immediately using antipsychotics.

Of those, Moride and colleagues reported, 10,969 (or 29.5%) started antipsychotic drugs during the study period.

For Cox proportional hazards modeling, they were matched with random sample of the members of the cohort who did not start the medications. Follow-up covered the year after starting antipsychotics.

Overall, the researchers found, 1.3% of those who started the antipsychotics had a heart attack within a year of beginning to take the medications.

In addition to the elevated risk in the first month, the risk was also elevated for the following 60 days, at a hazard ratio of 1.62, but the 95% CI missed statistical significance, ranging from 0.99 to 2.65.

The risk became lower as time went on: HR 1.36 for the first 90 days and HR 1.15 for the first 365 days. In both cases, the confidence intervals crossed unity, meaning the risk was not significantly different from the group that was not taking the drugs.

The researchers also conducted a self-controlled case series study among all 804 people who had a heart attack after starting antipsychotics, with a median follow-up of 47 months.

For that analysis, a participant was defined as having acute risk in the first 30 days after starting the drugs, and mid-term risk for the period from 31 to 60 and 61 to 90 days. The rest of the time they used the medications was defined as the residual risk period, and if they stopped the drugs they were in a withdrawal risk period for the next 90 days.

The reference period for the analysis - the “unexposed” period - was any time before a participant began taking the drugs or more than 90 days after stopping them.

The researchers found incidence rate ratios (IRR), compared with the unexposed periods, of 1.78 for the acute risk period and 1.67 for the 31- to 60-day period, with 95% CI from 1.26 to 2.52 and from 1.09 to 2.56, respectively.

The IRR for later periods and the withdrawal period did not reach significance.

Taken together, the findings suggest a risk increase that is “highest at the beginning of treatment and seems to decrease thereafter,” Moride and colleagues argued.

But because of the frequency with which antipsychotics are used in the demented population, the increased risk of heart attack “may have a major public health effect, which highlights the need for communicating such risk and for close monitoring of patients during the first weeks of treatment,” they concluded.

They cautioned that the drug plan records may not have captured cases in which antipsychotics were started in a secondary care setting, which could have resulted in an underestimation of the hazard ratios.

Gill and Seitz commented that the underlying mortality increase associated with the antipsychotic drugs remains an open question whose answer may advance the understanding of cardiovascular disease in general.

“Important lessons about the pathogenesis of cardiovascular disease may underlie the observed association between antipsychotic drug use and acute myocardial infarction,” they argued.

Primary source: Archives of Internal Medicine
Source reference: Pariente A, et al “Antipsychotic use and myocardial infarction in older patients with treated dementia” Arch Intern Med 2012; DOI: 10.1001/archinternmed.2012.28.

Additional source: Archives of Internal Medicine
Source reference: Gill SG , Seitz DP “From association to mechanism: Observational studies can help us understand fundamental drug actions” Arch Intern Med 2012; DOI: 10.1001/archinternmed.2012.682.