Initial HIV treatment with a widely used medication was associated with a small increase in the risk of suicidality in patients, researchers reported.

An analysis of four large randomized trials of efavirenz (Sustiva) showed that patients taking the drug had more than double the risk of suicidality than those who did not, according to Camlin Tierney, PhD, of the Harvard School of Public Health, and colleagues.

But the actual numbers remained small - just 62 events among more than 5,300 patients followed for a median of 96 weeks, Tierney and colleagues reported in the July 1 issue of Annals of Internal Medicine.

Suicidality, which was defined as suicidal thoughts or attempted or completed suicide, “occurred uncommonly but with increased frequency in patients treated with efavirenz,” the researchers reported.

Given that efavirenz is widely used, “the observed increased risk is clinically relevant,” they argued.

The issue has bothered clinicians for years. Efavirenz has known neurological effects early in treatment, often manifested in vivid and disturbing dreams but including depression and anxiety.

Indeed, the package insert for the drug warns of “rare but serious” psychiatric events including suicide, the researchers noted, although the warning cautions that a “causal relationship cannot be determined from postmarketing reports.”

On the other hand, they argued, the data in this analysis comes from randomized trials, which “increases the likelihood that these results represent a causal relationship between efavirenz and suicidality.”

The finding will not be “a surprise or a shock” to physicians treating people with HIV, commented Carl Fichtenbaum, MD, of the University of Cincinnati.

But the researchers have “put together a large body of data that really documents this very well,” he told MedPage Today.

“Oftentimes we don’t have the opportunity to look at adverse effects of medications because most of the research is really driven by industry, and it’s not in their best interests to look at adverse effects,” he added.

The drug is a non-nucleoside reverse transcriptase inhibitor and is one of the preferred agents recommended for initial HIV treatment by the International Antiviral Society-USA (IAS-USA).

The IAS-USA is expected to issue new guidelines next month, at the International AIDS Conference in Melbourne, but the preferred position of efavirenz is not likely to change.

Despite its preferred position, the relationship between efavirenz and suicidality is “not well-defined,” Tierney and colleagues noted.

To help clarify the issue, they looked at participant-level data from four large trials that compared HIV treatment regimens with or without efavirenz. Some details of the study were presented at the 2013 ID Week meeting.

The primary outcome was time to suicidality, but a secondary outcome was attempted or completed suicide.

In the efavirenz arms, there were 47 reported “suicidality” events in 5,817 person-years of follow-up, for a rate of 8.08 events per 1,000 person-years.

In the efavirenz-free arms, there were 15 such events in 4,099 person-years of follow-up, for a rate of 3.66 per 1,000 person-years.

Those numbers led to a hazard ratio for suicidality of 2.28 (95% CI 1.27-4.10, P=0.006).

Interestingly, the effect was stronger in the first weeks of therapy, consistent with the observation that the neurological effects of efavirenz occur early and wane over time.

Tierney and colleagues found that the HR for suicidality was 3.69 in the first 24 weeks of therapy and 1.54 thereafter; the first reached significance but the latter did not.

The pattern for attempted and complete suicide was similar—17 events in the efavirenz arms and five in the non-efavirenz arms—but the difference between the arms did not reach significance, the researchers reported.

On the other hand, eight of the nine suicides were in efavirenz arms, they found.

The findings suggest some important clinical implications, Fichtenbaum said. “Those who were previous injection drug users or those who have a history of psychiatric problems are, in particular, at higher risk,” he said. “Therefore, you have to think about whether this is the right medication when there are other choices available.” Tierney and colleagues cautioned that the studies did not use a standardized questionnaire about suicidality or depression, and psychiatric or suicidal history might have been downplayed by participants or not disclosed at all. Also, three of the four trials were open-label, so suicidal ideation might have been susceptible to reporting bias, although suicide itself was a postmarketing experience listed in the efavirenz prescribing information before the studies were conducted. Fichtenbaum said investigators might have been more alert to psychiatric problems in the efavirenz arms of those three studies and conversely might have been less conscious of the issue in the other arms. It’s also possible, Tierney and colleagues noted, that patients with a psychiatric history might not have been recruited because of worry over potential adverse events. ### Primary source: Annals of Internal Medciine