People who use
antiretroviral regimens containing efavirenz (Sustiva or Stocrin, also
in the Atripla coformulation) were
not at higher risk for impaired neurocognitive function, either overall or when
looking at specific functional domains, researchers reported on Thursday at the 20th International AIDS Conference in Melbourne.
is widely recommended for first-line HIV treatment, has a well-known
association with neuropsychiatric side-effects such as insomnia, vivid dreams,
hallucinations, dizziness and poor concentration. An American study reported last year found that suicidal thoughts and death by suicide
were more common among people who started treatment with efavirenz, though the
overall risk was low. The association between efavirenz and neurocognitive
impairment, such as problems with thinking and memory, is controversial and
prior studies have yielded conflicting results.
Andrea Antinori and
colleagues at the National Institute for Infectious Disease in Rome
conducted a retrospective, cross-sectional analysis of neurocognitive function
among HIV-positive people on combination antiretroviral therapy, aiming to
determine if those taking efavirenz-based regimens had worse performance.
The analysis included 859
participants at a single centre in Italy during 2000-2013. Nearly 80% were men,
the mean age was 46, and they had a median 13 years of education. The current
median CD4 T-cell count was approximately 480 cells/mm3, but the
nadir or lowest-ever level was just 190 cells/mm3, indicating substantial
All participants were on
ART with 69% having undetectable or very low viral load -- a proportion
Antinori noted reflects reality outside clinical trials. With regard to
potential cofactors for cognitive impairment, 18% had a history of injecting
drug use, 28% were coinfected with hepatitis C virus and 10% had anaemia.
At the time of
neuropsychological assessment, one-third of participants were taking an
efavirenz-containing ART regimen. Efavirenz users were more likely to be gay
men and on average they had less advanced HIV disease, higher CD4 counts, and
greater likelihood of having undetectable viral load, but were less likely to have
a history of drug injection or hepatitis C coinfection. The study also
controlled for clinical depression and other psychiatric conditions.
assessment involved a standardized series of 14 tests that measured five different
domains of function: concentration and speed of mental processing, mental
flexibility, memory, fine motor function and visual-spatial ability.
calculated Z-scores, or standardised values, for global neurocognitive
functioning and for each cognitive domain. Participants were classified as
having neurocognitive impairment if they scored more than one standard
deviation below average on at least two tests, or more than two standard deviations
below average on a single test.
The prevalence of neurocognitive
impairment did not differ significantly between people taking or not taking
efavirenz. Among participants on efavirenz, 32% were classified as impaired,
compared to 40% of those not using the drug -- not a statistically significant
difference. No increased risk of impairment was seen based on efavirenz use in
any of the specific cognitive domains.
In a univariate or
single-factor analysis, current efavirenz use was associated with a decreased risk of neurocognitive
impairment (odds ratio [OR] 0.71, or 29% lower risk), but this was no longer
significant after controlling for confounding factors (OR 1.02). Antinori
explained that this effect was probably due to the fact that efavirenz users
were less likely to have cofactors for cognitive impairment.
In a multivariate
analysis, the factors independently associated with increased likelihood of
neurocognitive impairment were older age, HIV disease severity, injection drug
use and hepatitis C coinfection. Higher education level and having a current CD4
count above 500/mm3 at
the time of assessment appeared to have a protective effect. Nadir CD4 count,
however, had no notable effect.
"In this large case series, efavirenz
exposure was not associated with an increased risk of statistical," the
researchers concluded. "Even though confounding by indication may play a
role, and reverse causality cannot be ruled out, our results suggest that presence of neurocognitive impairment among
persons treated with efavirenz-based combination ART is not more common than in
people not treated with efavirenz."
With the expiration of its patent protection, less
expensive generic versions of efavirenz will soon become available. Some
experts have suggested that efavirenz should no longer be considered a
preferred treatment option as newer drugs are more effective and better
tolerated. But efavirenz remains a safe and effective choice for many people,
and this study shows that neurocognitive problems are not a concern for those
who are able to tolerate the drug.