Efavirenz is a commonly used antiretroviral drug that causes neurologic side effects in more than 50% of patients.

To characterize efavirenz-associated neurologic symptoms in a randomized, controlled study of initial antiretroviral treatment.

Substudy of a randomized, double-blind, controlled trial of combination antiretroviral regimens (A5095) that was performed between March 2001 and January 2002.

Multicenter academic clinical trial units.

HIV-infected patients who were initiating therapy in the context of a controlled trial.

Neuropsychological performance measures, including the Digit Symbol Substitution Test and the Trail Making Test (Parts A and B); symptom questionnaires; standardized assessments of sleep quality, anxiety, and depression; and efavirenz plasma concentrations.

Twenty of 303 (6.6%) enrolled participants prematurely discontinued the study. Neuropsychological performance improved in both groups over time without significant differences between patients who were receiving efavirenz and those who were not. The efavirenz group experienced more neurologic symptoms at week 1 (P < 0.001) but not at weeks 4, 12, or 24. A sleep index revealed that participants receiving efavirenz had more “bad dreams” during the first week of therapy (P = 0.038). No significant changes in anxiety or depressed mood were noted. Changes in efavirenz-associated neurologic symptoms were correlated to efavirenz plasma concentrations at week 1 but not at later time points. Twelve (6%) patients receiving efavirenz stopped taking the drug before the end of the study because of central nervous system symptoms.

Participant selection may have been biased in favor of patients with fewer psychiatric complications. The study design permitted substitution of a new drug in place of efavirenz in cases of treatment-limiting toxicity.

In a large controlled trial, efavirenz use was associated with neurologic symptoms distinct from depression and anxiety that began early in therapy but resolved by week 4. Improvement in neuropsychological performance was comparable in patients who were receiving efavirenz and those who were not.

*For members of the A5097s Study Team, see the Appendix.