1. Home>
2. Specialties>
3. HIV/AIDS Clinical Care>
4. Summary and Comment

Studies Disagree on Frequency of Late CNS Side Effects from Efavirenz

Spanish researchers found that high plasma efavirenz levels correlated with CNS symptoms even after weeks or months of treatment. However, in a substudy of an ACTG trial, late neuropsychiatric symptoms were no more common among patients receiving efavirenz than among others.

High plasma efavirenz concentrations and cytochrome P450 2B6 genotype are risk factors for neuropsychiatric problems during the first few weeks of efavirenz therapy. In most patients, these problems are self-limited, but some patients complain of persistent CNS side effects, and others develop them only after weeks or months of treatment. How common are late CNS side effects, and are elevated plasma efavirenz levels still to blame in these cases? Two recent studies addressed these issues.

In the first, Spanish researchers correlated plasma efavirenz levels with neuropsychiatric symptoms in 17 patients who had been receiving the drug for a median of 18 months as part of a stable antiviral regimen. During the 18-month, open-label study, drug levels were checked several times, and patients were interviewed every 3 months about such problems as difficulty concentrating, depression, anxiety, and sleep disturbances. Overall, neuropsychiatric symptoms ascribed to efavirenz appeared in 10 patients and were severe enough in 4 that the drug was stopped. Mean plasma drug levels were significantly higher in patients with symptoms than in those without (5.10 vs. 2.79 µg/mL). A level of 2.74 µg/mL or higher (determined a median of 12 hours after dosing) could predict CNS toxicity with 91% sensitivity and 72% specificity. A multivariate analysis excluded body weight as a significant independent predictor of toxicity.

The second study involved extensive neuropsychiatric evaluations and symptom monitoring in a subset of patients enrolled in ACTG 5095, which had randomized treatment-naive subjects to receive one of three regimens — AZT/3TC + efavirenz, AZT/3TC/abacavir, or AZT/3TC/abacavir + efavirenz — in a blinded fashion. Evaluations occurred at baseline and at weeks 1, 4, 12, and 24. Overall, neuropsychiatric performance improved similarly in all groups over time. Efavirenz recipients complained of more symptoms (primarily bad dreams) at week 1 than other subjects did, but no differences were seen at later evaluations. A small but significant correlation between lower neuropsychiatric performance and high trough efavirenz level was found at weeks 4 and 12, but not at week 24. After the first week of treatment, no correlation could be established between subjective CNS symptoms and plasma efavirenz trough levels. In all, only 12 of 200 (6%) subjects receiving efavirenz requested a medication change because of CNS side effects.

Comment: An open-label study suggests that elevated plasma efavirenz levels continue to predispose to CNS side effects after months or even years of treatment in many patients, whereas a larger, blinded trial suggests that late CNS side effects seldom, if ever, occur. What is the clinician to do? The truth probably lies somewhere in between: Many late CNS symptoms ascribed to efavirenz might be unrelated to the drug, but it is certainly possible that high drug levels could continue to cause problems in some patients. The next question is whether these side effects can be managed by manipulating the dose of efavirenz. Many patients would probably do well on dosages less than the recommended 600 mg/day, given that most plasma drug levels in the Spanish study were well above the minimum therapeutic level of 1.0 µg/mL.

— Abigail Zuger, MD

Published in AIDS Clinical Care December 14, 2005

Citation(s):

Gutiérrez F et al. Prediction of neuropsychiatric adverse events associated with long-term efavirenz therapy, using plasma drug level monitoring. Clin Infect Dis 2005 Dec 1; 41:1648-53.

• Medline abstract (Free)

Clifford DB et al. Impact of efavirenz on neuropsychological performance and symptoms in HIV-infected individuals. Ann Intern Med 2005 Nov 15; 143:714-21.

• Original article (Subscription may be required)
• Medline abstract (Free)