Prevalence of clinical symptoms associated with highly active antiretroviral therapy in the Women's Interagency HIV Study.

BACKGROUND: The extended use of antiretroviral drugs among human immunodeficiency virus (HIV)-seropositive individuals underscores the need for a comprehensive evaluation of therapy-associated clinical symptoms. METHODS: Beginning in April 2000, 364 HIV-seronegative and 1256 HIV-seropositive women enrolled in a multicenter cohort study reported clinical symptoms that included abdominal pain, diarrhea, anorexia, nausea and/or vomiting, myalgias, fatigue, fever, body fat redistribution, dizziness, headaches, paresthesias, xerostomia, nephrolithiasis, and rash. We examined the prevalence of symptoms with respect to HIV infection and the use of highly active antiretroviral therapy (HAART), using data-correlation models. RESULTS: In the 6 months before a study visit, 49% of HIV-seronegative women, 67% of HIV-seropositive women not receiving therapy, and 69% of HIV-seropositive women receiving HAART reported any clinical symptom. The odds ratios (ORs) for reporting any symptom were 1.4 (95% confidence interval [CI], 1.1-1.8) for women who changed HAART regimens and 0.9 (95% CI, 0.7-1.1) for women reporting stable HAART use, compared with those reporting no therapy use. Significant findings (P<.05) for particular symptoms were an increased odds of diarrhea, nausea and/or vomiting, body fat redistribution, myalgias, and paresthesias, when data for women who changed HAART regimens were compared with those for women not receiving therapy. The OR for reporting any symptom was 1.5 (95% CI, 1.2-1.9) for women who switched HAART regimens and 1.6 (95% CI, 1.3-1.9) for women who discontinued HAART, compared with those reporting stable HAART use. CONCLUSIONS: Our findings confirm the high prevalence of clinical symptoms among HIV-seropositive women who changed HAART regimens. The high prevalence of symptoms among HIV-seronegative women and HIV-seropositive women not receiving therapy demonstrates that caution should be used when attributing the occurrence of symptoms entirely to HAART.