Home Search Abstracts Browse Sessions Program Committee E-mail Abstract Author View Session

Session 106 Poster Abstracts
Non-Mycobacterial Opportunistic Infections
Tuesday, 1:30 - 3:30 pm
Poster Hall


774    
Association of Herpes Zoster Incidence with HIV Infection and HAART Use in the Women's Interagency HIV Study
M J Glesby*1, D R Hoover2, T Tan3, Q Shi3, W Gao4, J DeHovitz5, A French6, T Maurer7, J Ru8, M Young9, and K Anastos4
1Cornell Univ, New York, NY, USA; 2Rutgers Univ., Piscataway, NJ, USA; 3Data Solutions, Bronx, NY, USA; 4Montefiore Med Ctr., Bronx, NY, USA; 5State Univ. of New York Downstate, Brooklyn, USA; 6Rush Univ. Med. Coll., Chicago, IL, USA; 7Univ. of California, San Francisco, USA; 8Univ. of Southern California, Los Angeles, USA; and 9Georgetown Univ., Washington, DC, USA

Background:  Herpes zoster (HZ) occurs at a wide range of CD4 counts in HIV-infected patients. While some studies found increased HZ incidence shortly after initiating HAART, the long-term effect of HAART on HZ is unknown.  We hypothesized that the risk of self-reported HZ in HIV-infected women, even with longer-term use of HAART, would remain higher than HIV-uninfected women in the Women's Interagency HIV Study (WIHS) cohort.

Methods:  The WIHS is a prospective cohort study of HIV-infected and uninfected high-risk women followed at 6-month intervals (semesters).  Beginning at the 2nd visit (4/95-8/96), we calculated the probability of reporting HZ at any study semester by HIV serostatus and CD4 stratum among HIV-infected women.  HIV seroconverters were excluded. Generalized estimating equations assessed the effect of age, race, log CD8, CD4, nadir pre-HAART CD4, HIV RNA, and HAART (modeled in an intent-to-treat fashion) on HZ incidence, adjusted for time between visits. 

Results:  We studied 489 HIV-uninfected women (4392 visits) and 2321 HIV-infected women (22,657 visits), of whom 1832 initiated HAART (18,265 visits).  The probability of HZ was 0.14% for HIV-uninfected women.  For HIV-infected women, the probability of HZ was substantially higher, particularly at lower CD4 counts:  1.2% for CD4 >750, 1.7% for CD4 500 to 749, 2.5% for CD4 350 to 499, 3.2% for 200 to 349, and 4.2% for CD4 <200 cells/mm3.  In multivariate analysis including all HIV-infected women, CD4 (odds ratio [OR] 0.90 per 100 cells/mm3, 95% confidence interval [CI] 0.84 to 0.94, p <0.0001) and log HIV RNA (OR 1.10 per log10 copies/mL, CI: 1.00 to 1.23, p = 0.06) were associated with HZ.  After adjusting for the effects of HAART on CD4 and HIV RNA, HAART use itself was not associated with HZ (OR 1.05, OR 0.79 to 1.39, p = 0.72).  Among HAART users, in a model adjusted for time on HAART, time between visits, and log HIV RNA, only CD4 was associated with HZ (OR 0.89 per 100 cells/mm3, CI:  0.83 to 0.95, p = 0.008).  Log CD8, race, age, and nadir CD4 were not statistically significant in these multivariate models.

Conclusions:  In HIV-infected women, HAART use was not associated with an increase in HZ incidence.  Higher CD4 was protective against HZ among all HIV-infected women and the subset receiving HAART.  Even women with CD4 >750, however, are at greatly increased risk of HZ compared to HIV-uninfected women.

Keywords: herpes zoster; women; antiretroviral therapy