The Effect of Highly Active Antiretroviral Therapy on Dermatologic Disease in a Longitudinal Study of HIV Type 1 Infected Women

Transcription

1 MAJOR ARTICLE HIV/AIDS The Effect of Highly Active Antiretroviral Therapy on Dermatologic Disease in a Longitudinal Study of HIV Type 1 Infected Women Toby Maurer, 1 Lori K. E. Rodrigues, 1 Niloufar Ameli, 2 Nittaya Phanuphak, 9 Stephen J. Gange, 4 Jack DeHovitz, 6 Audrey L. French, 8 Marshall Glesby, 7 Carol Jordan, 5 Ann Khalsa, 3 and Nancy A. Hessol 2 Departments of 1 Dermatology and 2 Medicine, University of California, San Francisco, and 3 University of Southern California, Los Angeles, California; 4 Department of Epidemiology, Johns Hopkins University Bloomberg School of Public Health, Baltimore, and 5 Montgomery County Health and Human Services, Silver Spring, Maryland; 6 State University of New York Health Science Center at Brooklyn, and 7 Weill Medical College of Cornell University, New York, New York; 8 CORE Center/Cook County Hospital, Rush Medical College, Chicago, Illinois; and 9 Thai Red Cross AIDS Research Centre, Bangkok, Thailand The effect of highly active antiretroviral therapy (HAART) on skin diseases was evaluated in 878 human immunodeficiency virus type 1 (HIV-1) infected women in the Women s Interagency HIV Study, a multicenter prospective study. HIV-1 infected women receiving HAART were less likely to have eczema, folliculitis, tinea pedis, and xerosis than were women who had not initiated HAART, independent of CD4 + cell count. Participants who had a prior history of a nadir CD4 + cell count of!200 cells/ml and recent CD4 + cell counts of cells/ml were more likely to have eczema and xerosis than were women with a nadir CD4 + cell count of 1200 cells/ml and recent CD4 + cell counts of 1349 cells/ml. An HIV-1 RNA load of 1100,000 copies/ml was associated with increased prevalence of herpes zoster infection (odds ratio, 6.10; 95% confidence interval, ). History of injection drug use was associated with a higher prevalence of onychomycosis, tinea pedis, and xerosis. Molluscum contagiosum was more prevalent among younger women. The skin is frequently affected in persons infected with HIV-1, and a wide variety of skin diseases develop over the course of HIV-1 disease progression [1 3]. The introduction of HAART has dramatically changed the natural course of HIV-1 infection, and numerous studies have demonstrated a strong reduction in HIV-1 related morbidity and mortality [4]. Several studies re- Received 28 July 2003; accepted 22 October 2003; electronically published 30 January Financial support: National Institute of Allergy and Infectious Diseases, with supplemental funding from the National Cancer Institute, the National Institute of Child Health & Human Development, the National Institute on Drug Abuse, and the National Institute of Craniofacial and Dental Research (grants U01-AI-35004, U01-AI-31834, U01-AI-34994, U01-AI-34989, U01-HD-32632, U01-AI-34993, U01- AI-42590, M01-RR00079, and M01-RR00083). Reprints or correspondence: Dr. Toby Maurer, Dept. of Dermatology, San Francisco General Hospital, 1001 Potrero Ave., Rm. 224, San Francisco, CA (tmaurer@itsa.ucsf.edu). Clinical Infectious Diseases 2004; 38: by the Infectious Diseases Society of America. All rights reserved /2004/ $15.00 port that the pattern of skin diseases observed in HIV-1 infected individuals is altered by HAART [1, 2, 5 9]. However, the majority of these studies evaluated dermatologic disease in men [1, 3, 7], and only a limited number of studies addressed the spectrum of dermatologic disease in HIV-1 infected women [2, 6, 9, 10]. Studying women with HIV-1 infection is important, because the proportion of women with this infection continues to increase in the United States. Although women accounted for 6.7% of all incident AIDS cases in 1986, by 1999, women accounted for 23% of new AIDS diagnoses and 32% of newly reported HIV-1 diagnoses [11]. The purpose of this study was to evaluate the prevalence of common skin diseases in HIV-1 infected women before and after the introduction of HAART. Our study used data from the Women s Interagency HIV Study (WIHS), one of the largest prospective studies of HIV-1 infected women in the United States. The initial recruitment in the WIHS occurred during 1994 HIV/AIDS CID 2004:38 (15 February) 579

2 1995, before the introduction of HAART in mid We have previously reported cross-sectional data from the baseline visit, whereby we found that the most-common skin diseases in the WIHS cohort were onychomycosis, xerosis, folliculitis, acne, eczema, herpes zoster, tinea pedis, warts, molluscum contagiosum, and seborrheic dermatitis [10]. In this study, we extended our previous observations by using extensive follow-up data from the WIHS cohort to evaluate the prevalence of skin conditions during the period in which HAART was introduced. PATIENTS AND METHODS Study participants. The WIHS is an ongoing prospective study of HIV-1 infection in women that is being conducted in 6 locations in the United States: New York City (2 sites); Washington, D.C.; Chicago; southern California; and the San Francisco Bay area. The WIHS methods and baseline cohort characteristics have been described previously [11]. In brief, from October 1994 through November 1995, 2628 women (2059 of whom were HIV-1 seropositive, and 569 of whom were HIV- 1 seronegative) were enrolled in the WIHS. Women were recruited from HIV-1 primary care clinics, hospital-based programs, research studies, community outreach sites, women s support groups, drug rehabilitation programs, HIV-1 testing sites, and referrals from enrolled participants. The institutional review boards of participating institutions approved the study protocol, and written informed consent was obtained from all patients. Every 6 months, WIHS participants were interviewed using a structured questionnaire, underwent physical and gynecological examinations, and provided multiple laboratory specimens. Quantification of the HIV-1 RNA load in plasma was performed using the isothermal nucleic acid sequence based amplification method (Nuclisens; Organon Teknika) in laboratories participating in the National Institutes of Health/National Institute of Allergy and Infectious Diseases (NIH/NIAID) Virology Quality Assessment Program. Lymphocyte subsets were quantified using standard flow cytometric methods in laboratories participating in the NIH/NIAID Flow Cytometry Quality Assessment Program. At each study visit, self-reported antiretroviral use in the period since the previous visit was assessed by interviewers. Women who reported use of 1 of the following drug combinations at the time of their study visit were considered to be receiving HAART: (1) 2 nucleoside reverse-transcriptase inhibitors (NRTIs) in combination with 1 protease inhibitor (PI) or 1 nonnucleoside reverse-transcriptase inhibitor (NNRTI), (2) 1 NRTI in combination with 1 PI and 1 NNRTI, (3) ritonavir and saquinavir in combination with 1 NRTI and 0 NNRTIs, or (4) abacavir in combination with 3 NRTIs in the absence of both PIs and NNRTIs. Combinations of zidovudine and stavudine with either a PI or an NNRTI were not considered to be HAART because of their contraindication [12]. Centrally trained examiners performed skin examinations. Results of skin tests were recorded on a standardized questionnaire, and codes were used to denote location, description, and diagnosis. Invalid data, including skin disease diagnoses with uncorrelated descriptors or locations, were excluded from the study. Diagnoses were made if skin lesions were present at the time of the examination. In these analyses, we included only those HIV-1 infected women who reported that they had received HAART. Furthermore, we restricted the analysis to women for whom we had complete clinical and demographic information, which included CD4 + cell count, HIV-1 RNA load, age, race, and history of injection drug use. Participants were evaluated during a total of 9 visits over a period of 4.5 years (1 October March 1999). After 5 years, the overall retention rate in the WIHS was 81% [13]. Statistical analysis. The presence or absence of the 10 most common skin diseases was measured at each visit, and the results were used as dichotomous outcome variables. We constructed separate logistic regression random-effects models to identify risk factors for the prevalence of each of the 10 skin diseases. Random subject effects were included to account for dependent observations from the same subject. Our primary predictor of interest was patients reported initiation of HAART since their most recent study visit. We also evaluated age at baseline (in 10-year age groups), race (dichotomized as African American [reference] or non African American), baseline history of injection drug use (dichotomized as yes or no [reference]), and HIV-1 RNA load (categorized as 100,000 copies/ ml [reference] or 1100,000 copies/ml). To assess the effect of recurrent skin disease on patient outcome, we included a predictor defined as the detection of skin disease during a previous WIHS skin examination. To evaluate the association between dermatologic skin diseases and CD4 + cell count, we used 2 separate modeling strategies. In the first model, we used 3 categories of CD4 + cell counts (i.e.,!200, , and 1349 [reference] cells/ml) from the visit immediately preceding the current visit. In the second model, we categorized study participants into 1 of 5 groups according to their nadir and current CD4 + cell counts (i.e., nadir and current CD4 + cell counts of!200 and!200,!200 and ,!200 and 1349, 1199 and , and 1199 and 1349 [reference] cells/ml). Random-effects models were fitted using PROC NLMIXED software (SAS Institute) [14]. RESULTS Of the 2059 HIV-1 infected women in the cohort, 1005 (48%) had initiated HAART during the study. Of these, 10% re- 580 CID 2004:38 (15 February) HIV/AIDS

3 ported no previous receipt of any antiretroviral treatment. Data on a total of 878 HIV-1 infected women for whom complete data were available were included in the analysis; this includes data from all pre- and post-haart study visits (table 1). Onychomycosis had the highest prevalence among those who had initiated HAART since their most recent study visit (7% of patients), followed by eczema (6%), folliculitis and xerosis (5% each), acne (3%), herpes zoster and tinea pedis (2% each), and cutaneous warts, molluscum contagiosum, and seborrheic dermatitis (1% each). The results of the random-effects models that did not include nadir CD4 + cell counts are summarized in table 2. HIV-1 infected women who had initiated HAART since their most recent study visit were statistically less likely to have eczema (OR, 0.62; 95% CI, ), xerosis (OR, 0.63; 95% CI, ), folliculitis (OR, 0.53; 95% CI, ), tinea pedis (OR, 0.46; 95% CI, ), or onychomycosis (OR, 0.66; 95% CI, ) than were women who had not initiated HAART. Molluscum contagiosum and xerosis were associated with CD4 + cell counts of!200 cells/ml ( P!.05). Eczema was associated with a CD4 + cell count of!350 cells/ml. Compared with HIV-1 loads of 100,000 copies/ml, an HIV-1 load of 1100,000 copies/ml was associated with an increased prevalence of herpes zoster (OR, 5.09; 95% CI, ). No other skin diseases were statistically associated with HIV-1 load. Younger women were more likely than older women to have molluscum contagiosum (OR, 0.50; 95% CI, ; table 2). Women with a prior history of skin disease were more susceptible to have a recurrence of the same skin disease at a later study visit ( P!.01; table 2). The occurrence of certain skin conditions was also influenced by history of injection drug use. Participants with a history of injection drug use at baseline had a significantly higher prevalence of onychomycosis (OR, 1.66; 95% CI, ), tinea pedis (OR, 1.11; 95% CI, ), and xerosis (OR, 1.57; 95% CI, ) than did women with no history of injection drug use. The results of the random-effects models that included nadir CD4 + cell counts were as follows: HIV-1 infected women who had initiated HAART since their most recent study visit were statistically less likely to have eczema (OR 0.62; 95% CI, ) and xerosis (OR, 0.63; 95% CI ) than were women who had not initiated HAART. Xerosis and eczema were significantly associated with recent CD4 + cell counts of!200 cells/ml ( P!.05) and!349 cells/ml ( P!.05), respectively. Women with nadir and recent CD4 + cell counts of!200 and cells/ml, respectively, were 2 times more likely to have eczema (OR, 1.91; 95% CI, ) and 2.1 times more likely to have xerosis (OR, 1.98; 95% CI, ) than were women with recent CD4 + cell counts of 350 cells/ml but whose prior CD4 + cell counts never decreased to!200 cells/ml. The other Table 1. Demographic and clinical characteristics of 878 HIV- 1 infected participants in the Women s Interagency HIV Study. Characteristic Value Age, years (3.8) (31.8) (48.6) (15.8) Race/ethnicity African American 446 (50.8) Hispanic 243 (27.7) White 163 (18.6) Asian/Pacific Islander 7 (0.8) Native American/Alaskan native 6 (0.7) Other 13 (1.4) History of injection drug use 306 (34.8) Duration of HAART, median years (IQR) 1.5 (1 2) CD4 + cell count, median cells/ml (IQR) a 264 ( ) HIV-1 RNA load, median copies/ml (IQR) a 15,500 ( ,500) NOTE. Data are no. (%) of patients, unless otherwise indicated. IQR, interquartile range. a Before HAART initiation. skin diseases did not show a statistically significant association with nadir CD4 + cell count. DISCUSSION This cohort offered a unique opportunity to examine the effect of HAART, CD4 + cell count, and virus load on the prevalence of skin disease among women who were examined at initiation of HAART and observed prospectively. Because of the years during which follow-up occurred especially the pre-haart years we have the advantage of being able to evaluate the effect of the nadir CD4 + cell count on the prevalence of skin disease. Previous studies that assessed the effect of HAART on cutaneous disease have been retrospective surveys and have focused primarily on male subjects [1, 3, 7]. Other studies have found that persons treated with HAART have fewer cutaneous diseases [1 3, 5 9]. Recent evidence has shown that exposure to antiretroviral medication makes HIV- 1 less fit for replication [15, 16]. Although individuals receiving antiretroviral drugs may not have their immune system fully reconstituted or undergo a change in their nadir CD4 + cell count and virus load parameters, they develop fewer opportunistic infections and remain healthy. The skin may behave in a similar fashion and have fewer disease manifestations [9]. The prevalence of cutaneous warts also did not change during receipt of HAART, nor was it associated with changing CD4 + cell counts [17]. In another study, there was no difference in HIV/AIDS CID 2004:38 (15 February) 581

4 Table 2. Summary of random-effects models in which covariates were evaluated for the prevalence of dermatologic diseases among HIV-1 infected patients in the Women s Interagency HIV Study, 1 October March Type of dermatologic infection Acne Eczema Folliculitis Herpes zoster Molluscum contagiosum Covariate OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P Age (per decade a ) 0.68 ( ) ( ) ( ).77 1 ( ) ( ).03 Race/ethnicity African American 0.98 ( ) ( ) ( ) ( ) ( ).89 Other History of IDU Yes 0.78 ( ) ( ) ( ) ( ) ( ).36 No Recent HAART b Yes 0.68 ( ) ( )! ( )! ( ) ( ).31 No Prior history of disease Yes 8 ( )! ( )! ( ) ( )! ( )!.01 No 1 (reference) 1 (reference) 1 (reference) 1 (reference) HIV-1 RNA copies/ml 1100, ( ) ( ) ( )! ( )! ( ) ,000 CD4 + cells/ml! ( ) ( )! ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) NOTE. IDU, injection drug use. a Within 10-year age groups (i.e., years, years). b Use of HAART since the most recent study visit. the prevalence of verruca vulgaris among patients with earlyor late-stage HIV-1 disease [18]. A previous study examined the effect of the nadir CD4 + cell count on cutaneous warts in HIV-1 infected men and noted that the resolution of warts was also dependent on the nadir CD4 + cell count [19]. That is, in male subjects with CD4 + cell counts that decreased to!50 cells/ml, warts did not resolve as quickly as they did in male subjects with nadir CD4 + cell counts of 150 cells/ml, independent of their subsequent CD4 + cell count and HIV-1 load. The nadir CD4 + cell count may play an important role in long-term immunologic recovery for HIV-1 infected people. The degree of immunologic damage may be different in those persons with low nadir CD4 + cell counts and may subsequently predict progression to AIDS. There is evidence to suggest that HIV-1 related immunodeficiency, as reflected in the nadir CD4 + cell count, is associated with poor recovery of CD4 + cells [20, 21]. As suggested in the literature, a delay in the initiation of HAART may limit phenotypic and functional immune restoration in patients with HIV-1 infection [22]. A previous study found that herpes zoster occurred in persons with CD4 + cell counts of cells/ml, but virus loads were not evaluated in this study [23]. Another study found that herpes zoster was not associated with the duration or stage of HIV-1 disease, but again, virus loads were not included in the analyses [24]. Persons with a history of injection drug use had a higher prevalence of xerosis, onychomycosis, and tinea pedis, independent of race and age. A high prevalence of dermatophytic infections has been noted in other studies of HIV-1 infected subjects [25, 26] and has been observed in individuals with HIV-1 infection and a history of injection drug use [18]. A few limitations of this study deserve mention. Women who initiated HAART in this cohort may have been sicker than women who did not initiate HAART. We tried to control for this by examining the effect of CD4 + cell counts and virus loads independent of HAART use. This study was not able to assess severity of disease with respect to HAART use. At least 1 study has noted that the severity of seborrheic dermatitis and psoriasis changed after initiation HAART, despite no change in prevalence [18]. Prospective studies might concentrate on the course of skin disease in individuals who are receiving HAART. This study did not account for any treatments administered for particular dermatologic conditions. It may have been the dermatologic treatments, rather than HAART, which accounted for reductions in the prevalences of folliculitis, eczema, tinea 582 CID 2004:38 (15 February) HIV/AIDS

5 Type of dermatologic infection Onychomycosis Seborrheic dermatitis Tinea pedis Warts Xerosis OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P OR (95% CI) P 1.19 ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( )! ( )! ( )! ( )! ( )! ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ) ( ).16 pedis, and xerosis. However, the other dermatologic diseases examined in this study, such as acne and seborrheic dermatitis, are easily treated diseases and did not show any changes in prevalence during this study. This study only considered those participants who had initiated HAART since their most recent study visit. It did not take duration of HAART therapy into account, because of the variability of treatment duration between study visits. The duration of HAART therapy may influence the prevalence of dermatologic disease. Finally, examination of several exposure-outcome associations revealed an increased risk of false-positive findings and, therefore, that assessment of the strength of these associations and confirmation of results in other studies is particularly important. An increasing number of HIV-1 infected individuals are living longer while receiving HAART. Dermatologic diseases in HIV-1 infected individuals are common and have become an important concern. As reflected in this study, HAART has decreased the prevalence of certain skin diseases, independent of CD4 + cell count or virus load. The decrease in the prevalence of skin disease can be an important motivator among persons receiving HAART with regard to therapy adherence. Our study shows the importance of nadir CD4 + cell count in the evaluation of HIV-1 infected individuals with dermatologic disease. The course of skin disease may be different in subgroups of patients in whom the nadir CD4 + cell count is!200 cells/ml and may provide valuable prognostic information to both patient and clinician. Although the nadir CD4 + cell count may play an important role in the long term clinical picture of HIV-1 infection, there may be limitations in the use of nadir CD4 + cell counts, particularly when patients are not undergoing regular measurement of CD4 + cell count or if they are not evaluated at the time that their CD4 + cell count decreases. Understanding exactly how HAART interacts with the immune system to prevent dermatologic manifestations or to reduce their severity and/or duration is an important area of future research. In addition, as the arsenal of HAART continues to grow, more investigation is needed to determine which drugs or drug combinations have the greatest impact in the prevention of skin diseases. Acknowledgments We thank the following principal investigators involved in the Women s Interagency HIV Study (WIHS) Collaborative Study Group for gathering and processing the data used in this article: Kathryn Anastos, New York City/Bronx Consortium; Howard Minkoff, Brooklyn, NY; Mary Young, Washington DC HIV/AIDS CID 2004:38 (15 February) 583

6 Metropolitan Consortium; Ruth Greenblatt, Connie Wofsy Study Consortium of Northern California; Alexandra Levine, Los Angeles County/Southern California Consortium; Mardge Cohen, Chicago Consortium; and Alvaro Munoz, WIHS Data Coordinating Center. References 1. Hicks CB, Myers SA, Giner J. Resolution of intractable molluscum contagiosum in a human immunodeficiency virus infected patient after institution of antiretroviral therapy with ritonavir. Clin Infect Dis 1997; 24: Aftergut K, Cockerell CJ. Update on the cutaneous manifestations of HIV infection: clinical and pathologic features. Dermatol Clin 1999; 17:445 71, vii. 3. Costner M, Cockerell CJ. The changing spectrum of the cutaneous manifestations of HIV disease. Arch Dermatol 1998; 134: Centers for Disease Control and Prevention. Update: trends in AIDS incidence United States, MMWR Morb Mortal Wkly Rep 1997; 46: Fowler MG, Melnick SL, Mathieson BJ. Women and HIV: epidemiology and global overview. Obstet Gynecol Clin North Am 1997; 24: Barton JC, Buchness MR. Nongenital dermatologic disease in HIVinfected women. J Am Acad Dermatol 1999; 40: Conant MA. HIV-related skin diseases clear with combination therapy. West J Med 1998; 169: Hengge UR, Franz B, Goos M. Decline of infectious skin manifestations in the era of highly active antiretroviral therapy. AIDS 2000; 14: Calista D, Morri M, Stagno A, Boschini A. Changing morbidity of cutaneous diseases in patients with HIV after the introduction of highly active antiretroviral therapy including a protease inhibitor. Am J Clin Dermatol 2002; 3: Mirmirani P, Hessol NA, Maurer TA, et al. Prevalence and predictors of skin disease in the Women s Interagency HIV Study (WIHS). J Am Acad Dermatol 2001; 44: Barkan SE, Melnick SL, Preston-Martin S, et al. The Women s Interagency HIV Study. WIHS Collaborative Study Group. Epidemiology 1998; 9: Guidelines for the use of antiretroviral agents in HIV-infected adolescents and adults. Department of Health and Human Services and Henry J. Kaser Family Foundations living document. Available at: http: // Accessed 14 July Hessol NA, Schneider M, Greenblatt RM, et al. Retention of women enrolled in a prospective study of human immunodeficiency virus infection: impact of race, unstable housing, and use of human immunodeficiency virus therapy. Am J Epidemiol 2001; 154: SAS Institute. SAS/STAT user s guide, version 8. Cary, NC: SAS Institute, Lederman MM, Valdez H. Immune restoration with antiretroviral therapies: implications for clinical management. JAMA 2000; 284: Deeks SG. Durable HIV treatment benefit despite low-level viremia: reassessing definitions of success or failure. JAMA 2001; 286: Uthayakumar S, Nandwandi R, Drinkwater T, Nayagam AT, Darley CR. The prevalence of skin disease in HIV infection and its relationship to the degree of immunosuppression. Br J Dermatol 1997; 137: Munoz-Perez MA, Rodriguez-Pichardo A, Camacho F, Colmenero MA. Dermatological findings correlated with CD4 lymphocyte counts in a prospective 3 year study of 1161 patients with human immunodeficiency virus disease predominantly acquired through intravenous drug abuse. Br J Dermatol 1998; 139: Rodrigues LK, Baker T, Maurer T. Cutaneous warts in HIV-positive patients undergoing highly active antiretroviral therapy. Arch Dermatol 2001; 137: Torriani FJ, Freeman WR, Macdonald JC, et al. CMV retinitis recurs after stopping treatment in virological and immunological failures of potent antiretroviral therapy. AIDS 2000; 14: Kaufmann GR, Bloch M, Finlayson R, Zaunders J, Smith D, Cooper DA. The extent of HIV-1 related immunodeficiency and age predict the long-term CD4 T lymphocyte response to potent antiretroviral therapy. AIDS 2002; 16: Lange CG, Valdez H, Medvik K, Asaad R, Lederman MM. CD4 + T- lymphocyte nadir and the effect of highly active antiretroviral therapy on phenotypic and functional immune restoration in HIV-1 infection. Clin Immunol 2002; 102: Friedman-Kien AE, Lafleur FL, Gendler E, et al. Herpes zoster: a possible early clinical sign for development of acquired immunodeficiency syndrome in high-risk individuals. J Am Acad Dermatol 1986; 14: Buchbinder P, Katz MH, Hessol NA, et al. Herpes zoster and human immunodeficiency virus infection. J Infect Dis 1992; 166: Gupta AK, Taborda P, Taborda V, et al. Epidemiology and prevalence of onychomycosis in HIV-positive individuals. Int J Dermatol 2000; 39: Child FJ, Fuller LC, Higgins EM, Du Vivier AW. A study of the spectrum of skin disease occurring in a black population in south-east London. Br J Dermatol 1999; 141: CID 2004:38 (15 February) HIV/AIDS