Recent studies have examined the experience of women and the potential for gender differences with respect to HIV progression and the acceptance, tolerance, adherence, and response regarding highly active antiretroviral therapy (HAART). Differences in CD4 cell count and viral load have not been reported in all studies. For any given CD4 cell count, women may be at a higher risk of HIV progression. Women appear to have an increased risk of progression to AIDS compared with men with the same viral load. They have lower initial viral loads than men in early-stage disease, but these catch up in advanced-stage disease. Because of depression and other psychological factors, women may be in greater need of supportive services, and this can affect the success of antiretroviral therapy. Women also have an increased risk of adverse drug reactions from HAART. Gender should be considered when prescribing therapy.
Knowledge of the natural history of HIV infection is based primarily on the studies of men, and the applicability of this information to women is unclear. In the era before HAART, evaluation of HIV disease progression focused mainly on survival. Retrospective epidemiologic studies have reported shorter survival periods and an increased relative risk of death for women with AIDS, compared with the relative risk of death for men with AIDS. This difference was believed to be due to the social issue of access to care, as opposed to biologic factors related to gender.
The majority of women with AIDS in the United States are unemployed, and 83% live in households with incomes below $10,000 per year. Only 14% are currently married, compared with 50% of all women in the United States aged 15 to 44 years who are married. Twenty-three percent of HIV-infected women live alone, 2% live in various facilities, and 1% are homeless. Approximately 50% have at least one child younger than 15 years. Similar to other population groups with AIDS in the United States, the majority of women with AIDS are from minority racial and ethnic groups, with African Americans accounting for 57%; Latinas, 20%; and whites, 23%.
In the pre-HAART era, women with symptomatic HIV disease had significantly higher levels of impairment in activities of daily living and instrumental activities of daily living, compared with HIV-infected men, in regard to personal care needs, informal care, and utilization of formal health care services. Despite such impairment, women did not seek care earlier in the course of disease. Use of help sources differed significantly by gender; women were more likely than men to rely on formal sources and more likely to receive help from extended family members (especially daughters). Hospital days to death, home care costs, and total cost of care were all higher for women. Both personal care needs and the structure of social support systems appear to differ by gender and may help explain observed differences in use of formal services. Before initiating HAART, women may be in greater need of support services because of their functional status impairments or gaps in the informal support systems available to them.
Socially, gender differences prevail in regard to medical access and health care. The unique social barriers to care and the psychologic, epidemiologic, and biologic aspects of HIV disease in women pose a number of challenges for individual providers as well as for the health care system. Care for women with HIV disease is often complicated by poverty as well as by the competing needs of children; therefore, innovative efforts must be made to provide both HIV care and women's health care to women in impoverished areas, in particular. Studies from the pre-HAART era found that women with AIDS received fewer services than men with AIDS. In addition, men under medical care were up to three times more likely to be offered zidovudine [AZT, Retrovir] monotherapy than were women, despite similar disease severity, insurance status, race, and injection drug use. Among those who were not drug users, men were twice as likely as women to receive either zidovudine or Pneumocystis carinii pneumonia [PCP] prophylaxis.
Estrogens and progestins appear to have immunoregulatory effects through production of cytokines. Luteinizing hormone [LH] may increase the percentage of circulating CD4 cells, while follicle-stimulating hormone [FSH] may decrease the percentage of CD4 cells but increase the percentage of CD8 cells. Combined oral contraceptive pill use was found to be associated with a trend toward a lower absolute CD4 cell count. High serum prolactin levels were found to be associated with symptomatic versus asymptomatic HIV infection, regardless of gender, CD4 cell count, or degree of viral suppression.
In a prospective study, nearly two-thirds of HIV-infected women reported irregular periods, and 15% noted spotting or abnormal bleeding. However, there was no statistical difference in these symptoms compared with those of HIV negative controls. In another prospective study requiring completion of monthly diaries by 802 HIV-infected women and 273 HIV negative women, infection was associated with very short menstrual cycles (less than 18 days). Yet, HIV status did not correlate with mean length or variability of standard cycles; in general, the possibility of having short cycles was only slightly increased. In a group of ovulating women in whom HIV RNA levels were measured weekly, the median HIV RNA level fell by 0.16 log10 from the early follicular phase to the midluteal phase. Women who did not ovulate had no change in median HIV RNA levels.
In the pre-HAART era, lower CD4 cell counts were reported during pregnancy, with a significant increase before and after delivery in healthy and HIV-infected women. Among HIV-infected women, the decline in CD4 cell counts during pregnancy was similar to that of noninfected women, but prepartum [before delivery] levels were increased and postpartum [after delivery] levels were reduced. In a prospective cohort study that compared pregnant HIV-infected women with nonpregnant HIV-infected women during follow-up, the rate of any AIDS-defining event was higher in pregnant women but did not reach statistical significance. Considering CD4 cell counts before conception, acceleration of disease progression is inconsistent among HIV-infected women who become pregnant during follow-up. In a prospective European study, pregnancy did not affect CD4 cell count decline after seroconversion.
When compared with HIV-seropositive men who abuse drugs, HIV-seropositive women who abuse drugs exhibited significantly poorer overall nutritional status (defined by low prealbumin levels) in the pre-HAART era. In addition, women had low retinol binding protein levels and low selenium levels. In women with CD4 cell counts below 200 CD4 cells/mm3, significantly lower levels of vitamin A, vitamin E, and selenium were observed, compared with the levels found in men. Such nutritional deficiencies may contribute to HIV progression.
The absolute number of CD4 cells indicates the extent of immune system damage that has already occurred in an HIV-infected person. The decline rate of CD4 T-lymphocyte count can vary but normally averages 40 to 80 CD4 cells/mm3 a year in untreated homosexual men. However, some patients may have relatively stable CD4 cell counts for years, while others may experience a rapid decline over a period of several months. In a cohort of Zambian (untreated) HIV-seropositive women, median annual CD4 cell decline was significantly higher than in Zambian men.
The absolute CD4 cell count and percentage were significantly higher in HIV-seronegative women than in heterosexual, HIV-seronegative men. At baseline evaluation, HIV-seropositive women have higher absolute CD4 cell counts and higher CD4:CD8 ratios than do men. This difference may persist for up to five years after HIV infection. Women also have higher CD4 cell counts at AIDS onset than men. Therefore, for any given CD4 cell count, women may be at a higher risk for HIV progression than men.
In men, there is an inverse correlation between plasma HIV RNA level and CD4 cell counts. However, at any CD4 cell count, plasma HIV RNA concentrations show wide interindividual variation. Those who display higher steady-state concentrations of plasma HIV RNA are at greater risk for disease progression.
Among HIV-infected children aged six years or older, the baseline geometric mean HIV RNA level was 13,261 copies/mL for girls versus 53,680 copies/mL for boys (P=.04). Among children aged nine years and older, repeated measures analysis showed a trend toward lower geometric mean HIV RNA values for girls than for boys: 8,710 copies/mL versus 30,902 copies/mL (P=.06). Yet, among children younger than six years, no gender difference in HIV RNA levels was observed. These results suggest that gender differences in HIV RNA levels may be mediated by factors associated with sexual maturation.
In a cohort of injection drug users in the pre-HAART era, HIV-seropositive women were found to have significantly lower HIV RNA levels both at baseline and at a later follow-up visit compared with men. Median viral load values for women were approximately half of those values for men, a finding also seen in other studies. The differences were consistent by branched chain assay [bDNA], polymerase chain reaction [PCR] assay, and quantitative microculture technique after controlling for CD4 cell count. HIV RNA levels were also consistently lower for women across various CD4 cell count strata (0 to 200, 201 to 499, and greater than 500 CD4 cells/mm3). Women with the same viral load values as men had a 1.6-fold higher risk of progression to AIDS.
A nested, case-controlled, longitudinal study of HIV-1 seroconverters also showed that women have an initial median viral load level that is much lower than that of men: 14,918 copies/mL versus 148,354 copies/mL (P=.001). Women experienced a much greater viral load increase over time (0.24 log per year versus 0.003 log per year, P=.002), so that after five years, the median viral load values became equivalent. These results are corroborated by other studies that show women with lower HIV RNA values in early-stage disease but equivalent HIV RNA values in advanced-stage disease.
The increased magnitude of HIV RNA levels was associated with an increased likelihood of isolating high-risk, oncogenic [cancer-causing] human papillomavirus [HPV] genotypes from cervical specimens. The incidence of abnormal Papanicolaou [Pap] test results was 81% in women with such isolates. This observation is similar to the increased risk of opportunistic infection in men with high viral loads.
When used in concert with viral load determinations, enumeration of CD4 cells enhances the accuracy with which the risk of disease progression and death in men can be predicted. CD4 cell count was highly predictive of survival in a cohort of HIV-infected women, even in the presence of a low viral load. Both CD4 cell count and HIV RNA level were predictive of survival. While lower viral loads and higher CD4 cell counts were also seen in the Swiss HIV Cohort Study, there were no gender differences in disease progression when CD4 cell counts and HIV RNA levels were controlled in Cox regression analyses.
Not all cohorts have observed gender differences in patients' CD4 cell counts and viral loads. Within the Johns Hopkins primary care inner-city population, women had the same CD4:HIV RNA ratio as did men. Investigators from the Community Programs for Clinical Research on AIDS (CPCRA) who conducted community-based studies also found no significant gender differences in viral load among patients enrolled in six clinical trials.
In Australia, women were shown to be significantly less likely than men to be receiving antiretroviral therapy, to be using prophylactic treatments, and to be optimistic about the benefits of therapy. Women felt isolated from both community and medical responses to HIV infection. Women with dependent children were more likely to accept antiretroviral therapy.
While most studies would indicate that the efficacy of the various reverse transcriptase inhibitors and/or protease inhibitors (PIs) is equal among women and men, some authors have observed lower rates of viral suppression. In New Orleans, women had an OR [odds ratio] of 1.55 (95% CI, 1.07 to 2.25) of having a viral load greater than 400 copies/mL [i.e., were much more likely to have detectable HIV in blood tests than men]. Women were also significantly more likely to have their HAART regimen changed and to experience depression during the study. In a study of 249 women receiving HAART, with [a median] baseline CD4 cell count of 182 CD4 cells/mm3 and viral load of 64,317 copies/mL, only 34% achieved durable viral suppression. These women did experience a significantly greater increase in CD4 cell count.
While women can respond to HAART, the toxicity may be different. In a retrospective review of HAART, women were found to have a statistically increased risk of adverse drug reactions compared with men (P=.008), with occurrence in 25% of men and 37% of women. A study of adverse effects of the PI ritonavir [Norvir] in 90 women compared with 996 men showed that women were more likely to have nausea, vomiting, malaise, fatigue, and numbness and tingling around the mouth but were less likely to have diarrhea. Women receiving the PI nelfinavir [Viracept] were more likely to have stomach pain, itching, and rash but less likely to have diarrhea.
Women can also experience body habitus changes and serum lipid abnormalities while receiving HAART (a single PI with two nucleoside reverse transcriptase inhibitors [NRTIs]). Changes included increase in abdominal size and breast size, weight gain of greater than 5 kg (11 lb), peripheral fat wasting, buttock fat wasting, and development of cervicodorsal fat pad. HAART was also associated with significant increases in total cholesterol, apolipoprotein B, and high-density lipoprotein cholesterol levels. When compared with men, women had significantly more fat accumulation but not fat depletion or the combination of fat accumulation and depletion.
Studies of adherence are gaining increased relevance, because it is apparent that HAART works only if the patient takes it, and medication-taking behavior is influenced by many factors. The association of adherence with viral load and CD4 T-lymphocyte counts was striking when adherence to a nelfinavir-based regimen was measured with electronic pill monitors. Adherence clearly determined the outcome of treatment with HAART. In British Columbia, using prescription refills as the measure for adherence to HAART, men had a nearly two-fold better rate of adherence than women. Within a multicenter, prospective women's cohort, levels of self-reported adherence were quite high (74% of women reporting greater than 95% adherence), yet only 39% of these women achieved undetectable viral loads. Significantly higher levels of crack cocaine use and depression were observed in women with poor and moderate self-reported adherence.
Anemia associated with hypermenorrhea developed in four of ten women who began treatment with HAART at a clinic in Denmark. All four were receiving ritonavir, and symptoms resolved in three of the women following substitution of a different PI. The women had normal menstrual function before therapy and did not experience intramenstrual [between cycles or "periods"] bleeding.
Several antiretroviral agents are known to interfere with the metabolism of oral contraceptives. Nevirapine [Viramune] induces those enzymes that metabolize ethinyl estradiol, leading to a clinically relevant decrease in the area under the concentration-time curve (AUC) of oral contraceptives that contain this agent. Similarly, the glucuronosyltransferase enzyme system [a family of enzymes that metabolize drugs] is induced by ritonavir and nelfinavir, also leading to a 40% and 47% decrease, respectively, in the AUC of ethinyl estradiol in women taking oral contraceptives. Indinavir [Crixivan] is associated with a 24% increase in levels of ethinyl estradiol and a 26% increase in levels of norethindrone in patients taking oral contraceptives, but the changes are not felt to be clinically significant.
There are many gender differences that could potentially affect HIV disease progression. Access to health care services or to antiretroviral medications may differ, as may the toxicity of antiretroviral drugs. Psychological factors, such as depression, may be different among HIV-infected men and women. Social and environmental factors, such as child rearing responsibilities, may interfere with the ability to access care and may affect disease progression. CD4 cell counts appear to be higher for women than for men who have the same degree of illness. In early stages of HIV disease, the level of HIV RNA in plasma appears to be lower but rises more quickly and eventually catches up to the levels for men in advanced stages of the disease.
Guidelines for initiation of antiretroviral therapy have been based on data derived primarily from men. As the question of when to initiate antiretroviral therapy is addressed, gender-specific recommendations may be appropriate. Further studies are needed to better understand gender differences and the effects on HIV disease progression, treatment, and response.
This article was reprinted from The AIDS Reader (2001;11:29-33), Copyright © 2001, Cliggott Publishing Company, a division of SCP Communications, Inc., and appeared subsequently on Medscape, www.medscape.com.
(Note: Publication information can be found at the end of the Abstract paragraph of the article on Medscape.)
Ross G. Hewitt, M.D., is a clinical associate professor in the department of medicine, State University of New York (SUNY) at Buffalo, and medical director of immunodeficiency services at Erie County Medical Center in Buffalo.
Nader Parsa, Ph.D., is an instructor in the department of social and preventive medicine at SUNY Buffalo.
Lawrence Gugino, M.D., is a clinical associate professor in the department of obstetrics and gynecology at SUNY Buffalo, and the women's HIV program director at Erie County Medical Center in Buffalo.
Chaisson, R.E. and others. Race, sex, drug use, and progression of HIV disease. New England Journal of Medicine 333(12): 751-756. September 1995.
Ehmann, W.C. and others. Relationship of CD4 lymphocytes to survival in a cohort of hemophiliacs with HIV. Journal of AIDS 7(10): 1095-1098. October 1994.
Fahey, J.L. and others. The prognostic value of cellular and serologic markers in infection with human immunodeficiency virus type 1. New England Journal of Medicine 322(3): 166-172. January 1990.
Hessol, N.A. and others. Progression of human immunodeficiency virus type 1 (HIV-1) infection among homosexual men in hepatitis B vaccine trial cohorts in Amsterdam, New York City and San Francisco, 1978-1991. American Journal of Epidemiology 139(11): 1077-1087. 1994.
Lemp, S.F. and others. Survival for women and men with AIDS. Journal of Infectious Diseases 166(1): 74-79. July 1992.
MacDonell, K.B. and others. Predicting progression to AIDS: combined usefulness of CD4 lymphocyte counts and p24 antigenemia. American Journal of Medicine 89(6): 706-712. December 1990.
Melnick, S.L. and others. Survival and disease progression according to gender of patients with HIV infection. Journal of the American Medical Association 272(24): 1915-1921. December 1994.
Morlat, P. and others. Women and HIV infection: a cohort study of 483 HIV-infected women in Bordeaux, France, 1985-1991. AIDS 6(10): 1187-1193. October 1992.
Munoz, A. and others. Acquired immunodeficiency syndrome (AIDS)-free time after human immunodeficiency virus type 1 HIV-1 seroconversion in homosexual men. American Journal of Epidemiology 130(3): 530-539. September 1989.
Sheppard, H.W. and others. The characterization of non-progressors: long term HIV-1 infection with stable CD4+ T-cell levels. AIDS 7(9): 1159-1166. September 1993.
Back to the SFAF BETA Spring 2001 contents page.