Hormonal Contraception and HIV: An Update

By Dr. Charles Morrison and Kim Best
August 2004

See Also:

Worldwide, about 150 million women use hormonal contraception. Use of hormonal contraception, particularly injectable progestins, is already high in areas of the world with high HIV prevalence and its use appears to be increasing rapidly.

Whether hormonal contraceptive use affects HIV acquisition, transmission, or disease progression continues to be an important area of research.

However, current knowledge concerning a potential relationship between hormonal contraception and HIV is insufficient to change current family planning practices. The World Health Organization's Medical Eligibility Criteria for Contraceptive Use states that women at risk of HIV infection or those whom are HIV-infected may use hormonal contraception with no restrictions.¹

There are biological and physiological reasons to believe that use of hormonal contraception could facilitate HIV acquisition.

Approximately 15 prospective studies have been conducted to consider a possible relationship between hormonal contraceptive use and HIV acquisition. About 12 of those studies have examined the relationship between oral contraceptive use and HIV acquisition; about five have addressed injectable progestin use and HIV acquisition. In contrast to cross-sectional studies, prospective studies reduce some sources of bias to study results. (For example, prospective studies can better identify when hormonal contraceptive use occurred and thus clarify the relationship between use and HIV acquisition.) Nevertheless, most of these prospective studies were not designed to directly consider the relationship between hormonal contraception and HIV acquisition; therefore, most have substantial methodological shortcomings.

Results of these studies have been mixed. The conclusions of two major recent reviews of research in this area conflict.² Some studies have shown an increased risk of HIV acquisition, while others have not. Notably, those studies that indicate increased risk tend to have been conducted among high-risk populations of women, such as sex workers.

One such study was a recently published 10-year prospective investigation of hormonal contraception and HIV acquisition conducted among sex workers in Mombasa, Kenya. This study found that women using the injectable contraceptive depot- medroxyprogesterone acetate (DMPA) and women using oral contraceptive (OC) pills were at a small to moderate increased risk (hazard ratio of 1.8 for DMPA and 1.5 for OCs) of HIV acquisition, compared with women using no hormonal contraception.³ It is important to note that if hormonal contraceptive use indeed increases the risk of HIV acquisition, that risk would be expected to be greater among women with more frequent exposure to HIV, such as sex workers.

Another prospective study of the relationship between the use of combined OCs or DMPA and HIV acquisition, which is substantially larger than the Mombasa study, is being conducted by FHI and collaborating institutions. The study, funded by the U.S. National Institute for Child Health and Human Development (NICHD), is being conducted in Uganda, Thailand, and Zimbabwe among some 6,200 HIV-negative, 18- to 35-year-old users of combined OCs, users of DMPA, and women not using hormonal contraception. Results, expected in 2005, should help clarify the question of the relationship between hormonal contraceptive use and HIV acquisition. Notably, this study has the advantage of being conducted primarily among family planning clients who are at lower risk of HIV infection and who are more similar to the majority of women who use hormonal contraception worldwide.

A theoretical concern exists that hormonal contraceptive use by HIV-infected women might increase genital tract shedding of HIV and thus increase transmission to uninfected partners and to newborns.

However, data are conflicting about such a relationship.

The only prospective study of the effect of hormonal contraceptive use on genital tract shedding of HIV, conducted among a population of family planning clients in Mombasa, Kenya, detected a significant but modest increase in cervical shedding of HIV-infected cells after initiation of various hormonal contraceptives, compared with before initiation. Yet, there was no difference in the amount of cell-free virus detected in genital secretions.⁴ While a certain degree of concern is warranted, more data are needed before a change in service delivery guidelines is made.

To further clarify the matter, FHI researchers are conducting a prospective study in Zimbabwe and Uganda, funded by NICHD, of the effect of combined OC and DMPA use on HIV genital shedding among approximately 250 women with acute or early HIV infection. The women will be compared with HIV-infected women not using hormonal contraception. Preliminary results are also expected in 2005.

Whether hormonal contraceptive use at or near the time of HIV infection affects disease progression is also of concern.

In a prospective study of HIV acquisition among 1,350 sex workers in Mombasa, Kenya -- 161 of whom were followed for a median of 34 months after being infected with HIV -- the use of DMPA at the time of infection was associated with a higher HIV viral set point. This finding suggests that use of DMPA at the time of infection may hasten HIV-related deterioration of the immune system and the natural course of HIV infection.⁵ In a subset of these HIV-infected sex workers (82 of whom used any hormonal contraception and 68 of whom did not), hormonal contraceptive use near the time of HIV acquisition was associated with infection with multiple strains of the virus.⁶ Infection with multiple strains appears to be related to a higher viral set point and faster CD4 decline, two key indicators of HIV disease progression.⁷

Since many of the studies of the possible impact of hormonal contraception on HIV disease progression have been conducted among sex workers in Kenya, and some of the findings have not been corroborated by other studies, further research among other populations of women in other geographic locations is needed before changes in service provision guidelines would be made.

Notably, hormonal contraceptives do not appear to protect against HIV or other sexually transmitted infections. Thus, while continuing to promote hormonal contraception for family planning when appropriate, providers should counsel hormonal contraceptive users who are also at risk of HIV to reduce their number of sexual partners and to use a condom correctly and consistently during each act of sexual intercourse.

References