Women With HIV Can Safely Use Most Contraceptive Methods
With few exceptions, women with HIV who decide to prevent or delay pregnancy can safely use almost any contraceptive method. Providers can help women with HIV choose and use a method that suits their needs and preferences in much the same way that they counsel other women. (For guidance specific to women with HIV, see Table 1.)
Current WHO Guidance Gives Women With HIV a Choice of Many Methods
|
Current guidance from WHO indicates that virtually all methods are safe for nearly every woman with HIV.
|
|
Having HIV, AIDS, or using ARVs poses no limitations on use of hormonal methods such as oral contraceptive pills (OCs), injectable contraceptives, and implants. Condoms, of course, have a special role, helping to prevent both pregnancy and STIs when they are used consistently and correctly. By following the standard precautions, health care personnel can provide all methods that require procedures—injectables, implants, IUDS, vasectomy, and female sterilization—without fears that they will become infected with HIV (see Web box, “Test Yourself: Safety Precautions and Infection Prevention in the Workplace”).
Most women with HIV can use IUDs. Some have questioned the safety of using the intrauterine device (IUD) among women with HIV. They have expressed concern that pelvic inflammatory disease (PID) (infection of the upper reproductive tract, usually caused by gonorrhea or chlamydia) might be more common in IUD users with HIV than in IUD users without HIV (121). Regardless of HIV status, it is not advisable to insert an IUD in any woman who has gonorrhea or chlamydia. Because there is a higher rate of PID in the first 20 days after IUD insertion than after the first 20 days (63), there is concern that the insertion procedure could introduce these disease organisms from the lower reproductive tract to the upper (248).
Special concern about women with HIV has proved unfounded, however. Evidence indicates that PID is not significantly more common among IUD users with HIV than among IUD users who are not infected with HIV (144, 198).
 |
|
A counselor talks to a pregnant woman with HIV in São Paulo, Brazil. Providers can help women think about their postpartum family planning options early in pregnancy.
(Photo: Sean Sprague/SpraguePhoto.com)
|
In 2004 the World Health Organization updated its guidance, based on recent research, and now advises that women with HIV can generally start using either a copper-bearing IUD or a hormonal IUD (248). Specific guidance includes:
- Women with HIV who do not have AIDS can generally have copper-bearing and hormonal IUDs inserted.
- Women with AIDS who are on ARVs and are clinically well generally also can have the IUD inserted.
- IUD insertion usually is not recommended for women who have AIDS and are not on ARVs, however. The IUD also is not usually recommended for women who are using ARVs but are not clinically well.
- If an IUD user becomes infected with HIV or if an IUD user with HIV develops AIDS, the IUD generally does not need to be removed. She should be monitored for signs of PID.
Female sterilization and vasectomy are safe for couples with HIV. Couples who want a permanent method of contraception can choose female sterilization or vasectomy, regardless of whether one or both of them have HIV. Women and men who are infected with HIV, including those who have AIDS, can safely undergo female sterilization or vasectomy. In some cases sterilization should be delayed. For example, women or men who have acute AIDS-related illness may have to wait until their condition is resolved before they can undergo the procedures. For people with AIDS special arrangements should be made to perform the procedure in a setting with a qualified provider who can carefully assess the specific person’s condition, including the need for general anesthesia, with appropriate equipment and support (99, 248). While some women and men with HIV find sterilization or vasectomy a good choice, providers need to be careful to avoid putting pressure on any client to have a permanent procedure, and informed choice must be assured (45).
Avoid spermicides. Spermicides containing nonoxynol-9 (N-9), and, by assumption, other spermicides that work in a similar manner, do not protect against HIV infection or STIs, and in fact, they may increase the risk of HIV infection in women who use these products very frequently, such as several times a day (11). Therefore, spermicides are not recommended for women at high risk of HIV infection or who have HIV infection. The concern about spermicides is that they may increase susceptibility to HIV infection or another strain of HIV. Re-infection with another strain of HIV may accelerate the progression of HIV disease (77).
The increased risk may occur because N-9 disrupts the membranes of cells in the lining of the vagina, possibly making entry easier for infectious organisms (255). Women who have multiple daily acts of intercourse should be advised to choose another method of contraception (226). Some folk medicine practices, such as douching with lime juice, may also damage the vaginal lining and should be discouraged (173).
Emergency contraception is safe for women with HIV. Emergency contraception can help prevent pregnancy after unprotected intercourse. It is an important option for all women, including women with HIV. Women can take emergency contraceptive pills (ECPs) up to five days after unprotected intercourse, although they are more effective the sooner they are taken after intercourse (249). If taken within three days (72 hours) after intercourse, ECPs reduce the risk of pregnancy by at least 75% (219). There are no data available on interactions between ECPs and ARVs. It is thought, however, that ARVs will not reduce the effectiveness of ECPs because ECPs contain higher doses of hormones than daily oral contraceptives. There is currently no evidence to justify increasing the ECP dosage for women on ARVs. Emergency contraception may be administered vaginally if the hormones taken orally cause nausea and vomiting (51, 140).
Many women, including women with HIV, lack sufficient knowledge about emergency contraception. Focus-group discussions among women with HIV in Australia, India, Kenya, South Africa, and Thailand found that many are not aware of the option (42). Providers can advise or remind women with HIV that emergency contraception can help prevent pregnancy in case of unprotected intercourse—for example, if a condom breaks or slips or is not used, if a woman starts her pack of OCs three days late or more or she forgets three or more pills in the first week, or if her IUD is expelled. Providers can give ECPs to clients to take home in advance in case the need arises. ECPs do not help to prevent STIs including HIV infection.
Women With HIV Have Special Questions About Hormonal Methods and IUDs
To help answer these questions, providers need up-to-date information on a number of topics (see companion INFO Reports, “Women and HIV: Questions Answered”). While research on many of these topics is limited, current knowledge is sufficient to support the guidance. Many of the concerns involve hormonal methods and are theoretical at this stage. Hormonal methods remain an important and effective contraceptive option for women with HIV.
- Acquisition. Hormonal methods and IUDs do not appear to increase most women’s risk of becoming infected with HIV if exposed to the virus (see below).
- Progression. Some studies, but not others, suggest that hormonal contraceptives may affect factors that influence the speed of progression of HIV disease (see below).
- Infectivity. Some studies, but not others, suggest that women with HIV could be more likely to transmit HIV if they use certain hormonal contraceptive methods (see below).
- Side effects. Side effects of hormonal contraceptives and IUDs do not appear to be different or more frequent in women with HIV than in uninfected women (see below).
- Drug interactions. Some ARVs may reduce the amount of hormone in the blood, but whether this decreases contraceptive effectiveness is not known. Hormonal contraceptives do not appear to reduce the effectiveness of ARVs (see below).
Acquisition: For Most, No Additional Risk With Hormonal Methods, IUDs
A woman who does not have HIV may ask a provider if hormonal methods or IUDs increase her chances of getting HIV. The most carefully conducted studies conclude that hormonal methods do not increase the risk of acquiring HIV among women in the general population (108, 143, 150). The largest and most rigorous study, involving over 4,500 women in Uganda and Zimbabwe, found that women using combined oral contraceptives (COCs) or the progestin-only injectable depo-medroxyprogesterone acetate (DMPA) and reporting no condom use were not more likely to become infected with HIV than users of other, nonhormonal contraceptives (143). Similarly, a study in South Africa found that the numbers of new cases of HIV were similar among women using either progestin-only injectables or COCs and among women not using any hormonal method, after adjusting for differences in sexual risk behaviors and the presence of STIs (150). Limited evidence also suggests that women using the copper-bearing IUD are not at greater risk of acquiring HIV (105, 128, 141, 197).
Among populations at high risk of HIV exposure, such as sex workers, some studies find that hormonal contraception increases the risk of HIV acquisition (116, 128). For example, sex workers in Kenya using COCs or DMPA had a 1.5 times and 1.8 times greater risk, respectively, of acquiring HIV than sex workers who were not using these methods, after adjusting for condom use and number of sexual partners (116).
Progression: Could Hormonal Methods Speed Up Disease?
Some studies, but not others, find that hormonal contraceptives may affect factors that speed the progression of HIV disease. A recent, carefully conducted study among 186 women in Uganda and Zimbabwe found no association between viral set point and use of either the DMPA or COCs at the time of HIV acquisition (142). The viral load “set point” refers to the point at which the amount of virus in a newly infected person levels off after the immune system initially attacks infected cells. Both a higher viral load set point and, to a lesser degree, infection with multiple subtypes of HIV, have been found to predict faster progression of the natural course of HIV infection (77, 115, 134, 160, 202). In contrast, a study of 161 sex workers in Mombasa, Kenya, found that women using DMPA at the time of HIV acquisition had a higher viral load set point, on average, and women using OCs or DMPA were more likely to be infected with multiple subtypes of HIV than women who used no contraceptive method at the time of infection (6).
Similarly, among women with established HIV infection, there is limited and conflicting evidence regarding whether starting a hormonal contraceptive affects disease progression. An analysis of U.S. data found no association between contraceptive use and changes in viral load over time among 177 women with established HIV infection who started OCs, DMPA, or Norplant® implants (23). A study of postpartum women with HIV in Kenya also found no significant immediate or longer-term effects of the use of OCs or DMPA on viral loads or CD4+ cell counts (see From Exposure to AIDS: The HIV Disease Continuum, for an explanation of CD4+ cells) (181). A recent randomized trial in Lusaka, Zambia, among postpartum women with HIV, however, found that progression to AIDS (as indicated by a fall in CD4+ cell counts to less than 200 cells) or death was more common among women randomly assigned to hormonal contraception than among those assigned to the copper-bearing IUD (209). A substantial number of women discontinued their randomized assigned methods, switched their contraceptive methods, or withdrew from the study. Further analysis suggests this is not likely to account for the finding, however (208).
More studies are needed to better understand whether there is a relationship between use of hormonal contraceptives and disease progression. A 2007 WHO technical meeting recommended further investigation and asked researchers currently studying cohorts of HIV-infected women to examine their data (254). At this point the evidence on disease progression is limited. Providers may want to advise women who ask about disease progression that researchers are looking into it but that there are no limitations on use of hormonal methods because of this or any other concern regarding HIV.
Infectivity: Limited and Unclear Evidence on Viral Shedding
Limited evidence suggests that women with high amounts of HIV in their genital secretions are more likely to transmit the virus to an uninfected partner during unprotected vaginal intercourse than women with low amounts of HIV in their secretions (7, 170). Evidence is conflicting as to whether COCs, DMPA, and possibly other hormonal contraceptives increase genital shedding, either indirectly by increasing a woman’s susceptibility to STIs, or directly by affecting the concentration of virus in genital secretions through shedding of HIV-infected cells from the cervix or vagina (31, 89, 109, 146, 231).
Some studies suggest that hormonal contraceptives might indirectly affect infectivity among women with HIV by increasing their susceptibility to some STIs (37, 73, 107). A study of Kenyan sex workers with HIV, after taking into account demographic characteristics and sexual behavior, found that those who used hormonal contraceptives were more likely to have chlamydia and cervicitis (117). It is thought that coinfection with an STI—that is, when a woman is infected with both HIV and another STI—can increase genital shedding of HIV, which might increase the risk of transmitting the virus to sexual partners (132).
Two studies in Kenya found that women who began using DMPA or COCs shed more HIV-infected cells from the cervix than before they used them (231) or when compared with women not using hormonal contraception (146). In contrast, studies have generally found no association between hormonal contraception and genital shedding of cell-free virus—that is, HIV that exists outside of cells within the bloodstream (109, 231). It is not fully understood whether there is a difference in infectiousness between HIV-infected cells and virus free of cells (100, 167, 193). The only available direct study of transmission, involving 156 women with HIV, found no association between the use of hormonal contraception and HIV transmission to uninfected male partners (59).
IUDs do not appear to increase the infectiousness of women with HIV. The same study that examined risk of transmission among oral contraceptive users found that copper-bearing IUDs did not increase the risk of HIV transmission from women with HIV to uninfected partners beyond the risk inherent in unprotected vaginal sex (59). The two studies that have looked at the prevalence of HIV-infected cells in the cervix found no greater shedding due to IUD use (146, 180).
While some concern about the effect of hormonal contraceptives on the risk of HIV transmission is warranted, the most important advice for women with HIV who use hormonal contraception and their uninfected partners is to continue using condoms (see Box: Dual Protection Strategies Help Prevent Pregnancy and STIs). Correct and consistent use of condoms minimizes the chances that, during sexual intercourse, the penis will come into contact with genital secretions that might contain HIV. Many factors influence how infectious a woman with HIV is, including stage of disease, whether she has a concurrent STI, and whether or not she is using ARVs.
Side Effects: Similar for Women With HIV and for Uninfected Women
 |
|
A health care worker in Chiang Mai, Thailand shows a client with HIV a wristwatch that can help her remember to take her ARV medicine on time. Clients who take ARVs and use oral contraceptive pills or injectable contraceptives should also remember to take pills or receive follow-up injections on time.
(Photo: © 2004 Melissa May, Courtesy of Photoshare)
|
Contraceptive side effects such as bleeding changes with hormonal methods or IUDs do not appear to be different in women with HIV and uninfected women. Observational studies of 41 women using implants in Thailand (213, 214) and 18 women using the hormonal levonorgestrel-releasing IUD (LNG-IUD) in Finland (89, 118) found that both methods were acceptable to women with HIV. Only a few women discontinued their method, none experienced pregnancy, and the frequency and type of side effects were similar to those experienced by women not infected with HIV in other studies. As for the copper-bearing IUD, a Kenyan study found that IUD users with HIV experienced no more PID or expulsions due to infection or pain than uninfected women in the 24 months following insertion (144). Studies of hormonal contraceptive use in the U.S. (232) and Kenya (117) and of DMPA and COCs specifically in Zambia (209) found contraceptive side effects in women with HIV to be similar in type and frequency to those among uninfected women in other studies.
Drug Interactions: Do ARVs Reduce the Effectiveness of Low-Dose Hormonal Contraceptives?
Limited evidence suggests that certain ARVs could alter blood levels of contraceptive hormones in women using low dose OCs (248). Theoretically, this could influence the effectiveness of these contraceptives. The biological basis behind this concern is that some classes of ARVs—the protease inhibitors (PIs) and the non-nucleoside reverse transcriptase inhibitors (NNRTIs)— speed up processing of hormonal contraceptives in the liver. This would lower the levels of estrogen and progestin in the blood (153, 195) (see Table 2).
Two small studies reported that the ARVs nevirapine and ritonavir could lower both estrogen and progestin levels enough to increase risk of contraceptive failure. Both studies evaluated the effect of just a single dose of COC; one pill containing 50μg of estrogen (plus a progestin) in one study (163) and one pill containing 35μg of estrogen (plus a progestin) in the other (135). No information is available on women taking a pill every day. Thus, it is not clear whether or how much contraceptive effectiveness would be reduced (153). A conclusive observational study of actual pregnancy rates among pill users would be difficult because inconsistent or incorrect pill use, which could lead to pregnancy, would be hard to distinguish from the effect of the ARV.
Despite the theoretical concern about contraceptive effectiveness, women taking ARVs still generally can use COCs. If a woman using ARVs wants to use COCs, she can be given a formulation with at least 30μg of estrogen, counseled about the importance of taking COCs every day (without missing pills), and encouraged to use condoms consistently. Correct and consistent condom use would help to make up for any decrease in effectiveness of the oral contraceptives as well as help to protect an uninfected sexual partner (195).
There is less concern that these ARVs could reduce the effectiveness of progestin-only injectable contraceptives (POIs) and implants. POIs provide high hormone levels, and, with both methods, the hormones are absorbed into the blood before they are metabolized by the liver, in contrast with OCs, which are first metabolized in the liver and then enter the bloodstream. The few studies available find that ARVs have little or no effect on hormone levels in DMPA users with HIV (30, 33, 153). At this time there are no studies on the effect of ARVs on the effectiveness of combined estrogen-progestin injectables or the progestin-only injectable NET-EN, but no effect is expected because injectable contraceptives do not pass first through the liver. Still, providers can emphasize returning on time for the next injection. This will help to ensure that hormone levels remain high enough to prevent pregnancy (195). Two studies are underway to further evaluate potential interactions between DMPA and ARVs (153).
Hormonal contraceptives do not appear to reduce the effectiveness of ARVs. A limited number of studies indicate that hormonal contraceptives do not influence the effectiveness of the ARVs tested. One study among women with HIV in the U.S. found no significant changes in the blood plasma levels of the ARVs nelfinavir, efavirenz, or nevirapine 12 weeks after women started using DMPA (33). Similarly, studies have found that use of COCs has no effect on plasma levels of the ARVs nevirapine or fortovase (131, 135). Studies in rats have found that combined injectables reduce the concentration of amprenavir by 20%. It is not known whether this effect occurs in humans (260).
|
Order this report
Download PDF
E-mail this page 
See Companion INFO Report: “Women and HIV: Questions Answered”