Neglect of Attention to Reproductive Health in Women With HIV Infection

Contraceptive Use and Unintended Pregnancies in the Swiss HIV Cohort Study

K Aebi-Popp; V Mercanti; C Voide; J Nemeth; A Cusini; B Jakopp; D Nicca; M Rasi; A Bruno; A Calmy; B Martinez de Tejada


HIV Medicine. 2018;19(5):339-346. 

In This Article


This study highlights the wide variety of contraceptive methods used by HIV–positive women of reproductive age in Switzerland. Male condoms remained the most frequently used contraceptive method, whereas the use of long–acting reversible contraceptives was very uncommon. One in six women using contraceptives experienced an unwanted pregnancy, with 42% occurring while using a combined hormonal pill. The majority of women opted for a pregnancy termination. Interestingly, one–third (35.5%) of women did not use any contraception, despite many being sexually active. In Switzerland, contraceptive counselling is partially integrated in HIV care, but women have to be referred to a gynaecologist for the prescription of hormonal and long–acting contraceptives, such as IUDs and injectables. This may result in a reduced access to effective contraception as women do not always engage in routine gynaecological care[13] or they are lost to follow–up for some time, especially after having a baby.[14] Condom use was common in our cohort (71%) and reported by 97% of all women on cART. Two–thirds reported condom use for the purposes of both contraception and the prevention of sexually transmitted diseases and 9% for contraception only. However, although condom use is effective for the prevention of sexually transmitted diseases, it is not efficient for the prevention of pregnancy and has high failure rates (up to 5.4 failures in 100 episodes of use compared with a failure rate of 0.6 for implants).[15]

In resource–poor settings with a high HIV prevalence, the proportion of HIV–positive women using contraceptive methods other than condoms is approximately 13–19% in West Africa and 14–58% in East Africa, compared with 51–72% in North Africa and western Asia.[16] Similar to other European countries, our study revealed relatively low use of contraceptive methods in Switzerland, apart from condoms. In a French study, 58.8% of women with HIV infection used condoms, 20.5% used other contraceptive methods, and 20.7% used traditional methods (e.g. body temperature) or no contraceptive method.[11] Similar results were published from Denmark, where 75% of women used a variety of types of contraception, with condoms accounting for 62% of those (of the 75%).[12] The pattern of contraceptive use differs in Switzerland between women living with HIV and HIV–uninfected women, of whom only 40% use condoms and 30% are on oral hormone contraception.[17]

ART and hormonal contraceptive methods can interact in several clinically meaningful ways. Cytochrome P450–inducing ART drugs, such as boosted PIs, reduce hormonal contraceptive effectiveness as a result of their effect on metabolism.[7–9,18] In our study, almost half of all women with unintended pregnancies were on a boosted PI regimen. Several pharmacokinetic studies and case reports implicate the nonnucleoside reverse transcriptase inhibitor EFV in hormonal contraceptive method failure.[6,7,19] In addition, a 12% failure rate of the levonorgestrel subdermal implant was reported among women on an EFV–based ART regimen. In our study, the fact that the rate of women on hormonal contraception was higher in ART–naïve women than in those on cART indicates uncertainty regarding the safe and reliable use of this form of contraception in combination with cART. Nevertheless, we were not able to prove that drug–drug interactions could explain unintended pregnancies because of the small number of women on hormonal contraception. In several recommendations, it is suggested that the contraceptive pill contain at least 35 μg of oestrogen if administered with cART based on a boosted PI. However, a recent study in over 5 million women showed that an oestrogen dose of 20 μg vs. 30–40 μg was associated with a lower risk of pulmonary embolism, ischaemic stroke and myocardial infarction.[20] Thus, it is crucial to consider the drug–drug interaction of cART and hormonal pills, as women with HIV infection should be able to choose the low–dose formulations to avoid any long–term cardiovascular risk.

Unintended pregnancies with or without induced abortions, including unsafe pregnancy terminations, can cause severe adverse outcomes in terms of maternal physical and psychological health.[21–23] A report on unintended pregnancies world–wide showed a relatively low average of 27 per 1000 women aged 18–44 years in Europe,[24] but as high as 80 per 1000 women in Africa of approximately the same age. The preventive effect of cART in terms of the onward transmission of HIV[25] and the scaling up of pre–exposure prophylaxis use world–wide[26,27] will inevitably lead to less condom use and expose young women to a higher risk of an unintended pregnancy. In our cohort, 72% of women were in a stable partnership, with most on cART with a suppressed viral load, and thus they could potentially have condomless sex without the risk of HIV transmission. This highlights the urgent need to offer reliable contraception to these women, which would probably be most effective as part of their integrated care in the HIV clinic.

The strength of this study was the nested survey in a national HIV cohort and the possibility to link the data. In addition, there are few reports of contraception and unintended pregnancies in HIV–positive women living in Europe. Limitations include the design of a self–report questionnaire, which may contain reporting bias. Disentangling failure rates caused by users and failure rates caused by drug–drug interactions and reduced drug efficacy was not feasible in this study because of small numbers.