INTRODUCTION — The management of the pregnant woman with HIV infection has evolved significantly over the past 25 years in light of advancements in drug development and a greater understanding of the prevention of perinatal HIV transmission. In the United States and Europe, the risk of HIV transmission from mother to infant had declined to historically low levels with the use of antiretroviral medications [1,2]. Contributions to this successful prevention effort include universal testing of pregnant women for HIV infection, the use of cesarean delivery (when appropriate), and avoidance of breastfeeding, when feasible. (See "Prenatal evaluation of the HIV-infected woman in resource-rich settings".)
This topic will address antiretroviral treatment (ART) in the HIV-infected pregnant woman, intrapartum management, and antiretroviral prophylaxis of her infant in resource-rich settings. In October 2016, the Department of Health and Human Services in the United States published updated guidelines on the evaluation and management of HIV-infected pregnant women . Our recommendations below are largely consistent with these guidelines. Many newer drug regimens recommended as preferred agents for nonpregnant adults have not yet been studied in pregnancy. Thus, recommended regimens for initiation of therapy are different from those for nonpregnant adults. These recommendations change regularly as safety and efficacy data become available.
Information regarding antepartum evaluation and teratogenicity and the pharmacokinetics of individual agents during pregnancy is found elsewhere. (See "Prenatal evaluation of the HIV-infected woman in resource-rich settings" and "Safety and dosing of antiretroviral medications in pregnancy".)
Information regarding the management of the HIV-infected pregnant female in resource-limited settings and the prevention of HIV transmission during breastfeeding is found elsewhere. (See "Prevention of mother-to-child HIV transmission in resource-limited settings" and "Prevention of HIV transmission during breastfeeding in resource-limited settings".)
RATIONALE FOR MATERNAL ART — The use of antiretroviral therapy (ART) during pregnancy involves two separate but related goals: reduction of perinatal transmission and treatment of maternal HIV disease . All pregnant HIV-infected women should receive combination ART, regardless of CD4 cell count or plasma HIV RNA copy number, to prevent perinatal transmission. The combination of maternal antepartum, maternal intrapartum, and infant prophylaxis is recommended to maximize infant pre-exposure and post-exposure prophylaxis and suppress maternal viral load below the limit of detection throughout pregnancy to decrease the risk of transmission. In resource-rich countries, with widespread use of ART during pregnancy, the incidence of perinatal transmission of HIV is estimated at less than 2 percent [1-3,6-9].
Decisions about the use of antiretroviral drugs during pregnancy should be made by the woman following discussion with her health care provider of the potential benefits and risks of therapy. Clinicians in the United States wishing to discuss management options may consider calling the National Perinatal HIV Consultation and Referral Service at the University of California, San Francisco: 1-888-448-8765.
HIV viremia and risk of infant infection — In general, the risk of perinatal transmission declines with decreasing levels of maternal HIV RNA (eg, <1000 copies/mL) although there is still some element of risk [10,11].
In a prospective cohort study, HIV RNA levels were serially measured in 1542 HIV-infected women who gave birth from 1990 to 2000 to assess the relationship between plasma viremia, the type of therapeutic intervention, and the risk of HIV perinatal transmission .The risk of HIV transmission declined with lower levels of maternal viremia at delivery. HIV transmission rates ranged from 1 percent for patients with a non-detectable viral load (<400 copies/mL) compared with 23 percent for those with a viral load >30,000 copies/mL. Even at low levels of viremia, the risk decreases with further reductions. In a study of 8075 HIV-infected women who received antepartum ART in France between 2000 and 2011, the odds of perinatal transmission were fourfold higher (95% CI 1.9-8.2) for women with a viral load of 50 to 400 copies/mL at delivery compared with those with a viral load <50 copies/mL (1.5 percent when a viral load was 50 to 400 copies/mL versus 0.3 percent when a delivery viral load was <50 copies/mL) . The transmission rate was 2.8 percent among women with a viral load >400 copies/mL.
These data support the use of ART with the goal of full viral suppression to decrease the risk of perinatal HIV transmission . These data also highlight the fact that women with a very low viral load can still transmit HIV to their infants, although the risk is low. This residual risk may be related to the presence of detectable HIV in the genital secretions. Thus, ART is recommended in all pregnant women for the prevention of perinatal transmission, regardless of the mother’s viral load.
Efficacy of ART in preventing transmission — Numerous studies have documented the efficacy of antiretroviral use in decreasing the risk of mother to child HIV transmission. In 1994, AIDS Clinical Trial 076 evaluated the safety and efficacy of zidovudine prophylaxis in preventing perinatal transmission among 477 HIV-infected pregnant women . Zidovudine was administered during the antepartum and intrapartum period to the mother and in the infant for the first six weeks of life. Women who were randomly assigned to the intervention arm had a significantly lower risk of transmitting HIV infection to their newborn (25 versus 8 percent) with minimal toxic side effects.
After this landmark trial conducted in the United States and France, many subsequent clinical studies of HIV perinatal transmission have been performed in resource-limited settings to determine if lower-cost medications could be used or if alternative strategies could be as effective. Much of these data are not directly applicable to resource-rich countries due to varying HIV genotypes, patient populations, and infant feeding practices. However, illustrative concepts have emerged from these clinical trials that are also instructive for management of patients in resource-rich settings (see "Prevention of mother-to-child HIV transmission in resource-limited settings", section on 'Efficacy of maternal ART in preventing transmission'):
●A combination regimen is more effective in reducing HIV transmission than a single-drug regimen [14,15].
●A three-part strategy (ie, antepartum, intrapartum, and infant prophylaxis) is more effective than giving medications only during the intrapartum and/or postpartum periods [16,17].
●Starting antiretroviral therapy (ART) earlier in pregnancy is more effective in reducing perinatal transmission than starting late in pregnancy [10,18,19]. In a study in South Africa of 730 pregnant women with advanced immune suppression starting ART during pregnancy, each additional week of maternal ART use reduced the odds of perinatal transmission by 8 percent . (See 'When to initiate ART during pregnancy' below.)
In resource-rich countries, with widespread use of ART during pregnancy, the incidence of perinatal transmission of HIV is estimated at less than 2 percent [1-3,6-9].
WHEN TO INITIATE ART DURING PREGNANCY — Many pregnant women who are known to be HIV-infected will already be taking antiretroviral therapy (ART). For treatment-naïve women, earlier initiation of ART is associated with increased likelihood of viral suppression by the time of delivery and decreased risk of transmission [10,19-22]. Thus, we recommend initiation of ART as soon as HIV is diagnosed in a pregnant woman or as soon as pregnancy is diagnosed in untreated women with an established HIV diagnosis.
Drug resistance testing is indicated prior to ART initiation for all individuals with detectable virus, but ART should be initiated prior to the availability of those results. Because a single mutation is associated with resistance to non-nucleoside reverse transcriptase inhibitors whereas multiple mutations are required for resistance to protease inhibitors, one of the preferred protease inhibitor-based ART regimens may be preferable for initial ART in treatment-naïve pregnant women in such circumstances due to lower likelihood of pre-existing resistance. (See 'Drug resistance testing' below and 'Antiretroviral selection and management' below.)
For pregnant women who are not yet on ART, the rate of virologic control after initiation of ART appears comparable to that in nonpregnant women. As an example, in a study of 519 treatment-naïve pregnant women initiating ART, 93 percent had achieved a HIV viral level <400 copies/mL by the third month of therapy . Available data additionally suggest that earlier initiation of ART is associated with increased likelihood of viral suppression by the time of delivery [19,21,22]. In a cohort of 671 ART-naïve pregnant women in the United States who initiated ART during pregnancy, a detectable viral load (>400 copies/mL) at delivery was documented in 13 percent and was associated with antiretroviral initiation at a later gestational age . Twenty-four percent of women who initiated the regimen during the third trimester had a detectable viral load.
Accordingly, earlier ART initiation is associated with a lower risk of perinatal transmission [10,20]. A study of 8075 women who received antepartum ART in France between 2000 and 2011 found an overall rate of transmission of 0.7 percent; no transmission occurred among 2651 infants born to women who received ART before conception, continued ART throughout pregnancy, and had a viral load <50 copies/mL at delivery . Regardless of viral load, the transmission rate increased from 0.2 percent for women receiving ART before conception to 0.4, 0.9, and 2.2 percent for women initiating ART in the first, second, or third trimester.
Some women may prefer delaying ART until after completion of the first trimester, when the fetus is less susceptible to the potential teratogenic effects of medications. However, later initiation of ART may be less effective in reducing in utero transmission of HIV . Furthermore, if ART is delayed beyond 28 weeks gestation, HIV RNA may not be fully suppressed by the time of delivery, which may increase the risk of perinatal transmission . As noted above, we recommend initiation of ART as soon as HIV is diagnosed in pregnancy regardless of trimester.
ANTIRETROVIRAL SELECTION AND MANAGEMENT
General principles — In pregnant women, antiretroviral regimen selection should take into account the resistance profile of the virus, the safety and efficacy of the drugs in the mother and fetus, the convenience and adherence potential of the regimen, the potential for drug interactions with other medications, and pharmacokinetic data in pregnancy. For women initiating antiretroviral therapy (ART) during pregnancy, regimen selection is similar to that for nonpregnant patients, with the exception that certain agents that are preferred in the general HIV-infected population (such as dolutegravir, elvitegravir, and tenofovir alafenamide) are not preferred in pregnancy because of limited clinical experience and/or lack of pharmacokinetic data to inform appropriate dosing during pregnancy. (See 'Preferred antiretroviral agents' below.) Treatment-experienced women on a suppressive antiretroviral regimen can usually continue it even if the agents are not specifically preferred during pregnancy, unless the regimen contains elvitegravir-cobicistat, stavudine, didanosine, or full-dose ritonavir. (See 'On ART with viral suppression' below.)
For both treatment-naïve and treatment-experienced women, it is important to be aware that pharmacokinetic changes in pregnancy may require increased dosing, more frequent dosing, or boosting of certain medications, especially protease inhibitors. (See 'Protease inhibitors' below and "Safety and dosing of antiretroviral medications in pregnancy", section on 'Protease inhibitors'.)
How and when to start ART during pregnancy is the topic of guidelines produced by the United States Department of Health and Human Services and is updated regularly. These guidelines can be accessed at http://aidsinfo.nih.gov/ContentFiles/PerinatalGL.pdf. In the United States, the National Perinatal HIV hotline can also be accessed at 1-888-448-8765.
The long-term virologic and immunologic efficacy of antiretroviral drugs in the HIV-infected pregnant woman has not been evaluated in randomized clinical trials, and few comparative data are available on the most efficacious agents to use. Most of the data that are available are focused on the safety and efficacy of antiretroviral medications in preventing infant transmission, as opposed to the efficacy of these drugs for the treatment of HIV in pregnant women . (See 'Efficacy of ART in preventing transmission' above.)
Selection of ART regimens in nonpregnant HIV-infected individuals is discussed elsewhere. (See "Selecting antiretroviral regimens for the treatment-naïve HIV-infected patient" and "Selecting an antiretroviral regimen for treatment-experienced HIV-infected patients who are failing therapy".)
Drug resistance testing — If HIV RNA is detectable on the baseline blood test, drug resistance testing should be performed to inform the optimal selection of an antiretroviral regimen . ART should be initiated in pregnant women prior to receiving the results of the resistance testing; if necessary, the regimen can be subsequently modified based on the results of the resistance assay. This is particularly important if the mother presents late for care, as the risk of perinatal transmission is associated with the duration of antenatal ART and level of viral suppression at delivery. (See 'When to initiate ART during pregnancy' above.)
Clinicians should be aware that drug resistance testing may not be feasible if the viral load is low (eg, <500 to 1000 copies/mL). (See "Overview of HIV drug resistance testing assays".)
Approach for specific populations
Treatment-naïve — Selecting a regimen that will be active against the patient’s virus based on the resistance profile is the primary objective. Within that constraint, we generally favor a regimen consisting a dual NRTI backbone with a protease inhibitor or integrase inhibitor as a third drug, using agents with a documented history of safety in pregnancy (table 1). Other factors that inform the regimen selection are patient factors such as adherence potential, tolerability, other comorbidities, and pharmacokinetic data in pregnancy. Some of the preferred regimens in nonpregnant adults lack data on pharmacokinetics during pregnancy; hence, preferred regimens for pregnant women differ somewhat from those for the general HIV-infected adult population.
The following describes our approach to regimen selection for a pregnant woman without evidence or suspicion of drug resistance, for whom a preferred NRTI combination backbone with a third preferred agent is appropriate. It is important to note that this is one approach, and there may be other appropriate reasons to choose a particular preferred agent over another. (See 'Preferred antiretroviral agents' below.)
Selecting the NRTI backbone — Tenofovir disoproxil fumarate-emtricitabine (TDF-FTC or the interchangeable TDF-lamivudine) and abacavir-lamivudine are the preferred NRTI combinations in pregnancy. (See 'Nucleoside reverse transcriptase inhibitors' below.)
We generally use TDF-FTC unless the patient has significant renal impairment (estimated glomerular filtration rate <60 mL/min/1.73 m2), in which case a TDF-containing combination should not be used and we instead prefer abacavir-lamivudine. Because abacavir-lamivudine should only be used in patients who are negative for the HLA-B5701 allele and testing may take several days to obtain the result, this regimen is less desirable for initiation during pregnancy, particularly if the patient presents late in gestation.
For women with hepatitis B virus coinfection, a TDF-containing combination is preferred because of its activity against hepatitis B virus. Regimen selection for patients who have hepatitis B virus coinfection and significant renal impairment, which would restrict use of tenofovir, should be done in consultation with an expert in treating hepatitis B virus.
There is limited clinical experience and very limited pharmacokinetic data during pregnancy with tenofovir-alafenamide, a formulation of tenofovir that has improved renal and bone safety compared with TDF; thus it is not recommend for routine initiation in pregnant women.
Further details on the use of these NRTI combinations are discussed elsewhere. (See 'Nucleoside reverse transcriptase inhibitors' below.)
Selecting the third drug — Among third drug options, there is the most clinical experience in pregnancy with the boosted protease inhibitor class, and one of these agents is generally our first choice. One consideration for deciding whether to use a protease inhibitor in pregnancy is the conflicting data regarding risk of preterm birth; for women who are at very high risk of preterm birth, such as those with a history of multiple spontaneous preterm births, we may avoid protease inhibitor-containing regimens. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Preterm birth'.)
Atazanavir-ritonavir is a good protease inhibitor option because it is administered once daily and thus minimizes problems with adherence. Additionally, there is now extensive experience with this drug in pregnancy, and it is well-tolerated, but it can cause asymptomatic maternal indirect hyperbilirubinemia. There has been some experience with darunavir-ritonavir use in pregnancy; although it must be given twice daily in pregnancy, it is well tolerated. Lopinavir-ritonavir is no longer considered a preferred agent in pregnancy and is an alternative regimen because of tolerability problems with nausea. (See 'Protease inhibitors' below.)
Raltegravir is another option as a third drug to use with the dual NRTI regimen and is associated with a rapid decline in viral load. It is particularly useful when there are concerns about drug interactions with protease inhibitor-based regimens. However, it must be administered twice daily in pregnancy. Dolutegravir is an alternative integrase inhibitor to use in pregnancy because of limited data, although emerging evidence on its use in pregnancy is reassuring. Elvitegravir is not recommended for use during pregnancy. (See 'Integrase inhibitors' below and 'Other antiretroviral agents' below.)
Efavirenz is no longer a preferred third agent because it has been associated with exacerbation of mental health disorders and central nervous symptoms. Nevertheless, efavirenz coformulated with TDF-FTC remains an appropriate option for women who cannot use protease inhibitors and would benefit from a once-daily, single pill regimen. (See 'Alternative agents' below.)
On ART with viral suppression — With rare exceptions, women who have achieved viral suppression on ART that was initiated before pregnancy should continue their current regimen during pregnancy, provided that they tolerate it, even if the agents are not one of the preferred antiretroviral drugs for use during pregnancy. Certain agents, in particular protease inhibitors, may warrant dosing changes during pregnancy to compensate for pharmacokinetic changes during that time. (See 'Protease inhibitors' below.)
The main exception to this approach is elvitegravir-cobicistat. Because of emerging data suggesting decreased drug levels during pregnancy and an associated risk of loss of virologic suppression, this agent is not recommended for initiation during pregnancy pending further data [3,25]. The optimal management of women who become pregnant while on a suppressive elvitegravir-cobicistat-containing regimen is uncertain. We suggest switching to a different regimen, choosing among preferred and alternative agents for use in pregnancy, guided by the treatment history and results of prior resistance testing. However, for women with barriers to switching (eg, a very strong preference for a once-daily regimen and inability to use a different one), it is reasonable to continue an elvitegravir-cobicistat-containing regimen after discussing the potential risks and the need for close viral monitoring. If loss of viral suppression occurs, then a switch to a more potent regimen is recommended.
Stavudine, didanosine, and full-dose ritonavir should also not be continued during pregnancy, even if they are a part of a virologically suppressive regimen. However, this is rarely a clinical issue, as none of these agents is commonly used and the combination of stavudine plus didanosine is also not recommended for the general HIV-infected population .
Although there has been concern about a potential risk of neural tube defects with efavirenz, enough reassuring data on first-trimester efavirenz exposures have accumulated so that recommendations from expert groups do not restrict efavirenz use in pregnancy [3,27,28]. Women who become pregnant while on a suppressive efavirenz-containing regimen should continue it. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Efavirenz'.)
On ART without viral suppression — Women who have not yet achieved viral suppression on ART initiated prior to pregnancy should be evaluated for reasons for virologic failure. In this case, the clinician must begin by asking the patient about their level of drug adherence. Nonadherence can result in incomplete viral suppression merely due to the lack of adequate drug level, even in the absence of drug resistance.
Drug resistance testing is also essential in this situation, and a new ART regimen should be tailored to the individual woman based on her resistance profile and individual medication tolerability. In such cases, agents that are not considered preferred antiretrovirals in pregnancy may be warranted (see 'Preferred antiretroviral agents' below and 'Other antiretroviral agents' below). Consultation with an expert in the management of drug-resistant HIV is recommended in such cases. If resistance to the current drugs in the ART regimen is not detected and suboptimal adherence is felt to have contributed to the failure to achieve viral suppression, a once-daily regimen would be particularly desirable. Alternatively, the original regimen could be continued along with intensive adherence counseling. (See 'Preferred antiretroviral agents' below.)
The virologic response must be monitored carefully. Women who do not show an appropriate virologic response require a repeat resistance panel test.
The general approach to HIV-infected individuals failing ART is discussed in detail elsewhere. (See "Selecting an antiretroviral regimen for treatment-experienced HIV-infected patients who are failing therapy".)
Treatment experienced but not currently on ART — It is important to obtain a history of all prior antiretroviral drug exposure and results of prior resistance testing (if done previously) in women who have a history of ART in the past. Past antiretroviral history and results of drug resistance testing should be used to guide the choice of ART regimen. In such cases, agents that are not considered preferred antiretrovirals in pregnancy may be warranted because of documented or presumed drug resistance. (See 'Preferred antiretroviral agents' below and 'Other antiretroviral agents' below.)
The virologic response must be monitored carefully. Women who do not show an appropriate virologic response require a repeat resistance panel test and should have careful assessment for adherence. If adherence appears adequate, they should also have a consultation with a clinician experienced in antiretroviral drug resistance. The recommended next-line therapy will vary based on the individual patient’s prior exposure and drug resistance profile. (See "Selecting an antiretroviral regimen for treatment-experienced HIV-infected patients who are failing therapy".)
Detectable viremia late in pregnancy — Women with a high viral load who present and/or initiate ART late in pregnancy (eg, after 28 weeks) may be at increased risk of transmission if viral suppression cannot be achieved by the time of delivery. In several case reports and series, the addition of raltegravir to a baseline ART regimen has been used among women with high viral loads presenting in late pregnancy because of its ability to rapidly decrease viral load within two weeks [29-33]. While intriguing, the comparative efficacy and safety of this strategy has not been evaluated and is thus not routinely recommended, but may be useful in individual circumstances. Regimen selection for the woman presenting late in pregnancy is the same as for other treatment-naïve or experienced pregnant women. Initiation of a regimen should not be delayed while awaiting results of drug resistance testing. (See 'Treatment-naïve' above and 'On ART without viral suppression' above and 'Treatment experienced but not currently on ART' above.)
Preferred antiretroviral agents — Preferred antiretroviral agents are those for which there are substantial experience and data documenting virologic efficacy, maternal and fetal safety, and tolerability during pregnancy . Preferred agents for pregnant women differ somewhat from those for the general adult population.
Nucleoside reverse transcriptase inhibitors — The preferred nucleoside reverse transcriptase inhibitor (NRTI) backbone combinations are :
●Tenofovir disoproxil fumarate-emtricitabine (TDF-FTC) or tenofovir disoproxil fumarate-lamivudine (TDF-3TC) – Either combination can be administered once daily, and TDF-FTC is available as a fixed-dose combination pill. These combinations are well-tolerated and commonly used in nonpregnant populations. In particular, this combination is preferred in women with hepatitis B virus coinfection. However, TDF should be used with caution in the setting of renal insufficiency because of its association with renal toxicity.
Although concerns have been raised  regarding a possible increased risk of very preterm birth and neonatal death with TDF when combined with lopinavir-ritonavir based on a trial conducted in Africa, we do not feel that the trial results are clear enough to stop using TDF as a preferred NRTI in pregnancy and are reassured by observational data on TDF use. Our stance is consistent with that of the British HIV Association and the United States Department of Health and Human Services Panel on Treatment of Pregnant Women Living with HIV and Prevention of Perinatal Transmission [3,35]. However, due to concerns about a potential increase in TDF levels with concomitant lopinavir-ritonavir, we avoid initiating TDF with lopinavir-ritonavir during pregnancy. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Tenofovir'.)
There has not been enough clinical experience during pregnancy with tenofovir-alafenamide, a formulation of tenofovir that has improved renal and bone safety compared with TDF, to recommend its routine initiation.
●Abacavir-lamivudine (ABC-3TC) – This combination is well tolerated, can be administered once daily, and can be used in patients with renal impairment. However, it should only be used in patients who test negative for HLA-B5701 because of the risk of a hypersensitivity reaction. Because the HLA-B5701 testing may take several days to result, it makes this regimen less desirable for initiation during pregnancy, particularly late pregnancy. Additionally, this combination should not be used with ritonavir-boosted atazanavir or efavirenz if the baseline viral load is >100,000 copies/mL because of a higher risk of virologic failure. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Abacavir'.)
Protease inhibitors — The preferred protease inhibitors (PI) are :
●Ritonavir-boosted atazanavir – This combination is administered once daily. Given pharmacokinetic changes during pregnancy, an increased dose during the second and third trimesters may be warranted for certain women, such as treatment-experienced women or those receiving concomitant tenofovir. Gastrointestinal side effects may also be seen with the use of this regimen, but since there is not a need for increased dosing in the first trimester, this may be a preferred regimen for women with early pregnancy nausea and vomiting. Atazanavir-ritonavir should not be used in a patient requiring proton pump inhibitors. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Atazanavir'.)
●Ritonavir-boosted darunavir – This combination is administered as 600 mg darunavir plus 100 mg ritonavir, each twice daily during pregnancy. Because of low trough levels with once-daily dosing, twice-daily dosing of darunavir is recommended during pregnancy, especially for treatment-experienced patients. In pharmacokinetic studies, a 17 to 33 percent reduction in darunavir plasma concentration during the third trimester compared with postpartum has been reported; whether the dose should be increased during pregnancy is under investigation. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Darunavir'.)
These protease inhibitors can also be boosted with cobicistat, but the pharmacokinetics and safety of protease inhibitor and cobicistat combinations in pregnancy have not been studied. Because of concern for decreased cobicistat drug levels, it is not recommended for initiation during pregnancy.
Protease inhibitor use during pregnancy has not been clearly associated with gestational diabetes. Nevertheless, some experts consider protease inhibitor exposure a risk factor for glucose intolerance and thus screen for it earlier in women on a protease inhibitor-containing regimen. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Gestational diabetes'.)
There are mixed data on the risk of preterm birth with the use of protease inhibitors during pregnancy. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Preterm birth'.)
Integrase inhibitors — The preferred integrase inhibitor for use in pregnancy is :
●Raltegravir – This agent is administered twice daily. Data on the integrase inhibitor raltegravir during pregnancy are limited but increasing and generally reassuring [36-38]. Raltegravir is an attractive option if drug interactions or toxicity preclude the use of other preferred agents. Additionally, because it can rapidly reduce viral load, raltegravir has been used in several case reports of women who presented late in pregnancy [31-33,38,39]. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Raltegravir'.)
Other antiretroviral agents
Alternative agents — Alternative agents include those that have proven virologic efficacy or have been well-studied in pregnancy but otherwise have certain drawbacks :
●Zidovudine-lamivudine (ZDV-3TC) – Despite extensive experience and trial data demonstrating its safety in pregnancy and efficacy in reducing the risk of perinatal HIV transmission , this combination is an alternative agent because of associated adverse effects (eg, anemia, headache) and the need for twice-daily dosing. Some but not all studies have suggested an increased risk of congenital heart defects with zidovudine exposure, and the clinical significance of this is unclear as many of the reported defects were ventricular septal defects detected by fetal ultrasound that can undergo spontaneous closure postnatally in many children [40,41]. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Zidovudine'.)
●Ritonavir-boosted lopinavir – This is an alternative protease inhibitor for use with a dual NRTI backbone during pregnancy. Experience with lopinavir-ritonavir during pregnancy is extensive [3,24,42], but the need for twice-daily dosing during pregnancy and adverse effects limit its use. In particular, its gastrointestinal effects may exacerbate the nausea associated with pregnancy. Several studies have shown decreased lopinavir plasma concentrations with standard dosing during pregnancy, and some experts recommend increasing the dose by 50 percent to achieve exposure similar to that in nonpregnant patients [43,44]. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Lopinavir-ritonavir'.)
●Dolutegravir – This is an alternative integrase inhibitor for use with a dual NRTI backbone during pregnancy, although data in pregnancy remain relatively limited. It is dosed once daily and is generally well tolerated. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Dolutegravir'.)
●Efavirenz – This is an alternative, once-daily NNRTI for use with a dual-NRTI backbone. Drawbacks to this agent include a risk for exacerbation of mental health disorders and central nervous system symptoms. The risks of central nervous system symptoms may detract from its use in women with a history of such symptoms. It also has significant drug-drug interactions with methadone, which may require altered methadone dosing if a patient initiates efavirenz during pregnancy. Although there has been concern about a potential risk of neural tube defects with efavirenz, enough reassuring data on first-trimester efavirenz exposures have accumulated so that recommendations from expert groups do not restrict efavirenz use in pregnancy [3,27,28]. It can be a useful agent for women who want a once-daily regimen and cannot take a protease inhibitor.
●Rilpivirine – This is an alternative NNRTI for use with a dual NRTI backbone during pregnancy, although data in pregnancy remain relatively limited. Its advantage is that it is coformulated in a single-pill once daily combination regimen with tenofovir and emtricitabine. Rilpivirine should not be used in individuals with baseline HIV viral load >100,000 copies/mL or CD4 cell count <200 cells/microL. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Rilpivirine'.)
Agents with insufficient data — There are many medications in the HIV armamentarium that have not been well evaluated in the pregnant patient. Occasionally these agents may be needed if the patient is infected with a drug resistant virus or has other reasons that preclude the use of agents with more data during pregnancy. In such cases, use of agents with less data in pregnancy may be warranted for maternal health reasons. Women who become pregnant on such regimens and are virologically suppressed should continue their regimen during pregnancy. The risks and benefits of such an approach need to be conveyed to the patient.
Not recommended — Elvitegravir and cobicistat have not been extensively evaluated in pregnancy, but they are not recommended for initiation during pregnancy because of emerging data suggesting a decrease in drug levels during the third trimester with an associated risk of loss of virologic suppression .
PREGNANCY COMPLICATIONS AND OUTCOMES WITH ANTIRETROVIRAL USE — There may be a small increased risk of preterm delivery and impaired fetal growth with the use of antiretroviral agents, particularly protease inhibitors during pregnancy, but the data from studies are conflicting. These are discussed in detail elsewhere. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Pregnancy outcomes with combination ART exposure'.)
Mode of delivery — In women with plasma HIV RNA ≤1000 copies/mL on ART, the overall incidence of transmission of HIV is low regardless of the mode of delivery (cesarean delivery versus standard vaginal delivery) or duration of membrane rupture, and a further decrease in transmission risk with cesarean delivery in such women is unclear [2,7,8,45-47]. Thus, we do not typically schedule cesarean deliveries for these women, unless indicated for obstetric reasons (table 2).
In contrast, in women whose viral loads remain >1000 copies/mL prior to 38 weeks gestation (eg, women not taking ART, women presenting late in pregnancy, or women not responding to their current ART regimen), performing cesarean delivery at 38 weeks is recommended, before the likelihood of labor onset and rupture of membranes . For these women, pre-labor cesarean delivery decreases the risk of transmission to the child. This was illustrated by a meta-analysis of 15 prospective cohort studies performed prior to the widespread use of antiretroviral agents during pregnancy, in which the incidence of HIV transmission to the infant was 8.4 percent (72 of 857 births) with cesarean delivery versus 16.7 percent (1280 of 7676 births) with vaginal delivery . The reduction in risk was similar when adjusted for use of zidovudine (versus nothing) and advanced HIV disease in the mother. A subsequent trial from the same time period that randomly assigned 436 HIV-infected women to cesarean or vaginal delivery produced similar results .
If such women present in labor or with ruptured membranes prior to scheduled cesarean delivery, management must be individualized and take into account the duration of the rupture of membranes/labor, current ART regimen, and HIV RNA level. If a woman is already in advanced labor or has prolonged rupture of membranes, the benefit of cesarean may be lost and consideration should be given to vaginal delivery.
Prior to the widespread use of ART during pregnancy, duration of membrane rupture was associated with risk of transmission . However, in a more recent prospective study of deliveries between 1996 and 2008, there were no cases of HIV transmission among 493 pregnant HIV-infected women on ART with HIV RNA <1000 copies/mL near delivery, 144 of whom had rupture of membranes more than four hours . Among the 146 women on ART who had HIV RNA >1000 copies/mL (55 of whom had cesarean while the others had vaginal delivery), perinatal transmission was 3.8 percent if duration of membrane rupture was less than four hours compared with 4.9 percent if four or more hours, and this difference was not statistically significant. Only viral load above 10,000 copies/mL was an independent risk factor for perinatal transmission.
Preterm premature rupture of membranes — When membrane rupture occurs before 37 weeks gestation, decisions about timing of delivery should be based on best obstetrical practices, taking into account risks of prematurity for the infant. The presence of HIV infection of the mother should not change management. Administration of antenatal corticosteroids to accelerate fetal lung maturity should be given if appropriate, as no data exist to suggest that these recommendations need be altered for HIV-infected women. When the decision is made to deliver, route of delivery should be according to obstetrical indications.
Intrapartum antiretrovirals — Women should continue taking their ART regimen as much as possible during labor and delivery or scheduled cesarean delivery. Additional administration of intrapartum intravenous zidovudine depends on the maternal HIV viral load near the time of delivery (table 2). If indicated, intravenous zidovudine should be given regardless of the presence of drug resistance to zidovudine.
●For HIV-infected women on ART with HIV RNA ≤1000 copies/mL consistently in late pregnancy and near the time of delivery and no concerns regarding adherence or resistance to the regimen, intravenous zidovudine is not specifically recommended as it does not appear to further reduce the risk of perinatal transmission in this setting [3,51,52].
●In contrast, for women with HIV RNA >1000 copies/mL near delivery, possible poor adherence, or unknown HIV RNA levels, intravenous zidovudine is recommended. For women scheduled for cesarean delivery with HIV RNA >1,000 copies/mL, intravenous zidovudine should be given three hours before cesarean delivery. HIV-infected women who present in labor and have not taken any antiretroviral medications should be given intravenous zidovudine immediately to prevent perinatal transmission.
Zidovudine crosses the placenta rapidly and can provide pre-exposure prophylaxis to the fetus. In the PACTG 076 trial, combined antepartum zidovudine, intravenous zidovudine during labor, and six weeks of infant zidovudine prophylaxis reduced perinatal transmission by 66 percent . There are no randomized clinical trials evaluating the benefit of intravenous zidovudine during labor among women using current combination antiretroviral regimens. However, among 7917 women from the French Perinatal Cohort who had HIV RNA <400 copies/mL at delivery, receipt of intravenous zidovudine intrapartum was not associated with a decreased risk of mother-to-child HIV transmission (transmission rates of 0.6 percent [42 of 7576] and 0 percent [0 of 341] with and without intrapartum zidovudine, respectively) . In contrast, zidovudine receipt was associated with lower transmission rates among women who had HIV RNA ≥1000 copies/mL (2.9 percent [45 of 1561] versus 7.5 percent [8 of 107] without intrapartum zidovudine).
Other intrapartum interventions — Obstetric management should minimize the duration of fetal exposure to maternal fluids and blood, and avoidance of fetal scalp electrode monitoring is recommended. The possible risks of interventions during management of labor should be weighed against the obstetrical indications and benefits.
INFANT PROPHYLAXIS — We recommend that all infants born to HIV-infected mothers receive antiretroviral post-exposure prophylaxis after birth to decrease the risk of HIV acquisition. Infant antiretroviral prophylaxis should be initiated as soon as possible, ideally within the first 6 to 12 hours of delivery. The type of prophylaxis depends on the virologic status of the mother (table 2).
Mothers with viral suppression (≤1000 copies/mL) — The risk of HIV infection is low for infants born to mothers who are on ART and have achieved a viral level ≤1000 copies/mL by the time of delivery. For such infants, four to six weeks of zidovudine prophylaxis is generally appropriate (table 3). The shorter duration is for infants of mothers who had a suppressed viral load throughout pregnancy and no adherence concerns. The risk of transmission may be greater in women who have low but detectable viral loads and some other risk factor for transmission (such as concurrent sexually transmitted infection, genital ulcer disease, or obstetric issue that threatens excess bleeding). In such cases, combination ART, as is given for infants with higher risk of infection, may be warranted. (See 'Mothers without viral suppression (>1000 copies/mL)' below.)
Mothers without viral suppression (>1000 copies/mL) — Because of a higher risk of HIV infection, infants should receive combination antiretroviral prophylaxis if their mothers have not received any ART during pregnancy, have received only intrapartum antiretrovirals, or have received antepartum ART but did not achieve viral suppression near delivery (ie, had a viral level >1000 copies/mL). Options for combination prophylaxis include a two-drug (zidovudine plus three doses of prophylactic-dose nevirapine given within 48 hours of birth, 48 hours after the first dose, and 96 hours after second dose) or a three-drug regimen (zidovudine, lamivudine, and treatment-dose [6 mg/kg] nevirapine) . We typically use the three-drug combination regimen at treatment doses for six weeks, if the infant can tolerate it (table 3). An alternative approach to this three-drug combination regimen, particularly if there are side effects or complications, is to discontinue the lamivudine and nevirapine components at two weeks if the HIV nucleic acid amplification test (NAAT) at birth was negative and continue zidovudine alone for the full six weeks.
The rationale for a combination antiretroviral regimen is to maximize the reduction in perinatal HIV transmission, but the optimal combination regimen is uncertain. Although evidence suggests that combination antiretroviral prophylaxis is better than zidovudine monotherapy for such high-risk infants, there are no comparative data suggesting that this particular three-drug combination (zidovudine, lamivudine, and treatment-dose nevirapine) is superior to the two-drug regimen discussed above (zidovudine plus prophylactic-dose nevirapine). The decision to administer a three-drug regimen should be made in consultation with a pediatric HIV specialist, preferably before delivery, if possible, and should consider the balance between the transmission risk and potential neonatal toxicities for the individual infant, with parental counseling on these issues.
The HIV Prevention Trials Network (HPTN) 040/Pediatric AIDS Clinical Trial Group 1043 study is the only randomized trial evaluating combination prophylaxis in high risk infants . It compared the efficacy of three infant prophylaxis regimens in preventing intrapartum mother-to-child transmission of HIV among infants to pregnant women who presented late to care and received only intrapartum zidovudine or no antiretroviral agents. The rate of intrapartum transmission of HIV was higher when infants received six weeks of zidovudine alone (4.8 percent) compared with six weeks of zidovudine plus three prophylaxis-level doses of nevirapine over eight days (2.2 percent) or six weeks of zidovudine plus two weeks of nelfinavir and lamivudine (2.4 percent). Neutropenia was more common with the triple drug regimen than the other two. Thus, the two-drug zidovudine plus nevirapine regimen offered the most favorable balance between efficacy and side effects. It is hypothesized but as yet unproven that replacing nelfinavir in the three-drug regimen with treatment-dose nevirapine, which has greater potency, and giving the regimen for a full six weeks would result in greater preventive efficacy that would offset the potential toxicity. Observational data suggest that this three-drug regimen (zidovudine, lamivudine, plus nevirapine) is associated with slightly lower hemoglobin levels and higher premature discontinuation rates compared with zidovudine only, but no major toxicity . Clinical trials evaluating the impact of very early treatment on viral reservoirs in HIV-infected infants by treating high-risk HIV-exposed infants with the three-drug regimen of zidovudine, lamivudine, and nevirapine at treatment doses will inform the pharmacokinetics and safety of this regimen in infants [55-57]. Data from these studies will guide future recommendations.
Furthermore, the threshold viral level at which to prompt the use of combination infant prophylaxis is not definitively known. Observational data suggest that the lowest rates of transmission are observed when maternal viral load is below the level of assay detection (eg, <50 copies/mL versus 400 to 1000 copies/mL) [10,20]. However, there has been no study to demonstrate relative efficacy and safety of combination prophylaxis compared with zidovudine only prophylaxis at these different thresholds of maternal viremia. We typically use 1000 copies/mL as the threshold for combination infant prophylaxis, but for women with lower but detectable viremia, we might use combination infant prophylaxis if other risk factors for HIV peripartum transmission are present. (See 'Mothers with viral suppression (≤1000 copies/mL)' above.)
In addition to a combination infant prophylaxis regimen, maternal intravenous zidovudine and a scheduled cesarean delivery is recommended for women who have detectable viremia (HIV RNA >1,000 copies/mL) near the time of delivery. (See 'Intrapartum management' above.).
Mothers with unknown HIV status — For the infant whose mother’s HIV status is unknown postpartum, rapid HIV testing of the mother or infant with a combination antigen-antibody assay is recommended as soon as possible. Infant combination antiretroviral prophylaxis as given for high-risk infants (see 'Mothers without viral suppression (>1000 copies/mL)' above) should be initiated immediately if the rapid test is positive, while awaiting confirmatory HIV testing . (See "Screening and diagnostic testing for HIV infection".)
Neonatal drug safety and pharmacology — It is important to note that pharmacokinetic data to allow appropriate dosing recommendations in neonates are available only for zidovudine, lamivudine, nevirapine, and nelfinavir, and in preterm infants, only for zidovudine and nevirapine. Furthermore, most of the available pharmacokinetic data for nevirapine are for lower prophylaxis doses as opposed to treatment-level dosing, and the desired trough levels for the two doses are different (the target trough level for prophylaxis is <100 ng/mL whereas trough level for treatment is 3,000 ng/mL).
Observational data on the use of three-drug prophylaxis regimens among 143 neonates (21 percent of whom were <37 weeks gestation and 40 percent of whom received zidovudine, lamivudine, and treatment-dose nevirapine as their regimen) demonstrated higher rates of non-specific signs and symptoms (10 percent) as well as premature drug discontinuation (10 percent) with three drugs compared with zidovudine alone (0 and 2 percent, respectively) but have not suggested major toxicity .
The use of lopinavir-ritonavir in four preterm infants (two sets of twins) was associated with heart block that resolved after discontinuation of the drugs; thus, this drug is not recommended for use in neonates for prophylaxis of mother-to-child transmission of HIV as it should not be administered to neonates before a postmenstrual age of 42 weeks and a postnatal age of at least 14 days . Although a trial evaluating antiretroviral prophylaxis in breastfeeding infants did not detect differences in side effects among those who received lopinavir-ritonavir versus lamivudine (grade 3 or 4 events in 51 and 50 percent), the study drugs in this trial were not initiated until day 7 of life and infants <2 kg were not enrolled .
POSTPARTUM CONTINUATION OF ANTIRETROVIRAL AGENTS — We recommend continuation of ART following pregnancy. This is consistent with recommendations in the United States and elsewhere to initiate and continue ART in all HIV-infected individuals, regardless of CD4 cell count or clinical state, to reduce the risk of disease progression and to prevent HIV sexual transmission, given findings from randomized clinical trials [60-62]. (See "When to initiate antiretroviral therapy in HIV-infected patients".)
In the postpartum period, adherence to an antiretroviral regimen may be particularly difficult. In a meta-analysis of studies of antiretroviral use during and after pregnancy, the pooled rate of adequate adherence (>80 percent of doses) in the postpartum setting was only 53 percent . This finding highlights the importance of careful adherence counseling and social support during the postpartum period in an effort to mitigate the risk of poor compliance among those who continue an antiretroviral regimen.
For women continuing ART postpartum, modification of the regimen may be appropriate if pregnancy-related factors led to selection of an ART regimen that contains agents that would not otherwise be used because of long term toxicity or convenience. Postpartum contraception should be ensured for women who continue on an efavirenz-containing regimen.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: HIV treatment in pregnant women" and "Society guideline links: HIV infection in infants and children".)
SUMMARY AND RECOMMENDATIONS
●In the United States and Europe, the risk of HIV transmission from mother to infant has declined to historically low levels with the use of antiretroviral medications. The combined use of maternal antepartum, maternal intrapartum, and infant antiretroviral prophylaxis maximizes infant pre-exposure and post-exposure prophylaxis to decrease the risk of HIV acquisition. (See 'Introduction' above.)
●Fully suppressive antiretroviral therapy (ART) for HIV-infected pregnant women has two separate but related purposes: reduction of perinatal transmission and treatment of maternal HIV disease. All pregnant HIV-infected women should receive combination ART, regardless of CD4 cell count or plasma HIV RNA copy number. (See 'Rationale for maternal ART' above and 'HIV viremia and risk of infant infection' above and 'Efficacy of ART in preventing transmission' above.)
●For treatment-naïve pregnant women, we recommend prompt initiation of ART regardless of gestational age (Grade 1B). Earlier initiation of an antiretroviral regimen is associated with increased likelihood of viral suppression by the time of delivery and thus decreased risk of transmission. (See 'When to initiate ART during pregnancy' above.)
●Selecting a regimen that will be active against the patient’s virus based on the resistance profile is the primary objective. Within that constraint, we generally favor agents with a documented history of safety in pregnancy (table 1). Other factors that inform the regimen selection are patient factors such as adherence potential, tolerability, other comorbidities, and pharmacokinetic data in pregnancy. (See 'General principles' above.)
●Treatment-naïve pregnant woman should be started on a regimen that consists of a dual nucleoside reverse transcriptase inhibitor (NRTI) backbone with either a protease inhibitor or integrase inhibitor as a third drug. The individual agents should be chosen from those that are preferred in pregnancy because of their substantial clinical use and demonstrated efficacy during pregnancy. ART should be initiated after obtaining resistance testing and can be adjusted after results of a resistance profile of the virus are obtained, if applicable. (See 'Treatment-naïve' above.)
●Treatment-experienced women on a virologically suppressive ART regimen that they tolerate can usually continue that regimen, even if the individual agents are not preferred in pregnancy. However, for patients who become pregnant while on an elvitegravir-cobicistat-containing regimen, we suggest switching to a different regimen because of low drug levels and the potential for virologic failure during pregnancy (Grade 2C). Stavudine, didanosine, and full-dose ritonavir should also not be continued during pregnancy, but these agents are rarely used in the general population. (See 'On ART with viral suppression' above.)
●In treatment-experienced women who are either not on an antiretroviral regimen or are on a failing regimen, antiretroviral selection will be mainly guided by the history of prior antiretroviral drug exposure and results of previous and current resistance testing. In such cases, use of agents that are not preferred in pregnancy may be warranted. (See 'On ART without viral suppression' above and 'Treatment experienced but not currently on ART' above.)
●Antiretroviral agents that are preferred in pregnancy include the NRTI combinations tenofovir disoproxil fumarate with emtricitabine or lamivudine as well as abacavir-lamivudine; the protease inhibitors ritonavir-boosted atazanavir and ritonavir-boosted darunavir; and the integrase inhibitor raltegravir. Some protease inhibitors may warrant dosing alterations during pregnancy given pharmacokinetic changes. (See 'Preferred antiretroviral agents' above.)
●There may be a small increased risk of preterm delivery and impaired fetal growth with the use of ART, particularly protease inhibitors, during pregnancy, but the data from studies are conflicting. (See "Safety and dosing of antiretroviral medications in pregnancy", section on 'Pregnancy outcomes with combination ART exposure'.)
●For women who have a viral load >1000 copies/mL prior to 38 weeks gestation (eg, women not taking ART, women presenting late in pregnancy, or women not responding to their current ART regimen), we recommend pre-labor cesarean delivery to reduce the risk of HIV transmission to the infant (table 2) (Grade 1A). The mode of delivery for women with viral load <1000 copies/mL on ART depends on obstetric indications alone. (See 'Mode of delivery' above.)
●Women should continue taking their ART regimen during labor and delivery or scheduled cesarean delivery. For women who have HIV RNA ≥1000 copies/mL or unknown viral levels in late pregnancy and around the time of delivery, we recommend intrapartum intravenous zidovudine to further reduce the risk of perinatal transmission (table 2) (Grade 1B). For women who have had HIV RNA <1000 copies/mL consistently in late pregnancy and around the time of delivery and have no concerns related to adherence or resistance to the regimen, intrapartum intravenous zidovudine is not associated with further reduction of HIV transmission. (See 'Intrapartum antiretrovirals' above.)
●We recommend that all infants born to HIV-infected mothers receive antiretroviral post-exposure prophylaxis after birth to decrease the risk of HIV acquisition (Grade 1B). Infant antiretroviral prophylaxis should be initiated as soon as possible, ideally within the first 6 to 12 hours of delivery. The precise prophylactic regimen depends on the mother’s use of antepartum antiretroviral agents and her viral load near delivery (table 3 and table 2). (See 'Infant prophylaxis' above.)
●Given recommendations for ART use in all HIV-infected patients, regardless of immune, clinical, or viral status, ART should be continued postpartum. Adherence to ART during the postpartum period may be particularly problematic for women, and it is important to provide careful adherence counseling and social support during the postpartum period to mitigate the risk of poor compliance. (See 'Postpartum continuation of antiretroviral agents' above and "When to initiate antiretroviral therapy in HIV-infected patients".)