Tuberculosis Prevention in Pregnant Women Living with HIV: A Review of the Current Management Strategies and Guidance

Introduction

Tuberculosis (TB) is the leading cause of infectious disease-related deaths globally, especially for people living with HIV.¹ Compared to people without HIV, people with HIV are approximately fourteen times more likely to develop TB disease, have poorer TB treatment outcomes, and experience three-fold higher mortality during TB treatment.² For women, the highest risk time to develop TB disease is during and immediately after pregnancy.^3,4 Among the 38.4 million people living with HIV globally, an estimated 1.3 million women living with HIV become pregnant annually.^5,6 Developing TB disease during pregnancy leads to adverse maternal and infant outcomes, including preterm birth, low birth weight, and fetal growth restriction.⁷

Pregnancy is also an independent risk factor for TB disease. Two large epidemiologic studies in Europe, with a combined sample size of over 840,000 pregnant women, found that the incidence of TB doubles in late pregnancy and early postpartum.^3,4,8–10 The synergistic immune changes of pregnancy and HIV infection result in a 10-fold higher risk of TB progression compared to pregnant women without HIV.¹¹ While antiretroviral therapy (ART) reduces the likelihood of developing active TB disease by up to 65%, people living with HIV continue to have an increased risk of active TB even when taking ART with CD4 cell counts >500.^12–14 The neglected global burden of TB in pregnancy has been increasingly recognized as a critical gap in maternal health programming.¹⁵

The World Health Organization (WHO) recommends TB preventive therapy (TPT) for all people living with HIV, including pregnant women. TPT significantly reduces the incidence of TB disease as well as all-cause mortality.^16–19 Up until 2011, the WHO recommended isoniazid preventive therapy (IPT) for everyone living with HIV.²⁰ But, in 2011, they updated their guidelines to replace IPT with the shorter regimen of three months of weekly isoniazid and rifapentine (3HP).²⁰ Multiple studies document better adherence and fewer side effects than IPT.^21–23 But, pregnant women were excluded from these trials and 3HP is, therefore, not currently recommended during pregnancy.

Several important knowledge gaps persist in our understanding of optimal TPT implementation during pregnancy, including safety of the TPT regimen, timing of initiation, and pharmacokinetic data, including drug interactions with ART. These gaps impact national and international guidelines. Of 44 President’s Emergency Plan for AIDS Relief (PEPFAR)-supported countries, only 64% included TB screening recommendations for pregnant women living with HIV in their national guidelines. While 80% of countries recommend TPT for pregnant women living with HIV, only 45% include specific TPT guidance for breastfeeding women living with HIV.²⁴ This review summarizes the current evidence on effective TPT strategies for pregnant women living with HIV and highlights key gaps in knowledge. Our objective was to synthesize the most up-to-date information about pregnancy considerations and management guidelines on TPT for pregnant women living with HIV. This unique review aims to provide a valuable resource for pregnancy-specific information on TPT for an international audience.

Methods

Study Design

In order to explore an in-depth yet flexible overview of TPT in pregnancy and highlight key studies that currently guide management to synthesize the existing knowledge, a narrative review study design was chosen for our review.

Search Strategy

We performed a search to gather relevant references using medical subject headings (MeSH) and keywords in PubMed. Our goal was to cast a wide search and to review each for relevance. While this approach did not require strict inclusion criteria, we included original research studies with any study design relevant to our search, clinical guidance letters relevant to our search, and related works regarding public health efforts surrounding the intersection of TPT, pregnancy, and HIV. The only manuscripts excluded were other reviews. The final search developed with MeSH and keywords was as follows: ((“Pregnant People”[Mesh] OR Pregnant*) AND (“HIV”[Mesh] OR Human Immunodeficiency Virus OR HIV)) AND (“Antiretroviral Therapy, Highly Active”[Mesh] OR Antiretroviral Therapy)) AND (“tuberculosis”[MeSH Terms] OR tuberculosis)) NOT (“Review” [Publication Type]). This search was completed on September 9, 2025, in PubMed. Works published from 1980 to 2026 were included for our review.

Study Selection

After review articles were excluded, the titles and abstracts for all articles were then screened for subjective relevance by two independent reviewers (SCD,SY). The full manuscripts of the abstracts that were deemed relevant to our review were then examined in-depth by both independent reviewers. The reference lists of each article were also reviewed and included if relevant. Our study selection methods can also be seen in Figure 1.

Figure 1 Flowchart for article selection for inclusion in the review.

Results

The initial search resulted in 369 results. After exclusion of review articles, 277 results remained. Of 277 abstracts screened, 111 were included. A total of 78 references published during the time period 1980 to 2025 were ultimately included in this review. We included six randomized controlled trials, 17 references related to clinical guidance, and 55 studies of various designs. A summary of the methodology and results can be found in Figure 1.

TPT Regimens In Non-Pregnant Populations

TPT regimen options vary in duration, but they all center around INH and rifamycins, primarily due to their bactericidal properties and/or ability to penetrate granulomas.^25,26 The US National Institutes of Health (NIH)/Office of AIDS Research recommends 3HP for people with HIV who have viral suppression and are receiving an efavirenz-, raltegravir-, or once-daily dolutegravir-based ART regimen.²⁷ The US Centers for Disease Control and Prevention (CDC) guidelines are similar but also include 3-months of daily INH and rifampin (3HR) as a preferred regimen. There are fewer data for 4-months of daily rifampin (4R) in people living with HIV, but the CDC lists this regimen as an alternative, as well as 6 or 9-months of daily INH (6H and 9H, respectively), primarily for when there is a drug interaction with rifamycins.²⁸ In a study that included participants in both high and low TB burden countries, 4R was shown to be non-inferior to 9H, and was associated with higher rates of completion and lower hepatotoxicity in adults living with and without HIV.²¹ Similarly, a post hoc analysis of two randomized trials including 270 people living with HIV (135 on ART) found that 4R was as effective as 9H for preventing TB, with increased completion and fewer adverse events.²⁹

The US CDC recommends administering TPT to people living with HIV with the following indications: (1) a positive screening test for TB infection (Tuberculin skin testing ≥5mm or positive IGRA) and no evidence of active TB disease; (2) recent, close contact with a person with pulmonary TB, regardless of CD4 count or LTBI screening test results. People living with HIV with CD4 <200 cells/mm³, a negative LTBI screening test, and no other indication for TPT should be retested once their CD4 is ≥200 and treated if indicated.

For high-burden for TB countries, the WHO guidelines from 2024 recommend 3HP as the preferred regimen for people with HIV, when drug–drug interactions permit use of rifamycins. They also include 3HR, 4R, 6H, 9H, and 1-month of daily INH and rifapentine (1HP) as alternative regimens. For those exposed to known multi-drug resistant TB, they recommend levofloxacin daily for 6 months.^8,9

Few of these regimens have been systematically studied in trials that include pregnant women. A summary of available TPT treatment regimens, including pregnancy- and ART- specific considerations, can be seen in Table 1. A summary of INH and rifamycin side effects, ART considerations, and breastfeeding safety guidelines can be seen in Table 2.

Table 1 Tuberculosis Preventive Therapy Regimens and Pregnancy Considerations

Table 2 Tuberculosis Preventive Therapy Medication Side Effects

Isoniazid Preventive Therapy

INH has been a cornerstone medication for both the treatment and prevention of TB.³⁰ It primarily inhibits bacterial cell wall synthesis following activation by the catalase-peroxidase enzyme KatG in Mycobacterium tuberculosis.²⁵ Here we review the adverse drug-related toxicities related to INH in pregnant women living with HIV on ART.³¹

Maternal, Pregnancy, and Infant Safety Outcomes

In general, INH is associated with hepatotoxicity, peripheral neuropathy, gastrointestinal symptoms, and fatigue.³⁰ In pregnancy, there is a heightened concern for INH-induced hepatotoxicity because pregnancy itself, is associated with an increased risk of liver injury through conditions such as hyperemesis gravidarum, intrahepatic cholestasis of pregnancy, pre-eclampsia, and acute fatty liver of pregnancy. A 2020 meta-analysis found a nonsignificant increased likelihood of hepatotoxicity among pregnant women exposed to IPT compared to those without IPT exposure (RR 1.64, 95% CI 0.78–3.44), although this risk was not stratified by HIV status.³² In the TB APPRISE trial, a multi-center randomized controlled study of IPT initiation during pregnancy (immediate) versus after pregnancy (deferred) in women living with HIV and on efavirenz-based ART, maternal hepatotoxicity did not differ between groups (rate difference, −0.89; 95% CI, −3.98 to 2.19).³³ While there are no specific guidelines on frequency of liver enzyme monitoring in pregnant women on IPT, the United States Food and Drug Administration (FDA) label for INH recommends baseline and at least monthly liver enzyme testing for “high-risk” populations.³⁴ Because INH reduces pyridoxine levels, pregnant women living with HIV should take pyridoxine supplementation of at least 25 to 50mg/day (up to 200 mg per day) to prevent peripheral neuropathy.³⁵

For pregnancy outcomes, there are mixed data for IPT. In the TB APPRISE trial, there was a higher incidence of a composite of adverse pregnancy outcomes, which included stillbirth or spontaneous abortion, low birth weight (<2500 g), preterm delivery (<37 weeks), and congenital anomalies, in the immediate versus the deferred IPT group (23.6% vs. 17.0%; CI, 0.8 to 11.9, p=0.01).^33,36 Their intention-to-treat analysis showed no significant differences in secondary maternal safety outcomes, including peripheral neuropathy, hepatotoxicity, and mortality.³³ A post-hoc analysis of the TB APPRISE trial found that IPT exposure during pregnancy was also associated with an increased risk of HIV exposed-uninfected infants becoming underweight, suggesting that in utero exposure to IPT may impact fetal growth.³⁷

Other trials of non-pregnant adults have allowed women to remain on IPT if they became pregnant after enrollment. The ACTG 5279/BRIEF-TB trial was an open label, randomized noninferiority trial of TB prevention using IPT or 1HP in non-pregnant persons with HIV, conducted across ten high TB burden (>60 cases per 100,000) countries: Botswana, Brazil, Haiti, Kenya, Malawi, Peru, South Africa, Thailand, USA, and Zimbabwe.³⁸ Pregnant women were excluded from enrollment. However, women who became pregnant in the IPT arm were allowed to continue therapy and remain enrolled in the study (therapy was discontinued for incident pregnancies in the 1HP arm). A secondary analysis compared frequency of composite adverse pregnancy outcomes (including spontaneous abortion, stillbirth, ectopic pregnancy; excluding induced abortion) in pregnancies with (n=39) versus without (n=89) IPT exposure.³⁹ Of the 39 pregnancies with IPT exposure, 19 (49%) completed IPT during pregnancy, with median exposure duration of 14.7 weeks (IQR 7.3, 24.3). Similar to TB APPRISE, a greater proportion of IPT-exposed pregnancies experienced the composite adverse pregnancy outcome compared with unexposed pregnancies in unadjusted analysis (RR 1.85; 95% CI 0.99–3.47, p=0.05). The risk remained significantly higher in the multivariate model adjusted for TBI status, maternal age, CD4 count, and ART regimen at enrollment (aRR 1.90; 95% CI 1.01–3.45, p=0.04). While IPT use had almost a twofold increase in spontaneous abortion and stillbirth, it was not associated with low birthweight (RR 1.01; 95% CI.29, 3.56) or preterm delivery (RR 0.87; 95% CI.32–2.42).³⁹ Interestingly, the effect of IPT exposure on non-live births was attenuated when the model was adjusted for variables collected at delivery (aRR 1.47 (95% CI 0.84, 2.55), largely due to decreased ART use at delivery and lower CD4 count.³⁹ Of note, maternal adverse events (eg. hepatotoxicity, peripheral neuropathy) were not included in this analysis.

In contrast to trial data, several large observational and programmatic analyses of pregnant women taking IPT found no association with adverse pregnancy outcomes. The TSHEPISO study was a prospective observational cohort study in South Africa of pregnant women living with HIV with and without TB disease.⁴⁰ Women without TB disease were offered IPT. Of 151 women with known pregnancy outcomes, 69 started IPT during pregnancy with 11 (16%) experiencing an adverse pregnancy outcome (defined as fetal demise, prematurity, LBW, or congenital anomaly) compared to 23 (28%) in the IPT-unexposed cohort. The adjusted odds of having an adverse pregnancy outcome was 2.5 times higher in the IPT-unexposed women compared with the IPT exposed women, indicating a protective benefit of IPT (95% CI 1.0–6.5, p=0.04).⁴⁰ Two maternal deaths occurred during the follow-up year, both in the IPT-unexposed group. One IPT-unexposed mother developed TB disease during the first year after delivery.⁴⁰ No infants in either group were diagnosed with TB, nor were there differences in infant hospitalizations. There was one infant death during the neonatal period in the IPT-exposed group, although cause of death was not discussed in the study.⁴⁰

Similarly, a large observational study using programmatic data in pregnant women living with HIV on ART from the Western Cape in South Africa had reassuring pregnancy outcomes.⁴¹ Of the 43971 pregnant women living with HIV, 16.6% received IPT during pregnancy, with 2647 exposed prior to 14 weeks’ gestation and 4663 exposed after 14 weeks’ gestation. Overall, IPT use during pregnancy was associated with a reduced risk of maternal TB disease by approximately 30%, with the greatest effect in those with CD4 ≤350 cells/μL (aHR, 0.51 [95% CI, 0.41–0.63]) vs. 0.93 [95% CI, 0.76–1.13] for CD4 count >350 cells/µL). Moreover, IPT use was associated with a reduced risk of a composite of adverse pregnancy outcomes, which included miscarriage, stillbirth, low birthweight, and neonatal death when compared to no IPT (aOR 0.83 [95% confidence interval, 0.78–0.87]). By trimester, the authors found that the risk of the composite pregnancy outcome was lower in women who received IPT after 14 weeks’ gestation (aOR 0.79 [95% CI, 0.74–0.85]) compared to those who received IPT prior to 14 weeks’ gestation (aOR 0.90 [95% CI, 0.82–0.99]), suggesting later initiation during pregnancy was safer.⁴¹ For maternal outcomes, IPT use was not associated with increased maternal mortality (aHR, 0.75 (95% CI, 0.46–1.22), but there was an increased association with severe liver injury (aHR, 1.51 (95% CI, 1.18–1.93) although the overall event number was low (n=127, 1.7%). This large-scale data provides reassuring evidence that IPT can be safely administered during pregnancy and in the context of concurrent ART (tenofovir-emtricitabine-efavirenz).⁴¹

A study in Botswana evaluated the effect of longer (36 months) IPT on all adults living with HIV, where they were given 6 months of INH, and subsequently randomized to receive INH for an additional 30 months versus placebo.⁴² Of 196 incidental pregnancies for which pregnancy outcome data were available, 103 (52.6%) were exposed to INH during pregnancy, 102 during the first trimester.⁴³ Pregnant women in the study had received a median of 341 days of IPT prior to pregnancy. Of the 196 pregnancies, 73 (37%) received combination ART, and the remainder received either short-course zidovudine or zidovudine/lamivudine prophylaxis. No cases of severe hepatitis were reported during pregnancy or postpartum. In multivariable analysis, neither IPT nor ART exposure during pregnancy was associated with composite adverse pregnancy outcomes, defined as preterm delivery, still birth, low birthweight, spontaneous abortion, neonatal mortality or congenital anomaly (INH aOR 0.6, 95% CI 0.3–1.1; ART aOR 1.8 (95% CI: 0.9–3.6).⁴³ Similarly, duration of IPT prior to pregnancy was not associated with increased odds of adverse pregnancy outcomes either.⁴³

Finally, a retrospective cohort study from Kenya compared birth outcomes (preterm birth, low birth weight, congenital anomalies, perinatal death) among a cohort of women with HIV, with and without IPT exposure.⁴⁴ Of 576 mother-infant pairs evaluated, 152 (26.4%) were exposed to IPT during pregnancy, 145 were on non-nucleoside reverse transcriptase inhibitor (NNRTI)-based ART, 2 on protease inhibitor (PI)-based ART, and 5 on integrase strand transfer inhibitor (INSTI)-based (dolutegravir) ART. There was a trend towards fewer composite adverse birth outcomes in those with antenatal IPT exposure (26% vs. 33%; P = 0.08). Antenatal IPT was not associated with low birth weight, congenital anomalies, or perinatal deaths. In fact, preterm birth was lower among women with antenatal IPT exposure (21% vs. 30%; P = 0.03) even after adjusting for maternal age and viral load (aOR 0.62, 95% CI 0.40–0.98; P = 0.04).⁴⁴

Pharmacokinetics

Pregnancy-specific physiology has not been shown to alter INH disposition in most studies, but data are mixed. The IMPAACT P1026s study was a pharmacokinetic study designed to evaluate antiretroviral and anti-TB drug levels in pregnant women living with HIV and active TB disease using second, third trimester, and postpartum sampling.⁴⁵ For INH, concentrations were lower during pregnancy compared to postpartum.⁴⁵ In the third trimester, 32% and 26% of participants did not reach INH C _max and AUC_0-24 targets, respectively. However, median exposures for INH during pregnancy did meet therapeutic targets so no dose modifications were indicated. The clinical impact of lower INH exposure for TB prevention remains uncertain.^8,27,46

Interestingly, 54% of study participants in IMPAACT 1026s were slow NAT2 acetylators, which is higher than most previously reported cohorts.^45,47 Slow NAT2 acetylators metabolize INH more slowly, resulting in higher INH exposure and significantly increased hepatotoxicity risk.^48–50 To differentiate the effects of pregnancy versus acetylator status on INH exposures, the study team performed within-participant comparisons and found comparable results between the within-participant comparison and the larger mixed-effect model. They concluded that the difference in exposures is likely attributable to pregnancy-mediated factors.⁴⁵ In TB APPRISE, they found that NAT2 genotype was the main determinant of INH concentrations, with a five-fold difference between rapid and slow metabolizers. However, after adjusting for maternal weight and NAT2 and CYP2B6 genotype, they found that pregnancy independently increases INH clearance by 26%.⁵¹

Rifamycin-Containing Regimens

Recognizing the safety and adherence challenges with prolonged IPT regimens, shorter treatment courses that contain rifamycins are now recommended as first-line TPT regimens for non-pregnant adults by both the US CDC and the WHO.^8,9,28 Rifamycins are effective in inhibiting bacterial DNA-dependent RNA polymerase, thereby preventing protein synthesis and causing cell death. Rifampin has been reported to cross the placenta and appears in fetal cord blood, although the human fetal effects are not known. The rifampin FDA label urges caution in pregnancy as an increase in spina bifida and cleft palate has been seen in pregnant mice and rabbit models. Importantly, these outcomes were observed when administered during times of organogenesis and at rifampin concentrations higher than what is observed in human TPT studies.^52,53 Nonetheless, rifampin has been used as part of active TB treatment for decades with no clear definitive associations with teratogenicity.⁵⁴ Rifampin is generally well tolerated, although potential risks, such as drug interactions, must be carefully weighed against benefits of therapy.⁵⁵ The CDC, Department of Health and Human Services (DHHS), and WHO guidelines report that 4R may be used in pregnancy despite no safety or efficacy studies specifically in this population.⁸ Importantly, for people living with HIV, rifamycin-containing regimens (including 3HP, 3HR or 4R) may significantly reduce antiretroviral drug concentrations through induction of cytochrome P450 enzymes.

Maternal, Pregnancy, and Infant Safety Outcomes

There are fewer data on rifamycin-containing TPT regimens in pregnancy than IPT. IMPAACT 2001 was a Phase I/II trial conducted in five high TB burden countries and enrolled 50 pregnant women (20 with HIV on efavirenz-based ART), all of whom took 3HP.⁵⁶ The study reported no drug-related serious adverse maternal, pregnancy or infant safety outcomes, treatment discontinuations, or maternal or infant TB disease.⁵⁶ The DOLPHIN Moms study was a Phase II trial that randomized pregnant women living with HIV on dolutegravir-based ART in South Africa to receive either 3HP or 1HP. The study enrolled 215 pregnant participants, which allowed a precise safety estimate for adverse pregnancy outcomes for pregnant women living with HIV taking either 1HP or 3HP.⁵⁷ Overall, women administered either 1HP or 3HP had a low rate of composite adverse outcome events (defined as premature discontinuation of TPT, grade 3+ liver function tests, stillbirth, preterm delivery, and low birthweight). While there was a higher prevalence of adverse outcomes in women taking 1HP versus 3HP, the number of adverse pregnancy outcomes in the 1HP arm was similar to what had been reported from the TB APPRISE study in women who initiated IPT during pregnancy. These results provide promising data that rifapentine-containing TPT regimens, and in particular, 3HP, can be safely given to pregnant women with HIV.⁵⁷ The 1HP regimen is not recommended at this time as there is not enough evidence for safety. Pregnancy data from trials of rifapentine-containing regimens for tuberculosis treatment are similarly reassuring.⁵⁸

There are no pregnancy-specific data for 3HR in pregnant women living with HIV. Of note, the Infectious Diseases Society of America (IDSA) and Office of AIDS research Advisory Council guidelines state that 3HR is an acceptable alternative regimen based on data extrapolation, with moderate evidence supporting use.²⁷

Pharmacokinetics

The previously described IMPAACT P1026s study found that rifampin exposure during pregnancy was similar to postpartum, and median AUC values met therapeutic targets. Compared to women without HIV, however, pregnant women living with HIV on efavirenz-based ART were found to have 42% lower rifampin AUC₀₋₂₄ and 35% lower C_max in the third trimester.⁴⁵ No dosing changes were recommended.

Similar to IMPAACT 1026s, IMPAACT 2001 also found that rifapentine clearance remained the same from pregnancy to postpartum in women living with HIV who were given 3HP for TPT. The clearance of rifapentine was 30% higher in women with HIV compared to those without HIV, but drug exposure levels remained within the range seen in non-pregnant populations with and without HIV. The study concluded there was no need for dose adjustment of rifapentine in pregnant for women taking 3HP with efavirenz-based ART.⁵⁶ New data from supports that 3HP can be given with once-daily dolutegravir without dose adjustment.⁵⁷

TPT for Multi-Drug Resistant TB in Pregnancy

There are no guidelines for MDR TPT in pregnancy. For non-pregnant adults, WHO guidance from 2024 recommends 6 months of daily levofloxacin (6Lfx) for contacts exposed to MDR/rifampicin-resistant TB, including individuals with HIV on ART. Pregnant women are currently excluded from this recommendation.⁵⁹ The IDSA and the Office of AIDS Research Advisory Council cite that fluoroquinolones are not recommended in pregnancy because of arthropathy seen in animal studies.^27,60 Human studies, however, have not found an increased risk of birth defects, congenital musculoskeletal abnormalities, preterm delivery, or miscarriage in pregnancies exposed to fluoroquinolones.⁶¹ The ongoing BREACH-TB trial (NCT04062201) will evaluate one-month of bedaquiline preventive therapy for individuals exposed to either rifampin resistant or drug susceptible TB and will include pregnant women living with HIV.⁶²

Impact of TPT on ART Medications

In addition to the impact of pregnancy physiology on drug disposition, pharmacokinetic and pharmacodynamic interactions between ART drugs and anti-tuberculosis medications represent a critical consideration for pregnant women living with HIV.

The TB APPRISE study provided comprehensive pharmacokinetic data on drug–drug interactions between INH and efavirenz in pregnant women living with HIV in high TB incidence settings.⁵¹ The investigators found that genetic polymorphisms, particularly in NAT2 and CYP2B6, significantly influenced both INH and efavirenz pharmacokinetics during pregnancy.⁵¹ Rapid acetylators of INH had lower drug exposures, potentially affecting the efficacy of TPT. Pregnancy has also been associated with increased activity of CYP2B6, which could lead to lower levels of drugs metabolized by the CYP2B6 pathway, including efavirenz.⁶³ In the study, the team observed that isoniazid decreased efavirenz clearance by 7% in CYP2B6 normal metabolizers and 13% in slow and intermediate metabolizers. These findings underscore that pregnant women receiving concurrent efavirenz-based ART and IPT may require individualized dosing considerations based on genetic factors, though routine genotyping remains impractical in most high-burden settings.³⁷

Rifamycins, including rifampicin and rifapentine, are potent inducers of cytochrome P450 enzymes and can significantly reduce exposures of several antiretroviral agents.⁶⁴ Rifamycins, as a class, are contraindicated with protease-inhibitor based ART and nevirapine, a non-nucleoside reverse transcriptase inhibitor.⁶⁵ Emerging data on integrase inhibitor-based regimens, particularly dolutegravir, have important implications for pregnant women. The ACTG A5372 study examined the interaction of 1HP with dolutegravir. They concluded that, when 1HP is administered, twice-daily dolutegravir 50 mg is needed to maintain trough concentrations comparable to standard once-daily dosing without rifapentine. The study did not include pregnant women.⁶⁶ In the DOLPHIN Moms study, 25 women taking 1HP and 25 women taking 3HP underwent sparse pharmacokinetic sampling to assess the effect of rifapentine on dolutegravir levels during pregnancy. When given with twice-daily dolutegravir, both arms had median dolutegravir troughs that remained at levels consistent with viral suppression in non-pregnant populations. For 3HP, even standard once-daily dolutegravir dosing achieved adequate median dolutegravir troughs with high rates of viral suppression and treatment completion.⁵⁷ Full results of the study are anticipated this year.

Optimal TPT Regimen in Pregnancy

To understand the decision algorithm around choosing the optimal TPT in pregnancy, Rosen et al conducted a comprehensive modeling study examining five TPT strategies for pregnant people with HIV in South Africa: No TPT; 6H during pregnancy (Immediate 6H); 3HP during pregnancy (Immediate 3HP); 6H postpartum (Deferred 6H); or 3HP postpartum (Deferred 3HP).⁶⁷ Their analysis suggests that when TPT confers higher stillbirth (≥4%) and low birthweight (≥20%) risks, immediate TPT would produce the most combined maternal and fetal/infant deaths, even with low maternal CD4 cell count and high TB incidence. If immediate TPT, however, yields a <4% or <20% increase in stillbirth or low birthweight, respectively, immediate TPT would produce fewer combined deaths than deferred TPT.⁶⁷ In other words, immediate TPT during pregnancy can reduce maternal, fetal, and infant deaths when the risk of adverse pregnancy outcomes from TPT is low. If TPT increases the risk of stillbirth or low birth weight over those cutoffs, deferring TPT until after delivery may be safer. This modeling analysis explicitly incorporated hepatotoxicity risk into their clinical decision framework, acknowledging the theoretical increased risk of drug-induced liver injury during pregnancy. The authors also note that 3HP may be preferred over the 6H due to better adherence of a shorter regimen therefore reducing hepatotoxicity risk overall and heightening effectiveness. The study did not, however, address potential drug–drug interactions with 3HP and ART regimens.⁶⁷

Implementation And Public Health Perspectives

Choosing a regimen is just one aspect of optimizing TB prevention in pregnancy. Implementation can also be a major hurdle for any preventive health issue. In 2011, the Lesotho Ministry of Health found that integrating active TB case finding into maternal and child health settings improved IPT uptake as well. All pregnant women were screened for symptoms of active TB. IPT was initiated in 78.5% (124/158) of pregnant women with HIV who screened negative for active TB and had no contraindications; a marked increase from the period before active case finding was initiated.^56,68 Similar to Lesotho, a study in Malawi found that integrating IPT into HIV care resulted in improved IPT uptake, even in pregnant women.⁶⁹ The overall IPT initiation rate was 70% (95% CI: 46%–86%) among non-pregnant adults with HIV, with a comparable rate of 67% (95% CI: 32%–90%) among pregnant women.⁶⁹

But context is important. An analysis from three large hospitals in Cape Province, South Africa, identified geographic inequities in IPT uptake among pregnant women living with HIV, with rural residence associated with lower IPT uptake.⁷⁰ A study from Matlosana, South Africa, found that comprehensive interventions to build trust among pregnant women living with HIV may increase the uptake of TB preventive treatment in both antepartum and postpartum settings.⁷¹ These studies suggest that successful IPT implementation is possible within existing healthcare frameworks, but community engagement, access to care, and trust-building are essential. Key considerations for integration of TB screening into routine HIV and antenatal care services include training healthcare providers on pregnancy-specific protocols and establishing monitoring systems for adverse events, including those that are pregnancy-specific.⁷²

Summary Of Guidelines For TPT In Pregnant Women With HIV

The WHO recommends TPT for all people living with HIV, including pregnant women. The regimens recommended for pregnancy are 6H, 9H, 4R, or 3HR, despite safety concerns with IPT and limited pregnancy-specific data for rifampin-containing regimens.⁷³ It is also mentioned that 3HP may be safe although more evidence is needed. The 2024 WHO consolidated guidelines continue to recommend IPT for pregnant women with HIV, because they contend that systematic deferral of IPT to postpartum would deprive women of protective benefit when they are most vulnerable to TB. They strongly encourage baseline liver function tests when IPT is given in pregnancy, especially if there are other risk factors for liver toxicity (eg. hepatitis B). They also recommend vitamin B6 supplementation for anyone taking a regimen with INH. The WHO acknowledges that rifampin can be given in pregnancy though there are limited data to make a recommendation on rifapentine, specifically.^8,9 They emphasize the importance of considering drug–drug interactions between rifamycin-based TPT in people with HIV on ART. Table 1 summarizes the 2024 WHO Consolidated Guidelines for TPT regimens. Notably, this was published prior to the DOLPHIN Moms study data was available.

The US CDC recommends that pregnant women who are at risk for developing TB disease, including those living with HIV, should be tested for TB infection. In general, they recommend delaying treatment for TB infection in most pregnant women. But if pregnant women with HIV have a positive TB infection test and a low CD4 count, then treatment should not be delayed, even in the first trimester.⁷⁴ Similar to the WHO, regimens that are recommended in pregnancy include 4R, 3HR, 6H, and 9H, though they do not specifically comment on the safety of those regimens for women living with HIV on ART. They also recommend pyridoxine supplementation if taking INH.⁷⁴ For pregnant women who have detectable virus, a close household contact with pulmonary TB or a recent positive test for TB infection, IPT (6H or 9H) is recommended as first-line with 3HR and 4R listed as acceptable alternatives.²⁷ For pregnant women with HIV who are receiving effective ART and have none of the risk factors listed above, TPT may be deferred until after delivery.

The lack of high-quality safety data, however, is more challenging for countries with a high TB burden. In 2025, South Africa’s National Essential Medicine List Committee (NEMLC) and Expert Review Committee (ERC) changed their guidelines to recommend provision of TPT to pregnant women living with HIV be based on CD4 count: CD4 counts ≤200 cells/mm³ with ART initiation should receive 12 months of IPT after exclusion of TB disease; CD4 counts >200 cells/mm³ should defer IPT until postpartum.⁷⁵ This decision has been controversial as South Africa has the highest burden of TB in the world, with an incidence of pulmonary tuberculosis of 852 cases per 100,000 in 2018.⁷⁶ Because of potential safety concerns with IPT and insufficient safety data on 3HP, most pregnant women with HIV in South Africa are currently not offered TPT.

TPT Clinical Management Strategy

A reasonable approach to TPT in pregnant women living with HIV can be seen in Figure 2. In areas with TB transmission <500/100,000, initiate TPT immediately if a pregnant woman with HIV has had a TB infection test convert from negative to positive in the past two years, or has had recent exposure to a person with infectious active TB. If the patient has neither indication for immediate TPT, then the health care provider and the pregnant woman should engage in shared medical decision-making regarding timing of TPT. If she is on effective ART with adequate CD4 counts, it is reasonable to defer TPT until two to three months after delivery.⁷⁴ If she is not on ART, then she should initiate ART as soon as possible, be evaluated for active TB, and initiate TPT if indicated.

Figure 2 A reasonable approach to TPT in pregnant women living with HIV proposed by the authors given the available evidence.

Strengths, Limitations, and Future Directions

Our review has several strengths and limitations. Our study design of narrative review was chosen for flexibility to identify articles that were relevant to the TPT regimens currently in guidelines for people living with HIV. This may not include all relevant literature on the topic. However, given the limited number of TPT regimens, our search strategy was thorough, geographically diverse, and considered complete when we were unable to identify new publications about this topic that specifically noted pregnant women. Additionally, we utilized only one database (PubMed) for article selection which may have also limited our results. However, we did include manuscripts with any study design (ie. observational vs interventional), including data that has been recently presented and published in conference hearings, which allowed us to provide the most current data on use of these regimens in pregnancy.

Our review highlights the lack of clinical trial data to inform guidelines regarding safety of TPT in pregnant women living with HIV.^77,78 Studies that include pregnant and postpartum women are critical to determine safe management options for women with HIV and for infants who are breastfeeding. Additionally, studies including neonates and follow up of children born to mothers with HIV on TPT are needed to assess true safety of the mother and child dyad.

Conclusions

This review offers a comprehensive look at the current recommendations regarding TPT in pregnant women living with HIV. However, significant gaps remain for the safety and pharmacokinetics of TPT, including drug–drug interactions with ART. Emerging evidence suggests that TPT can be administered safely and effectively during pregnancy. But for pregnant women to continue to protect themselves and their babies from the adverse effects of maternal TB, they must be included in trials of new TB prevention drugs and regimens. Of note, drug–drug interactions regarding neonatal safety during breastfeeding were beyond the scope of this article. It is an important area of ongoing research, and further understanding is necessary for pregnant and postpartum women to safely complete TB treatment courses. For pregnant women to continue to protect themselves and their babies from the adverse effects of maternal TB, they must be included in trials of new TB prevention drugs and regimens. Efforts to standardize and expand TPT in pregnant women living with HIV are a necessary component to reducing TB-related maternal morbidity and mortality globally.