Isoniazid preventive therapy: do the risks outweigh the benefits in pregnant women with HIV?

Isoniazid preventive therapy (IPT) during and after pregnancy in women living with HIV on antiretroviral therapy (ART) in tuberculosis (TB) endemic areas in Africa, Asia and Haiti resulted in serious adverse events quite possibly attributable to isoniazid with no significant reduction in TB cases, participants heard last week at the Conference on Retroviruses and Opportunistic Infections (CROI 2018) in Boston.

Adverse pregnancy outcomes were significantly higher among women getting IPT during pregnancy than after (23% (106) and 17% (77), respectively, p = .009), Dr Amita Gupta presenting on behalf of the IMPAACT P1078/TB Apprise study team stated.

Of the six maternal deaths, four were due to liver failure and all occurred after delivery (at 5, 5, 7, 12, 19 and 40 weeks).



An antibiotic that works by stopping the growth of bacteria. It is used with other medications to treat active tuberculosis (TB) infections, and on its own to prevent active TB in people who may be infected with the bacteria without showing any symptoms (latent TB). 




A pill or liquid which looks and tastes exactly like a real drug, but contains no active substance.

person years

In a study “100 person years of follow-up” could mean that information was collected on 100 people for one year, or on 50 people for two years each, or on ten people over ten years. In practice, each person’s duration of follow-up is likely to be different.


A detailed research plan that describes the aims and objectives of a clinical trial and how it will be conducted.

This phase IV randomised, double-blind, placebo-controlled trial, the first to focus on TB prevention in pregnant and postpartum women living with HIV and at high risk of developing TB, comprising 956 women enrolled from August 2014 to April 2016 included 13 sites in eight countries.

These findings challenge current World Health Organization (WHO) guidelines. Dr James McIntyre of Anova Health Institute, South Africa, chairing a press conference, supported the study team’s recommendation to re-evaluate the guidelines weighing the risks and benefits of starting isoniazid in pregnant women living with HIV.

TB surpasses HIV as the primary infectious disease killer globally. Having HIV and being pregnant increases the risk of developing TB. Maternal TB can result in death or adverse pregnancy outcomes as well as infant TB and HIV transmission.

WHO, based on strong evidence, recommends IPT plus ART for people living with HIV. However, for pregnant and postpartum women the quality of evidence is weak. Pregnant women are excluded from IPT/TB prevention trials. Additionally, retrospective data show that isoniazid is associated with increased liver damage in pregnant and postpartum women.

Dr Gupta noted that the safety, efficacy and optimal timing of IPT for pregnant women living with HIV on ART are unknown.

Based on the hypothesis that IPT can be safely begun during pregnancy the team chose to compare overall maternal safety and toxicity of immediate (begun in pregnancy) to deferred (begun postpartum) isoniazid in pregnant women living with HIV enrolled at or after 14 weeks through 34 weeks gestation.

Women were randomised 1:1 to arm A (immediate IPT) begun at entry (isoniazid 300mg daily for 28 weeks then placebo), and arm B (deferred IPT) began on placebo until week 12 after delivery then received isoniazid 300 mg daily for 28 weeks. The mother-infant pairs were followed up until 48 weeks postpartum.

Monthly safety evaluations were undertaken. The primary safety endpoint was first treatment-related maternal adverse events at or greater than grade three or permanent drug discontinuation due to toxicity.

Secondary outcomes included maternal liver damage, maternal and infant death, TB, adverse pregnancy outcomes and infant adverse events.

Of the 956 women enrolled, 93% were black with a median age of 29 years. The median CD4 cell count was 493 cells/mm3.  Approximately a third were enrolled between 14 and 24 weeks into the pregnancy and 66% between 24 and 34 weeks.

A total of 776 (81%) had a viral load under 200 copies/ml. All were on ART with 814 (85%) on an efavirenz-based regimen of which one was on efavirenz alone and another on efavirenz and nevirapine combined. Thirty per cent (284) had a positive interferon gamma release assay (IGRA) (a blood test to determine the presence of latent TB). Three per cent (31) had a previous history of TB. Median follow-up was 58.6 weeks.

Fifteen per cent reached the primary safety end point, 74 and 73 in arms A and B, respectively. This signifies incidence rates of 15.4 and 14.9 per 100 person-years with an incidence rate difference of 0.5, (95% CI: -4.4-5.4).

Approximately 30% had all-cause grade three or above maternal adverse events, 145 in arm A and 136 in arm B with incidence rates of 35.4 and 31.3 per 100 person-years, respectively. The incidence rate difference was 4.2 (95% CI: -3.6-12.0).

Forty-five women permanently discontinued the study drug because of liver function adverse events (35 discontinued because of protocol-defined toxicity; nine due to non protocol-defined low grade toxicity and one died due to toxicity).

All-cause liver damage occurred in 29 women (6%) in arm A and 34 (7%) in arm B. The incidence rates per 100 person-years were 5.8 and 6.7, respectively. The incidence rate difference was -0.9 (95% CI: -4.0-2.2).

Dr Gupta stated that there were no significant differences in maternal safety by treatment arm. However, higher levels of liver damage were seen after delivery regardless of whether on IPT or type of ART.

The six maternal deaths (two in the immediate IPT arm, four in the deferred IPT arm) were in three countries (Zimbabwe, Botswana and Tanzania). The women were between 24 and 38 years old with CD4 cell counts ranging from 402 to 609 cells/mm3.).  All women were on efavirenz/tenofovir/emtricitabine started before IPT.

A review of the first two maternal deaths by the Data and Safety Monitoring Board resulted in a request for a participant letter with explicit information regarding the signs and symptoms of liver damage and the risk of death from isoniazid and ART. Consequently 77 (8%) withdrew consent prior to study completion.

Forty-two per cent (380) of infants experienced grade 3 or 4 adverse events with no significant differences seen in infant safety by treatment arm.

Four of the six maternal TB cases were culture confirmed, of which one was isoniazid resistant (arm A). There was one case of infant TB. No significant differences were seen in maternal or infant TB by treatment arm.

Timing of IPT did not affect TB risk.


Gupta A et al. Randomized trial of safety of isoniazid preventive therapy during or after pregnancy, 25th Conference on Retroviruses and Opportunistic Infections (CROI 2018), Boston, abstract 142LB, 2018.

View the abstract on the conference website.

Watch the webcast of this presentation on the conference website.

Update. Following the conference presentation, this study was published in a peer-reviewed journal:

Gupta A et al. Isoniazid Preventive Therapy in HIV-Infected Pregnant and Postpartum Women. New England Journal of Medicine 381:1333-1346, 2019.

DOI: 10.1056/NEJMoa1813060