NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.

Antiretroviral Therapy for HIV Infection in Adults and Adolescents: Recommendations for a Public Health Approach: 2010 Revision. Geneva: World Health Organization; 2010.

Cover of Antiretroviral Therapy for HIV Infection in Adults and Adolescents

Antiretroviral Therapy for HIV Infection in Adults and Adolescents: Recommendations for a Public Health Approach: 2010 Revision.

Show details


15.1. Recommendations for HIV-infected pregnant women

  1. Start ART in all pregnant women with HIV and a CD4 count of ≤350 cells/mm3, irrespective of clinical symptoms.
    (Strong recommendation, moderate quality of evidence)
  2. CD4 testing is required to determine if pregnant women with HIV and WHO clinical stage 1 or 2 disease need to start ARV treatment or ARV prophylaxis for PMTCT.
    (Strong recommendation, low quality of evidence)
  3. Start ART in all pregnant women with HIV and WHO clinical stage 3 or 4, irrespective of CD4 count.
    (Strong recommendation, low quality of evidence)
  4. Start one the following regimens in ART-naive pregnant women eligible for treatment:
    • AZT + 3TC + EFV;
    • AZT + 3TC + NVP;
    • TDF + 3TC (or FTC) + EFV;
    • TDF + 3TC (or FTC) + NVP.
    (Strong recommendation, moderate quality of evidence)
  5. Do not initiate EFV during the first trimester of pregnancy.
    (Strong recommendation, low quality of evidence)

In making these recommendations, the ART and PMTCT panels placed high value on ensuring that treatment begins early for pregnant women with HIV, improving maternal and child-health outcomes and avoiding MTCT, over and above concerns about cost or feasibility.

When to start

On the question of when to start, no studies specific to pregnant women were identified in the systematic review prepared for this guideline revision. Evidence from the general population supports strong recommendations for the timing of initiation in terms of reduction of mortality, disease progression, serious adverse events, the risk of TB and the risk of HIV transmission (sexual and mother to child). As with the recommendation on when to start in the general population, the panel recognized the uncertainty around the prognostic value of some WHO clinical stage 2 conditions, and data from modelling and observational studies indicating that more than 50% of HIV-infected patients with WHO clinical stage 2 have a CD4 count of ≤350 cells/mm3. The panel therefore recommended that all pregnant women with WHO clinical stages 1 and 2 should have access to CD4 testing in order to decide when to start treatment.

What to start

On the question of what to start, no GRADE evidence profiles were prepared as no RCTs were identified that compared the use of different ARV regimens in pregnant women. Cohort studies report a reduction of HIV transmission and death.(100) There is no evidence to suggest an increase in maternal serious adverse events and there are no studies specifically evaluating maternal response to ART. Registry data on the use of TDF in pregnancy show no signals to raise concern, and there is no evidence to suggest that TDF + 3TC (or FTC) is not an acceptable alternative to AZT + 3TC.(101,102)

As discussed in the section on EFV, there is very low quality conflicting evidence on the risks of EFV causing neural tube defects, with the overall rates of birth defects reported in association with EFV, NVP and TDF similar to rates reported in congenital defects registries of general populations. However, data are currently insufficient to determine whether there is an increased risk of rare anomalies such as neural tube defects with first-trimester EFV exposure.

The review of NVP safety in pregnant women with CD4 counts between 250 and 350 cells/mm3 did not confirm an increased risk of serious adverse events. However, while data from two prospective cohorts indicate no association between NVP and liver enzyme elevation, pregnancy itself was associated with an increased risk of any liver enzyme elevation and that this association was present, regardless of prior ART and NVP exposure history.(103) The panel concluded that the benefits of using NVP in pregnancy outweighed the risks. The panel was unable to conclude from the evidence reviewed whether there were benefits associated with the use of EFV compared to NVP in pregnant women after the first trimester and with higher or unknown CD4 cell counts, although more than half of the panel members preferred EFV in these situations.

15.2. Recommendations for women with prior exposure to antiretrovirals for PMTCT

  1. Initiate a non-NNRTI-based ART in women who have received single-dose nevirapine (sdNVP) alone or in combination with other drugs without an NRTI tail within 12 months of initiating chronic ART. If an NNRTI-based regimen is started, perform viral load testing at 6 months and, if there are >5000 copies/ml, switch to a bPI-based regimen.
  2. Initiate a standard NNRTI-based ART regimen in women who have received sdNVP alone or in combination with other drugs with an NRTI tail within 12 months of initiating chronic ART and perform viral load-testing at 6 months. If the viral load is >5000 copies/ml, changing to a bPI is recommended.
  3. Initiate a standard NNRTI-based ART regimen in women who have received sdNVP (alone or in combination with other drugs) more than 12 months before starting therapy (with or without a NRTI tail) if possible. The viral load should be evaluated at 6 months and if it is >5000 copies/ml a change in the bPI-based regimen is required.

Initiate a standard NNRTI regimen in women who have received ARV drugs such as AZT alone, without sdNVP, for PMTCT.

Table 11ART regimens recommended for women with prior exposure to PMTCT regimen

Previous ARV exposure for PMTCTRecommendations for initiation of ART when needed for treatment of HIV for maternal health
sdNVP1 (+/- antepartum AZT) with no AZT/3TC tail2 in last 12 monthsInitiate a non-NNRTI regimen

PI preferred over 3 NRTI
sdNVP (+/- antepartum AZT) with an AZT/3TC tail in last 12 monthsInitiate an NNRTI regimen

If possible, check viral load3 at 6 months and if >5000 copies/ml, switch to second-line ART with PI
sdNVP (+/- antepartum AZT) with or without an AZT/3TC tail over 12 months agoInitiate an NNRTI regimen

If possible, check viral load3 at 6 months and if >5000 copies/ml, switch to second-line ART with PI
Option A4

Antepartum AZT (from as early as 14 weeks of gestation)

sdNVP at onset of labour*

AZT + 3TC during labour and delivery*AZT + 3TC tail for 7 days postpartum*

* sd-NVP and AZT + 3TC can be omitted if mother receives >4 weeks of AZT antepartum
Initiate an NNRTI regimen

If possible, check viral load3 at 6 months and if >5000 copies/ml, switch to second-line ART with PI

If no sdNVP was given, start standard NNRTI (viral load does not need to be checked unless clinically indicated as no sdNVP received)
All triple ARV regimens (including Option B), irrespective of duration of exposure and time since exposure

Option B4

Triple ARV from 14 weeks gestation until after all exposure to breast milk has ended

AZT + 3TC + LPV/r



TDF + [3TC or FTC] + EFV
Initiate standard NNRTI regimen

If EFV-based triple ARV was used for prophylaxis and no tail (AZT + 3TC; or TDF + 3TC; or TDF + FTC) was given when triple ARV was discontinued after cessation of breastfeeding (or delivery if formula feeding), check viral load3 at 6 months and if >5000 copies/ml, switch to second-line ART with PI

Single-dose nevirapine (sdNVP) is one 200-mg tablet of NVP.


A tail is the provision of two NRTIs, typically AZT/3TC, for a minimum of 7 days following sdNVP or the cessation of any NNRTI-based regimen with the objective of minimizing NNRTI resistance.


If VL is not available, continue NNRTI regimen and monitor clinically (and immunologically if available).


Options A or B are viewed as equally effective for PMTCT in women who do not require therapy for their own health and are recommended options in the 2010 update of Use of antiretroviral drugs for treating pregnant women and preventing HIV infection in infants.


The long half-life of NVP and its low genetic barrier to resistance means that detectable drug levels persist for 2–3 weeks in the presence of active viral replication following a single maternal dose. (104106) EFV also has a long half-life, with detectable drug levels for more than 21 days following discontinuation.(107) This has clinical relevance in pregnancy when ARVs are provided solely for prophylaxis against perinatal transmission and discontinued after delivery or after breastfeeding. In a meta-analysis of 10 studies, the prevalence of NVP resistance 4 to 8 weeks following sdNVP was 35.7% and the prevalence of NVP resistance in infants who became infected despite prophylaxis was 52.6%.(108) In most women, resistant virus can no longer be detected 6 to 12 months after exposure. However, low levels of viral resistance can persist for longer periods and in some cases can remain present in latently infected cells.(109111)

Data suggest that women starting NNRTI-based therapy within 6–24 months of sdNVP exposure have higher rates of viral failure than those without sdNVP exposure. A definite relationship between time from sdNVP exposure to starting NNRTI-based therapy has been observed but varied between studies from 6 months to 24 months, with a definite improvement in response if >12 months since sdNVP exposure and start of therapy.(112119) A tail regimen for a minimum of 7 days is recommended following sdNVP or if NNRTI-based triple therapy ART is used for the prevention of perinatal transmission and subsequently stopped. NNRTI resistance rates of 0% to 7% at 2 to 6 weeks postpartum have been reported with the use of various tail regimens. (120125)

15.3. Recommendations for HIV/HBV coinfection

  1. Start ART in all HIV/HBV-coinfected individuals who require treatment for their HBV infection, (chronic active hepatitis), irrespective of the CD4 cell count or the WHO clinical stage.
    (Strong recommendation, low quality of evidence)
  2. Start TDF and 3TC (or FTC)-containing antiretroviral regimens in all HIV/HBV coinfected individuals needing treatment.
    (Strong recommendation, moderate quality of evidence)

In developing these recommendations, the panel placed high value on promoting HBV diagnosis and more effective treatment of HIV/HBV coinfection. The systematic review of this topic did not find RCTs which addressed critical HIV outcomes (death, disease progression, serious adverse events) and the GRADE profile reported only on outcomes related to HBV (HBV viral load and HBV drug resistance).

Liver biopsy and HBVDNA are not usually available in the large majority of resource-limited settings. A global definition of chronic active hepatitis for resource-limited settings based on clinical and available laboratory parameters is under discussion.

When to start

On the question of when to start ART in HIV/HBV coinfection, there are no RCTs comparing early versus late initiation of ART. However, observational data demonstrate that individuals with HIV/HBV coinfection have a threefold to sixfold increased risk of developing chronic HBV infection, an increased risk of fibrosis and cirrhosis and a 17-fold increased risk of death compared to HBV-infected individuals without HIV infection.(126,127 ) Similarly, observational data support a reduction in liver-related disease with earlier and HBV-active combination ART.(128)

What to start

On the question of what ART to start in HIV/HBV coinfection, there are data from one RCT supporting the use of at least two agents with activity against HBV (TDF plus 3TC or FTC) in terms of improved HBV viral load response and reduced development of HBV drug resistance. (129,130 )

15.4. Recommendations for HIV/tuberculosis coinfection

  1. Start ART in all HIV-infected individuals with active TB, irrespective of the CD4 cell count.
    (Strong recommendation, low quality of evidence)
  2. Start TB treatment first, followed by ART as soon as possible afterwards (and within the first eight weeks).
    (Strong recommendation, moderate quality of evidence)
  3. Use efavirenz (EFV) as the preferred NNRTI in patients starting ART while on TB treatment.
    (Strong recommendation, high quality of evidence)

In making these recommendations, the panel placed high value on the reduction of early mortality from HIV/TB coinfection, the potential for reduction of TB transmission when all individuals with HIV are started on ART earlier, and improved morbidity/mortality, reduction of TB recurrence and improved management of TB for coinfected HIV/TB patients.

When to start

On the question of when to initiate ART in TB infection, one RCT (SAPIT study) provides moderate evidence for the early initiation of ART in terms of reduced all-cause mortality and improved TB outcomes.(131) Trial participants were grouped into “integrated” (immediate and end of TB drug initiation phases combined) and “sequential” treatment arms. Mortality was 55% lower in the integrated treatment arm (5.1/100 person-years) compared to the sequential treatment arm (11.6 per 100 person-years), which was terminated. The trial is continuing to examine the outcomes of starting ART immediately or starting at the completion of the initiation phase of TB treatment. Until further data are available, it is recommended that ART be initiated as soon as TB therapy is tolerated. Ideally, this may be as early as 2 weeks and not later than 8 weeks.

There are limited data on the initiation of ART in patients with TB and CD4 counts of >350 cells/ mm3.

Impact on TB transmission and incidence

ART has been reported to reduce TB rates by up to 90% at the individual level and by approximately 60% at the population level, and to reduce TB recurrence rates by 50%.(13,132,133) Modelling suggests that the initiation of ART for all those with HIV/TB coinfection, if accompanied by high levels of coverage and ART adherence, reduces the number of TB cases, TB mortality rates and TB transmission at the population level.(134)

What to start

The recommendations from the 2006 ART guidelines are maintained. Specifically, EFV is recommended because of less interaction with rifampicin compared to NVP. For those HIV/TB coinfected individuals who are unable to tolerate EFV, an NVP-based regimen or a triple NNRTI (AZT + 3TC + ABC or AZT + 3TC + TDF) are alternative options. In the presence of rifampicin, no lead-in dose of NVP is required.(50,135138). Similarly, if patients temporarily change from NVP to EFV because they need to take rifampicin-containing TB therapy and subsequently switch back to NVP on completion of TB treatment, no lead-in dosing of NVP is required.(60,61)

15.5. Rifabutin


Drug interactions between rifampicin and boosted protease inhibitors (bPIs) prohibit the concomitant use of standard therapies for both HIV and TB. Rifampicin induces the cytochrome P450 enzyme system, lowering standard-dose bPI plasma concentrations by 75–90%. All bPIs (at standard doses) are contraindicated with rifampicin. LPV/r or SQV/r may be used with an adjusted, superboosted dose of RTV (LPV/r 400 mg/400 mg BID or SQV/r 400 mg/400 mg BID) or doubling the standard LPV/r daily dose (LPV/r 800 mg/200 mg BID) but this is associated with high levels of toxicity, and requires close clinical and laboratory monitoring. The recommendation to use LPV/r 800 mg/200 mg BID is based on low-quality evidence and is associated with a similar level of toxicity to LPV/r 400 mg/400 mg BID. However, this option may be more feasible in RLS, as LPV/r is widely available but RTV as a sole formulation is not.(139142)

There is no comparable recommendation for ATV/r, a WHO-preferred bPI (143). Unlike rifampicin, rifabutin has minimal effect on bPI plasma concentrations.


In a systematic review conducted for this guideline update, ten clinical trials (five RCTs and five cohort studies) were identified, which assessed the efficacy and safety of rifabutin in TB infection with or without HIV infection. The five RCTs were included in a Cochrane review, which found no differences in TB cure or relapse rates between rifampicin and rifabutin.(28)

In the five cohort studies, 313 individuals received rifabutin and ART, of whom 125 received a PI. Due to methodological issues, no rigorous efficacy assessment from these studies was possible, but there was no sign of rifabutin inferiority in comparison with rifampicin.

Taken together, these studies report comparable safety and efficacy of rifabutin and rifampicin. However, evidence from RCTs comes largely from HIV-uninfected individuals, and data on the use of rifabutin with ART are limited to first-generation, usually unboosted, PIs. A further limitation is that the evidence in HIV-infected individuals receiving a bPI and rifabutin is based on only 125 patients. In addition, the clinical experience with rifabutin for TB disease in resource-limited settings is limited, especially in the context of the bPIs currently recommended by WHO.

Clinical considerations


The recommended dose of rifabutin in the presence of a bPI is 150 mg three times per week. (144) However, it should be noted that this dose has been reported to result in inadequate rifabutin levels and acquired rifabutin resistance.(145 ) Rifabutin is contraindicated if administered with the new NNRTI etravirine plus a bPI (37% reduction of etravirine levels).

Adverse events

The most common adverse events associated with rifabutin are neutropenia, leucopenia, elevations of hepatic enzymes, rash and upper gastrointestinal complaints, and, more rarely, uveitis. In the systematic review, discontinuation attributable to adverse events was uncommon. This review revealed one case report of uveitis in combination with a bPI.(146)

Monitoring and programmatic implications

This systematic review indicates that a bPI and rifabutin coadministration will not require intensive monitoring and can be used in primary care settings. However, the DOTS strategy promotes daily administration of TB therapy, preferably in FDCs.(147 ) Intermittent dosing of rifabutin will complicate the programmatic roll-out of TB therapy and precludes the development of rifabutin-containing FDCs. Further research is needed into the pharmacokinetics of rifabutin 75 mg once-daily in the presence of bPIs. Meanwhile, the ability to use standard bPI doses outweighs the inconvenience of intermittent dosing.

Copyright © 2010, World Health Organization.

All rights reserved. Publications of the World Health Organization can be obtained from WHO Press, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland (tel.: +41 22 791 3264; fax: +41 22 791 4857; e-mail: tni.ohw@sredrokoob). Requests for permission to reproduce or translate WHO publications – whether for sale or for noncommercial distribution – should be addressed to WHO Press, at the above address (fax: +41 22 791 4806; e-mail: tni.ohw@snoissimrep).

Bookshelf ID: NBK138524


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...