University of California, San Francisco Logo

University of California, San Francisco | About UCSF | Search UCSF | UCSF Medical Center

Pregnancy and Mother-to-Child Transmission
HIV InSite's Coverage of the 14th Conference on Retroviruses and Opportunistic Infections
Claire Townsend, MSc, Centre for Paediatric Epidemiology and Biostatistics, Institute of Child Health, University College London

Published March 26, 2007
Back to Main Conference Coverage Page

Maternal and Perinatal Issues
Safety of ART in Pregnancy and Postpartum Period
Safety of Specific Antiretroviral Drugs in Pregnancy
Response to ART in Pregnancy
Mother-to-Child Transmission
Pharmacokinetic Studies in Pregnancy

Maternal and Perinatal Issues

Abrams launched the discussion about maternal and pediatric care by highlighting the priorities in both resource-rich and resource-limited settings.(1) She pointed out that the success of efforts for the prevention of mother-to-child transmission (PMTCT) of HIV in resource-rich countries means that the focus is now on the efficacy, safety, and dosing of antiretroviral regimens, and on the treatment of women with drug-resistant virus. In resource-limited settings, however, improving and expanding coverage of PMTCT programs and tackling the issue of whether to breast-feed, and if so, for how long, remain the priorities.

Safety of ART in Pregnancy and Postpartum Period

Several posters based on larger trials or cohorts of HIV-infected adults, mostly in sub-Saharan Africa, focused on a subset of pregnant women receiving antiretroviral therapy (ART) for PMTCT and explored pregnancy and perinatal outcomes. Findings were generally positive, with low rates of adverse outcomes such as birth defects, infant mortality, maternal and pediatric anemia, and disease progression following postpartum treatment cessation.

Namale and colleagues reported on the outcomes of 198 pregnancies from the DART trial of HIV-infected adults in Uganda and Zimbabwe.(2) All women in the cohort took ART, mostly zidovudine + lamivudine with tenofovir or nevirapine. Among pregnancies with known outcomes, 55% resulted in live birth, 7% in stillbirth, and 38% in abortion (half of which were spontaneous). Four women died, 2 in pregnancy and 2 postpartum. Of the 91 live-born infants, 11% were born prematurely (<37 weeks), 2% had congenital abnormalities, 5% died neonatally, and none were HIV infected.

A poster from the Kisumu Breastfeeding Study (KiBS) in Kenya described the prevalence of anemia among 254 women receiving ART (lamivudine/zidovudine and either nevirapine or nelfinavir) from 34 weeks' gestation until 6 months postpartum.(3) Surprisingly, they found that the proportion of women with anemia (hemoglobin count ≤9.4 g/dL) decreased substantially over time, both in women with advanced disease and in those with earlier HIV disease. However, 5% of women developed grade III or IV anemia during the intervention, half of whom discontinued zidovudine for that reason.

Taha and colleagues described hematologic and hepatic parameters in HIV-uninfected infants given single-dose nevirapine and zidovudine for 1 week; some were also given nevirapine or nevirapine + zidovudine up to 14 weeks of age.(4) The investigators found the prevalence of anemia to be very low (<2%) in all groups at 3 months of age. At 6 and 9 weeks, prevalence was higher in infants exposed to nevirapine or nevirapine + zidovudine up to 14 weeks, but hemoglobin levels recovered after cessation of treatment. Hepatic factors also were explored, and no particular cause for concern was identified.

Joao and colleagues explored birth weight and preterm delivery in 645 pregnant women in Brazil who started ART in pregnancy.(5) The ART regimens included nelfinavir monotherapy, nevirapine monotherapy, and zidovudine alone or zidovudine + lamivudine. Newborns whose mothers were on ART had slightly lower birth weights than those whose mothers were not on ART, but the proportion of newborns with birth weight of <2,500 g was similar. Preterm delivery was 2-fold higher in the nelfinavir group compared with the nevirapine group.

The effects on disease progression of treatment interruption after pregnancy were explored in the Women and Infants Transmission Study (WITS) in the United States.(6) Among 206 ART-naive women with CD4 counts of ≥350 cells/µL, the authors found that progression to AIDS and changes in CD4 count and HIV viral load over the following year were similar regardless of whether treatment was stopped after delivery or continued postpartum. Progression to U.S. Centers for Disease Control and Prevention (CDC) class B events occurred more frequently among women who stopped treatment than among those who continued (25.5% vs 11.5%), but this difference was not statistically significant.

Safety of Specific Antiretroviral Drugs in Pregnancy

Several studies addressed concerns about specific antiretroviral drugs, including nevirapine, efavirenz, boosted saquinavir, and atazanavir.

Phanuphak and colleagues presented two posters based on a prospective study in Bangkok, Thailand.(7,8) Pregnant women were given zidovudine, lamivudine, and nevirapine from 14 weeks if their CD4 count was <200 cells/µL at the time, and from 28 weeks if their CD4 count was ≥200 cells/µL. If toxicities were detected, nevirapine was stopped and the women switched to efavirenz or a protease inhibitor (PI). One poster explored nevirapine-associated toxicities in 494 women, 98 of whom had CD4 counts between 250 and 350 cells/µL.(7) The study confirmed that higher rates of hepatotoxicity (grades I to IV) and rash were found in women with CD4 counts ≥350 cells/µL. However, no excess of hepatotoxicity was detected in women with CD4 counts between 250 and 350 cells/µL. The other poster explored the use of efavirenz after the first trimester of pregnancy in 606 women (including some enrolled in the MTCT-Plus Initiative), 38 of whom had switched from nevirapine to efavirenz because of toxicity.(8) Mean duration of efavirenz exposure in pregnancy was 7 weeks. Preterm birth (<37 weeks' gestation) was reported in 10% and low birth weight (<2,500 g) was reported in 16%. There were no birth defects. These rates were no higher than those reported for 553 pregnant women from the same cohort who remained on nevirapine-based regimens.

Natha and et al reviewed the notes of 31 HIV-infected women in London who received atazanavir in pregnancy (33 pregnancies); 20 of the pregnancies occurred with women who conceived on atazanavir.(9) Among the 13 cases in which atazanavir was initiated during gestation, pretreatment serum bilirubin and alanine aminotransferase (ALT) levels were within normal limits. Of the 25 women who delivered, 23 had elevated bilirubin levels (>17 µm/L) at delivery, but infant outcomes were good: none required phototherapy, had birth defects recorded, or were HIV infected. Adequate concentrations of atazanavir were recorded in most women.

The safety of boosted saquinavir in pregnancy was discussed by Hanlon and colleagues, who recorded hepatotoxicity (grades I-IV) in 45 women who began zidovudine, lamivudine, and boosted saquinavir in the third trimester of pregnancy.(10) About a third developed abnormal transaminase levels within 2 to 4 weeks of ART initiation, and ART was changed in 6 patients as a result of hepatotoxicity. All transaminase levels normalized within 6 weeks of delivery.

Response to ART in Pregnancy

A number of studies explored response to treatment in HIV-infected pregnant women starting ART for the first time in pregnancy.

Two posters explored time to undetectable viral load following initiation of ART. Data from the European Collaborative Study on 240 ART-naive women who started ART in pregnancy showed that about three quarters of them achieved viral suppression before delivery.(11) The authors found that women on nevirapine-based ART achieved viral suppression (<50 copies/mL) faster than those on PI-based ART. They also found that black women reached viral suppression faster than white women. Similarly, a study in Canada by Schalkwyk and colleagues found that 80% of 114 women achieved viral suppression by the time of delivery, with a mean time to viral suppression of 58 days.(12) However, they found no difference in time to undetectable viral load between women on PI-based regimens and women on nonnucleoside reverse transcriptase inhibitor-based regimens.

In a report from the KiBS in Kenya, Awino and colleagues compared viral load and CD4 counts for 203 pregnant women at treatment initiation (34 weeks' gestation), delivery, and 6 months postpartum.(13) The proportion of women with viral loads of ≥10,000 copies/mL declined from about 80% to 9% at delivery, then rose to 17% at 6 months postpartum. CD4 counts showed improvements throughout the study period. More than 80% of women had good adherence, and, of these, 84% had viral loads <10,000 at 6 months postpartum.

Mother-to-Child Transmission

Mother-to-child transmission (MTCT) rates in both resource-rich and resource-limited areas were reported, and the cost effectiveness of providing ART for preventing MTCT also was discussed.

Townsend and colleagues presented data from national obstetric and pediatric HIV surveillance in the United Kingdom and Ireland. They reported that MTCT rates had declined from a high of 26% in 1993 to about 1% in 2004.(14) Among 3,703 pregnancies, risk factors for transmission included vaginal or emergency cesarean section delivery, mothers not receiving ART in pregnancy, premature delivery, and female gender for infants. Mono or dual therapy was independently associated with a 7-fold reduction, and regimens consisting of 3 or more drugs were associated with a 10-fold reduction in the risk of transmission, compared with no ART. In women who received ART in pregnancy, elective cesarean section was associated with a 3-fold decreased risk of transmission compared with vaginal delivery.

Kuhn et al investigated the association between a gene for a CCR5 ligand (CCL3-L1) and the risk of MTCT in 4 South African cohorts.(15) DNA samples were taken from 852 HIV-infected pregnant women and their infants; two thirds of women received single-dose nevirapine. The authors reported that uninfected infants tended to have more copies of the gene than infected infants, suggesting a protective effect of the CCL3-L1 gene. However, the association was restricted to infants whose mothers did not receive single-dose nevirapine.

MTCT rates in a cohort of 146 pregnant women in Mozambique were described by Jamisse and colleagues.(16) All women had CD4 counts of <350 cells/µL, started nevirapine-based ART in pregnancy, and continued ART postpartum. Infant formula was provided, but women could choose whether to breast-feed. Of 149 infants, 11% died neonatally, and 7 (7%) of 106 who were tested for HIV were found to be infected.

In a poster based on the DREAM program in 6 African countries, outcomes among 341 pregnant women were described.(17) All women were given ART in pregnancy (mostly nevirapine, lamivudine, and either zidovudine or stavudine) and for 6 months postpartum, and all engaged in exclusive breast-feeding. At 1 month of age, 1.2% of infants were infected, with an additional 0.8% infected by 6 months. The neonatal mortality rate was found to be relatively low, and the authors concluded that antenatal and postpartum ART appears to be safe and effective at reducing MTCT in an African setting.

The cost effectiveness of ART for PMTCT was demonstrated in a separate poster based on the DREAM program.(18) The analysis included 1,862 pregnant women. The cost per infection averted was calculated at $518 ($22 per disability-adjusted life year [DALY] saved). Taking into account savings to the public sector (relating to the treatment of HIV-infected children), intervention costs decreased to $149 per infection averted ($6 per DALY saved).

Pharmacokinetic Studies in Pregnancy

Several studies added to the evidence base with regard to the pharmacokinetics of antiretrovirals (largely PIs) in pregnancy, an area in which research is urgently needed to guide clinical practice.

Ferreira et al demonstrated a lack of placental transfer of enfuvirtide in an ex vivo human placental cotyledon perfusion model, using placentas from full-term uncomplicated pregnancies and at concentrations double the normal therapeutic level, concluding that it is unlikely that this drug could lead to toxicity in the fetus.(19)

Within the TARGET study, Bonora and colleagues investigated the human placental transfer of tenofovir.(20) In a study of 30 women, using paired maternal and cord blood samples taken at the time of cesarean section, the cord blood to maternal blood drug concentration of tenofovir was 0.99, representing efficient transplacental passage, most likely through simple diffusion, and without increased drug concentrations in fetal blood. In another study of tenofovir pharmacokinetics, Burchett et al presented data from PACTGP 1026s, suggesting that, although the tenofovir area under the plasma concentration-time curve (AUC) is decreased in the pregnant vs postpartum and nonpregnant states, no dosing modification is needed in pregnancy, with three quarters of the 19 women exceeding the defined AUC threshold target in the third trimester.(21)

Read and colleagues presented the first available data on the pharmacokinetics of the new nelfinavir tablet (625 mg twice daily), comparing pharmacokinetic parameters in the third trimester and postpartum period.(22) Their findings that the nelfinavir AUC fell below target in 45% of subjects in late pregnancy vs 5% in postpartum suggest that dosage modification may be necessary in pregnancy.

Pharmacokinetic data on the new saquinavir/ritonavir tablet (1,000/100 mg twice daily) in 14 pregnant women were presented by Burger and colleagues, who concluded that there was adequate pharmacokinetic activity in all women (33-week gestation data: Cmin 1.1; Cmax 3.5; AUC 17.6; T half 3.3).(23) Atazanavir pharmacokinetics were investigated by Ripamonti et al, who studied 17 pregnant women and reported atazanavir exposure to be steady in the third trimester and similar to that in the nonpregnant state (thus, no pregnancy dosage adjustment is needed).(24) Atazanavir was found to cross the placenta at a 13% ratio (ie, limited but therapeutic placental transfer).

Data from routine lopinavir (lopinavir/ritonavir 400/100 mg twice daily) drug monitoring in 36 pregnant women on triple-ARV regimens were presented by Khuong-Josses et al, who concluded that standard doses seem to provide adequate exposure levels even in late pregnancy, with the soft-gel formulation and newer tablets having similar plasma concentrations in the third trimester.(25)

However, a French case-control study of 100 HIV-infected pregnant women matched to nonpregnant women found that lopinavir levels were significantly lower in the pregnant women than in the controls.(26)


  1. Abrams E. Optimizing Maternal and Pediatric Care. Oral abstract 1g.
  2. Namale L, Zalwngo E, Chidziva E, et al. Pregnancy and Pregnancy Outcome among Women in the DART Trial. Abstract 746.
  3. Thomas T, Masaba R, Ndivo R, et al. Prevalence of Anemia among Women in a Trial of HAART including Zidovudine for PMTCT in Kisumu, Kenya. Abstract 748.
  4. Taha T, Kafulafula G, Li Q, et al. Effect of ART Regimens on Selected Hematologic and Hepatic Parameters of HIV-1-uninfected African Infants. Abstract 749.
  5. Joao E, Calvet G, Sidi L, et al. Virologic Control and Infant Outcomes among Pregnant Women Exposed to Different ART Regimens during Pregnancy. Abstract 757.
  6. Watts D, Mofenson L, Lu M, et al. Treatment Interruption after Pregnancy and Disease Progression: A Report from the Women and Infants Transmission Study. Abstract 751.
  7. Phanuphak N, Apompong T, Limpongsanurak S, et al. Toxicities from Nevirapine-based ART Regimens in Pregnant Women with CD4 Count between 250 and 350 Cells/mm3. Abstract 752.
  8. Phanuphak N, Apompong T, Limpongsanurak S, et al. Pregnancy Outcomes of Women Receiving Efavirenz after the First Trimester. Abstract 745.
  9. Natha M, Hay P, Taylor G, et al. Atazanavir Use in Pregnancy: A Report of 33 Cases. Abstract 750.
  10. Hanlon M, O'Dea S, Clarke S, et al. Maternal Hepatotoxicity with Boosted Saquinivir as Part of Combination ART in Pregnancy. Abstract 753.
  11. Patel D, Cortina Borja M, Thorne C, et al; European Collaborative Study. Time to Undetectable Viral Load after HAART Initiation in HIV-infected Pregnant Women in Europe. Abstract 758.
  12. van Schalkwyk J, Chaworth-Musters T, Maan E, et al. Time to HIV Suppression with HAART in Pregnancy. Abstract 759.
  13. Awino J, Zeh C, Bondo P, et al. CD4 and Viral Load Response and Adherence among ART-naive Women in a Trial of HAART for PMTCT in Kisumu, Kenya. Abstract 760.
  14. Townsend C, Cortina-Borja M, Peckham C. MTCT of HIV in the UK and Ireland, 1990-2004. Abstract 761.
  15. Kuhn L, Schramm D, Donninger S, et al. African Infants' CC Chemokine Ligand 3-like 1 Gene Copy Numbers Influence the Risk of Perinatal HIV Transmission Only in the Absence of Maternal Nevirapine. Abstract 755.
  16. Jamisse L, Balkus J, Farquhar C. Perinatal HIV Transmission with HAARTduring Late Pregnancy and Post-partum. Abstract 756.
  17. Palombi L, Germano P, Liotta G, et al. Safety and Efficacy of Maternal HAART in the Prevention of Early and Late Postnatal HIV-1 Transmission in Mozambique. Abstract 747.
  18. Orlando S, Peroni M, Benedetti S, et al. Cost-effectiveness of Using HAART in PMTCT on the DREAM Program. Abstract 762.
  19. Ferreira C, Ceccaldi P, Visseaux B, et al. Lack of Placental Transfer of Enfuvirtide in an ex vivo Human Placental Cotyledon Perfusion Model. Abstract 737.
  20. Bonora S, Gonzalez de Requena D, Chiesa E, et al. Transplacental Passage of Tenofovir and other Antiretrovirals at Delivery. Abstract 738a.
  21. Burchett S, Best B, Mirochnick M, et al. Tenofovir Pharmacokinetics during Pregnancy, at Delivery and Postpartum. Abstract 738b.
  22. Read J, Best B, Stek A. Nelfinavir Pharmacokinetics (625-mg Tablets) during the Third Trimester of Pregnancy and Post-partum. Abstract 740.
  23. Burger D, Eggink A, Van der Ende I, et al; SARA study group. The Pharmacokinetics of Saquinavir in New Tablet Formulation + Ritonavir (1000/100 mg Twice Daily) in HIV-1-infected Pregnant Women. Abstract 741.
  24. Ripamonti D, Cattaneo D, Airoldi M, et al. Atazanavir-based HAART in Pregnancy. Abstract 742.
  25. Khuong-Josses M, Boussaïri A, Palette C, et al. Lopinavir Drug Monitoring in 36 Pregnant Women. Abstract 743.
  26. Peytavin G, Pierre-François S, Cassard B, et al. Reduced Lopinavir Exposure during Pregnancy: A Case Control Study. Abstract 579.