Recommendations for the Use of Antiretroviral Drugs in Pregnant Women with HIV Infection and Interventions to Reduce Perinatal HIV Transmission in the United States

The information in the brief version is excerpted directly from the full-text guidelines. The brief version is a compilation of the tables and boxed recommendations.

  •   Table of Contents

Download Guidelines

General Principles Regarding Use of Antiretroviral Drugs during Pregnancy

Combination Antiretroviral Drug Regimens and Maternal and Neonatal Outcomes

Last Updated: December 7, 2018; Last Reviewed: December 7, 2018

Panel's Recommendations Regarding Combination Antiretroviral Drug Regimens and Maternal and Neonatal Outcomes
Panel's Recommendations
  • Clinicians should be aware of a possible increased risk of adverse neonatal outcomes (e.g., preterm delivery) in pregnant women who are receiving antiretroviral therapy. However, given the clear benefits of such regimens for both a woman’s health and the prevention of perinatal transmission, HIV treatment should not be withheld for fear of altering pregnancy outcomes (AII).
Rating of Recommendations: A = Strong; B = Moderate; C = Optional
Rating of Evidence: I = One or more randomized trials with clinical outcomes and/or validated laboratory endpoints; II = One or more well-designed, nonrandomized trials or observational cohort studies with long-term clinical outcomes; III = Expert opinion

Table 5. Results of Studies Assessing Associations Between Antiretroviral Regimens and Preterm Delivery
Study Location(s); Dates of Study Total Number of Pregnancies/ Total Number on ARV Drugs Types of ARV Regimens Compared (Numbers) Association Noted Between ARV Regimens and Preterm Delivery Notes
European Collaborative Study and Swiss Mother and Child HIV Cohort Study;
1986–20001
3,920/896
  • Mono (573)
  • Multi-no PI (215)
  • Multi-PI (108)
  • YES (compared with no ARV)
  • Multi: 1.82 (1.13–2.92)
  • Multi-PI: 2.60 (1.43–4.7)
  • Increase in PTD if ARV was initiated before pregnancy versus in third trimester.
United States;
1990–199830
3,266/2,123
  • Mono (1,590)
  • Multi (396)
  • Multi-PI (137)
  • NO (compared with mono)
  • Multi: 0.95 (0.60–1.48)
  • Multi-PI: 1.45 (0.81–2.50)
  • 7 prospective clinical studies.
European Collaborative Study;
1986–200452
4,372/2,033
  • Mono (704)
  • Dual (254)
  • Multi (1,075)
  • YES (compared with mono/dual)
  • Multi in pregnancy: 1.88 (1.34–2.65)
  • Multi pre-pregnancy: 2.05 (1.43–2.95)
  • N/A
United States;
1990–200241
2,543/Not given Early (25 Weeks):
  • Mono (621)
  • ≥2 ARVs without PI or NNRTI (198)
  • Multi-NNRTI or Multi-PI (357)
Late (≥32 Weeks):
  • Mono (932)
  • ≥2 ARVs without PI or NNRTI (258)
  • Multi-NNRTI or Multi-PI (588)
  • NO (compared with mono)
  • No association between any ARV and preterm delivery
  • PTD decreased with receipt of any ARV, ART that contained ZDV, and other ARV regimens compared with no ARV.
United States;
1990–20023
1,337/999
  • Mono (492)
  • Multi- no PI (373)
  • Multi-PI (134)
  • YES (compared with Mono and Multi-no PI)
  • Multi-PI: 1.8 (1.1–3.03)
  • Multi-PI reserved for those with advanced disease and those who experienced virologic failure while on other multi-ARV regimens.
Brazil, Argentina, Mexico, Bahamas;
2002–200538
681/681
  • Mono/Dual NRTI (94)
  • Multi-NNRTI (257)
  • Multi-PI (330)
  • NO (compared with Mono/Dual NRTI)
  • No association between any ARV regimen and PTD
  • All patients were on ARV for ≥28 days during pregnancy.
  • Pre-eclampsia/eclampsia, cesarean delivery, diabetes, and low BMI were associated with PTD.
Meta-Analysis, Europe and United States;
1986–20044
11,224/Not given
  • Multi-no PI (including Dual) or Multi-PI (2,556)
  • YES (only comparing Multi-PI with Multi-no PI)
  • PI vs. Multi-no PI: 1.35 (1.08–1.70)
  • 14 studies, 5 in PTD ARV comparison.
  • No overall increase in PTD with antepartum ARV.
  • PTD increased in those on ARV pre-pregnancy and in first trimester compared with later use.
Italy;
2001–20065
419/366
  • Multi-PI second trimester (97)
  • Multi-PI third trimester (146)
  • YES
  • Multi-PI second trimester:
    2.24 (1.22–4.12)
  • Multi-PI third trimester:
    2.81 (1.46–5.39)
  • Multivariate association also with HCV.
United States;
1989–20046
8,793/6,228
  • Mono (2,621)
  • Dual (1,044)
  • Multi-no PI (1,781)
  • Multi-PI (782)
  • YES (compared with Dual)
  • Multi-PI: 1.21 (1.04–1.40)
  • Lack of antepartum ARV also associated with PTD.
  • PTD and LBW decreased over time.
United Kingdom, Ireland;
1990–20057
5,009/4,445
  • Mono/Dual (1,061)
  • Multi-NNRTI or Multi-PI (3,384)
  • YES (compared with Mono/Dual)
  • Multi-PI or Multi-NNRTI: 1.51 (1.19–1.93)
  • Similar increased risk with Multi-PI or Multi-no PI.
  • No association with duration of ARV use.
Germany, Austria;
1995–20018
183/183
  • Mono (77)
  • Dual (31)
  • Multi-NNRTI (54)
  • Multi-PI (21)
  • YES (compared with Mono)
  • Multi-PI: 3.40 (1.13–10.2)
N/A
United States;
2002–200725
777/777
  • Mono (6)
  • Dual (11)
  • Multi-no PI (202)
  • Multi-PI (558)
  • NO (compared PI with all non-PI)
  • Multi-PI: 1.22 (0.70–2.12)
  • All patients started ARV during pregnancy.
  • Study analyzed only spontaneous PTD.
Swiss Mother and Child HIV Cohort Study;
1985–20079
1,180/941
  • Mono (94)
  • Dual (53)
  • Multi-PI or Multi-no PI (409)
  • Multi-PI (385)
  • YES (compared with no ARV)
  • Multi: 2.5 (1.4–4.3)
  • No association of Mono/Dual with PTD compared with no ARV.
  • No confounding by duration of ARV or maternal risk factors.
Botswana;
2006–200810
530/530
  • Multi-NRTI, ABC plus ZDV plus 3TC (263)
  • Multi-PI, LPV/r plus ZDV plus 3TC (267)
  • YES
  • Multi-PI vs. Multi-NRTI: 2.03 (1.26–3.27)
  • Secondary analysis of data from randomized, controlled clinical trial of ARV begun at 26–34 weeks for prevention of perinatal transmission.
  • All CD4 cell counts >200 cells/mm3
Botswana;
2007–201044
4,347/3,659
  • ARV, regimen unspecified (70)
  • Mono (2,473)
  • Multi (1,116), 91% Multi-NNRTI
  • NO
  • No association between multi-ART and very PTD (<32 weeks' gestation)
  • Observational; multi-ART before conception associated with very SGA and maternal hypertension during pregnancy.
Spain;
1986–201026
519/371
  • Mono/Dual NRTI (73)
  • All Multi (298)
  • Multi-PI (178)
  • NO (compared with No ARV plus Mono/Dual)
  • Spontaneous PTD not associated with Multi-ARV or Multi-PI before or during pregnancy
  • PTD associated with Multi-ARV given in second half of pregnancy and with prior PTD.
Botswana;
2009–2011
11
9,504/7,915
  • Mono (4,625)
  • All Multi (3,290)
  • Multi-PI (312)
  • YES (Multi-ARV before and during pregnancy compared with Mono): 1.2 (1.1–1.4) and 1.4 (1.2–1.8)
  • YES (Multi-PI compared with Multi-no PI before pregnancy): 2.0 (1.1–3.6)
  • ART group classified by initiation before and during pregnancy.
France; ANRS French Perinatal Cohort
1990–200912
8,696/8,491
  • Mono (950)
  • Dual (590)
  • Multi-PI (2,414)
  • YES (Multi compared to Mono): 1.69 (1.38–2.07)
  • YES (before conception compared to during pregnancy): 1.31 (1.11–1.55)
  • Patients on ART before and during pregnancy had increased rates of PTD.
United States;
2000–201143
183/183
  • Multi-PI (183)
  • NO (no control group without ART)
  • Rate of PTD 18.6%
  • SGA rate: 31.2%.
  • Patients on NNRTI-based ART less likely to have SGA: 0.28 (0.1–0.75).
United States;
2007-201013
1,869/1,810
  • Mono/Dual (138)
  • Multi-NRTI (193)
  • Multi-NNRTI (160)
  • Multi-PI (1,319)
  • YES (compared with no ARV in first trimester)
  • Multi-PI in first trimester vs. none in first trimester
  • PTD 1.55 (1.16–2.07); spontaneous PTD 1.59 (1.10–2.30)
N/A
Latin America;
2002–201214
1,512/1,446
  • No ART or ART <28 days (66)
  • Mono/Dual (130)
  • Multi-no PI (409)
  • Multi-PI (907)
  • YES (when on ARVs at conception): PTD 1.53 (1.11–2.09)
  • ART for treatment rather than prophylaxis was associated with increased rates of LBW (<2,500 g) infants: 1.8 (1.26–2.56).
  • Multi-no PI associated with decreased risk of LBW (0.33 [0.14–0.74]) and stillbirth (0.11 [0.04–0.34]). 
  • Multi-PI associated with decreased risk of stillbirth: 0.14 (0.05–0.34).
Uganda;
2009–201253
356/356
  • Multi-NNRTI, EFV (177)
  • Multi-PI, LPV/r (179)
  • NO (no control group without ART)
  • Trend in increased incidence of PTD among women starting ART 24–28-week GA was NS: aOR = 1.76 (0.96–3.23).
Italy;
1997–201354
158/158
  • Mono/Dual (27)
  • Multi-no PI (17)
  • Multi-PI (114)
  • NO (no control group without ART)
  • PTD rate was 17% for this cohort.
  • Trend towards association of PTD with longer duration of ART: 2.82 (0.35–8.09).
Canada;
1988–201115
589/530
  • No ART (59)
  • Mono (77)
  • Multi-no PI (166)
  • Multi-non-boosted PI (220)
  • Multi-boosted PI with RTV (144)
  • YES (Multi-boosted PI compared to Multi-non-boosted PI): 2.01 (1.02–3.97)
  • NO (non-PI regimens compared to Multi-non-boosted PI): 0.81 (0.4–1.66)
  • Highest risk of PTD was among women not taking ART compared to non-boosted PI group: 2.7 (1.2–6.09).
United Kingdom;
2007–201231
493/493
  • Multi-PI, LPV/r (306)
  • Multi-PI, ATV/r (187)
  • NO (comparing 2 PI-based regimens): aOR = 1.87 (0.93–3.75)
  • Rate of PTD was 13% among women who conceived on ART and 14% among women who started ART during pregnancy.
  • In a multivariate analysis, a history of PTD was associated with recurrent PTD: aOR =  5.23 (1.91–14.34).
Republic of the Congo;
2007–201239
188/188
  • Multi-no PI, EFV (31)
  • Multi-no PI, NVP (146)
  • NO (comparing EFV 13% vs. NVP 10%)
  • Rate of PTD was 11%, with no difference between study groups.
  • LBW increased in EFV group (33% vs. 16%, P = 0.04).
  • Stillbirth rate was 4% (8/188).
Tanzania;
2004–201116
3,314/2,862
  • No ART (452-excluded)
  • Mono (1,768)
  • Multi (1,094)
  • YES (Multi before pregnancy vs. Mono): 1.24 (1.05–1.47)
  • Very PTD, YES (Multi before pregnancy vs. Mono): 1.42 (1.02–1.99)
  • NO (Multi during pregnancy compared to Mono): 0.85 (0.7–1.02)
  • Rate of PTD was 29%; women who conceived on ART were more likely to have PTD compared to women on ZDV monotherapy.
  • Pregnancy-induced hypertension associated with PTD: 1.25 (1.03–1.51).
67 Countries and US Territories, APR;
1989–
201340
14,684/14,684
  • ARV with ZDV (12,780)
  • ARV without ZDV (1,904)
  • NO (any ZDV-ARV vs. non-ZDV ARV exposure): 1.0 (0.9–1.2)
  • PTD rate 12%.
  • LBW rate 16%, RR of LBW with ZDV-ART vs. non-ZDV ART RR = 1.2 (1.0–1.3), P = 0.02.
  • Stillbirth rate: 1.5%, RR = 0.8 (0.5–1.1).
Texas, United States;
1984–201432

1,004/792
  • No ART (177)
  • Mono, Dual, or Multi-no PI (230)
  • Multi-PI (597)
  • NO (no-PI ART vs. PI ART): 0.9 (0.5–1.5)
  • Rate of PTD: 13% to 21%.
  • Rate of SGA: 19% to 23%, OR = 1.3 (0.8–1.9).
India, Malawi, South Africa, Tanzania, Uganda, Zambia, Zimbabwe, PROMISE Trial;
2011–201434

3,490/3,096
  • Mono (1,386)
  • All Multi (2,710)
  • Multi-PI with ZDV (1,385)
  • Multi-PI with TDF (325)
  • YES (Multi ≥14 weeks vs. Mono)
  • Rate of PTD: 21% on Multi-PI with ZDV ART compared to ZDV-Mono (P < 0.001).
  • Rate of very PTD: 6% in Multi-PI with TDF ART and 3% in Multi-PI with ZDV ART (P = 0.04).
  • LBW was more common in Multi-PI with ZDV ART compared to ZDV Mono (23% vs. 12%, P < 0.001) and in Multi-PI with TDF compared to ZDV Mono (17% vs. 9%, P = 0.004).
United States and Puerto Rico, SMARTT;
2007–201617

1,864/1,658
  • Multi (1,658)
  • YES: (Multi-PI vs. No ART): 1.59 (1.1–2.3)

  • PI-based ART exposure in first trimester was associated with increased risk of spontaneous PTD compared with no first-trimester ART.
South Africa;
2011–201424
3,723/3,547
  • Dual (974)
  • Multi (2,573)
  • NO 
  • Dual: 0.2 (0.08–0.5)
  • Multi: 0.3 (0.1–0.9)
  • PTD rate regardless of ART: 22% to 23%.
  • LBW rate on ART: 9% to 15%. Risk of LBW: Dual 0.06 (0.02–0.2) and Multi 0.12 (0.04–0.4).
  • SGA rate on ART: 7% to 9%. Risk of SGA: Dual 0.37 (0.1 to 1.5) and Multi 0.3 (0.07 to 0.9).
  • Stillbirth rate on Dual (1.2%) and Multi (2.2%). Risk of stillbirth: Dual 0.08 (0.04–0.2) and Multi 0.2 (0.1–0.3).
Botswana;
2012–201418

11,932/10,592
  • Multi-PI (398)
  • Multi-NNRTI (4,597)
  • YES
  • Multi-PI: 1.36 (1.06–1.75)
  • Multi-NNRTI: 1.14 (1.01–1.29) 
  • SGA rates were significantly higher in Multi-PI ART (27.7% and 20.4%) and NVP-based ART (24.9% and 28.2%) compared to EFV-based ART (16.9%).
  • Stillbirth rates were higher in NVP-based ART: 2.31 (1.64–3.26). 
19 Countries, 5 Continents;
2002–201335

23,490 (meta-analysis of 10 studies)
  • Mono, Dual, or Multi-no PI
  • Multi-PI 
  • YES
  • Multi-PI: 1.3 (1.04–1.6), I2 = 47%

  • 5 of 10 studies demonstrated increased risk of PTD with an aOR range of 1.2–4.14.
South Africa;
2011–201428
1,461/1,159
  • Dual (424)
  • Multi (735)
  • YES
  • Multi: 1.65 (1.17-2.33)
  • ART before pregnancy: 1.72 (1.33-3.01)
  • PTD rate was 25%.
  • Similar rates of PTD observed among women on ART before pregnancy and women starting ART during pregnancy.
Netherlands;
1997-201533
2,184/1,392
  • Multi (1,392)
  • PI-based and non-PI based ART
  • NO
  • 1.39 (0.99–1.94); comparing women on ART before pregnancy to those who started ART during pregnancy
  • PTD rate was 14.7%.
  • SGA rate was 23.8% overall; significantly higher in women taking ART before pregnancy (27.3%) vs. those starting ART during pregnancy (21.5%): aOR = 1.35 (1.0–1.9).
  • PI-based ART before pregnancy associated with SGA: 1.49 (1.1–2.1).
South Africa, SAPMTCTE;
2012–201320
2,599/2,269
  • Dual (873)
  • Multi (1,396)
  • YES
  • 1.2 (1.0–1.5) compared to infants who were not exposed to HIV
  • 1.7 (1.1–2.5) in infants exposed to ART from conception
  • PTD rate was 12.9%; women with HIV who were not on ARVs had higher rates of PTD than women without HIV.
  • LBW rate was 13.0%; HIV-exposed infants more likely to be LWB: 1.6 (1.3–1.9).
  • SGA rate was 16.9%; HIV-exposed infants more likely to have SGA: 1.3 (1.1–1.6).
Multiple Countries;
1993–201427
37,877 (meta-analysis of 17 studies)
  • Multi with TDF
  • Other ART without TDF
  • NO
  • RR = 0.9 (0.81–0.99), I2 = 59%; women on Multi with TDF had lower rates of PTD compared to women on other ART without TDF
  • PTD rate over 4 studies was 20.3%.
  • Stillbirth rate over 3 studies was 4.4%; stillbirth rate was lower among TDF-exposed patients: 0.6 (0.43–0.84).
United Kingdom/Ireland;
2007–201555
6,073/6,073
  • Multi-PI (4,184)
  • Multi-NNRTI (1,889)
  • YES
  • Multi-PI associated with PTD: 1.56 (1.19-2.04)
  • Multi-PI before conception with CD4 count <350 cells/mm3, 1.99 (1.02–3.85) and 1.9 (1.01–3.57) and with CD4 count >350 cells/mm3, 1.61 (1.07–2.43)
  • PTD rate was 10.4%.
  • SGA rate was 20.4%.
South Africa;
2010–201529
4,435/2,549
  • Multi-NNRTI, EFV plus TDF plus FTC/3TC (1,481)
  • Multi-NNRTI, other EFV-based ART (187)
  • Multi-NNRTI, NVP-based ART (343)
  • ZDV (528)
  • NO
  • NVP-based ART aOR = 0.66 (0.27–1.63) (NS) and other EFV-based ART (aOR 0.72; 95% CI, 0.24±2.12) vs. EFV plus TDF plus FTC/3TC.
  • PTD rate was 10.4%.
  • SGA rate was 10.4%.
  • LBW rate was 9.6%.
North America;
2007–201319
4,646/1,621
  • Multi-PI (1,621), TDF plus FTC plus LPV/r, TDF plus FTC plus ATV/r, ZDV plus 3TC plus LPV/r
  • YES
  • TDF plus FTC plus ATV/r vs. ZDV plus 3TC plus LPV/r: aOR = 0.69 (0.51–0.94)
  • PTD rate was 19%.
  • LBW rate was 19.6%.
Note: The data presented in the column Association Noted between ARV Regimens and Preterm Delivery represent the published results of the study in the corresponding row. Depending on the study designs, these are adjusted and unadjusted odds ratios and relative risks.

Key to Acronyms: 3TC = lamivudine; ABC = abacavir; aOR = adjusted odds ratio; ART = antiretroviral therapy; ARV = antiretroviral; ATV/r = atazanavir/ritonavir; BMI = body mass index; CD4 = CD4 T lymphocyte; dual = 2 ARV drugs; EFV = efavirenz; FTC = emtricitabine; GA = gestational age; HCV = hepatitis C virus; LBW = low birth weight; mono = single ARV drug; multi = 3 or more ARV drugs; multi-PI = combination ART with PI; LPV/r = lopinavir/ritonavir; NNRTI = non-nucleoside reverse transcriptase inhibitor; NRTI = nucleoside reverse transcriptase inhibitor; NS = nonsignificant; NVP = nevirapine; OR = odds ratio; PI = protease inhibitor; PROMISE = Promoting Maternal and Infant Survival Everywhere; PTD = preterm delivery; RR = relative risk; RTV = ritonavir; SAPMTCTE = South African Prevention of Mother-to-Child Transmission Evaluation; SGA = small for gestational age; SMARTT = Surveillance Monitoring for ART Toxicities; TDF = tenofovir disoproxil fumarate; ZDV = zidovudine

Download Guidelines