Summary
Background
Although a growing body of evidence supports zero risk of sexual HIV transmission from a person with sustained virological suppression, known as U=U (undetectable equals untransmittable), data have been insufficient to determine whether this is also true for vertical HIV transmission. We conducted a systematic review and meta-analysis to quantify vertical transmission risk by maternal HIV viral load (mHVL) and to evaluate the applicability of U=U to perinatal and postnatal HIV transmission.
Methods
In this systematic review and meta-analysis, we searched PubMed, Embase, Web of Science, Cochrane Library, Cumulative Index of Nursing and Allied Health Literature, the WHO Global Health Library, and abstracts from the International AIDS Society Conference and the Conference on Retroviruses and Opportunistic Infections (2016–24) for studies published from Jan 1, 1989, to Dec 31, 2024, reporting the relationship between mHVL near birth (to estimate perinatal transmission risk by 6 weeks) or during breastfeeding (to estimate monthly postnatal transmission risk by mHVL within the past 6 months) and vertical transmission. We pooled risks of perinatal and postnatal transmission across prespecified mHVL categories. We also conducted comparative analyses to determine the adjusted relative risk (aRR) of transmission by mHVL using Poisson meta-regression. The protocol for this analysis is registered on the International Prospective Register of Systematic Reviews (PROSPERO; CRD42019146768).
Findings
147 studies were included in the analysis; 138 studies contributed to perinatal analyses and 13 studies contributed to postnatal analyses. Data on 82 723 mother–child pairs were included across all analyses. Pooled perinatal transmission risks were 0·2% (95% CI 0·2–0·3) with a mHVL of <50 copies per mL, 1·3% (1·0–1·7) with 50–999 copies per mL, and 5·1% (2·6–7·9) with ≥1000 copies per mL. aRRs of perinatal transmission were 6·3 (3·9–10·3) with a mHVL of 50–999 copies per mL and 22·5 (13·9–36·5) with ≥1000 copies per mL versus <50 copies per mL. In subgroup analyses, in five studies reporting on 4675 women receiving pre-conception antiretroviral therapy (ART) with a mHVL of <50 copies per mL near birth, there were zero (0%, 0·0–0·1) perinatal transmissions. Monthly postnatal transmission risks were 0·1% (0·0–0·4) with recent mHVL <50 copies per mL and 0·5% (0·1–1·8) with a mHVL of ≥50 copies per mL.
Interpretation
Perinatal transmission with a mHVL of <50 copies per mL is ≤0·2% overall. Zero transmissions were observed among women receiving ART before pregnancy with a mHVL of <50 copies per mL near birth, supporting U=U in pregnancy and birth. Postnatal transmission was very low—but not zero—among women with a recent mHVL of <50 copies per mL. Current data, largely from studies lacking frequent mHVL monitoring or modern first-line ART regimens, are insufficient to assess U=U during breastfeeding.
Funding
National Institutes of Health, WHO, and Massachusetts General Hospital.
Introduction
Globally, an estimated 1·3 million women with HIV become pregnant each year.1 Vertical HIV transmission risk has declined from 30% to <5% with earlier maternal HIV diagnosis, the scaling up of antiretroviral therapy (ART), and the use of infant antiretroviral prophylaxis.2 However, in 2022, there were still an estimated 130 000 new paediatric HIV infections.3Vertical transmission can occur in utero (ie, transplacental), intrapartum (through infant exposure to maternal body fluids), or postnatally (through breastmilk).4 Some guidelines recommend caesarean birth for women not on ART or with an elevated maternal HIV viral load (mHVL) before birth to reduce the risk of HIV transmission during birth.5 In-utero and intrapartum transmission together comprise perinatal transmission, which has declined as antenatal access to ART has improved, leading to an increasing proportion of infant infections globally that are attributable to breastfeeding.6 Breastfeeding has many health benefits for both mothers and infants, and replacement feeding is associated with increased infant mortality in settings where there is limited access to formula milk and clean water.7 In 2023, WHO issued a policy brief stating that the risk of vertical transmission with a suppressed mHVL is <1%, and recommending breastfeeding among women on ART.8 Although WHO and many national guidelines encourage breastfeeding alongside maternal ART, guidelines in high-income countries have recommended against breastfeeding by women with HIV for decades. However, in the past 2–7 years, US and European guidelines have shifted to support shared decision making about infant feeding modality if maternal virological suppression is sustained.9, 10
Research in context.
Evidence before this study
There is a large and growing body of evidence pointing towards an undetectable HIV viral load being associated with no risk of sexual transmission of HIV, supporting the U=U (undetectable equals untransmittable) campaign that helps to improve quality of life and reduce stigma for individuals living with HIV, but such data have not been comprehensively analysed for perinatal or postnatal HIV transmission. We searched PubMed, Web of Science, Cochrane Library, the Cumulative Index of Nursing and Allied Health Literature, and the WHO Global Health Library, and abstracts from the International AIDS Society Conference and the Conference on Retroviruses and Opportunistic Infections (2016–24) for studies published in English, French, and Spanish from Jan 1, 1989, to Dec 31, 2024, combining search terms for systematic reviews (eg, “systematic review” and “meta-analysis”) and vertical HIV transmission (eg, “HIV”, “vertical transmission”, and “mother-to-child transmission”). Previous individual studies have shown maternal HIV viral load (mHVL) to be a strong predictor of perinatal and postnatal transmission. However, we did not identify any existing systematic reviews and meta-analyses of vertical HIV transmission that have directly considered the role of mHVL when quantifying risks of vertical transmission or that have systematically assessed the applicability of U=U to pregnancy and breastfeeding.
Added value of this study
In this systematic review and meta-analysis, we have addressed a gap in the literature by pooling data from about 80 000 mother–child pairs to estimate the absolute risks of perinatal and postnatal transmission, stratified by recent mHVL, to guide discussions between patients and providers and to inform modelled estimates of vertical HIV transmission. We found that starting antiretroviral therapy (ART) before pregnancy and sustaining a mHVL of <50 copies per mL through birth was associated with zero observed risk of perinatal transmission of HIV, supporting the applicability of U=U to pregnancy. Postnatal transmission was also associated with mHVL throughout breastfeeding. Although data to inform this relationship were more limited, a low transmission rate was observed even with a mHVL of <50 copies per mL. Furthermore, we found no statistically significant difference in vertical transmission between caesarean birth and vaginal birth across studies among individuals with a mHVL of <1000 copies per mL in the ART era.
Implications of all the available evidence
Our findings—that mHVL is strongly correlated with perinatal and postnatal transmission risks and that U=U for perinatal transmission when ART is begun before pregnancy and when a mHVL of <50 copies per mL is sustained through birth—highlight the importance of early HIV diagnosis and sustained engagement of pregnant people in HIV care throughout pregnancy and breastfeeding to eliminate vertical transmission. More data on postnatal transmission with frequent viral load monitoring are needed to assess the applicability of U=U to breastfeeding; current data suggest that risk is very low but not zero. Our results also support guidelines that recommend caesarean birth only if mHVL is ≥1000 copies per mL.
A large body of evidence supports a zero risk of sexual HIV transmission from a person with sustained virological suppression, which is also known as U=U (undetectable equals untransmittable).11, 12, 13 U=U is a public health message that aims to empower people with HIV by raising awareness about the efficacy of ART and decreasing fear of HIV transmission. Although U=U is increasingly being accepted for sexual HIV transmission, data have been insufficient to determine whether it also applies to vertical transmission.14 Our objective was to quantify the relationship between mHVL and vertical HIV transmission risk to support patient and provider decision making, inform mathematical modelling, and evaluate U=U for perinatal and postnatal transmission.
Methods
Search strategy and selection criteria
For this systematic review and meta-analysis, we searched PubMed, Embase, Web of Science, Cochrane Library, the Cumulative Index of Nursing and Allied Health Literature, the WHO Global Health Library, and abstracts from the International AIDS Society Conference and the Conference on Retroviruses and Opportunistic Infections (2016–24) for references published and indexed from Jan 1, 1989, to Dec 31, 2024, that were written in English, French, or Spanish (appendix p 4). Two reviewers (among CMD, OU, EC, EP, PRT, and ALC) independently screened each abstract and full text. Conflicts were settled by consensus with involvement of a third reviewer (CMD or ALC) as needed.
Articles were included if they contained sufficient data on ten mother–child pairs or more to estimate risks for either perinatal transmission (infant infection detected by around 6 weeks after birth, based on when most studies reported transmission) or postnatal transmission (infant infection from around 6 weeks after birth to the end of breastfeeding) in specified mHVL categories. We assessed all studies for potential overlap (ie, studies that reported data from the same cohort over the same time period) and included only the overlapping study that provided data on the most participants.
For perinatal transmission analyses, mHVL categories were bounded by predetermined thresholds based on evolving viral load assay cutoffs over time and thresholds of clinical relevance: 50 copies per mL (studies reporting data for <20–40 copies per mL were also included in the <50 copies per mL category), 400 copies per mL, 1000 copies per mL, and 10 000 copies per mL. We excluded studies that did not report mHVL at or near birth (defined as ≥34 weeks of gestation, from within 4 weeks before birth to 1 week after birth, or qualitively described as near birth if exact timing was not reported).
For postnatal transmission, because data were known to be sparse, we pre-established the mHVL thresholds of 50 copies per mL (with <20–40 copies per mL nested within this category), 400 copies per mL (with <300 copies per mL nested within <400 copies per mL), 1000 copies per mL, and a qualitative threshold of detectable versus undetectable mHVL, using the assay threshold of quantification as reported in each study. We excluded studies that did not report sufficient data to calculate both the numerator of infants who acquired HIV during an interval and the denominator of infant person-time at risk during breastfeeding by recent mHVL (eg, if there was insufficient information on duration of breastfeeding or engagement in care at the time of infant HIV status assessment by mHVL within the preceding 6 months).
Data extraction and quality assessment
For each included study, one reviewer (among CMD, OU, EC, EP, PRT, and ALC) extracted data into an Excel template (version 16). A second reviewer (among CMD, OU, EC, EP, PRT, and ALC) verified all extracted data, and disagreements were handled through consensus and involvement of a third reviewer (CMD or ALC) if necessary. We extracted data on study setting (ie, country, income classification by World Bank criteria, and UNAIDS region),15 study type, and dates of study start and end. We also extracted cohort characteristics, including maternal antiretroviral use (categories included ART containing three or more drugs or a two-drug integrase-strand transfer inhibitor [INSTI]-based regimen, fewer than three antiretrovirals, no antiretrovirals, single-dose nevirapine, or mixed), ART regimen (ie, non-nucleoside reverse transcriptase inhibitor [NNRTI]-based, INSTI-based, protease inhibitor-based, or mixed when multiple regimens were used), timing of ART start (ie, pre-conception, post-conception, or unreported), proportion of women with caesarean birth, proportion of infants ever breastfed, and uptake of infant prophylaxis.
For perinatal transmission analyses, we extracted data on the number of children who acquired HIV by around 6 weeks after birth (number of events) among all liveborn children (number at risk) by mHVL category as measured closest to birth. For postnatal transmission analyses, we extracted data on the number of children who acquired HIV within 6 months of mHVL measurement in a prespecified category (number of events) among all children known to be breastfeeding since the last mHVL measurement as person-time at risk. The period of 0–6 weeks after birth was omitted from time at risk to avoid overlap with the perinatal transmission assessment. We adapted a quality assessment instrument for prevalence studies by Hoy and colleagues16 to assess the risk of bias in each reference (appendix p 5).
Statistical analysis
For each mHVL category, we estimated perinatal transmission as pooled risks with calculated 95% CIs using an inverse variance heterogeneity model with double arcsine transformation.17 We estimated the monthly risk of postnatal transmission by pooling the number of transmission events per person-month at risk across studies using an inverse variance heterogeneity model. Heterogeneity across studies was assessed using the I2 statistic—values greater than 50% indicated substantial heterogeneity. We interrogated causes of heterogeneity between studies for the pooled risk estimates by conducting subgroup analyses by setting, year of study start, maternal antiretroviral use, ART regimen, and timing of ART start. We also assessed potential publication bias using Doi plots.18 All pooled risk analyses were performed in Meta XL (version 5.3).19
To determine the relative risk of perinatal and postnatal transmission in different mHVL categories, we compared the risk of transmission by category limited to a subset of studies that reported on transmission risks in multiple mHVL categories from the same population, using a Poisson regression model with study-level random effects.20, 21 Studies that reported transmission risk for both a reference category (eg, <50 copies per mL or <400 copies per mL) and at least one other category were included. For perinatal transmission, we also performed comparative analyses with meta-regression that adjusted for study-level maternal antiretroviral use and setting (ie, low-income and middle-income country, high-income country, or mixed) based on clinical judgement. These adjustments were not included for the postnatal transmission regressions because of the small number of studies.
Finally, we conducted a secondary analysis to investigate the effect of caesarean birth on risk of transmission. We calculated the absolute difference in risk within mHVL categories by directly comparing the pooled transmission risks. We also estimated the relative risk of transmission associated with caesarean birth (compared with vaginal birth) using a Poisson regression model with study-level random effects and interactions between the mHVL category and an indicator for caesarean birth.
The protocol for this analysis is registered on the International Prospective Register of Systematic Reviews (PROSPERO; CRD42019146768).
Role of the funding source
The funders of the study had no role in study design, data collection, data analysis, or writing of the manuscript.
Results
We identified 4461 unique records, of which 147 studies were included in the meta-analysis (figure, appendix pp 6–11): 138 studies contributed data on perinatal transmission and 13 studies contributed data on postnatal transmission. All studies contributed to the pooled risk analyses, 92 studies contributed to the perinatal transmission comparative analyses, and eight studies contributed to postnatal transmission comparative analyses. Data on 82 723 mother–child pairs were included across all analyses.
Figure.
PRISMA diagram of included studies
Several databases were searched and relevant articles were screened for overlap, appropriate study type, and useable data. *Overlapping studies are those in which data from the same cohort were published in multiple studies; the study contributing the greatest number of participants to the analysis was included, and all other studies including data from the same cohort over the same time window were excluded. †Other reasons for exclusion included articles published in a language other than English, Spanish, or French; articles that had data on fewer than ten participants; and articles that reported only on perinatal transmission involving mothers with undetectable viral load without specifying the viral load threshold for undetectable. Some records provided data for both the perinatal and postnatal transmission analyses.
Included studies spanned 44 countries and all seven UNAIDS regions (appendix p 12).15 94 (64%) studies were initiated before 2010 (appendix p 13). For perinatal transmission estimates, reporting on breastfeeding practices and the use, type, and duration of infant prophylaxis were too infrequent and heterogeneous to allow for dedicated subgroup analyses. 90 (61%) studies had low risk of bias, 55 (37%) had moderate risk, and two (1%) had high risk (appendix pp 6–11). Some of the mHVL thresholds examined showed potential publication bias (appendix p 14).
The pooled risks of perinatal transmission for mHVL near birth were 0·2% (95% CI 0·2–0·3) with <50 copies per mL, 1·3% (1·0–1·7) with 50–999 copies per mL, and 5·1% (2·6–7·9) with ≥1000 copies per mL, irrespective of maternal ART use (table 1). When limited to women receiving ART at birth, the pooled risks were similar: 0·2% (0·2–0·3) with <50 copies per mL, 1·4% (1·0–1·8) with 50–999 copies per mL, and 4·3% (2·6–6·4) with ≥1000 copies per mL, with less heterogeneity across studies. Perinatal transmission risks with a mHVL of 50–399 copies per mL (1·3%, 0·6–2·2) were similar to those with 50–999 copies per mL (1·3%, 1·0–1·7). Risks were highest with a mHVL of ≥10 000 copies per mL near birth: 12·7% (4·9–21·7) overall and 5·0% (1·2–10·0) for women on ART. Lower mHVL was consistently associated with lower pooled perinatal transmission risk across the range of mHVL categories examined (table 1, appendix p 15).
Table 1.
Pooled estimates of perinatal transmission by maternal HIV viral load at birth and use of ART at birth
|
Mother–child pairs |
Mother–child pairs with maternal ART*use at birth |
|||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Studies (n) | Participants (n) | Events (n) | Pooled estimate, % (95% CI) | I2 | Studies (n) | Participants (n) | Events (n) | Pooled estimate, % (95% CI) | I2 | |
| <50 copies per mL | 56 | 20 002 | 44 | 0·2% (0·2–0·3) | 0% | 48 | 18 246 | 41 | 0·2% (0·2–0·3) | 0% |
| 50–999 copies per mL | 17 | 3610 | 44 | 1·3% (1·0–1·7) | 0% | 12 | 3560 | 46 | 1·4% (1·0–1·8) | 0% |
| ≥1000 copies per mL | 32 | 5617 | 294 | 5·1% (2·6–7·9) | 69·3% | 16 | 1387 | 60 | 4·3% (2·6–6·4) | 37·3% |
| ≥50 copies per mL | 41 | 8017 | 215 | 2·7% (2·0–3·5) | 20·3% | 27 | 5428 | 102 | 1·9% (1·6–2·3) | 0% |
| <1000 copies per mL | 45 | 33 564 | 212 | 0·6% (0·3–0·9) | 69·9% | 32 | 22 793 | 101 | 0·5% (0·3–0·6) | 24·4% |
| 50–399 copies per mL | 10 | 3587 | 49 | 1·3% (0·6–2·2) | 52·0% | 4 | 2692 | 32 | 1·2% (0·7–1·8) | 23·2% |
| <400 copies per mL | 36 | 29 668 | 149 | 0·5% (0·2–0·8) | 69·1% | 25 | 17 592 | 63 | 0·4% (0·3–0·5) | 0% |
| ≥400 copies per mL | 16 | 7415 | 361 | 4·5% (1·5–8·1) | 90·6% | 8 | 1294 | 36 | 2·9% (2·0–3·8) | 0% |
| <10 000 copies per mL | 26 | 17 397 | 266 | 1·3% (0·1–2·9) | 89·7% | 11 | 6685 | 42 | 0·6% (0·1–1·4) | 23·6% |
| ≥10 000 copies per mL | 15 | 3604 | 498 | 12·7% (4·9–21·7) | 94·6% | 3 | 260 | 13 | 5·0% (1·2–10·0) | 25·4% |
ART=antiretroviral therapy.
Combination ART containing three or more drugs or a two-drug integrase-strand transfer inhibitor-based regimen.
In subgroup analyses, in five studies reporting on 4675 women who started ART before conception and had a mHVL of <50 copies per mL near birth, there were zero (0%, 95% CI 0·0–0·1) instances of perinatal transmission (appendix p 16). Among women who started ART during pregnancy, pooled perinatal transmission risk with a mHVL near birth of <50 copies per mL was 0·4% (0·2–0·6). Among all studies included in this meta-analysis (not only limited to studies reporting outcomes disaggregated by timing of ART start), there were no reported perinatal transmissions when women were receiving ART before conception and had a mHVL of <50 copies per mL near birth (details on all transmission events reported with a mHVL of <50 copies per mL near birth are provided in the appendix [pp 17–19]).
Substantial heterogeneity was observed across most of the pooled perinatal transmission analyses. A major contributor to this heterogeneity was study setting, with higher observed transmission risk in low-income and middle-income countries versus high-income countries (appendix pp 20–23); transmission risk did not differ substantially by study year or risk of bias (appendix pp 24–31). Pooled transmission risks were similar with INSTI-based, NNRTI-based, or protease inhibitor-based ART at any given mHVL (appendix p 32). We could not directly compare these regimens given the small number of studies that evaluated more than one ART regimen within the same population.
In comparative analyses limited only to studies in which perinatal transmission risk was reported from more than one mHVL category within the same study population, the unadjusted relative risks of transmission were 6·4-fold (95% CI 3·9–10·4, p<0·0001) higher with mHVL near birth of 50–999 copies per mL, and 23·6-fold (14·6–37·9, p<0·0001) higher with mHVL near birth with ≥1000 copies per mL versus <50 copies per mL (table 2). When using an assay with a cutoff of 400 copies per mL, unadjusted relative risks of transmission were 3·5-fold (2·1–5·9, p<0·0001) higher with a mHVL near birth of 400–999 copies per mL, and 9·6-fold (7·3–12·8, p<0·0001) higher with ≥1000 copies per mL versus <400 copies per mL. All mHVL category comparisons remained significant when adjusting for study setting and maternal ART use (appendix p 33).
Table 2.
Comparison of perinatal HIV transmission by mHVL at birth when data on multiple mHVL categories were reported within the same study
| Studies* | Unadjusted relative risk (95% CI) | p value | Adjusted relative risk (95% CI)† | p value | |
|---|---|---|---|---|---|
| Comparison: <50 copies per mL, 50–999 copies per mL, and ≥1000 copies per mL | |||||
| <50 copies per mL | .. | 1 (ref) | .. | 1 (ref) | .. |
| 50–999 copies per mL | 15 | 6·4 (3·9–10·4) | <0·0001 | 6·3 (3·9–10·3) | <0·0001 |
| ≥1000 copies per mL | 13 | 23·6 (14·6–37·9) | <0·0001 | 22·5 (13·9–36·5) | <0·0001 |
| Comparison: <400 copies per mL, 400–999 copies per mL, and ≥1000 copies per mL | |||||
| <400 copies per mL | .. | 1 (ref) | .. | 1 (ref) | .. |
| 400–999 copies per mL | 5 | 3·5 (2·1–5·9) | <0·0001 | 3·3 (2·0–5·7) | <0·0001 |
| ≥1000 copies per mL | 6 | 9·6 (7·3–12·8) | <0·0001 | 9·0 (6·5–12·4) | <0·0001 |
| Comparison: <50 copies per mL and ≥50 copies per mL | |||||
| <50 copies per mL | .. | 1 (ref) | .. | 1 (ref) | .. |
| ≥50 copies per mL | 54 | 8·8 (6·6–11·8) | <0·0001 | 8·8 (6·6–11·8) | <0·0001 |
| Comparison: <400 copies per mL and ≥400 copies per mL | |||||
| <400 copies per mL | .. | 1 (ref) | .. | 1 (ref) | .. |
| ≥400 copies per mL | 17 | 8·1 (6·3–10·5) | <0·0001 | 8·0 (7·9–8·1) | <0·0001 |
| Comparison: <1000 copies per mL and ≥1000 copies per mL | |||||
| <1000 copies per mL | .. | 1 (ref) | .. | 1 (ref) | .. |
| ≥1000 copies per mL | 29 | 8·8 (6·9–11·1) | <0·0001 | 8·6 (6·7–10·9) | <0·0001 |
| Comparison: <10 000 copies per mL and ≥10 000 copies per mL | |||||
| <10 000 copies per mL | .. | 1 (ref) | .. | 1 (ref) | .. |
| ≥10 000 copies per mL | 13 | 4·7 (4·0–5·5) | <0·0001 | 4·7 (4·0–5·5) | <0·0001 |
mHVL=maternal HIV viral load.
The number of studies is not reported for the reference maternal HIV RNA category, as that number differs for each level of comparison. Therefore, the number of studies included in a given comparison is listed with the comparator, rather than the reference category.
Estimates were adjusted by study setting and maternal antiretroviral therapy use.
In a secondary analysis of mode of birth, with a mHVL near birth of <1000 copies per mL, pooled perinatal transmission risks were similar with vaginal birth (0·4%, 95% CI 0·2–0·5) and caesarean birth (0·5%, 0·3–0·8; appendix p 34). When mHVL near birth was ≥1000 copies per mL, pooled risks were 8·5% (1·1–18·3) with vaginal birth and 3·4% (0·0–15·4) with caesarean birth. In comparative analyses limited to a subset of studies that reported transmission risks in a given mHVL category by mode of birth in the same study population, these differences were not statistically significant. The relative risks of transmission with caesarean birth versus vaginal birth were 1·6 (0·8–3·1, p=0·19) with a mHVL of <1000 copies per mL, and 0·9 (0·3–3·3, p=0·91) with a mHVL of ≥1000 copies per mL (table 3). Findings were similar when restricted only to studies in which women were receiving ART (appendix p 35).
Table 3.
Risk of perinatal transmission with caesarean birth versus vaginal birth by maternal HIV viral load at birth
| Studies |
Caesarean birth |
Vaginal birth |
Relative risk of vertical HIV transmission with caesarean birth vs vaginal birth (95% CI) | p value | |||||
|---|---|---|---|---|---|---|---|---|---|
| At risk | Events | % | At risk | Events | % | ||||
| Comparison: caesarianvsvaginal birth at <50 copies per mL, 50–999 copies per mL, and ≥1000 copies per mL | |||||||||
| <50 copies per mL | 7 | 7062 | 10 | 0·1 | 4870 | 10 | 0·2 | 0·7 (0·3–1·7) | 0·42 |
| 50–999 copies per mL | 4 | 2205 | 30 | 1·4 | 708 | 8 | 1·1 | 1·2 (0·6–2·7) | 0·62 |
| ≥1000 copies per mL | 1 | 444 | 13 | 2·9 | 95 | 3 | 3·2 | 0·9 (0·3–3·3) | 0·91 |
| Comparison: caesarianvsvaginal birth at <50 copies per mL, 50–399 copies per mL, 400–999 copies per mL, and ≥1000 copies per mL | |||||||||
| <50 copies per mL | 7 | 7062 | 10 | 0·1 | 4870 | 10 | 0·2 | 0·7 (0·3–1·7) | 0·42 |
| 50–399 copies per mL | 2 | 1794 | 23 | 1·3 | 598 | 7 | 1·2 | 1·1 (0·5–2·6) | 0·82 |
| 400–999 copies per mL | 2 | 373 | 7 | 1·9 | 68 | 1 | 1·5 | 1·3 (0·2–10·6) | 0·81 |
| ≥1000 copies per mL | 1 | 444 | 13 | 2·9 | 95 | 3 | 3·2 | 0·9 (0·3–3·3) | 0·91 |
| Comparison: caesarianvsvaginal birth at <400 copies per mL, 400–999 copies per mL, and ≥1000 copies per mL | |||||||||
| <400 copies per mL | 6 | 9816 | 39 | 0·4 | 6270 | 32 | 0·5 | 0·7 (0·5–1·2) | 0·19 |
| 400–999 copies per mL | 2 | 373 | 7 | 1·9 | 68 | 1 | 1·5 | 1·4 (0·2–11·4) | 0·74 |
| ≥1000 copies per mL | 1 | 444 | 13 | 2·9 | 95 | 3 | 3·2 | 0·9 (0·3–3·2) | 0·91 |
| Comparison: caesarianvsvaginal birth at <50 copies per mL and ≥50 copies per mL | |||||||||
| <50 copies per mL | 7 | 7062 | 10 | 0·1 | 4870 | 10 | 0·2 | 0·7 (0·3–1·7) | 0·47 |
| ≥50 copies per mL | 4 | 2662 | 43 | 1·6 | 805 | 11 | 1·4 | 1·3 (0·6–2·4) | 0·52 |
| Comparison: caesarianvsvaginal birth at <400 copies per mL and ≥400 copies per mL | |||||||||
| <400 copies per mL | 6 | 9816 | 39 | 0·4 | 6270 | 32 | 0·5 | 0·7 (0·5–1·2) | 0.20 |
| ≥400 copies per mL | 3 | 1092 | 25 | 2·3 | 227 | 6 | 2·6 | 0·9 (0·4–2·2) | 0·80 |
| Comparison: caesarianvsvaginal birth at <1000 copies per mL and ≥1000 copies per mL | |||||||||
| <1000 copies per mL | 5 | 3886 | 25 | 0·6 | 3176 | 13 | 0·4 | 1·6 (0·8–3·1) | 0·19 |
| ≥1000 copies per mL | 1 | 444 | 13 | 2·9 | 95 | 3 | 3·2 | 0·9 (0·3–3·3) | 0·91 |
| Comparison: caesarianvsvaginal birth at <10 000 copies per mL and ≥10 000 copies per mL | |||||||||
| <10 000 copies per mL* | 1 | 3359 | 22 | 0·7 | 2524 | 13 | 0·5 | 1·3 (0·6–2·5) | 0·49 |
| ≥10 000 copies per mL | 1 | 186 | 10 | 5·4 | 32 | 2 | 6·3 | 0·9 (0·2–3·9) | 0·85 |
Study-level random effects not used in Poisson regression for relative risks because only one study is included in this category.
Among the 13 studies included in the pooled postnatal transmission risk analysis (appendix p 36), the monthly risks of transmission while breastfeeding were 0·1% (95% CI 0·0–0·2) with a recent mHVL of 50 copies per mL and 0·3% (0·0–1·8) with a recent mHVL ≥50 copies per mL (appendix p 37). Similarly, monthly postnatal transmission risks were 0·1% (0·0–0·2) with a recent mHVL of <400 copies per mL and 0·7% (0·0–1·9) with a recent mHVL of ≥400 copies per mL. In comparative analyses limited to studies that contained data on postnatal transmission risk from more than one mHVL category from the same population, the monthly risks of transmission while breastfeeding were 0·1% (0·0–0·4) with a recent mHVL of 50 copies per mL and 0·5% (0·1–1·8) with a recent mHVL ≥50 copies per mL (table 4). In these studies, the unadjusted relative risks of transmission were 6·3 (1·2–34·6, p=0·034) with recent mHVL of ≥50 copies per mL versus <50 copies per mL, 4·3 (1·6–12·0, p<0·0049) with ≥400 copies per mL versus <400 copies per mL, and 6·9 (1·9–25·7, p<0·0001) with ≥1000 copies per mL versus <1000 copies per mL. Details of postnatal transmissions that occurred with a most recent mHVL of <50 copies per mL are reported in the appendix (p 38).
Table 4.
Comparison of monthly postnatal HIV transmission by most recent maternal HIV viral load within 6 months preceding paediatric HIV infection among studies including multiple maternal HIV viral load categories
| Person-months at risk | Events (n) | Studies (n) | Pooled estimate, % (95% CI) | Unadjusted relative risk*(95% CI) | p value | |
|---|---|---|---|---|---|---|
| Comparison: <50 copies per mL and ≥50 copies per mL | ||||||
| <50 copies per mL† | 3382 | 2 | NA | 0·1% (0·0–0·4) | 1 (ref) | .. |
| ≥50 copies per mL | 1101 | 4 | 3 | 0·5% (0·1–1·8) | 6·3 (1·2–34·6) | 0·034 |
| Comparison: <400 copies per mL and ≥400 copies per mL | ||||||
| <400 copies per mL‡ | 8356 | 8 | NA | 0·1% (0·0–0·3) | 1 (ref) | .. |
| ≥400 copies per mL | 1926 | 15 | 4 | 0·4% (0·1–1·5) | 4·3 (1·6–12·0) | <0·0049 |
| Comparison: <1000 copies per mL and ≥1000 copies per mL | ||||||
| <1000 copies per mL | 21 384 | 3 | NA | 0% (0·0–0·2) | 1 (ref) | .. |
| ≥1000 copies per mL | 16 349 | 12 | 4 | 0·2% (0·0–1·3) | 6·9 (1·9–25·7) | <0·0001 |
| Comparison: undetectable and detectable maternal viral load | ||||||
| Undetectable§ | 8388 | 8 | NA | 0·1% (0·0–0·3) | 1 (ref) | .. |
| Detectable | 2832 | 19 | 6 | 0·5% (0·2–1·3) | 5·1 (2·1–12·4) | <0·0038 |
NA=not applicable.
Sample size was insufficient to allow for adjusted analyses.
Included studies using an assay cutoff of either <20 copies per mL or <50 copies per mL.
Included studies using an assay cutoff of either <300 copies per mL or <400 copies per mL.
Viral load threshold for undetectable was <20 copies per mL, <50 copies per mL, <300 copies per mL, or <400 copies per mL across various studies based on the assay threshold of quantification as reported in each study.
Discussion
We found that mHVL was strongly correlated with perinatal transmission risk, with a 6·3-fold increased risk of transmission with a mHVL near birth of 50–999 copies per mL and 22·5-fold increased risk with ≥1000 copies per mL versus <50 copies per mL. There were no instances of perinatal transmission in which mothers started ART before pregnancy and had a mHVL of <50 copies per mL near birth, supporting the application of U=U to perinatal transmission.
Our analysis adds to previous reviews by examining the effect of mHVL on vertical transmission risk, which to our knowledge has not been done previously. In a 2019 UNAIDS review, overall perinatal transmission risks among women receiving ART ranged from 0·3% (with ART started pre-conception) to 1·4% (with ART started during pregnancy). Although these risks were not stratified by mHVL, they are similar to the 0·2% and 1·4% perinatal transmission risks we found for women on ART with a mHVL of <50 copies per mL or 50–999 copies per mL near birth.22 Our analysis adds to the 2019 UNAIDS review by disaggregating risks of transmission with pre-conception ART by mHVL, finding that transmission risk was 0% when women starting ART pre-conception had a mHVL near birth of <50 copies per mL. The UNAIDS review also identified a monthly postnatal transmission risk of 0·02–0·1% among women starting ART before or during pregnancy, although these risks also were not stratified by mHVL. These risks were close to our mHVL-based estimates at all but the highest mHVL category of ≥10 000 copies per mL. Another review of postnatal HIV transmission among women on ART found a cumulative postnatal vertical transmission risk at 12 months of 2·9% (95% CI 0·7–5·2), which translates to a monthly probability of 0·2%, which again is similar to our findings.23
This study also adds to a growing body of literature suggesting that U=U for perinatal transmission when ART is started before pregnancy and a mHVL of <50 copies per mL is sustained through birth. We found that in five studies reporting on 4675 women who started ART pre-conception, there were no perinatal transmissions when mHVL was <50 copies per mL near birth.24, 25, 26 Similarly, a study from the French Perinatal Cohort27 (excluded from our analysis due to a 90-day interval between mHVL monitoring at birth) found that, among 5482 women who started ART pre-conception and had a mHVL of <50 copies per mL before birth, there were no cases of perinatal transmission. Importantly, for U=U to hold true in pregnancy, mHVL needs to be sustained at <50 copies per mL on ART from conception through to birth, as transmission has been described even when the mHVL is 50–399 copies per mL or with a mHVL of <50 copies per mL at birth when ART was started after conception.27 These findings highlight the importance of early ART initiation, sustained engagement in care, and structures to support ART adherence among women of childbearing potential to eliminate vertical transmission.
We also found that postnatal transmission was strongly associated with higher mHVL, but there were insufficient data to confirm that U=U applies to breastfeeding. There were considerably fewer studies that provided data on postnatal versus perinatal transmission risks, in part because settings with the greatest access to serial mHVL monitoring are also those in which breastfeeding among women with HIV was discouraged until 2018 or later. Many studies were also excluded due to lack of reporting on infant time at risk (eg, insufficient information about breastfeeding duration or the relationship between recent mHVL and infant status assessment, in part due to high rates of postpartum disengagement from care). Although some transmissions were observed following mHVL <50 copies per mL, intervals between observed mHVL and documented infant diagnosis were long. Viraemia could have occurred between last mHVL observation and infant infection, as some studies report viraemia in up to a third of postpartum women.28 Alternatively, postnatal transmission might occur even with virological suppression, which would be consistent with rare reports of postnatal transmission with a mHVL of <50 copies per mL, perhaps due to cell-associated HIV transmission in breastmilk.29, 30, 31, 32, 33 Heterogeneous reporting of infant postnatal prophylaxis use across studies limited the ability to directly assess its role in further reducing postnatal HIV transmission risks. Furthermore, few studies included in the postnatal transmission analyses used INSTI-based ART, so breastfeeding transmission risks with virological suppression on currently recommended regimens remain unclear. The very low 0·1% monthly risk of postnatal transmission with a recent mHVL of <50 copies per mL still supports evolving guidance that encourages a supportive, shared decision-making approach to caring for women with HIV who want to breastfeed.9 However, more data are needed from studies in which breastfeeding women are monitored frequently and from a range of geographical settings with varying breastfeeding practices to further assess the applicability of U=U during breastfeeding.
In our analysis, we found that there was no significant difference in transmission risk by mode of birth at a given mHVL. Guidelines in many high-income countries recommend elective caesarean birth to prevent intrapartum transmission, although mHVL thresholds at 36 weeks of gestation differ (eg, in the USA, recommended if mHVL ≥1000 copies per mL; in the UK, considered if 50–399 copies per mL, and recommended if ≥400 copies per mL; in Sweden, recommended if ≥200 copies per mL).9, 34 Before the widespread use of highly effective three-drug ART, caesarean birth reduced intrapartum transmission, but studies of women receiving ART have not shown a similar benefit.35, 36, 37 For example, among women who received ART during pregnancy in Canada, vertical transmission rates were 0·6% (six of 946) with vaginal birth and 1·4% (ten of 699) with caesarean birth, and benefits of caesarean birth were only observed among women who did not receive ART (p=0·016).38 Similarly, in the French Perinatal Cohort, there was no reduction in perinatal transmission risk with caesarean versus vaginal birth if mHVL was <1000 copies per mL.39 Given the maternal complications and morbidity associated with caesarean birth,40 as well as the unclear benefit of this practice in the modern ART era, guidelines that recommend caesarean birth among women on ART with a mHVL of <1000 copies per mL near birth should be reconsidered.
There were several limitations to this analysis. First, postnatal transmission data were mostly limited to studies in which intervals between mHVL observation and infant HIV testing were long. More data on postnatal transmission risks by mHVL are needed from settings with access to frequent mHVL monitoring and INSTI-based ART. Second, we defined perinatal transmission as HIV infection recognised by around 6 weeks after birth. This definition might overestimate risk in some settings by also including early postnatal transmission. Finally, the included studies showed substantial heterogeneity. We accounted for some of this heterogeneity by conducting extensive subgroup analyses and meta-regression. However, in the absence of participant-level data, we could not adjust for timing of ART initiation in pregnancy, maternal ART regimens, exclusive breastfeeding practices, and infant prophylaxis regimens and duration, which could substantially influence transmission risks but were infrequently described in the included studies.
We found that vertical HIV transmission risks are strongly correlated with mHVL. Perinatal and postnatal transmission risks with maternal HIV virological suppression are very low, and population-level risks are also declining in settings that are approaching epidemic control. Access to modern ART regimens and support systems that enable women to rapidly achieve and sustain virological suppression are crucial to eliminating vertical HIV transmission. Consistent with the 2023 WHO policy brief,8 our data support U=U for perinatal transmission and highlight the importance of early HIV diagnosis and ART initiation among pregnant women. More data on the relationship between maternal viral load and postnatal transmission (particularly with INSTI-based ART regimens) are needed to assess whether the concept of U=U applies during breastfeeding. However, women should be empowered through shared decision making to choose the infant feeding modality that best suits their families given the low risks of postnatal transmission with sustained virological suppression.
Contributors
Data sharing
All of the data included in this analysis are publicly available, as this study is a meta-analysis of published studies. All data relevant to the study are included in the Article or included in the appendix.
Declaration of interests
LM serves as a paid consultant to WHO on the safety of antiretroviral drugs in pregnancy. CMD and ALC received support from WHO to adapt the results from this analysis for presentation to the WHO HIV Guidelines Development Group. All other authors declare no competing interests.
Acknowledgments
This Article was supported by the Eunice Kennedy Shriver National Institute of Child Health and Human Development (R37 HD079214 [ALC] and K08 HD101342 [CMD]), the National Institute of Mental Health (R01 MH115793 [MK]), the National Institute of Allergy and Infectious Diseases (T32 AI007433 [CMD]), WHO (CMD and ALC), the Massachusetts General Hospital Parental Scholars Award (CMD), and the James and Audrey Foster Massachusetts General Hospital Research Scholars award (ALC). The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding sources.
Acknowledgments
Conceptualisation: CMD and ALC. Data curation: CMD, OU, EC, EP, and PRT. Formal analysis: CMD, OU, JG, FS, EC, and PRT. Funding acquisition: CMD. Investigation: CMD, OU, EC, EP, PRT, and LM. Methodology: CMD, JG, and FS. Project administration: OU, EC, and EP. Resources: ALC. Supervision: CMD and ALC. Validation: CMD and DC. Visualisation: CMD and DC. Writing of the original draft: CMD. Review and editing of manuscript: CMD, OU, JG, FS, EC, EP, PRT, DC, RS, MK, LM, and ALC. All authors have seen and reviewed the final text and agreed with the decision to submit the manuscript for publication.
Supplementary Material
References
- 1.WHO Mother-to-child transmission of HIV. https://www.who.int/teams/global-hiv-hepatitis-and-stis-programmes/hiv/prevention/mother-to-child-transmission-of-hiv
- 2.WHO Global guidance on criteria and processes for validation: elimination of mother-to-child transmission of HIV, syphilis and hepatitis B virus. Aug 4, 2022. https://www.who.int/publications/i/item/9789240039360
- 3.UNAIDS The path that ends AIDS—2023 UNAIDS Global AIDS update. 2023. https://thepath.unaids.org/wp-content/themes/unaids2023/assets/files/2023_report.pdf
- 4.Cockbain B, Fidler S, Lyall H. Preventing perinatal HIV acquisition; current gaps and future perspectives. Curr Opin HIV AIDS. 2024;19:293–304. doi: 10.1097/COH.0000000000000881. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Amin O, Powers J, Bricker KM, Chahroudi A. Understanding viral and immune interplay during vertical transmission of HIV: implications for cure. Front Immunol. 2021;12 doi: 10.3389/fimmu.2021.757400. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Bamford A, Foster C, Lyall H. Infant feeding: emerging concepts to prevent HIV transmission. Curr Opin Infect Dis. 2024;37:8–16. doi: 10.1097/QCO.0000000000000986. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Kuhn L, Sinkala M, Semrau K, et al. Elevations in mortality due to weaning persist into the second year of life among uninfected children born to HIV-infected mothers. Clin Infect Dis. 2010;50:437–444. doi: 10.1086/649886. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.WHO The role of HIV viral suppression in improving individual health and reducing transmission. July 23, 2023. https://www.who.int/publications/i/item/9789240055179
- 9.Panel on Treatment of HIV During Pregnancy and Prevention of Perinatal Transmission . Department of Health and Human Services; June 12, 2025. Recommendations for the use of antiretroviral drugs during pregnancy and interventions to reduce perinatal HIV transmission in the United States.https://clinicalinfo.hiv.gov/sites/default/files/guidelines/documents/perinatal-hiv/guidelines-perinatal.pdf [Google Scholar]
- 10.Crisinel PA, Kusejko K, Kahlert CR, et al. Successful implementation of new Swiss recommendations on breastfeeding of infants born to women living with HIV. Eur J Obstet Gynecol Reprod Biol. 2023;283:86–89. doi: 10.1016/j.ejogrb.2023.02.013. [DOI] [PubMed] [Google Scholar]
- 11.Broyles LN, Luo R, Boeras D, Vojnov L. The risk of sexual transmission of HIV in individuals with low-level HIV viraemia: a systematic review. Lancet. 2023;402:464–471. doi: 10.1016/S0140-6736(23)00877-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Calabrese SK, Mayer KH, Marcus JL. Prioritising pleasure and correcting misinformation in the era of U=U. Lancet HIV. 2021;8:e175–e180. doi: 10.1016/S2352-3018(20)30341-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.UNAIDS Undetectable=untransmittable—Public health and HIV viral load suppression. March 22, 2024. https://www.unaids.org/sites/default/files/media_asset/undetectable-untransmittable_en.pdf
- 14.Waitt C, Low N, Van de Perre P, Lyons F, Loutfy M, Aebi-Popp K. Does U=U for breastfeeding mothers and infants? Breastfeeding by mothers on effective treatment for HIV infection in high-income settings. Lancet HIV. 2018;5:e531–e536. doi: 10.1016/S2352-3018(18)30098-5. [DOI] [PubMed] [Google Scholar]
- 15.UNAIDS Regions. https://www.unaids.org/en/regionscountries/regions
- 16.Hoy D, Brooks P, Woolf A, et al. Assessing risk of bias in prevalence studies: modification of an existing tool and evidence of interrater agreement. J Clin Epidemiol. 2012;65:934–939. doi: 10.1016/j.jclinepi.2011.11.014. [DOI] [PubMed] [Google Scholar]
- 17.Doi SAR, Barendregt JJ, Khan S, Thalib L, Williams GM. Advances in the meta-analysis of heterogeneous clinical trials I: the inverse variance heterogeneity model. Contemp Clin Trials. 2015;45:130–138. doi: 10.1016/j.cct.2015.05.009. [DOI] [PubMed] [Google Scholar]
- 18.Furuya-Kanamori L, Barendregt JJ, Doi SAR. A new improved graphical and quantitative method for detecting bias in meta-analysis. Int J Evid Based Healthc. 2018;16:195–203. doi: 10.1097/XEB.0000000000000141. [DOI] [PubMed] [Google Scholar]
- 19.Barendregt JJ, Doi SA. MetaXL user guide. http://www.epigear.com/index_files/MetaXL%20User%20Guide.pdf
- 20.Spittal MJ, Pirkis J, Gurrin LC. Meta-analysis of incidence rate data in the presence of zero events. BMC Med Res Methodol. 2015;15:42. doi: 10.1186/s12874-015-0031-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Böhning D, Mylona K, Kimber A. Meta-analysis of clinical trials with rare events. BMJ. 2015;57:633–648. doi: 10.1002/bimj.201400184. [DOI] [PubMed] [Google Scholar]
- 22.Stover J, Glaubius R, Mofenson L, et al. Updates to the Spectrum/AIM model for estimating key HIV indicators at national and subnational levels. AIDS. 2019;33(suppl 3):S227–S234. doi: 10.1097/QAD.0000000000002357. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Bispo S, Chikhungu L, Rollins N, Siegfried N, Newell ML. Postnatal HIV transmission in breastfed infants of HIV-infected women on ART: a systematic review and meta-analysis. J Int AIDS Soc. 2017;20 doi: 10.7448/IAS.20.1.21251. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Townsend CL, Byrne L, Cortina-Borja M, et al. Earlier initiation of ART and further decline in mother-to-child HIV transmission rates, 2000–2011. AIDS. 2014;28:1049–1057. doi: 10.1097/QAD.0000000000000212. [DOI] [PubMed] [Google Scholar]
- 25.Mandelbrot L, Tubiana R, Le Chenadec J, et al. No perinatal HIV-1 transmission from women with effective antiretroviral therapy starting before conception. Clin Infect Dis. 2015;61:1715–1725. doi: 10.1093/cid/civ578. [DOI] [PubMed] [Google Scholar]
- 26.Brunet C, Reliquet V, Jovelin T, et al. Effectiveness and safety of saquinavir/ritonavir in HIV-infected pregnant women: INEMA cohort. Med Mal Infect. 2012;42:421–428. doi: 10.1016/j.medmal.2012.07.010. [DOI] [PubMed] [Google Scholar]
- 27.Sibiude J, Le Chenadec J, Mandelbrot L, et al. Update of perinatal human immunodeficiency virus type 1 transmission in France: zero transmission for 5482 mothers on continuous antiretroviral therapy from conception and with undetectable viral load at delivery. Clin Infect Dis. 2023;76:e590–e598. doi: 10.1093/cid/ciac703. [DOI] [PubMed] [Google Scholar]
- 28.Myer L, Dunning L, Lesosky M, et al. Frequency of viremic episodes in HIV-infected women initiating antiretroviral therapy during pregnancy: a cohort study. Clin Infect Dis. 2017;64:422–427. doi: 10.1093/cid/ciw792. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Flynn PM, Taha TE, Cababasay M, et al. Association of maternal viral load and CD4 count with perinatal HIV-1 transmission risk during breastfeeding in the PROMISE postpartum component. J Acquir Immune Defic Syndr. 2021;88:206–213. doi: 10.1097/QAI.0000000000002744. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Shapiro RL, Hughes MD, Ogwu A, et al. Antiretroviral regimens in pregnancy and breast-feeding in Botswana. N Engl J Med. 2010;362:2282–2294. doi: 10.1056/NEJMoa0907736. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Malaba TR, Nakatudde I, Kintu K, et al. 72 weeks post-partum follow-up of dolutegravir versus efavirenz initiated in late pregnancy (DolPHIN-2): an open-label, randomised controlled study. Lancet HIV. 2022;9:e534–e543. doi: 10.1016/S2352-3018(22)00173-4. [DOI] [PubMed] [Google Scholar]
- 32.Davis NL, Corbett A, Kaullen J, et al. Antiretroviral drug concentrations in breastmilk, maternal HIV viral load, and HIV transmission to the infant: results from the BAN Study. J Acquir Immune Defic Syndr. 2019;80:467–473. doi: 10.1097/QAI.0000000000001941. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Giuliano M, Andreotti M, Liotta G, et al. Maternal antiretroviral therapy for the prevention of mother-to-child transmission of HIV in Malawi: maternal and infant outcomes two years after delivery. PLoS One. 2013;8 doi: 10.1371/journal.pone.0068950. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Gilleece Y, Tariq S, Awosusi F, et al. British HIV Association; 2020. British HIV Association guidelines for the management of HIV in pregnancy and postpartum 2018 (2020 third interim update)https://www.bhiva.org/file/5f1aab1ab9aba/BHIVA-Pregnancy-guidelines-2020-3rd-interim-update.pdf [DOI] [PubMed] [Google Scholar]
- 35.European Collaborative Study Mode of delivery in HIV-infected pregnant women and prevention of mother-to-child transmission: changing practices in western Europe. HIV Med. 2010;11:368–378. doi: 10.1111/j.1468-1293.2009.00800.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.International Perinatal HIV Group The mode of delivery and the risk of vertical transmission of human immunodeficiency virus type 1—a meta-analysis of 15 prospective cohort studies. N Engl J Med. 1999;340:977–987. doi: 10.1056/NEJM199904013401301. [DOI] [PubMed] [Google Scholar]
- 37.Kennedy CE, Yeh PT, Pandey S, Betran AP, Narasimhan M. Elective cesarean section for women living with HIV: a systematic review of risks and benefits. AIDS. 2017;31:1579–1591. doi: 10.1097/QAD.0000000000001535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38.Forbes JC, Alimenti AM, Singer J, et al. A national review of vertical HIV transmission. AIDS. 2012;26:757–763. doi: 10.1097/QAD.0b013e328350995c. [DOI] [PubMed] [Google Scholar]
- 39.Warszawski J, Tubiana R, Le Chenadec J, et al. Mother-to-child HIV transmission despite antiretroviral therapy in the ANRS French perinatal cohort. AIDS. 2008;22:289–299. doi: 10.1097/QAD.0b013e3282f3d63c. [DOI] [PubMed] [Google Scholar]
- 40.Briand N, Jasseron C, Sibiude J, et al. Cesarean section for HIV-infected women in the combination antiretroviral therapies era, 2000–2010. Am J Obstet Gynecol. 2013;209:335.e1–335.e12. doi: 10.1016/j.ajog.2013.06.021. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
All of the data included in this analysis are publicly available, as this study is a meta-analysis of published studies. All data relevant to the study are included in the Article or included in the appendix.
