Share this article on:

Use of combination neonatal prophylaxis for the prevention of mother-to-child transmission of HIV infection in European high-risk infants

Chiappini, Elenaa; Galli, Luisaa; Giaquinto, Carlob; Ene, Luminitac; Goetghebuer, Tessad; Judd, Alie; Lisi, Catiusciaf; Malyuta, Ruslang; Noguera-Julian, Antonih; Ramos, Jose Tomasi; Rojo-Conejo, Pabloj; Rudin, Christophk; Tookey, Patl; de Martino, Maurizioa; Thorne, Clairelfor the European Pregnancy and Paediatric HIV Cohort Collaboration (EPPICC) study group in EuroCoord

doi: 10.1097/QAD.0b013e32835cffb1
Epidemiology and Social

Objectives: To evaluate use of combination neonatal prophylaxis (CNP) in infants at high risk for mother-to-child transmission (MTCT) of HIV in Europe and investigate whether CNP is more effective in preventing MTCT than single drug neonatal prophylaxis (SNP).

Design: Individual patient-data meta-analysis across eight observational studies.

Methods: Factors associated with CNP receipt and with MTCT were explored by logistic regression using data from nonbreastfed infants, born between 1996 and 2010 and at high risk for MTCT.

Results: In 5285 mother–infant pairs, 1463 (27.7%) had no antenatal or intrapartum antiretroviral prophylaxis, 915 (17.3%) had only intrapartum prophylaxis and 2907 (55.0%) mothers had detectable delivery viral load despite receiving antenatal antiretroviral therapy. Any neonatal prophylaxis was administered to 4623 (87.5%) infants altogether; 1105 (23.9%) received CNP. Factors significantly associated with the receipt of CNP were later calendar birth year, no elective caesarean section, maternal CD4 cell count less than 200 cells/μl, maternal delivery viral load more than 1000 copies/ml, no antenatal antiretroviral therapy, receipt of intrapartum single-dose nevirapine and cohort. After adjustment, absence of neonatal prophylaxis was associated with higher risk of MTCT compared to neonatal prophylaxis [adjusted odds ratio (aOR) 2.29; 95% confidence interval (95% CI) 1.46–2.59; P < 0.0001]. Further, there was no association between CNP and MTCT compared to SNP (aOR 1.41; 95% CI 0.97–2.5; P = 0.07).

Conclusion: In this European population, CNP use is increasing and associated with presence of MTCT risk factors. The finding of no observed difference in MTCT risk between one drug and CNP may reflect residual confounding or the fact that CNP may be effective only in a subgroup of infants rather than the whole population of high-risk infants.

Supplemental Digital Content is available in the text

aDepartment of Sciences for Woman and Child's Health, University of Florence, Anna Meyer Children's University Hospital, Florence

bUniversity of Padua, Italy

cVictor Babes Hospital, Bucharest, Romania

dHôpital St Pierre, Brussels, Belgium

eMedical Research Council Clinical Trials Unit, London, UK

fDepartment of Statistics, University of Florence, Italy

gPerinatal Prevention of AIDS Initiative, Odessa, Ukraine

hHospital Sant Joan de Déu-Universitat de Barcelona, Barcelona

iHospital Universitario de Getafe, Madrid

jHospital 12 de Octubre, Madrid, Spain

kUniversität Basel, Switzerland

lUniversity College London, UK.

Correspondence to Elena Chiappini, Department of Science for Woman and Child's Health, University of Florence, Anna Meyer Children's University Hospital, Florence, Italy. E-mail:

Received 3 August, 2012

Revised 13 November, 2012

Accepted 23 November, 2012

Supplemental digital content is available for this article. Direct URL citations appear in the printed text and are provided in the HTML and PDF versions of this article on the journal's Website (

Back to Top | Article Outline


Mother-to-child transmission (MTCT) of HIV infection has diminished to less than 2% in resource-rich countries as a result of widespread use of preventive interventions [1–4]. These include universal prenatal testing, administration of antiretroviral drugs to the mother during pregnancy and intrapartum and to the newborn, scheduled caesarean section delivery and avoidance of breastfeeding [1–4]. However, cases of vertical transmission continue to occur. Possible reasons include failures to administer appropriate and timely prophylaxis, for example because of lack of HIV testing, late presentation or seroconversion in pregnancy, the emergence of antiretroviral-resistant viruses and poor maternal adherence to antiretroviral therapy (ART) [5]. Most guidelines recommend consideration of combination neonatal prophylaxis (CNP), that is, two or more antiretroviral drugs in combination, in certain circumstances in which there is high risk of MTCT [6,7], and increasing use of CNP has been documented in some European countries [8–10]. However, the optimal prophylactic regimen and the additional efficacy of CNP in reducing MTCT risk are not known [6,7]. In a recent randomized controlled trial in nonbreastfed infants, a lower transmission rate with CNP use versus single drug neonatal prophylaxis (SNP) was observed after excluding in-utero transmissions [11]. However, this trial included a highly selected population of infants whose mothers were diagnosed at delivery and untreated during pregnancy, and no data are currently available regarding other risk groups.

We conducted an individual patient-data meta-analysis from a collaboration of observational studies with the aim of evaluating use of CNP in a population of infants at high risk of MTCT of HIV in Europe. Moreover, we attempted to test the hypothesis that CNP is more effective in reducing MTCT risk than one drug prophylaxis in high-risk mother–child pairs, after adjusting for antenatal ART use and other possible confounding factors. Secondary objectives were to characterize the subset of mother–child pairs at high risk for MTCT in Europe.

Back to Top | Article Outline


Inclusion criteria

Children born to diagnosed HIV-infected mothers between 1 January 1996 and 30 June 2010 at high risk for acquiring HIV infection, defined according to U.S. Guidelines [7], were included: those born to mothers who received antenatal and intrapartum antiretroviral drugs but had suboptimal viral suppression at delivery (defined as a detectable viral load (>50 copies/ml) documented in the last 8 weeks of pregnancy and/or at delivery), received only intrapartum antiretroviral drugs, and received no antenatal or intrapartum antiretroviral drugs. All infants had to have been followed up since birth and to have known HIV infection status. Breastfed infants were excluded.

Back to Top | Article Outline

Data collection and creation of pooled dataset

The present study was conducted within the European Pregnancy and Paediatric HIV Cohort Collaboration (EPPICC), a network of European cohorts of prospectively observed mother–child pairs within EuroCoord ( [12]. Eight cohorts including mother–child pairs from seven countries participated: the UK and Ireland's National Study of HIV in Pregnancy and Childhood (NSHPC) and Collaborative HIV Paediatric Study (CHIPS) [2]; the Italian Register for HIV infection in children (ITLR) [3]; the Madrid Cohort of HIV-infected Children [13]; the Catalan Cohort of HIV-infected Children (CoRISPE-Cat) [14]; the ‘Victor Babes’ Hospital Cohort, Bucharest, Romania [15]; the Swiss Mother and Child HIV Cohort Study (MoCHiV) [16] and the European Collaborative Study (ECS) on HIV-infected pregnant women and their children [1]. The ECS was considered as two separate studies, the Western-ECS and Ukraine-ECS, because of the specific approach to prevention of MTCT (PMTCT) in the latter, a lower income setting, namely based on intrapartum single-dose nevirapine (sdNVP) up to 2003, and use of combined ART (cART) for PMTCT only since 2008. Individual cohort methods and characteristics are described in detail elsewhere [1–4,13–17].

Cohorts provided anonymous data according to a standard operating procedure, submitted using the HIV Cohorts Data Exchange Protocol ( Duplicates were removed before merging the datasets into the final study dataset (final merger January 2011). The dataset included maternal and infant demographic data, birth year, sex, birth weight, gestational age, maternal viral load and CD4 cell count at delivery or closest to delivery (within last 8 weeks of pregnancy), use of antenatal ART, intrapartum prophylaxis with intravenous zidovudine (ZDV) or sdNVP, neonatal prophylaxis (including type/number of drugs, timing and duration), mode of delivery, description of infant risk factors (as defined below) and infant infection status. Data not available within some cohort studies included information regarding neonatal prophylaxis duration (not available for the NSHPC/CHIPS) and maternal viral load (not available for the Ukraine-ECS). Information about other potential maternal risk factors (i.e. current illicit drug use, drug dosages, adherence and resistance profile) was not available. Each participating study was responsible for ensuring that ethics approval for the analysis was in place and for compliance with local and national data protection requirements.

Back to Top | Article Outline


HIV infection was diagnosed by the persistence of HIV antibodies after 18 months or, before 18 months, by detection of viral markers (by DNA or RNA PCR) on at least two occasions (excluding cord blood) [7]. Uninfected children were defined as those with negative HIV antibodies and/or negative PCR on at least two occasions. Neonatal prophylaxis was defined as any course of one or more antiretroviral drugs administered with a prophylactic purpose and initiated within the first 72 h of life [7]. CNP was defined as combination of two or more antiretroviral drugs (i.e. sdNVP plus ZDV or ZDV plus lamivudine plus lopinavir/ritonavir) given for PMTCT. Maternal viral load at delivery was defined as the closest measurement prior to delivery within 8 weeks.

Back to Top | Article Outline

Statistical analysis

Proportions were compared using χ 2 or Fisher's exact test and medians using Wilcoxon Mann–Whitney U tests. All significant tests were two-sided. Logistic regression models were fitted to obtain odds ratios (ORs) and 95% confidence intervals (CIs). Maternal CD4 cell count was categorized as less than or at least 200 cells/μl and maternal viral load as less than 50, 50–999 and at least 1000 copies/ml. Factors potentially associated with receipt of neonatal prophylaxis and of CNP were explored in logistic regression analyses. Factors initially included in the univariable analyses were birth year, sex, preterm delivery (categorized as ≤32, 33–36, ≥37 weeks), cohort (categorized as NSHPC/CHIPS, ITLR, Ukraine-ECS, others), intrapartum intravenous ZDV use, intrapartum sdNVP use, maternal viral load and CD4 cell count at delivery, antenatal ART (categorized as none, monotherapy, dual therapy and cART with three or more drugs), maternal age (categorized as < or ≥21 years) and elective caesarean section delivery. These factors were included in multivariable models if significantly associated with risk of receiving neonatal prophylaxis or CNP in univariable analysis (if P < 0.05).

To test the hypothesis that CNP is more effective in reducing MTCT risk than one drug prophylaxis in high-risk situations, univariable and multivariable logistic regression models were performed. Variables included in the univariable analysis were as above, plus neonatal prophylaxis (categorized as none, one drug, CNP, unknown). These factors were included in the multivariable model if significantly associated with MTCT in univariable analysis. Since several European and US guidelines recommend the inclusion of ZDV in the neonatal prophylactic regimen [6,7], a subanalysis was performed including only ZDV-containing regimens.

Heterogeneity between studies in the association between use of CNP and MTCT risk was tested using a χ 2 test of homogeneity of OR, and in logistic regression models. Significant heterogeneity between studies was apparent when the Ukraine-ECS was included in the model of MTCT risk factors (χ 2 = 10.47; P = 0.015), which disappeared on its exclusion (χ 2 = 0.91; P = 0.633). Therefore, the Ukraine-ECS was excluded from the final model.

Back to Top | Article Outline

Missing data analysis

There was a substantial proportion (34%) of missing data on maternal viral load at delivery, partly owing to there being no measurements available for the Ukraine-ECS. However, excluding data from Ukraine-ECS, maternal delivery viral load data were still lacking for 22% of mother–infant pairs. As this variable is the main factor influencing MTCT [7], missing data analysis was performed using the ICE (multiple imputation by the Multivariate Imputation by Chained Equations in R system of chained equations) method, which creates multiple imputed datasets from an original dataset with missing values [18]. Five imputation datasets were generated and the multivariable analyses repeated on these.

Back to Top | Article Outline

Study power analysis

A retrospective analysis showed that there was 93.0% power to detect a difference in MTCT of 2% between infants receiving neonatal prophylaxis with one drug versus those receiving CNP, at α = 0.05; n1 = 2919; n2 = 1015 and an MTCT rate in group 1 of 3.4%.

Statistical analyses were performed using the STATA/SE version 10.0 software package (Stata Corporation, College Station, Texas, USA).

Back to Top | Article Outline


Population characteristics

A total of 5285 mother–infant pairs were defined as at high risk for MTCT and included in the study. Overall, for 1463 (27.7%) mother–infant pairs no antenatal or intrapartum antiretroviral prophylaxis was used, for 915 (17.3%) only intrapartum prophylaxis was used and for 2907 (55.0%) antenatal ART was received but mothers had detectable viral load around delivery; 80% (n = 2324) of these latter mothers had received 4 weeks antenatal ART or less, 10% (n = 302) more than 4 weeks and 10% (n = 281) had unknown ART duration. Characteristics of the study subjects, by cohort, are summarized in Table 1. More than two-thirds of infants originated from two national multicentre cohorts, 41% (n = 2184) from the NSHPC/CHIPS (UK and Ireland) and 23% (n = 1214) from the ITLR (Italy), with the remaining third from the Madrid Cohort (77 infants; 1.5%), the CoRISPE-Cat study (375 infants; 7.1%), the ‘Victor Babes’ Hospital Cohort (45 infants; 0.9%), MoCHiV (168 infants; 3.2%), Ukraine-ECS (826 infants; 15.6%) and Western-ECS (396 infants; 7.5%). Overall, 45% of women received cART antenatally and 13% received sdNVP intrapartum, of whom three-quarters were from the Ukraine-ECS. Ten percent of women were severely immunosuppressed (CD4 cell count <200 cells/μl), and overall half delivered by elective caesarean section (Table 1).

Table 1

Table 1

Back to Top | Article Outline

Use of neonatal prophylaxis

Neonatal prophylaxis was administered to 4623/5285 (87.5%) infants, of whom 3518 (66.6%) received one drug and 1105 (23.9%) received CNP; most infants on CNP received three drugs (n = 677; 61.3%), with the remaining 428 (38.7%) receiving two. Information regarding neonatal prophylaxis duration was available for 1260 (35.8%) infants receiving one drug and for 404 (36.6%) infants receiving CNP [median duration 5 weeks (interquartile range, IQR 4–6) and 6 weeks (IQR 4–6), respectively; P = 0.281]. The proportion of neonates receiving CNP increased from 19.1% (164/860) in 1996–2000 and 19.6% (468/2386) in 2001–2005 to 34.3% (471/1376) in 2006–2010 (P < 0.0001, χ 2 test for trend), largely because of increasing use of three-drug CNP over the period (Fig. 1). CNP use also differed by cohort, being higher in NSHPC/CHIPS (30.7%) and other cohorts (31.4%), than in the ITLR (7.4%) and Ukraine-ECS (13.1%) (Table 1; χ 2 = 257.3; P < 0.0001).

Fig. 1

Fig. 1

Among infants receiving neonatal prophylaxis with one drug, 89.5% received ZDV (n = 3152), 8.9% (n = 312) received sdNVP, 10 received another NRTI and 67 had an unspecified drug. Overall, the proportion receiving CNP was 24.1% (681/2140) among those whose mothers received antenatal ART (all with detectable delivery viral load according to inclusion criteria), 22.9% (191/834) among those whose mothers received no antenatal or intrapartum antiretroviral prophylaxis and 23.9% (231/967) among those whose mothers received only intrapartum prophylaxis (χ 2 = 0.563; P = 0.095). The proportion of infants receiving CNP by gestational age was 22.8% (807/3545) in term infants, 25.6% (190/742) in those born at 33–36 weeks and 32.9% (86/261) in severely preterm infants (≤32 weeks) (χ 2 = 14.40; P < 0.0001) (Fig. 2).

Fig. 2

Fig. 2

Overall, the most frequently adopted CNP regimens were ZDV plus lamivudine (n = 149; 13.5%), ZDV plus sdNVP (n = 250; 22.7%) and ZDV plus sdNVP plus lamivudine (n = 615; 55.8%), of the remaining 89 infants on CNP, 48 (4.4%) received protease inhibitors. Among the severely preterm infants who had received CNP (86/261; 32.9%), 56 (65%) received ZDV plus lamivudine plus sdNVP; 29 (33.7%) received two drugs (i.e. ZDV+sdNVP; ZDV plus lamivudine; lamivudine plus sdNNNVP) and lopinavir/ritonavir plus ZDV and lamivudine was used only in one child (1%).

Back to Top | Article Outline

Factors associated with receipt of neonatal prophylaxis and combination neonatal prophylaxis

In univariable analyses, factors significantly associated with receipt of any or single drug neonatal prophylaxis among the 5149 children with available information (136 children with missing information on neonatal prophylaxis excluded) were calendar birth year, intrapartum prophylaxis with intravenous ZDV and maternal viral load (data not shown). In multivariable analysis, calendar birth year (aOR 1.43 per year; 95% CI 1.35–1.52; P < 0.0001), receipt of intrapartum prophylaxis with intravenous ZDV (aOR 18.50; 95% CI 10.31–33.19; P < 0.0001) and detectable maternal viral load at delivery (aOR 5.42; 95% CI 2.01–14.05; P < 0.0001 for 50–999 copies/ml and aOR 3.62; 95% CI 1.63–8.03; P = 0.002 for ≥1000 copies/ml versus <50 copies/ml) remained independently associated with receipt of neonatal prophylaxis.

Among the 4623 children receiving neonatal prophylaxis, factors significantly associated with receipt of CNP compared to single drug neonatal prophylaxis in multivariable analyses were calendar birth year, cohort, lack of intrapartum sdNVP, maternal detectable viral load, no antenatal ART, maternal CD4+ T-lymphocyte count less than 200 lymphocytes/μl and lack of elective caesarean section delivery, but not lack of intrapartum prophylaxis with intravenous ZDV prophylaxis (Table 2). Among children receiving CNP, factors associated with receipt of three versus two drugs were more recent birth year (aOR 1.46; 95% CI 1.38–1.54; P < 0.0001), lack of antenatal ART (aOR 0.21; 95% CI 0.07–0.68; P = 0.009), receipt of intrapartum prophylaxis with intravenous ZDV (aOR 3.38; 95% CI 2.30–4.46; P < 0.0001) and younger gestational age (aOR 0.44; 95% CI 0.24–0.77; P = 0.005).

Table 2

Table 2

Back to Top | Article Outline

Factors associated with mother-to-child transmission of HIV infection

Crude MTCT rates were 5.6% (95% CI 4.9–6.3; 250/4459) in the Western European cohorts and 16.0% (95% CI 13.5–18.4; 132/826) in the Ukraine-ECS. Crude MTCT rates were 3.4% (95% CI 2.7–4.0), 6.3% (95% CI 4.8–7.6) and 17.7% (95% CI 13.9–21.5) for one-drug neonatal prophylaxis, CNP and no neonatal prophylaxis, respectively (Table 3); in the Ukraine-ECS cohort, crude MTCT rates were 18.7% (95% CI 15.4–22.2; 93/506), 13.9% (95% CI 6.3–21.6; 11/79) and 25.7% (95% CI 17.5–33.9; 28/109) for these neonatal prophylaxis groups, respectively. For all subsequent MTCT analyses, Ukraine-ECS was excluded because of significant heterogeneity, leaving a total of 4459 children. Subanalysis by risk factor group was performed. Crude MTCT rates were 1.8% (39/2140) and 4.2% (29/681; aOR 1.97; 95% CI 1.14–3.39; P = 0.014) in one drug and CNP groups, respectively, among infants whose mothers received antenatal ART; 7.0% (18/257) and 5.9% (8/134; aOR 0.86; 95% CI 0.28–2.64; P = 0.804) in those whose mothers received no antenatal or intrapartum antiretroviral prophylaxis; 8.0% (42/523) and 13.7% (27/198) (aOR 1.57; 95% CI 0.81–3.08; P = 0.178) among those whose mothers received only intrapartum prophylaxis.

Table 3

Table 3

In univariable analysis, factors associated with increased MTCT risk were earlier calendar birth year, female sex, preterm delivery, cohort, lack of intrapartum prophylaxis with intravenous ZDV, higher maternal viral load, lack of antenatal ART, lack of elective caesarean delivery and having no neonatal prophylaxis (Table 3). In multivariable analysis, female sex, gestational age 32 weeks or less, maternal viral load at delivery more than 1000 copies/ml, lack of antenatal ART, lack of elective caesarean section and having no neonatal prophylaxis remained significantly associated with transmission (Table 3). Not having any neonatal prophylaxis was associated with a more than two-fold increased risk of MTCT while probability of acquisition was not significantly different between infants receiving CNP and those receiving neonatal prophylaxis with one drug (OR 1.41; P = 0.07) (Table 3). A subanalysis was performed including only ZDV-based neonatal prophylactic regimens (n = 3831), which yielded similar results (aOR 1.49; 95% CI 0.95–2.12 for MTCT risk in infants receiving CNP versus those receiving ZDV only). Missing data analysis also corroborated these findings (Appendix 1,

Back to Top | Article Outline


We have demonstrated increased use of CNP in Europe in this study of infants at elevated MTCT risk, although with some geographic differences. By 2006–2010, one-third of neonates born to mothers not in receipt of antenatal ART, or delivering with detectable viral load despite treatment, received CNP. Guidelines in the United States and Europe have suggested that high-risk infants may benefit from modifications in the standard ACTG076 treatment protocol, including use of CNP since 2005 [6,7], based on expert opinion.

Clinical trial evidence of the superiority of CNP over ZDV monoprophylaxis became available in 2011: the NICHD-HPTN040/PACTG1043 study demonstrated that among infants not exposed to antenatal ART and with negative PCR in the first days of life, ZDV use for 6 weeks plus three NVP doses, or use of ZDV for 6 weeks with lamivudine plus nelfinavir for 2 weeks, significantly reduced transmission risk compared with ZDV monotherapy (respective intrapartum transmission rates of 2.2, 2.5 and 4.9%) [11].

In our study, factors associated with receipt of CNP, besides cohort and calendar birth year, were those also associated with higher risk for MTCT, as expected given current guidelines, such as high maternal delivery viral load, no antenatal ART, severe maternal immunodeficiency and lack of elective caesarean section delivery. We also observed an association between intrapartum sdNVP use and greater probability of receiving CNP, most likely reflecting sdNVP use in which the mother was diagnosed intrapartum. Despite the fact that most drugs used are not currently licensed for use in preterm infants in Europe and poor data are available regarding pharmacokinetic profile and tolerability [19–21], CNP was used here quite extensively in preterm infants and in particular it should be noted that CNP use was significantly more common in severely preterm infants (<32 weeks; 33%) than in other groups.

As expected given our selected study population, MTCT rates were higher than those reported within Western European cohorts overall, at 6% compared with around 1–2% [1–3], and consistent with rates of 7–8% reported among subgroups of HIV-positive pregnant women with late presentation and/or poor access to services [22,23]. The largest subgroup were infants whose mothers had received antenatal ART (in nearly 75% of cases cART), but exhibited ongoing viral replication at delivery, with a transmission rate of 2.7%. Around 20–30% of treated pregnant HIV-positive women had detectable viral load at delivery in recent studies, associated with severe maternal immunosuppression, high baseline viral load, shorter ART duration, injecting drug use and young age [24–26]. In our study, although some treated women with detectable viral load may have experienced viral rebound owing to nonadherence and/or drug resistance, most (80%) received less than 4 weeks of ART and thus probably had insufficient duration of ART to reduce viral load to less than 50 copies/ml before delivery. Our findings concur with established evidence that MTCT risk is increased wherein delivery maternal viral load is high, wherein no antenatal ART is received, among severely preterm infants and in girls [27–29]. With respect to intrapartum interventions, elective caesarean section delivery reduced MTCT risk by around half (aOR 1.29) but intrapartum prophylaxis with intravenous ZDV use was not associated with reduced transmission.

Overall, 12% of neonates received no neonatal prophylaxis, with an aOR of 2.29 for transmission risk versus those receiving single drug neonatal prophylaxis (predominantly ZDV). Infants receiving CNP had a higher transmission rate (6.3%) than children receiving one drug prophylaxis (3.4%), and CNP receipt was associated with risk factors for MTCT, including a four-fold increased probability of receipt in infants whose mothers had delivery viral load more than 1000 copies/ml. We used multivariable analyses to control for confounding by indication, but even after controlling for known risk factors, we obtained an aOR of 1.4 (P = 0.07) for MTCT risk in the CNP group, versus single drug neonatal prophylaxis. The expected effect would be in the opposite direction based on results from PACTG1043, suggesting that our estimate remained confounded, potentially because of both residual and unmeasured confounding. However, our population is very different from the highly selected population included in the PACTG1043 trial, which exclusively enrolled infants born to untreated mothers diagnosed at delivery. We observed a lower MTCT rate in infants receiving CNP rather than one drug in the group of infants whose mothers received no antenatal or intrapartum antiretroviral prophylaxis (5.9 versus 7.0%), although study power was insufficient to disclose any significant difference. This subgroup was similar to the population included in the PACTG1043 trial. Taken together these results suggest a benefit of CNP in infants whose mothers have not received any prophylaxis, such as those diagnosed late in pregnancy or intrapartum. On the contrary, in our study the subgroup of infants whose mothers received antenatal ART but had a detectable delivery viral load did not appear to benefit from CNP, having a significantly higher MTCT rate of 4.2% compared with 1.8%.

We were unable to investigate differences between specific CNP regimens. Most CNP regimens involved ZDV and/or lamivudine and/or sdNVP, with very limited use of protease inhibitor-based combinations, reflecting lack of appropriate formulations and dosing information. Lopinavir/ritonavir exposure in utero and neonatally was recently associated with a transient adrenal dysfunction and cardiac toxicity [30], resulting in the recommendation that this should be avoided in newborns. In the PACTG1043 trial there was significantly more toxicity in the nelfinavir-containing arm than the ZDV plus sdNVP arm [11], and the large interpatient variability in nelfinavir exposure reported highlights the challenges of protease inhibitor dosing in the context of neonatal prophylaxis [20].

The MTCT rate for CNP above is similar to the overall transmission rates in the PACTG1043 trial's ZDV/NVP and triple combination neonatal prophylaxis arms (7.1 and 7.4%, respectively), although the rate in our single drug group is lower than the 11.1% in the ZDV trial arm [11], probably reflecting our inclusion of the group with nonsuppressive antenatal ART exposure.

Observational data have well recognized limitations, which should be considered in interpreting our results: our study population was heterogeneous, and data regarding resistance profile, adherence, timing of maternal diagnosis, drug dosing and toxicity were lacking [31–32]. It is likely that unmeasured confounders influenced some results, including the higher MTCT in infants receiving CNP rather than single drug prophylaxis. Furthermore, in our study the lack of information about timing of infection may explain lack of improved efficacy of CNP. As we lacked comprehensive data on early PCR results, we could not exclude transmissions that occurred in utero and our analyses were limited to total transmissions in this nonbreastfed population. In the PACTG1043 trial, 66% of the infected infants were infected in utero, with an in-utero transmission rate of 5.4% [11].

In conclusion, CNP is increasingly used in Europe among neonates at high risk for MTCT. Discrepancies in the use of CNP by cohort between cohorts may be ascribed to the fact that European guidelines are heterogeneous.

Although CNP may be a good strategy for PMTCT in specific subgroups of infants at high risk for acquisition of HIV, such as those born to mothers diagnosed intrapartum and/or without antenatal ART, as suggested by the PACTG1043 trial [11], we were unable to confirm this with our observational, ‘real-life’ data. Considering the lack of data on pharmacokinetics, the potentially increased toxicity of CNP and the risk of resistance for infants infected despite PMTCT interventions, further studies of CNP are needed. Our data confirm the benefit of neonatal postexposure prophylaxis on MTCT in this high-risk population. However, more than one in 10 of these high-risk infants did not receive neonatal prophylaxis, representing a missed prevention opportunity [23]. To optimize PMTCT in Europe, strategies to improve maternal viral suppression rates at delivery are needed. Updated guidelines recommend earlier initiation of ART in pregnancy for women not on ART at conception [6–7], whereas recent findings in Europe indicate that improved management of HIV-positive women of childbearing age receiving treatment before conception is also needed [33,34].

Back to Top | Article Outline


Contributions: E.C., L.G. and M.deM. were responsible for the study concept and design. E.C. and C.L. were responsible for undertaking the analyses; E.C. acts as guarantor for the analyses and has full access to the dataset. E.C. and C.T. drafted the manuscript, and all members of the Writing Committee critically reviewed the manuscript. E.C., M.deM., L.E., L.G., A.J., R.M., A.N.-J., J.T.R., P.R.-C., C.R., C.T. and P.T. provided data for the study. All members of the Writing Committee (listed authors) participated in discussions about the design of the study, the choice of statistical analyses and interpretation of the findings.

The following persons contributed to this EPPICC analysis (listed by cohort name): CoRISPE-cat, Spain (J. Almeda, O. Calavia, D. Carnicer-Pont, J. Casabona, M.T. Coll, J. Escribano, M. Espiau, C. Fortuny, L. García, A. Martín-Nalda, J Masip, L. Mayol, M. Méndez, A. Mur, A. Noguera-Julian, C. Rodrigo, P. Soler-Palacín, A. Soriano, T. Vallmanya), European Collaborative Study (Thorne, R. Malyuta, I. Semenenko, I. Grosch Wörner, H.J. Scherpbier, M. Kreyenbroek, M.H. Godfried, F.J.B. Nellen, K. Boer, L. Navér, A.B. Bohlin, E. Belfrage, S. Lindgren, J. Levy, P. Barlow, Y. Manigart, M. Hainaut, T. Goetghebuer, B. Brichard, J. De Camps, N. Thiry, G. Deboone, H. Waterloos, A. MÛr, A. Payà, M.A. López-Vilchez, R. Carreras, N.H. Valerius, V. Rosenfeldt, O. Coll, G.P. Taylor, E.G.H. Lyall, Z. Penn, T. Niemieç, M. Marczynska, S. Dobosz, J. Popielska, A. Oldakowska, T. Pilipenko, A. Zayuz, S. Posokhova, T. Kaleeva, A. Shelyag, S. Servetsky, A. Stelmah, G. Kiseleva, O.A. Zalata S. Solokha, M.P. Grazhdanov, N. Bashkatova, I. Raus, O.V. Yurchenko, Z. Ruban, O. Gloushenko), Italian Register (Elena Chiappini, Luisa Galli, Maurizio de Martino, Pier-Angelo Tovo, Clara Gabiano), Madrid Cohort, Spain (J.T. Ramos, P. Rojo-Conejo), MoCHiV, Switzerland (C. Rudin), UK and Ireland National Study of HIV in Pregnancy and Childhood/Collaborative HIV Paediatric Study (P. Tookey, J. Masters, I. Shakes, H. Haile-Selassie, C. Townsend, A. Judd, C. O’Leary, D.M. Gibb, K. Doerholt), Victor Babes Cohort, Romania (L. Ene, D. Duiculescu).

This study was funded by the European Union Seventh Framework Programme (FP7/2007–2013) under EuroCoord grant agreement number 260694 and the PENTA Foundation.

Back to Top | Article Outline

Conflicts of interest

There are no conflicts of interest.

Back to Top | Article Outline


1. Boer K, England K, Godfried MH, Thorne C. European Collaborative StudyMode 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.
2. Townsend CL, Cortina-Borja M, Peckham CS, de Ruiter A, Lyall H, Tookey PA. Low rates of mother-to-child transmission of HIV following effective pregnancy interventions in the United Kingdom and Ireland, 2000–2006. AIDS 2008; 22:973–981.
3. Chiappini E, Galli L, Lisi C, Gabiano C, Giaquinto C, Giacomet V, et al. Italian Register for HIV Infection in ChildrenRisk of perinatal HIV infection in infants born in Italy to immigrant mothers. Clin Infect Dis 2011; 53:310–313.
4. Magder LS, Mofenson L, Paul ME, Zorrilla CD, Blattner WA, Tuomala RE, et al. Risk factors for in utero and intrapartum transmission of HIV. J Acquir Immune Defic Syndr 2005; 38:87–95.
5. Mepham SO, Bland RM, Newell ML. Prevention of mother-to-child transmission of HIV in resource-rich and -poor settings. BJOG 2011; 118:202–218.
6. de Ruiter A, Mercey D, Anderson J, Chakraborty R, Clayden P, Foster G, et al. British HIV Association and Children's HIV Association guidelines for the management of HIV infection in pregnant women. HIV Med 2008; 9:452–502.
7. Perinatal HIV Guidelines Working Group. Recommendations for Use of Antiretroviral Drugs in Pregnant HIV-1-Infected Women for Maternal Health and Interventions to Reduce Perinatal HIV Transmission in the United States, 14 September 2011.
8. England K, Thorne C. for the European Collaborative StudyUse of neonatal antiretroviral prophylaxis for prevention of mother-to-child transmission of HIV is decreasing in Western Europe. Clin Infect Dis 2009; 48:1797–1800.
9. Chiappini E, Galli L, Tovo PA, Gabiano C, Buffolano W, Cellini M, et al. for the Italian Register for HIV infection in childrenUse of postnatal antiretroviral prophylaxis for prevention of mother-to-child transmission of HIV is increasing in Italy. Clin Infect Dis 2009; 49:1618–1619.
10. Haile-Selassie H, Townsend C, Tookey P. Use of neonatal postexposure prophylaxis for prevention of mother-to-child HIV transmission in the UK and Ireland, 2001–21008. HIV Med 2011; 2:422–427.
11. Nielsen-Saines K, Watts H, Gonçalves Veloso H, Bryson Y, Joao E, Pilotto JH, et al. and NICHD/HPTN 040 Study Group. Phase III Randomized Trial of the Safety and Efficacy of 3 Neonatal ARV Regimens for Prevention of Intrapartum HIV-1 Transmission: NICHD HPTN 040/PACTG 1043 (Paper # 124LB). In 18th CROI. Boston; 27 February to 6 March 2011.
12. de Wolf F, Sabin C, Kirk O, Thorne C, Chene G, Porter K. for EuroCoordDeveloping a multidisciplinary network for clinical research on HIV infection: the EuroCoord experience. Clin Investig 2013; 2:255–264.
13. González-Tomé MI, Ramos Amador JT, Guillen S, Solís I, Fernández-Ibieta M, Muñoz E, et al. Spanish Cohort of HIV-infected Mother-Infant PairsGestational diabetes mellitus in a cohort of HIV-1 infected women. HIV Med 2008; 9:868–874.
14. Noguera A, Fortuny C, Muñoz-Almagro C, Sanchez E, Vilaseca MA, Artuch R, et al. Hyperlactatemia in human immunodeficiency virus-uninfected infants who are exposed to antiretrovirals. Pediatrics 2004; 114:e598–e603.
15. Muila A, Georgescu A. Retrospective observational study on risk factors of HIV-1 infection on a group of infected pregnant women and new-born children from ‘Dr Victor Babes’ Clinical Hospital for Infectious and Tropical Diseases, Bucharest [in Romanian]. Romanian Revue Infect Dis 2009; XII:154–163.
16. Rudin C, Spaenhauer A, Keiser O, Rickenbach M, Kind C, Aebi-Popp K. Brinkhof MW; Swiss HIV Cohort Study (SHCS); Swiss Mother & Child HIV Cohort Study (MoCHiV). Antiretroviral therapy during pregnancy and premature birth: analysis of Swiss data. HIV Med 2011; 12:228–235.
17. Thorne C, Semenenko I, Pilipenko T, Malyuta R. Ukraine European Collaborative Study GroupProgress in prevention of mother-to-child transmission of HIV infection in Ukraine: results from a birth cohort study. BMC Infect Dis 2009; 9:40.
18. White IR, Royston P, Wood AM. Multiple imputation using chained equations: issues and guidance for practice. Stat Med 2011; 30:377–399.
19. Nikanjam M, Chadwick EG, Robbins B, Alvero C, Palumbo P, Yogev R, et al. IMPAACT P1030 TeamAssessment of lopinavir pharmacokinetics with respect to developmental changes in infants and the impact on weight band-based dosing. Clin Pharmacol Ther 2012; 91:243–249.
20. Mirochnick M, Nielsen-Saines K, Pilotto JH, Pinto J, Veloso VG, Rossi S, et al. NICHD HPTN 040/PACTG 1043 PROTOCOL TeamNelfinavir and Lamivudine pharmacokinetics during the first two weeks of life. Pediatr Infect Dis J 2011; 30:769–772.
21. Zhao W, Cella M, Della Pasqua O, Burger D, Jacqz-Aigrain E. on behalf of Pediatric European Network for Treatment of AIDS (PENTA) 15 study groupPopulation pharmacokinetics and maximum a posteriori probability Bayesian estimator of abacavir: application of individualized therapy in HIV-infected infants and toddlers. Br J Clin Pharmacol 2012; 73:641–650.
22. European Collaborative Study in EuroCoord, Bailey H, Townsend C, Cortina-Borja M, Thorne C. Insufficient antiretroviral therapy in pregnancy: missed opportunities for prevention of mother-to-child transmission of HIV in Europe. Antivir Ther 2011; 16:895–903.
23. Peters V, Liu KL, Dominguez K, Frederick T, Melville S, Hsu HW, et al. Missed opportunities for perinatal HIV prevention among HIV-exposed infants born 1996–2000, Pediatric Spectrum of HIV Disease Cohort. Pediatrics 2003; 111:1186–1191.
24. Floridia M, Ravizza M, Pinnetti C, Tibaldi C, Bucceri A, Anzidei G, et al. Italian Group on Surveillance on Antiretroviral Treatment in PregnancyTreatment change in pregnancy is a significant risk factor for detectable HIV-1 RNA in plasma at end of pregnancy. HIV Clin Trials 2010; 11:303–311.
25. Katz IT, Shapiro R, Li D, Govindarajulu U, Thompson B, Watts DH, et al. Risk factors for detectable HIV-1 RNA at delivery among women receiving highly active antiretroviral therapy in the women and infants transmission study. J Acquir Immune Defic Syndr 2010; 54:27–34.
26. Tariq S, Townsend CL, Cortina-Borja M, Duong T, Elford J, Thorne C, Tookey PA. European Collaborative Study; National Study of HIV in Pregnancy ChildhoodUse of zidovudine-sparing HAART in pregnant HIV-infected women in Europe: 2000–2009. J Acquir Immune Defic Syndr 2011; 57:326–333.
27. Galli L, Puliti D, Chiappini E, Gabiano C, Tovo PA, Pezzotti P, de Martino M. Writing CommiteeLower mother-to-child HIV-1 transmission in boys is independent of type of delivery and antiretroviral prophylaxis: the Italian Register for HIV Infection in Children. J Acquir Immune Defic Syndr 2005; 40:479–485.
28. Thorne C, Newell ML. European Collaborative StudyAre girls more at risk of intrauterine- acquired HIV infection than boys?. AIDS 2004; 18:344–347.
29. European Mode of Delivery Collaboration. Elective caesarean-section versus vaginal delivery in prevention of vertical HIV-1 transmission: a randomised clinical trial. Lancet 1999; 353:1035–1039
30. Simon A, Warszawski J, Kariyawasam D, Le Chenadec J, Benhammou V, Czernichow P, et al. ANRS French Perinatal Cohort Study GroupAssociation of prenatal and postnatal exposure to lopinavir-ritonavir and adrenal dysfunction among uninfected infants of HIV-infected mothers. JAMA 2011; 306:70–78.
31. Demas PA, Webber MP, Schoenbaum EE, Weedon J, McWayne J, Enriquez E, et al. Maternal adherence to the zidovudine regimen for HIV-exposed infants to prevent HIV infection: a preliminary study. Pediatrics 2002; 110:e35.
32. Lahoz R, Noguera A, Rovira N, Català A, Sánchez E, Jiménez R, Fortuny C. Antiretroviral-related hematologic short-term toxicity in healthy infants: implications of the new neonatal 4-week zidovudine regimen. Pediatr Infect Dis J 2010; 29:376–379.
33. European Collaborative Study. Factors associated with HIV RNA levels in pregnant women on nonsuppressive highly active antiretroviral therapy at conception. Antivir Ther 2010; 15:41–49.
34. Huntington SE, Bansi LK, Thorne C, Anderson J, Newell ML, Taylor GP, et al. UK Collaborative HIV Cohort (UK CHIC) Study and the National Study of HIV in Pregnancy and Childhood (NSHPC)Treatment switches during pregnancy among HIV-positive women on antiretroviral therapy at conception. AIDS 2011; 25:1647–1655.

children; mother-to-child transmission; prophylaxis

Supplemental Digital Content

Back to Top | Article Outline
© 2013 Lippincott Williams & Wilkins, Inc.