Background
Initially established in 1986 to monitor paediatric patients with AIDS, the United Kingdom and Ireland’s National Study of HIV in Pregnancy and Childhood (NSHPC) has conducted comprehensive population-based surveillance of HIV infection in pregnant women since 1990. Observational data collected through the NSHPC can be used to examine trends in the diagnosis and management of HIV in pregnant women and related pregnancy outcomes. Data from this and other studies have demonstrated that vertical transmission risk reduction strategies, such as the use of antiretroviral therapy (ART) during pregnancy, delivery management and avoidance of breastfeeding, can result in an appreciable reduction of the mother-to-child transmission (MTCT) rate from approximately 20 to 0.5 % among populations living in resource-rich settings [
1‐
8]. In more resource-limited regions, similarly low vertical transmission rates have not yet been achieved, as MTCT prevention efforts still confront numerous challenges, including access to and availability of ART [
9‐
11].
Guidelines issued by the British HIV Association (BHIVA) and the World Health Organization (WHO) for the management of HIV infection in women during pregnancy recommend combination ART for maternal treatment needs and to reduce the risk of MTCT [
12‐
14]. For pregnant women with HIV infection who require ART for their own health, based on CD4+ T-cell count or disease stage, immediate initiation or continuation of ART is advised. National recommendations vary regarding the timing of initiation of ART to reduce the risk of MTCT during pregnancy in women with HIV who do not need ART for their own health [
12‐
15]. In the United Kingdom, current BHIVA guidelines recommend initiating ART by 24 weeks of gestation for this group of women, with a range of factors to be considered in deciding when to begin therapy, including maternal CD4+ T-cell count, baseline viral load and potential for preterm delivery [
12]. As with any drug taken during pregnancy, antenatal exposure to ART agents raises concerns regarding toxicity, teratogenicity and an increased risk of adverse pregnancy outcomes, including preterm delivery, low birth weight and congenital abnormalities [
16‐
19].
The incidence of pregnancy in women with HIV infection accessing HIV clinical care substantially increased between 2000 and 2009 in the United Kingdom, and subsequently stabilised. Increasing numbers of women with HIV infection are initiating or maintaining ART during pregnancy [
3,
7,
20,
21]. Lopinavir, a protease inhibitor that is co-formulated with ritonavir (LPV/r), is commonly used as an anchor drug as part of an ART regimen during pregnancy. Previous investigations into the association between the use of protease inhibitor–containing ART during pregnancy and preterm delivery have yielded mixed results [
18,
22‐
28]. A 2013 systematic review of 9 studies of ART in pregnant women with HIV infection (
n=2675, in total) found that LPV/r use was associated with reduced maternal viral load at the time of delivery and MTCT rates <3 %, and identified no unique safety concerns regarding preterm delivery or low birth weight [
29]. However, the majority of patients included in this analysis lived in resource-limited regions, limiting the generalisability of the findings to populations living in other settings.
Using data from the NSHPC for pregnancies that were due to deliver between 2003 and 2012, the current study examined the use of LPV/r among pregnant women with HIV infection in the United Kingdom and Ireland to explore its association with maternal and foetal outcomes, including viral suppression, gestational age at delivery, prevalence of low birth weight at delivery, MTCT rates and reported adverse infant outcomes.
Methods
The NSHPC is a population-based surveillance study of HIV in women and children in the United Kingdom and Ireland that collects maternal, perinatal and paediatric data through 2 parallel, confidential and voluntary reporting schemes: an obstetric scheme for the notification of pregnancies in women with HIV infection and a paediatric scheme through which infants exposed to HIV and children who have HIV are reported; full details are published elsewhere [
4]. Pregnancies reported through the obstetric scheme were included in this retrospective analysis of individual patient data if the mother was identified as having a positive test for HIV infection by the time of delivery, had an expected date of delivery between 1 January 2003 and 31 December 2012 and received LPV/r at any time during her pregnancy. LPV/r was used in approximately 37 % of all pregnancies reported to the NSHPC in this period. Pregnancies that did not have a reported outcome by 30 June 2013 were excluded. Pregnancy was used as the unit of observation. HIV testing guidelines called for polymerase chain reaction (PCR) testing shortly after birth, and 2 weeks and 2 months after cessation of infant prophylaxis. Antibody testing at 18 months was also recommended.
Ethics approval
The National Study of HIV in Pregnancy and Childhood has London Multi-Centre Research Ethics Committee (NHS) approval (MREC/04/2/009). Individual patient consent is not sought or required for reporting to the NSHPC, which is an observational surveillance study subject to review by appropriate national information governance bodies (NIGB Ref: PIAG/BPSU 2–10(a)/2005, IG Toolkit reference: Version 12 2014/15 EE133902-FOPHS-PPAPPIASG; now under review with the Health Research Authority’s Confidentiality Advisory Group). This analysis was based on data already collected through the routine NSHPC study protocols and, per study approvals, individual patient consent or further permissions are not required. Furthermore, no disaggregated data were shared with anyone outside the NSHPC, and the report is totally anonymised.
Definitions
Maternal age was the mother’s age at conception. Trimesters were defined as follows: first, <13 completed gestational weeks; second, between 13 and 26 completed gestational weeks; and third, ≥27 completed gestational weeks. Low infant birth weight was defined as <2500 g.
Baseline CD4+ T-cell counts and viral loads (HIV RNA measurement reported as copies/mL) were the first reported measurements in the first or second trimester of pregnancy. CD4+ T-cell counts and viral loads at delivery were reported as the measurements taken closest to delivery during the third trimester or within 7 days after delivery. If delivery was preterm with no maternal viral load or CD4+ T-cell count reported in the third trimester, the last maternal measurement taken up to 28 days before and 7 days after delivery was used. Caesarean deliveries were classified as elective if they were performed before the initiation of labour and the rupture of membranes. A delivery was classified as preterm if it occurred before 37 completed gestational weeks. Infant HIV infection status was classified as uninfected or infected on the basis of reported PCR or HIV antibody results, or indeterminate for infants whose infection status had not yet been reported. Viral suppression was defined as an HIV RNA measurement <50 copies/mL.
Congenital abnormalities were classified according to the WHO International Classification of Diseases, 10th revision, using information provided by clinicians at infant notification or at follow-up.
Statistical analysis
Statistical analysis was performed using STATA v12.0 software (StataCorp, College Station, TX, USA), and standard descriptive statistics were used to summarise the data, with proportions, medians and interquartile ranges (IQRs) presented. Comparisons were assessed using a χ
2 test for categorical variables; 95 % confidence intervals (CIs) were calculated for rates.
Discussion
The effectiveness of LPV/r-containing regimens during pregnancy was analysed using observational data from comprehensive national surveillance of 4864 pregnancies with LPV/r exposure in 4118 women with HIV infection in the United Kingdom and Ireland over a 10-years period. The results show that LPV/r-containing regimens during pregnancy were effective in maintaining or increasing maternal CD4+ T-cell counts, inducing or maintaining suppressed maternal viral loads and substantially reducing the risk of MTCT.
The low MTCT rates and the declining trend over time observed in this analysis are comparable with findings from previous analyses of NSHPC data [
3,
4,
8], in which a steady decline in the MTCT rate over time has been documented, from a high of 18.5 % in 1990–1993 to 1.0 % in 2003–2006, and more recently to 0.46 % in 2010–2011. Similar improvements in MTCT rates have also been reported in other populations of women living in resource-rich areas during similar time periods [
5‐
7].
The declining MTCT rates across Europe partly reflect the earlier use of ART in pregnancy among women not on treatment at conception, together with the increasing proportion of women who are already on suppressive regimens at conception. In this analysis, approximately one-fifth of pregnancies were conceived whilst the woman was receiving LPV/r-based ART, with an undetectable viral load in early pregnancy in 82 % of women. This is consistent with an earlier finding from a European cohort (women delivering in 2000–2011 in 9 Western European countries) where 19 % of women on ART at conception had a viral load of >200 copies/mL [
30]. In the present analysis, practically all women with an undetectable viral load who conceived whilst on LPV/r-containing regimens maintained viral suppression throughout the pregnancy, whereas 69 % of women who initiated ART during pregnancy achieved an undetectable viral load by delivery.
In the absence of obstetric indications for a caesarean delivery, viral suppression before the end of pregnancy allows for a vaginal delivery [
12]. In this analysis spanning from 2003 to 2012, slightly more than one-third of pregnancies were delivered vaginally, despite nearly three-quarters having an undetectable viral load by delivery. This finding may partly reflect the study period: since 2005, BHIVA guidelines have included planned vaginal delivery for women with suppressed viral load at term as an option and have recommended this approach since 2012 [
12,
31]. However, recent findings from the NSHPC have also shown that women with undetectable viral loads delivering at maternity units with higher caseloads of women with HIV infection were significantly more likely to deliver vaginally than those delivering at less experienced units [
32]; similar regional differences have also been observed in France [
33].
Overall, nearly three-quarters of pregnancies were in women with delivery viral load below 50 copies/mL, and for most of the remainder, there was at least partial viral suppression (<1000 copies/mL). In France, approximately 78 % of women with HIV infection delivering in 1997–2010 had viral loads <400 copies/mL [
33], and 76 % of a Spanish cohort delivering in 2004–2007 had an undetectable viral load (≤50 copies/mL) [
5]. Factors associated with an increased probability of having a detectable viral load at delivery in a recent European pooled analysis included low CD4+ T-cell count (<200 cells/mm
3), young maternal age, HIV diagnosis in the third trimester or intrapartum and injection drug use [
15]. In the current analysis, among the small number of pregnancies with higher viral loads (≥1000 copies/mL) at delivery, ART was not initiated until the third trimester in 60 % of women compared with 18 % in the study population as a whole. MTCT rates were approximately 2 % for the mother-infant pairs where LPV/r therapy was not initiated until the third trimester, with no decline observed over time. Although only 14 % of mother-infant pairs overall were in this group, they accounted for 43 % of the MTCTs, highlighting the need to diagnose HIV infection and initiate ART earlier in some groups of women.
By 31 December 2014, 120 of the 517 previously indeterminate singletons were known to be uninfected, as well as another 7 twins or triplets. No new infections had been reported.
The current analysis of antenatal LPV/r exposure was associated with a similar incidence and rate of adverse pregnancy outcomes, including preterm delivery, low birth weights and incidence of congenital abnormalities as noted in other study populations. The congenital abnormality rate (2.9 %) in this infant population with antenatal exposure to LPV/r, regardless of overall exposure time, is similar to the rates previously reported for the entire NSHPC cohort and the uninfected general population from the region [
34,
35]. Analysis of data from the Antiretroviral Pregnancy Registry (APR) has also not identified any relationship between birth defects and LPV/r use in 2458 pregnancies with exposure to LPV/r [
36]. The rate of birth defects in the APR study population (2.2–2.4 %) was also comparable to the rate observed in the reference population.
Some studies have indicated an association between the use of protease inhibitor—containing ART during pregnancy and preterm delivery, whilst others have not [
18,
22‐
28]. The preterm delivery rate in our study (12.7 %) is somewhat lower than reported in some earlier analyses of other European observational cohorts, although a number of studies have noted declines in preterm delivery rates in more recent years [
15,
37]; these results may partly reflect the change in mode of delivery policy, leading to reduction of iatrogenic late preterm deliveries due to maternal HIV infection. The stillbirth rate here was 9.2 per 1000 infants, which is higher than the background rate in the United Kingdom (5.8 per 1000 infants in 2003 and 4.7 per 1000 infants in 2013) [
38,
39], but slightly lower than the previously reported rate for the NSHPC as a whole from 1990 to 2006 (11.0 per 1000 infants) [
4].
A strength of this study is that it includes data from a large unselected population of women and children with HIV infection collected through a long-term comprehensive active surveillance programme with high response rates. A potential limitation of the study, given that data are provided by antenatal care providers, is that early miscarriages and terminations are likely to be under-reported for women who did not receive care from an antenatal care clinic/provider early in their pregnancy. The difference in the proportion of miscarriages and terminations reported for women conceiving whilst on LPV/r and those initiating LPV/r in pregnancy is likely to be related to this limitation.
The low MTCT rates found in this study, which correlate with MTCT rates reported in other studies, suggest that the WHO goal of the elimination of vertical transmission of HIV in Europe [
40] is achievable and within target for populations from resource-rich areas, although countries from Eastern Europe and Central Asia face additional barriers [
9,
41]. Likewise, a review of 8 globally diverse studies that included a large proportion of patients from Africa noted MTCT rates of 0 to 2.8 % in women who received LPV/r therapy during pregnancy [
29].
Acknowledgements
Medical writing support was provided by John E. Fincke, PhD, and Lisa M. Havran, PhD, at Complete Publication Solutions, LLC (North Wales, PA); this support was funded by AbbVie. National surveillance of obstetric and paediatric HIV is undertaken through the NSHPC in collaboration with Public Health England and Health Protection Scotland. We also acknowledge the contribution of the midwives, obstetricians, genitourinary physicians, paediatricians, clinical nurse specialists and all other colleagues who report to the NSHPC through the British Paediatric Surveillance Unit of the Royal College of Paediatrics and Child Health, and the obstetric reporting scheme established under the auspices of the Royal College of Obstetricians and Gynaecologists. We thank Emmanouil Bagkeris and Heather Bailey who assisted with data extraction and analysis, and Helen Peters (Data Manager/Statistician), Icina Shakes and Kate Francis (Study Assistants) and Angela Jackson (Research Assistant) for their essential contributions to the NSHPC. We also acknowledge the invaluable contribution of our colleague and greatly missed friend Janet Masters, NSHPC data manager and co-ordinator throughout the period covered by this analysis, who died in December 2012. Tribute at
www.ucl.ac.uk/nshpc/jmasters.
Competing interests
Two of the authors are employees of AbbVie. As such, AbbVie played a role in the interpretation of data and reporting of the results. PAT and CT received research support from AbbVie Inc. to perform this analysis; they have also both received grants/research support from the European Union FP7 and the UK National Screening Committee. PAT has received grants/research support from the Health Protection Agency (UK) and the Welton Foundation. CT has received grants/research support from the Wellcome Trust, ViiV and UNICEF. JvW and MN are employees of AbbVie Inc. and may hold AbbVie stock or stock options.
Authors’ contributions
PAT, CT, JvW, and MN were responsible for conception and design of the study, interpretation of the data, were involved in drafting the manuscript and revising it critically for important intellectual content, have given final approval of the version to be published, and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. PAT and CT were also involved in the collection and analysis of the data. All authors read and approved the final manuscript.