SEARCH

SEARCH BY CITATION

Keywords:

  • Antiretroviral therapy;
  • HIV;
  • pregnancy outcome;
  • surveillance

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Contribution to authorship
  9. Ethics approval
  10. Acknowledgements
  11. References

Objective  To describe the changing demographic profile of diagnosed HIV-infected pregnant women over time and trends in pregnancy outcome, uptake of interventions and mother-to-child transmission.

Design  National surveillance study.

Setting  UK and Ireland.

Population  Diagnosed HIV-infected pregnant women, 1990–2006.

Methods  Active surveillance of obstetric and paediatric HIV conducted through the National Study of HIV in Pregnancy and Childhood.

Main outcome measures  Maternal characteristics, pregnancy outcome, use of antiretroviral therapy, mode of delivery and mother-to-child transmission.

Results  A total of 8327 pregnancies were reported, increasing from 82 in 1990 to 1394 in 2006, with an increasing proportion from areas outside London. Injecting drug use as the reported risk factor for maternal HIV acquisition declined from 49.2% (185/376) in 1990–1993 to 3.1% (125/4009) in 2004–2006 (P < 0.001), while the proportion of women born in sub-Saharan Africa increased from 43.5% (93/214) in 1990–1993 to 78.6% (3076/3912) in 2004–2006 (P < 0.004). Reported pregnancy terminations decreased from 29.6% (111/376) in 1990–1993 to 3.4% (135/4009) in 2004–2006 (P < 0.001). Most (56.4%, 3717/6593) deliveries were by elective caesarean section, with rates highest in 1999 (66.4%, 144/217). Vaginal deliveries increased from 16.6% (36/217) in 1999 to 28.3% (321/1136) in 2006 (P < 0.001). Use of antiretroviral therapy in pregnancy increased over time, reaching 98.4% (1092/1110) in 2006, and the overall mother-to-child transmission rate declined from 18.5% (35/189) in 1990–1993 to 1.0% (29/2832) in 2004–2006.

Conclusions  The annual number of reported pregnancies increased dramatically between 1990 and 2006, with changing demographic and geographic profiles and substantial changes in pregnancy management and outcome.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Contribution to authorship
  9. Ethics approval
  10. Acknowledgements
  11. References

Following the introduction of routine antenatal HIV testing throughout the UK and Ireland from 1999 onwards,1,2 the proportion of HIV-infected pregnant women diagnosed before delivery rose from around 60% in 1999 to over 90% since 2003.3,4 As a result of high uptake of interventions to reduce the risk of mother-to-child transmission (antiretroviral therapy, elective caesarean section delivery and avoidance of breastfeeding), transmission rates in many parts of Europe have declined to less than 2% in recent years.5–7 Meanwhile, the prevalence of HIV has continued to rise and more than doubled among women giving birth in England and Scotland from 0.09% to 0.23% between 2000 and 2006, following a more gradual increase during the 1990s.3,4

HIV infection in pregnant women and children in the UK and Ireland has been monitored continuously since the late 1980s through unique, population-based, active surveillance schemes, allowing the effects of changes in policy and practice to be observed on a national scale. Reports are sought for all pregnancies diagnosed up to the time of delivery, regardless of timing of booking for antenatal care or uptake of interventions. In this analysis, we explore the changing demographic profile of HIV-infected pregnant women over the course of the HIV epidemic, along with trends in pregnancy outcome, uptake of interventions and mother-to-child transmission rates.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Contribution to authorship
  9. Ethics approval
  10. Acknowledgements
  11. References

Surveillance of obstetric and paediatric HIV in the UK and Ireland is carried out through the National Study of HIV in Pregnancy and Childhood (NSHPC). Pregnancies in women with diagnosed HIV infection are notified through an active obstetric surveillance scheme, run under the auspices of the Royal College of Obstetricians and Gynaecologists. In a parallel reporting scheme, children with HIV infection and infants born to infected women are notified through the British Paediatric Surveillance Unit8 or directly to the NSHPC. Analyses were based on pregnancies between 1990 and 2006, in women diagnosed with HIV infection before delivery and reported to the NSHPC by June 2007.

Data collected included maternal demographic characteristics and clinical status, pregnancy outcome and delivery details, antiretroviral therapy and child’s infection status. Year of delivery was used for births and year of expected date of delivery for other outcomes. For some analyses, years were grouped into five time periods according to availability of interventions: 1990–1993 (pre-antiretroviral therapy), 1994–1996 (zidovudine monotherapy9), 1997–1999 (highly active antiretroviral therapy [HAART] and elective caesarean section10), 2000–2003 and 2004–2006 (HAART and routine antenatal screening1).

Exposure to antenatal antiretroviral therapy was categorised as untreated, monotherapy, dual therapy or HAART (three or more antiretroviral drugs) with or without a protease inhibitor (PI). Information on indication for treatment (whether antiretroviral therapy was prescribed only for prevention of mother-to-child transmission or also because the woman required it for her own health) was requested from 2006. Gestational age was measured in completed weeks. Miscarriage was defined as foetal death up to 23 weeks gestation and stillbirth as foetal death at ≥24 weeks gestation. Delivery at <37 weeks was defined as premature and at <32 weeks as very premature. Low birthweight was defined as <2500 g and very low birthweight as <1500 g. Maternal clinical status was classified as symptomatic if AIDS or HIV-related symptoms were reported during pregnancy.11 Maternal CD4 cell count and HIV plasma viral load were only routinely reported from 1998 onwards. CD4 count closest to delivery and viral load closest to delivery and up to 7 days postpartum were used in the analyses. Viral loads reported as ‘<400 copies/ml’ (n = 78) were reclassified as 200 copies/ml. Mode of delivery was classified as vaginal, elective caesarean section or emergency caesarean section, which included those carried out after rupture of membranes and/or onset of labour, as well as for other obstetric indications (e.g. pre-eclampsia and foetal distress). Detailed maternal information (timing of initiation of antiretroviral therapy, maternal clinical status, viral load and CD4 counts) was collected through the obstetric scheme and was therefore not available for births reported only through the paediatric scheme (∼11%).

For the purpose of these analyses, infants were classified as ‘uninfected’ if a negative polymerase chain reaction (PCR) test was reported after 1 month of age or a negative HIV antibody test after 18 months of age in non-breastfed infants, and ‘infected’ if a positive PCR was reported at any time or a positive antibody test after 18 months of age.

Reporting rates by geographical area were obtained by dividing the annual number of pregnancies in HIV-infected women reported in each country and English strategic health authority by the number of resident women aged 15–49 years, derived from 2000 census data for the UK12 and 2002 census data for Ireland.13 Maps were drawn using R, version 2.4.1.14

Data were managed in an Access 2002 database (Microsoft Corp., Redmond, WA, USA) and analysed using Stata 9.0 (Stata Corp., College Station, TX, USA). Means were compared using t tests and medians using Wilcoxon tests. Trends in medians were assessed using Cuzick’s nonparametric test for trend across ordered groups (function ‘nptrend’ in Stata).15 Logistic regression models were fitted to obtain odds ratios (ORs) and 95% confidence intervals (CIs). P values for trends were obtained using ordinary logistic regression models; those in Table 1 were adjusted for multiple comparisons using a Bonferroni correction.

Table 1.  Maternal characteristics and pregnancy outcome by time period for 8327 pregnancies
 1990–1993, n (%)1994–1996, n (%)1997–1999, n (%)2000–2003, n (%)2004–2006, n (%)Total, nP value (trend)
  • P values are for trends over time, comparing each category against all others combined. P values were obtained using logistic regression and adjusted for multiple comparisons using a Bonferroni correction.

  • *

    Includes 179 pregnancies due to end in 2004–2006 for which outcome was still pending as of June 2007.

  • **

    Includes women with missing exposure information.

  • ***

    Includes women whose partner acquired HIV through injecting drug use (n = 195).

  • ****

    Other ethnicities include black Caribbean, black other, Asian and mixed.

  • *****

    Six infants born in the Channel Islands.

Pregnancy outcome (n = 8327) 
Live birth235 (62.5)260 (80.0)456 (79.2)2623 (86.3)3405 (84.9)6979 
Stillbirth1 (0.3)2 (0.6)7 (1.2)24 (0.8)43 (1.1)77>0.50
Miscarriage15 (4.0)5 (1.5)32 (5.6)119 (3.9)158 (3.9)329>0.50
Termination of pregnancy111 (29.5)48 (14.8)70 (12.2)155 (5.1)135 (3.4)519<0.005
Other/outcome not known*14 (3.7)10 (3.1)11 (1.9)120 (3.9)268 (6.7)423<0.005
Exposure category (n = 8327) 
Other**191 (50.8)231 (71.1)485 (84.2)2873 (94.5)3884 (96.9)7664 
Injecting drug use associated***185 (49.2)94 (28.9)91 (15.8)168 (5.5)125 (3.1)663<0.001
Region of birth (n = 8009) 
UK/Ireland105 (49.1)101 (32.8)143 (25.0)450 (15.0)496 (12.7)1295<0.004
Europe (excluding UK/Ireland)9 (4.2)13 (4.2)21 (3.7)69 (2.3)121 (3.1)233>0.50
Sub-Saharan Africa93 (43.5)179 (58.1)381 (66.6)2309 (76.9)3076 (78.6)6038<0.004
Elsewhere7 (3.3)15 (4.9)27 (4.7)175 (5.8)219 (5.6)4430.016
Ethnic origin (n = 8157) 
White167 (57.6)127 (39.3)150 (26.1)427 (14.1)492 (12.5)1363<0.003
Black African115 (39.7)183 (56.7)388 (67.5)2357 (78.0)3140 (79.5)6183<0.003
Other****8 (2.8)13 (4.0)37 (6.4)236 (7.8)317 (8.0)611<0.003
Age at delivery (n = 8244) 
14–24 years107 (29.9)63 (19.9)101 (17.7)557 (18.5)704 (17.7)1532<0.003
25–34 years235 (65.6)234 (73.8)379 (66.4)1940 (64.3)2463 (61.9)5251<0.003
35–46 years16 (4.5)20 (6.3)91 (15.9)519 (17.2)815 (20.5)1461<0.003
Clinical status (n = 7253) 
No HIV-related symptoms258 (78.2)210 (74.7)428 (80.6)2413 (89.0)3026 (89.0)6335 
HIV-related symptoms/AIDS72 (21.8)71 (25.3)103 (19.4)299 (11.0)373 (11.0)918<0.001
Region of pregnancy report (n = 8321*****) 
London191 (50.8)210 (65.0)437 (76.0)1728 (56.9)1803 (45.0)4369 
England outside London65 (17.3)51 (15.8)72 (12.5)823 (27.1)1711 (42.7)2722 
Scotland, Wales, Northern Ireland91 (24.2)49 (15.2)30 (5.2)90 (3.0)166 (4.1)426 
Ireland29 (7.7)13 (4.0)36 (6.3)397 (13.1)329 (8.2)804 
Total (n = 8327)376325576304140098327 

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Contribution to authorship
  9. Ethics approval
  10. Acknowledgements
  11. References

All pregnancies

There were 8327 pregnancies in 6788 HIV-infected women; 5540 women had one pregnancy reported during the study period, 1010 two, and 238 three or more. The annual number of reported pregnancies increased 17-fold, from 82 in 1990 to 1394 in 2006, the steepest rise occurring between 1999 and 2003 (Figure 1). There were 116 multiple births; in one twin pair, one infant was stillborn, and there were two twin stillbirths (five stillborn infants altogether). Pregnancy outcomes are shown in Table 1. The 423 pregnancies with other or unknown outcome included two maternal deaths (one suicide and one death from HIV encephalopathy, both in the mid-1990s) and 23 ectopic pregnancies (all ending before 13 weeks gestation); 110 women were known to have left the British Isles before delivery and another 109 were otherwise lost to follow up; information on pregnancy outcome was still pending for 179 pregnancies due to end in 2004–2006.

image

Figure 1. Number of pregnancy reports by timing of maternal diagnosis (before or during pregnancy) and year, 1990–2006. EDD, expected date of delivery. Note: Prevalence of HIV in women giving birth in England and Scotland increased from 0.09% in 2000 to 0.23% in 2006.4

Download figure to PowerPoint

Maternal characteristics

There were marked changes over time in the demographic profile of the women (Table 1). The proportion of women reported to have probably acquired HIV through injecting drug use or from a drug-using partner declined 10-fold, from around half in 1990–1993 to less than 5% in 2004–2006 (Table 1). Meanwhile, the proportion of women born in sub-Saharan Africa increased from under half in 1990–1993 to almost 80% in 2004–2006, with a corresponding change in maternal ethnic origin (Table 1). The proportion of women born in Asia rose from 0% (0/214) in 1990–1993 to 2.2% (87/3912) in 2004–2006 (trend: P = 0.003). As a proportion of all women with either country of birth or ethnicity reported, those born in the Caribbean or of black Caribbean ethnicity increased from 1.4% (4/293) in 1990–1993 to 3.7% (147/3989) in 2004–2006 (trend: P = 0.051). Among women born abroad, information on date of arrival in the British Isles was only available for 55.6% (3735/6714); among these, median time between arrival and delivery was 3.3 years (interquartile range [IQR]: 1.4–6.1 years). Seven women (four born in the British Isles), aged between 15 and 20 years at delivery, were known to have acquired infection vertically from their own mothers.

Median maternal age at delivery increased over time, from 27.2 years (IQR: 24.4–30.1 years) in 1990–1993 to 30.2 years (IQR: 26.4–34.0 years) in 2004–2006 (trend, P < 0.001). There was no evidence of an increase in the proportion of teenage pregnancies (age 14–19 years) over time: 2.8% (10/358) of pregnancies were in teenagers in 1990–1993, 3.9% (22/571) in 1997–1999 and 3.4% (134/3982) in 2004–2006 (trend: P = 0.276).

Up to 1999, the majority of pregnancies were in women who already knew their HIV status before they became pregnant (67.0%, 856/1277) (Figure 1). As routine antenatal screening was introduced throughout the British Isles, the proportion of women diagnosed during pregnancy rose to 60.1% (1566/2606) between 2001 and 2003 and then declined subsequently (Figure 1). Among women who knew their HIV status at conception (and with information on where their diagnosis was made), the proportion diagnosed in a previous pregnancy doubled from 18.0% (24/133) in 2000 to 36.1% (181/502) in 2006 (trend: P < 0.001). Overall, only 2.9% (120/4112) of antenatal diagnoses were made in the last 2 weeks of pregnancy.

The prevalence of AIDS or HIV-related symptoms in pregnancy decreased over time, from over 20% in the early 1990s to 11% since 2000 (P < 0.001) (Table 1).

Geographic origin of reports

There were also changes in the geographic origin of reports. The proportion of pregnancies reported from London increased significantly from about 50% in 1990–1993 to 76% in 1997–1999 (trend: P < 0.001) (Table 1). Subsequently, although the number of reports from London continued to rise until 2004, when they stabilised, the proportion reported from elsewhere in England increased significantly from 12.5% in 1997–1999 to 42.7% in 2004–2006 (trend: P < 0.001) (Table 1). Figure 2 shows annual reporting rates (number of reports per million women aged 15–49 years) across the UK and Ireland over time. By 2004–2006, rates in most areas of the UK were similar to those in London in 1990–1999. In Ireland, reporting rates stabilised following a substantial rise in the number of reports from only 13 in 1999 to almost 150 in 2003.

image

Figure 2. Pregnancy reporting rates in the UK and Ireland by time period and country/region.

Download figure to PowerPoint

Terminations

The proportion of reported pregnancies ending in termination decreased 10-fold from around 30% in 1990–1993 to 3% in 2004–2006 (Table 1). Termination was 2.6 times more common in women diagnosed with HIV before pregnancy than in those diagnosed antenatally (OR = 2.65, 95% CI: 2.18–3.23, P < 0.001), but similar declines occurred in both groups (34.7% in 1990–1993 to 4.1% in 2004–2006 for diagnosis before pregnancy, trend: P < 0.001; 20.2% to 2.7% for diagnosis during pregnancy, trend: P < 0.001). There was no association between HIV exposure category or maternal age and termination, after adjusting for year and timing of diagnosis (P > 0.50 in both cases). Median gestational age at termination was 10 weeks (IQR: 9–13 weeks) for women diagnosed before pregnancy and 15 weeks (IQR: 12–19 weeks) for those diagnosed during pregnancy (P < 0.001). About 30% (101/335) of previously diagnosed women having terminations were taking antiretroviral therapy at conception. Three terminations were carried out after 24 weeks: one at 25 weeks for major congenital heart defects, one at 30 weeks for anencephaly and one at 29 weeks because of concerns about the woman’s mental health.

Miscarriages

Four percent (329/8327) of reported pregnancies resulted in a miscarriage; 80.9% (266/329) occurred before 20 weeks gestation, 10.3% (34/329) at 20–21 weeks and 8.8% (29/329) at 22–23 weeks. The overall miscarriage rate remained constant over time (trend: P = 0.951), as did the rate of late miscarriages (≥20 weeks) (trend: P = 0.890).

Live births and stillbirths

The following results refer to the 85% of pregnancies (7056/8327) resulting in a live birth or stillbirth.

Antiretroviral therapy

Changes in the use of antiretroviral therapy over time are shown in Figure 3. Following the introduction of zidovudine for the prevention of mother-to-child transmission in 1994,9 uptake of antiretroviral therapy (mostly monotherapy) increased rapidly. Although monotherapy has continued to be offered in some circumstances, it was overtaken by HAART in the late 1990s; by 2006, 98.4% (1092/1110) of diagnosed pregnant women received antiretroviral therapy at some time in pregnancy, 94.5% (1032/1092) of whom received HAART and 4.4% (48/1092) monotherapy.

image

Figure 3. Use of antiretroviral therapy and mother-to-child transmission rates (with 95% CI) by year of delivery, 1990–2006. HAART, highly active antiretroviral therapy; MTCT, mother-to-child transmission; PI, protease inhibitor.

Download figure to PowerPoint

Type of HAART regimens changed over time, with an increase in the use of PIs among HAART-treated women from 36.1% (43/119) in 1999 to 73.2% (755/1032) in 2006 (P < 0.001) (Figure 3). Concerns about toxicity of nevirapine (a non-nucleoside reverse transcriptase inhibitor) in women with high CD4 counts16 may have led to women with less advanced disease being prescribed PI-containing regimens. Evidence supporting this includes higher median CD4 counts in women on PIs in 2006 than those on other HAART regimens (420 cells/mm3 versus 330 cells/mm3, P < 0.001); women on PIs were also more likely to have been diagnosed during pregnancy (48.1%, 363/755, versus 28.9%, 80/277, P < 0.001) than before.

Among pregnancies in previously diagnosed women, timing of initiation of antiretroviral therapy (before or during pregnancy) was available for 87.3% (2942/3369); once combination therapy became widely available, the proportion on treatment at conception rose rapidly and then stabilised at around 45% (1188/2544) between 1999 and 2006 (trend: P = 0.936). In 2006, 39.9% (125/313) of previously diagnosed women on HAART were reported not to require treatment for their own health and only to be taking it for prevention of mother-to-child transmission; in contrast, among those diagnosed during pregnancy and taking HAART, 69.2% (155/224) were doing so only to prevent mother-to-child transmission.

CD4 counts and viral load

CD4 count in pregnancy was reported for 75.9% (4901/6459) of births between 1998 and 2006. Median CD4 count increased significantly, from 315 cells/mm3 (IQR: 225–425 cells/mm3) in 1998 to 395 cells/mm3 (IQR: 270–570 cells/mm3) in 2006 (trend, P < 0.001). Viral load was reported for 78.8% (5092/6459) of births between 1998 and 2006: the proportion with viral load <50 copies/ml increased from 29.0% (27/93) in 1998 to 69.0% (609/882) in 2006 (trend: P < 0.001), while median viral load in the remainder decreased from 4494 copies/ml (IQR: 1423–13 400 copies/ml) to 425 copies/ml (IQR: 131–5507 copies/ml) in 2006 (trend: P < 0.001).

Mode of delivery

Between 1995 and 2006, most deliveries were by elective caesarean section (56.4%, 3717/6593), increasing from 38.3% (36/94) in 1997 to a high of 66.4% (144/217) in 1999 (trend: P < 0.001) (Figure 4). Subsequently, the proportion of elective caesarean sections declined to around 50% (555/1136) in 2006, as vaginal deliveries increased from 16.6% (36/217) to 28.3% (321/1136) between 1999 and 2006 (trend: P < 0.001), mostly due to a rise in planned vaginal deliveries (Figure 4). The proportion of emergency caesarean sections increased from 17.1% (37/217) in 1999 to 22.9% (260/1136) in 2006 (trend: P = 0.001); among these, the proportion occurring at 37 weeks gestation or later increased from 48.6% (18/37) in 1999 to 60.6% (155/256) in 2006 (trend: P = 0.021).

image

Figure 4. Mode of delivery by year, 1995–2006.

Download figure to PowerPoint

Gestational age, birthweight, stillbirth and neonatal death

Median gestational age was 38 weeks (IQR: 38–39 weeks); 14.2% (975/6874) of deliveries were premature (<37 weeks) and 3.6% very premature (<32 weeks). Median birthweight was 3050 g (IQR: 2720–3370 g); 14.1% (827/5865) of infants were of low birthweight (<2500 g) and 3.1% (179/5865) of very low birthweight (<1500 g). There were no significant trends over time in the proportion of infants who were premature (P = 0.564), very premature (P = 0.182), of low birthweight (P = 0.197) or of very low birthweight (P = 0.340). There were 77 stillbirths (1.1%) and 30 neonatal deaths (0.4% of live births); no trends over time were detected (P = 0.304 and P = 0.460, respectively).

Mother-to-child transmission

Infection status was reported for 86.9% (6063/6979) of live births (twins were counted as one birth, and none were discordant). The overall mother-to-child transmission rate declined significantly from 18.5% (35/189, 95% CI: 13.3–24.8%) in 1990–1993 to 1.0% (29/2832, 95% CI: 0.7–1.5) in 2004–2006 (Figure 3).

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Contribution to authorship
  9. Ethics approval
  10. Acknowledgements
  11. References

The annual number of reported pregnancies in HIV-infected women increased substantially between 1990 and 2006, the sharpest rise occurring between 1999 and 2003, when universal screening strategies were adopted throughout the UK and Ireland2,17 and during which time HIV prevalence rose in women giving birth.4 Reductions in mother-to-child transmission rates5,7 and in overall HIV-associated mortality and morbidity following the introduction of HAART18 have contributed to an increase in birth rates in HIV-infected women.19,20 In this population, we observed an increasing number of pregnancies in previously diagnosed women and overall decline in pregnancy terminations. The lower termination rate in women diagnosed during pregnancy, compared with those diagnosed previously, could be due to differential reporting of terminations between the two groups but could also reflect a lack of opportunity for termination among women diagnosed later in pregnancy, as suggested by the later median gestational age at termination in this group. We did not detect an increase in the rate of miscarriage over time. Miscarriages occurring before the first antenatal appointment and terminations carried out in specialist clinics are likely to be underreported because the obstetric surveillance system is focused on women attending for antenatal care; nevertheless, these data provide an indication of trends over time.

In the early 1990s, the epidemic in pregnant women was driven mainly by injecting drug use, while in later years most women originated from areas with generalised HIV epidemics, particularly sub-Saharan Africa. This pattern has also been observed for the UK HIV epidemic more widely.4 Pregnancies in young women who acquired HIV vertically from their own mothers were reported; this group is likely to increase in the future as girls infected perinatally in the early years of the HIV epidemic reach child-bearing age.21 Furthermore, an increasing proportion of infected children are now likely to survive into adulthood as a result of advances in the management of paediatric HIV.22

There were also changes in the geographic origin of reports within the British Isles. Although reporting rates were highest in London, the proportion of reports from elsewhere increased markedly. This reflects both differences in overall prevalence (e.g. in 2006, an estimated 0.42% of women giving birth in London were HIV infected, compared with 0.14% in the rest of England4) and also differences in antenatal detection rates over time. Routine antenatal testing was introduced earlier in London than elsewhere, and uptake was initially higher.17 Information on asylum status is not collected in this study, but the dispersal of asylum seekers from the South East of England23 could also have contributed to the increase in reports from outside London.4,24 Concerns about the impact of this dispersal policy on continuity of HIV care have been voiced.25,26 These findings highlight the need for a wide range of HIV specialist services throughout the UK and Ireland, particularly in areas where HIV-infected pregnant women were not previously seen and specialist care may not be readily available.

Although the use of antenatal zidovudine monotherapy declined substantially following the introduction of combination therapy, it is still used for some women who do not require HAART for their own health, in accordance with the British HIV Association guidelines.27 In recent years, most women received HAART in pregnancy, and in 2006, about half were reported to require it for their own health. Among previously diagnosed women, about 45% were already on therapy when they became pregnant.

Women diagnosed in pregnancy and those with higher CD4 counts were more likely to be on PIs than those diagnosed before pregnancy and with lower CD4 counts, suggesting that concerns about nevirapine hepatotoxicity in women with CD4 counts >350 cells/mm3 contributed to the rise in PI-based HAART from 2003 onwards.16 The trend towards higher CD4 counts and lower viral loads is likely to reflect not only the beneficial effect of antiretroviral therapy, but also earlier diagnosis of HIV infection resulting from antenatal screening.

As we have previously reported,28 prematurity rates in women on HAART in this population are about 1.5 times (adjusted OR = 1.51, 95% CI: 1.19–1.93) higher than in those on monotherapy or dual therapy. However, this has not led to an overall significant increase in prematurity rates among diagnosed women over time, probably because of the changing demography and better health of the diagnosed population.

The increase in vaginal deliveries in recent years occurred in the context of guidelines suggesting that women on HAART who achieve viral suppression can opt for a vaginal delivery.27 This recommendation may also have contributed to the increase in emergency caesarean sections, resulting from unexpected complications arising during planned vaginal deliveries. The increase in the proportion of emergency procedures being carried out at term supports this interpretation. Although planned caesarean section deliveries usually take place when HIV specialist midwives and obstetricians are on duty, spontaneous vaginal deliveries and emergency caesarean sections may occur unexpectedly, with staff less prepared for dealing with HIV-associated deliveries. While the importance of providing HIV-infected women with a range of treatment and delivery options should not be underestimated, the impact of such policies should be monitored closely, both locally and nationally. Nonetheless, overall mother-to-child transmission rates have remained consistently below 2% since 1999, a remarkable achievement given the diverse and unselected nature of this population. A detailed analysis of factors associated with transmission of infection in the HAART era is reported separately.29

This national surveillance study, which relies on the cooperation of a large number of health professionals, is an active reporting system with high response rates (>90%).30 Comparison of the number of pregnancies reported to the NSHPC with national unlinked anonymous seroprevalence data suggests that case ascertainment for pregnancies ending in a live birth is extremely high.4

This report describes the changing epidemiology of HIV in pregnancy in the UK and Ireland and highlights the success of routine antenatal HIV screening policies, both through increased detection of HIV in pregnancy and high uptake of antiretroviral therapy, leading to very low mother-to-child transmission rates. Ongoing established surveillance makes it possible to assess the impact of changes in guidelines and clinical practice on the uptake and effectiveness of interventions and provides timely evidence to help optimise obstetric and perinatal management and outcomes in this diverse population.

Funding

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Contribution to authorship
  9. Ethics approval
  10. Acknowledgements
  11. References

The NSHPC was previously funded by AVERT and the Department of Health and is currently funded by the Health Protection Agency. C.L.T. is funded by a UK Medical Research Council Training Fellowship (ref: G0501895). This work was undertaken at the UCL Institute of Child Health which received a proportion of funding from the Department of Health’s NIHR Biomedical Research Centres funding scheme.

Contribution to authorship

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Contribution to authorship
  9. Ethics approval
  10. Acknowledgements
  11. References

C.L.T. and P.A.T. participated in the data collection and drafted the paper. C.L.T. carried out the statistical analyses with support from M.C.-B. All authors participated in developing the concept of the paper and interpreting the results. All authors commented on all drafts of the paper and approved the final version. P.A.T. is responsible for the NSHPC and is the guarantor.

Ethics approval

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Contribution to authorship
  9. Ethics approval
  10. Acknowledgements
  11. References

Ethics approval for the NSHPC was renewed following review by the London Multi-Centre Research Ethics Committee in 2004 (ref. MREC/04/2/009).

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Contribution to authorship
  9. Ethics approval
  10. Acknowledgements
  11. References

National surveillance of obstetric and paediatric HIV is undertaken through the NSHPC in collaboration with the Health Protection Agency Centre for Infections and Health Protection Scotland. We gratefully 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 run under the auspices of the Royal College of Obstetricians and Gynaecologists. We thank Janet Masters who co-ordinates the study and manages the data, Icina Shakes for administrative support, and Marie-Louise Newell, Claire Hankin and Barbara Willey for their helpful comments on drafts of this paper.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Methods
  5. Results
  6. Discussion
  7. Funding
  8. Contribution to authorship
  9. Ethics approval
  10. Acknowledgements
  11. References