Prevention of human immunodeficiency virus mother–to

Abstract

The objective of the study was to investigate the HIV-mother-to-child transmission (MTCT) rate in Israel. This was a retrospective study of HIV-infected pregnant women, mainly immigrants from Ethiopia, in six Israeli AIDS centres, in 2000–2005. Medical records of mothers and newborns were evaluated for HIV status, treatment and MTCT rates. Three hundred pregnancies of 241 HIV-infected women, resulting in 304 live births, were studied. In 86/241(36%) women, HIV diagnosis was made during the current pregnancy or shortly after labour. Thirty others were diagnosed during previous pregnancies. Highly active antiretroviral therapy (HAART) was prescribed in 76% of pregnancies. The mean viral load before labour was 23,000 ± 100,000 copies/mL with a mean CD4 of 406 ± 223 (range 4–1277) cells/mm3. Caesarian sections were preformed in 175/300 pregnancies (103/175 with viral load <1000 copies/mL). During labour, azidothymidine (AZT) was given to 80% and nevirapine to 8% of the women. Eighty-eight percent of the neonates received AZT for six weeks. The overall HIV-MTCT rate was 3.6%. MTCT correlated significantly with delayed HIV diagnosis, low CD4, lack of HAART during pregnancy and lack of perinatal treatment. HIV treatment of mothers and their newborns throughout pregnancy, labour and perinatal period are crucial for effective prevention of MTCT, emphasizing the need for early HIV screening, diagnosis and treatment.

Keywords

HIV AIDS mother-to-child transmission highly active antiretroviral therapy Ethiopia Israel

Introduction

Mother-to-child transmission of the human immunodeficiency virus (HIV-MTCT) is the cause of most paediatric HIV infections worldwide. It may reach 40% or more without intervention.^1–5 In the era of highly active antiretroviral therapy (HAART), HIV-MTCT has been dramatically decreased to less than 2% and virtual elimination of HIV in infants by 2010 has been set as a goal for the European region.^6,7 However, in Africa and in other undeveloped countries, HIV vertical transmission is still a major medical problem.⁸

HIV vertical transmission can take place in utero, but mostly occurs during delivery or postnatally through breastfeeding. Efforts to reduce HIV vertical transmissions have, therefore, included specific interventions during pregnancy, labour and the postnatal period.⁵ HIV-MTCT prevention programmes should include three parts: (1) HAART given to all women regardless of their HIV status during pregnancy as well as treatment during and after labour to all mothers and neonates; (2) Elective caesarean section for specific indications (e.g. viral load >1000 copies/mL); (3) Refraining from breastfeeding.^9–11 The implantation of such a programme, including HIV screening during pregnancy for an early diagnosis differs between countries and among populations within the same country.

In Israel, the vast majority of HIV-infected females are immigrant Jews from Africa (Ethiopia). A smaller group immigrated from east Europe, mainly the former USSR. HAART as well as all other medical interventions, including feeding formula for neonates, are available in Israel with no charge for all HIV-infected patients. Therefore, AIDS clinics in Israel are unique for treating African patients with HIV (subtype C) with all western medical facilities.

In the present study, we evaluated 300 pregnancies of HIV-infected women from six Israeli AIDS centres, in the HAART era (2000–2005). Our objectives were to determine the MTCT rate among our patients (mainly Ethiopian immigrants), the predictors for HIV vertical transmission and the implantation of the HIV-MTCT prevention programme in Israel.

Methods

Study participants

From January 2000 to October 2005 all HIV-infected pregnant women and their infants in six Israeli AIDS centres were retrospectively evaluated. Mothers’ files were studied for mode and time of HIV diagnosis, HIV status during pregnancy and labour, treatment during pregnancy, mode of delivery and MTCT rates. Infants’ files were evaluated for postpartum therapy, breastfeeding and outcome.

Procedures

Plasma HIV viral load was determined by a commercial kit (COBAS Ampliprep/Amplicor version 1.5, standard format; Roche Molecular System, Branchburg, NJ, USA). Infants’ cellular HIV-1 DNA was quantitated by home-made polymerase chain reaction (Central Virology Laboratory, Public Health Services, Ministry of Health, Sheba Medical Center, Tel Hashomer, Israel). All infants were evaluated at day 2 and thereafter at months 1 and 4. HIV– enzyme-linked immunosorbent assay testing was performed at the age of 18 months.

CD4 cells were counted by fluorescence activated cell sorting using fluorescein isothiocynate CD4 monoclonal Abs (IQ Products, Groningen, The Netherlands).

Statistical analysis

For uni-variables comparison we used the Student's t-test for quantitative variables, and the chi-square test or Fisher's exact test for qualitative variables. Data are presented as mean ± standard deviation. P ≤ 0.05 was considered significant.

Results

Mothers’ characteristics

Between January 2000 and October 2005, 300 pregnancies carried by 241 HIV-infected women in six Israeli AIDS centres were studied retrospectively. Maternal characteristic are summarized in Table 1. During the study period, 188 women delivered once, 48 delivered twice, four delivered three times and one woman delivered four times. Thirteen of our women had previously given birth to an HIV-infected child, and another four had a child who died in infancy with suspected AIDS. HIV diagnosis was made during current pregnancy in 86 (36%) out of 241 women (70 during pregnancy and 16 during or shortly after labour), 30 other women were diagnosed with HIV during previous pregnancy (before January 2000). Thus, 116 (48%) women in our cohort were diagnosed with HIV during or immediately after pregnancy. HIV diagnosis of 125 (52%) women was made 5 + 3.22 (range 1–14) years before current pregnancy, 68 of them received HAART before pregnancy for HIV-related (not pregnancy) indications.

Table 1

Demographic characteristics of 241 HIV-infected pregnant women

Number of pregnancies	300
Age (range) years	30±6 (17–46)
African origin (Ethiopian)	207 (86%)
Mode of HIV acquisition	Heterosexual: 229 (96%)
	IV DU: 10 (4%)
	Blood products: 1
First pregnancy	41 (17%)
Mean number of children per mother (range) before current pregnancy	2.3±1.6 (0–9)
Time of delivery (gestation weeks)	38±2

IV DU = intravenous drug users

Treatment during pregnancy and maternal adverse events

HAART was prescribed in 229 (76%) of 300 pregnancies. In 100 pregnancies therapy was required only for MTCT prevention and was stopped after delivery. Treatment with HAART was recommended to 197/229 (86%), with dual therapy (azidothymidine [AZT] and lamivudin) to 20/229 (9%) and with monotherapy (AZT alone) to 12/229 (5%). Severe adverse events related to ART (requiring hospitalization or change in protocol) were documented in 14 (6%) of all treated women. Adverse events included diarrhoea (six cases), anaemia related to AZT (three cases), diabetes mellitus (two cases), nevirapine-associated rash (one case), Steven-Jhonson syndrome (one case) and liver function tests disturbances (two cases). All events were resolved with drug change and no deaths were reported. In 71 (24%) pregnancies no antiretroviral (ARV) were prescribed because of late diagnosis, lack of compliance or patients refusal. In all 45/71 (63%) of these women were diagnosed with HIV during or after current pregnancy (16 of them during labour).

The mean HIV-viral load before labour in the entire cohort was 23,000±100,000 copies/mL (Log 4.3±5; range: <400– 10⁶). In 162 pregnancies, viral load was below level of detection (<400 copies/mL). Undetectable viral loads were more common, though not statistically significant, among patients who were treated with HAART rather than with mono or dual therapy (142/197 (72%) versus 20/32 (62%); respectively). The mean CD4 count before labour was 406±223 (range 4–1277) cells/mm³. In 42 pregnancies, the CD4 count was below <200 cells/mm³.

Mode, time of delivery and treatment during labour

The median time of delivery was 38±2.2 (range 24–42) weeks of gestation. Nineteen (6.3%) pregnancies ended prematurely (weeks 24–35). During labour, AZT (IV) was given to 242 (80%) of our pregnant women and an additional single dose of nevirapine to 25 (8%). Caesarian sections (electives and non-electives) were performed in 175 (58%) of 300 pregnancies, HIV viral load below 1000 copies/mL was observed in 103 (59%) of them. Only minor postoperative complications such as minor wound infection and minor bleeding were documented. One hundred and twenty-five (42%) deliveries were vaginal, HIV viral load was below 1000 copies/mL in 74 (59%) of them.

Neonatal treatment and outcome

The 300 pregnancies reported here, resulted in 304 live births and two still births. Of 304 neonates 269 (88%) received AZT for six weeks and 48 (16%) received an additional single dose of nevirapine. As many as 16 (5.3%) neonates were breastfeed. In our cohort, 37 (12%) neonates required treatment in the neonatal intensive care unit. Perinatal deaths (still births and deaths within the first 28 days of live) were documented in six (2%) neonates, and three (1%) other infants had died within the first eight months of life (only one of them was infected with HIV). Fifteen (5%) babies were born with birth defects, six cardiac, two neurological, one congenital diaphragmatic hernia and six others.

HIV-MTCT rate

The rate of HIV-MTCT in the entire cohort was 3.6% (11 out of 300 pregnancies). Eight infected infants were born to mothers of African origin. Vertical transmission decreased from 10 of 162 pregnancies between the years 2000–2002 (MTCT rate of 6.1%) to one of 138 pregnancies in 2003–2005 (MTCT rate of 0.7%; P = 0.01). HIV-MTCT rate correlated with the time of HIV diagnosis of the mothers. Thus, HIV vertical transmission was observed in 9.3% (8/86) of women diagnosed with HIV during current pregnancy/labour compared with 1.4% (3/214) of transmission in women diagnosed with HIV in the prepregnancy period (P = 0.0009).

Table 2 demonstrates the risk factors for HIV-MTCT in the 11 HIV-infected neonates. As can be seen in the table, in addition to late HIV diagnosis mother's younger age (P = 0.03), the absence of HAART during pregnancy (P < 0.0001) and high viral load during labour (P < 0.0001) are significant risk factors for HIV-MTCT. CD4 cells count at labour was not significantly associated with HIV-MTCT, although low CD4 counts <200 cells/mm³ correlated with HIV-MTCT (P = 0.05). Perinatal interventions were negatively associated with HIV-MTCT. Thus, therapy with AZT (IV) in labour (P < 0.001), neonatal treatment with syrup AZT for six weeks (P = 0.0004) and the absence of breastfeeding (P < 0.001) significantly correlated with the absence of HIV-MTCT (Table 2). The HIV-MTCT rate in our entire cohort, among babies delivered by vaginal delivery was insignificantly higher than in those delivered by caesarean sections (5.6% versus 2.2%; Table 3). In our cohort there were 123 deliveries of mothers with viral load >1000 copies/mL at labour, 51 delivered vaginally and 72 underwent caesarean section. A higher though not statistically significant HIV-MTCT rate was observed with vaginal deliveries 7/51 (13.7%) compared with caesarean sections 4/72 (5.5%). In deliveries of mothers with viral load <1000 copies/mL, no differences in HIV-MTCT rates were observed regarding the mode of delivery. The HIV-MTCT for vaginal deliveries with viral load >1000 copies/mL was significantly higher (P = 0.001) than for vaginal deliveries with viral load of <1000 copies/mL (13.7% and 0%; respectively; Table 3).

Table 2

Characteristics of mothers and infants, according to HIV-MTCT status

	Transmitters	Non-transmitters	P
Number	11	289
Mothers’ age (years)	26±5.5	30±5.8	0.03
Mothers’ years from HIV diagnosis	0.5±1.5	3.5±3.5	0.006
HIV diagnosis in current pregnancy or labour	8 (73%)	78 (27%)	0.002
HIV diagnosis in labour	3 (27%)	13 (4.5%)	0.01
Absence of HAART during pregnancy	10 (90%)	61 (21%)	<0.0001
CD4 cells/mm³ (in labour)	353±389	407±216	NS (0.47)
CD4 <200 cells/mm³ (in labour)	4 (36%)	38 (13%)	P = 0.05
Viral load copies/mL (in labour)	165,000±197,000	23,000±99,000	<0.0001
IV AZT (in labour)	3 (27%)	239 (82%)	<0.0001
Neonatal therapy with AZT	5 (45%)	262 (91%)	0.0004
Infants breast fed	6 (54%)	10 (3.4%)	<0.0001

NS = not significant; IV AZT = intravenous azidothymidine; MTCT = mother-to-child transmission; HAART = highly active antiretroviral therapy

Table 3

HIV-MTCT and mode of delivery

Mode of delivery	Viral load at delivery	Number of pregnancies	Number of transmissions
Caesarian section^*	<1000 copies/mL	103	None
Caesarian section^*	<1000 copies/mL	72	4 (5.5%)
Vaginal delivery^†	<1000 copies/mL	74	None
Vaginal delivery^†	>.1000 copies/mL	51	7 (13.7%)

P= not significant for caesarean sections above versus below viral load of 1000 copies/mL

†

P = 0.001 for vaginal deliveries with viral load above versus below 1000copies/mL MTCT= mother-to-child transmission

Discussion

The implantation of HIV-MTCT prevention programme in Israel is quiet effective since a relative low rate (3.6%) of HIV-MTCT was observed in our cohort. Moreover, the rate of HIV-MTCT has dramatically decreased (to 0.7%) in the last three years of our retrospective study. Early HIV diagnosis is a prerequisite for the application of all interventions to reduce HIV-MTCT. Therefore, it is a crucial component of any prevention programme. In our study as much as 48% of the women were diagnosed with HIV during pregnancy. Moreover, three of 11 children were vertically infected by women who were not diagnosed with HIV before delivery, emphasizing the importance of antenatal testing even in a country with a low HIV prevalence like Israel. Routine voluntary HIV-antibody testing of pregnant women on an opting-out basis, regardless of risk factors or ethnic origin, has been recommended by most guidelines and was implanted by many countries with a high acceptance rate.^12,13 We believe that offering HIV-antibody testing to all pregnant women in Israel will enable us to identify all infected women during early stages or pregnancy without stigmatizing women of high-risk groups.

Numerous other risk factors for HIV-MTCT have been documented in the past,^11,13,14 and several different interventions to prevent HIV-MTCT have been suggested with efficacy demonstrated for HAART prophylaxis, avoidance of breastfeeding and elective caesarean section.¹⁴ In our cohort, the absence of any ART during pregnancy, high HIV viral load and severe immune deficiency in labour correlated with HIV-MTCT, whereas the treatment of mothers with intravenous AZT during labour and of neonates with syrup AZT for the first six weeks of life negatively correlated with HIV-MTCT (Table 2). Although the differences in transmission rates between women treated during pregnancy with HAART and those treated with mono or dual therapy were not statistically significant, in the HAART era, combination therapy with HAART (as opposed to mono or dual therapy) should be offered to all HIV-infected pregnant women regardless of their viral load or CD4 counts.

Breastfeeding is associated with an approximate doubling of the HIV-MTCT risk.¹³ Indeed in our study, breastfeeding highly correlated (P = 0.0001) with HIV-MTCT rate (Table 2). Since infant formula is available to all infants of HIV-infected mothers in Israel, avoidance of breastfeeding is highly recommended.

The preferred mode of delivery for HIV-infected pregnant women in the HAART era has been an issue of debate.^13,15,16 In the European mode of delivery randomized trial the HIV-MTCT rate among women allocated to vaginal delivery was 10.5% versus 1.8% in the elective caesarean section group.¹⁵ Similar results were reported in a large meta-analysis involving 15 American and European observational studies. ^13,16 However, the added benefit of elective caesarean section on reducing HIV-MTCT in women with low plasma HIV viral load (<1000 copies/mL) treated with HAART is uncertain.¹³ In our study, vaginal delivery for women with a viral load of more than 1000 copies/mL correlated with higher rates of HIV-MTCT as compared with caesarean section deliveries (Table 3). For those women, elective caesarean section should be recommended. Among women with a HIV viral load of less than 1000 copies/mL no vertical transmission was documented in our study regardless of mode of delivery (Table 3). For the latter women there is no proven benefit for caesarean section, thus vaginal delivery should be considered.

Although the vast majority of infants born to HIV-infected mothers in the HAART era are protected from vertical transmission, they are exposed to ARV drugs, for which there is only limited information on neonatal toxicity. In our study, 37 of 304 (12.2%) newborns to HIV-infected women (24 of these infants were exposed to HAART during pregnancy) required treatment in the neonatal intensive care unit, this is higher than the rate for non-HIV deliveries in Israel, which is about 8%.¹⁷ Further studies are needed to define the prevalence and risk factors of neonatal morbidity in the HAART era.

In conclusion, implantation of HIV-MTCT prevention programme in six Israeli centres was successful and a low rate (0.7%) of vertical transmission was observed in the last three years of our study. Nevertheless, an early screening and diagnosis of HIV-infected pregnant women remains a major goal for our HIV-MTCT prevention programme in order to improve antenatal care, and to further decrease HIV transmission.

References

Quian

, Visonti

, Gutierrez

RS.

Detection of HIV infection in pregnant women by rapid testing: a successfully strategy to reduce its vertical transmission.

Rev Chilena Infectol 2005; 22: 321–6

McIntyre

Prevention of mother-to-child transmission of HIV: treatment options.

Expert Rev Anti Infect Ther 2005; 3: 971–80

Newell

ML.

Current issues in the prevention of mother-to-child transmission of HIV-1 infection.

Trans R Soc Trop Med Hyg 2006; 100: 1–5

Tonwe-Gold

, Ekouevi

, Viho

Antiretroviral treatment and prevention of peripartum and postnatal HIV transmission in West Africa: evaluation of a two tiered approach.

PLoS Med 2007; 4: e257

Chou

, Smiths

, Huffman

LH.

Prenatal screening for HIV: a review of the evidence for the U.S. Preventive Services Task Force.

Ann Intern Med 2005; 143: 38–54

Newell

, Thorne

Antiretroviral therapy and mother-to-child transmission of HIV-1.

Expert Rev Anti Infect Ther 2004; 2: 717–32

Ostergren

, Malyuta

Elimination of HIV infection in infants in Europe - challenges and demand for response.

Semin Fetal Neonatal Med 2006; 11: 54–7

Coffie

, Ekouevi

, Chaix

ML.

Maternal 12-month response to antiretroviral therapy following prevention of mother-to-child transmission of HIV type 1, Ivory Coast, 2003-2006.

Clin Infect Dis 2008; 46: 611–21

Public health Task Force. Recommendation for Use of Antiretroviral Ddrugs in Pregnant HIV-1 Infected Women for Maternal Health and Intervention to Reduce Perinatal HIV-1 Transmission in the United States, 2005. http://AIDSinfo.nih.gov

10.

Foster

, Lyall

EG.

HIV in pregnancy: evolution of the clinical practice in the UK.

Int J STD AIDS 2006; 17: 660–7

11.

Buchholz

, Beichert

, Marcus

German-Austrian recommendations for HIV therapy in pregnancy and in HIV-exposed newborn - update 2005.

Eur J Med Res 2006; 11: 359–76

12.

Naver

, Lindgren

, Belfrage

Children born to HIV-1 infected women in Sweden in 1982-2003, trends in epidemiology and vertical transmission.

J Acquir Immune Defic Syndr 2006; 42: 484–9

13.

Thorne

, Newell

ML.

HIV.

Semin Fetal Neonat Med 2007; 12: 174–81

14.

D'Iappolito

, Read

, Korellitx

Missed opportunities for prevention of mother to child transmission immunodeficiency virus type 1 in Latin America and the Caribbean.

Pediatr Infect Dis J 2007; 26: 649–53

15.

The European mode of delivery collaboration.

Elective caesarean section versus vaginal delivery in preventive vertical HIV-1 transmission: a randomized clinical trial.

Lancet 1999; 353: 1035–9

16.

The international perinatal HIV group.

Mode of delivery and vertical transmission of HIV-1: a meta analysis from fifteen prospective cohort studies.

N Engl J Med 1999; 340: 977–87

17.

Israeli Ministry of Health. The Department for Mother, Children and Adolescence. www.health.gov.il

Prevention of human immunodeficiency virus mother–to–child transmission in Israel