Abstracts Vol. 1 [International Conference on AIDS (11th: 1996: Vancouver, Canada)]

Track C: Epidemiology and Public Health
Conclusions: The vertical transmission of HIV infection may be acquired during pregnancy
or after birth.The results from this study support that the HIV transmission in utero before
late pregnancy and/or delivery in our population appears to be very low. Prevention of
perinatal transmission during delivery may be one means of reducing maternal-fetal HIV
infection.
Yupa Uiwijitaroon, Blood Transfusion Cente, Faculty of Medicine, Khon Kaen University,
Khon Kaen, Thailand 40002, Fax [phone]
Tu.C.2583
PERINATAL HIV- I TRANSMISSION AMONG WOMEN SEROCONVERTING DURING
PREGNANCY, BANGKOK,THAILAND
Ro npisuthiBon Anuvat2, Siriwasin W1, Shaffer N3,4, Chaisilwattana P,Young N3'4, Wasi
C 1, Kaewchaiyo G2, Kalish M4, Bhiraleus P 1, Chinayon P2, Mastro TD.3,4.I Siriraj Hospital,
Mahidol University, Bangkok; 2Rajvithi Hospital, MOPH, Bangkok; 3HIV/AIDS Collaboration,
NonthaburiThailand; 4CDC, Atlanta, USA.
Objectives: To evaluate the immunologic and virologic characteristics of women with acute
HIV- I infection during pregnancy and to determine whether they are at increased risk of
perinatal transmission.
Methods: At 2 large hospitals in Bangkok with a background HIV- I seroprevalence of 2%
and with routine HIV counseling and testing at first antenatal clinic (ANC) visit and at late
third trimester pregnant women documented to have seroconverted during pregnancy (I
st test negative and 2nd test positive) were enrolled as a special subgroup in a prospective
mother-child HIV- I transmission study Infection outcome, CD4 profiles, and plasma viral
load (NASBA, Organon Teknika) for seroconverting pregnant women (SC) were compared
with those testing HIV positive at I st ANC (POS).
Results: From Nov 92 through Mar'94, 342 HIV-infected women were enrolled during
pregnancy, including I9 SC.The perinatal transmission rate was I16.7% (3/18) among SC and
24.7% (65/263) among POS (p=0.4).The median absolute CD4 counts for the 2 groups
(all p>0.5) were:
3rd Tri Delivery 6 Months 12 Months
SC 500 460 550 530
POS 460 445 490 460
Viral load at delivery was similar between the 2 groups (median 5,500 vs. 4,500 copies/ml;
p=0.5).
Conclusions: In this sample, women seroconverting during pregnancy had similar immunologic and virologic profiles as women identified as HIV-infected at first ANC visit and had a
similar risk of perinatal transmission.
Anuvat Roongpisuthipong, c/o HIV/AIDS Collaboration, 88/7 Sol Bamrasnaradura,Tivanon
Road, Nonthaburi I 1000,Thailand.Tel: [phone] I, Fax: [phone]
Tu.C.2584
AN EXPLANATION FOR THE LOW NUMBER OF PAEDIATRIC HIV INFECTIONS IN
AMSTERDAM,THE NETHERLANDS
Van den Hoek, I Anneke R, van Haastrecht HJA, Spijkerman IJB, Coutinho RA. Municipal
Health Service, Department of Public Health, Amsterdam
Objective: In the western world, injecting drug using women (IDU) are an important
source for paediatric HIV infection. In Amsterdam the HIV prevalence among IDU, since
1986, is ca 30% and the estimated cumulative number of HIV infected IDU women is ca
500. Nevertheless, at the end of 1994, only 7/I 629 AIDS cases were in children. In order to
gain an insight into this low number of paediatric infections, we studied the pregnancy histories of women participating in the Amsterdam cohort study among injecting and non-injecting drug users.
Methods: Since July 1994 2 6 female drug users participating in the Amsterdam cohort
study were interviewed about pregnancies and outcomes.
Results: IDU and non-IDU: 64% of the women were Dutch; 8 I% (175) had a history of IDU.
I 06 women (49%) reported a total of I 64 children. IDU had less often children than nonIDU (45% vs 68%). Of the 1 75 IDU 37% were HIV infected and of the 4 I non-IDU 7%.
IDU only: 78 out of 175 (45%) IDU reported a total of 115 children.The mean age at intake
into this study was 34 (sd 5.5).The mean age at the birth of their last child was 24.5 (sd 5),
which indicates that most of these women had their children before or in the beginning of
the AIDS epidemic. Since July 94, I 6 IDU reported pregnancy of whom 4 were infected
with HIV: 4 (I HIV pos) gave birth, 2 (I HIV pos) were still pregnant at the last visit, 5 (2
HIV pos) spontaneously miscarried and 5 (0 HIV pos) got an induced abortion.Women
who got pregnant were significantly younger, tended to be less often Dutch and current
injectors and more often nulliparous.The live birth rate among HIV pos IDU was 10.3 per
1000 person years and for HIV neg IDU 21.2 per I000 person years, and can be considered very low (life birth rate in the Netherlands: 87.7/1000 women aged 25-40 years).
Conclusion: The relatively old age of drug using women in Amsterdam together with the
low birth rate in HIV positives explain for a large part the low number of paediatric HIV
infections.
Anneke van den Hoek, Municipal Health Service, Nieuwe Achtergracht 100, 10 I8 WT
Anmsterdam the Netherlands. Telephone: ~3 I-20-5555 384 Fax: [phone]
Tu.C.2585
HOSPITALISATION OF CHILDREN BORN TO HIV INFECTED MOTHERS
Thoee Claire, the European Collaborative Study
Objectives: To describe the pattern of in-patient hospitalisation among children born to
HIV infected women in I10 centres of the European Collaborative Study (ECS).
Methods: Dta on III5 rhildien enrolled between 987 and December 1 995 and followed prospectively since birth according to a standard protocol were analysed, including
number of hospitalisations, duration of stay and dischaige diagnoses.
Results: This analysis included 148 infected and 758 uninfected children and 209 of indeter
minate HIV status. Mean length of follow-up was 36.8 months (range 0 to 1I 17.3 months).
Three hundred and twenty-eight children (29.4%) had been hospitalised as in-patients on
649 occasions (I.98/admitted child) and median length of stay per admission was 10 days

Tu.C.2583 - Tu.C.2587
(range I to 370 days). For the total cohort, 2/100 child-days were spent in hospital in the
first year of life (7.6/100 child-days for infected and I / 100 child-days for uninfected children). Infection status was associated with hospital admission with 99/148 (66.9%) infected
and 198/758 (26.1%) uninfected children admitted (p<0.02). Infants of mothers with AIDS
and whose mothers used drugs during pregnancy were more likely to be admitted to hospital, irrespective of their infection status. Infected children were hospitalised for the first
time at younger ages than those uninfected (p<0.02): 47/99 (47.5%) infected children were
diagnosed during or after their first hospitalisation. By 2 years of age, an estimated 2 1.5% of
infected children will have been hospitalised compared to 9.8% of uninfected children, rising
to 54.6% and 27.3% respectively by 5 years of age. Fifty-eight (8.9%) admissions were for
social reasons (17/58 for infected children), accounting for 25.4% of all bed days.
Conclusions: This novel information highlights the utilisation of in-patient hospital care by
both infected and uninfected children within this European population.The use of hospital
resources for social admissions by both infected and uninfected children born to HIV infected women provides an argument for better support and substitute care provisions for families affected by HIV
Claire Thorne, Epidemiology Unit, Institute of Child Health, 30 Guilford St, London,
WCIN I EH Tel: [phone] x2105 Fax: 017 I 813 8233 email: [email protected]
Tu.C.2586
PASSIVE HYPERIMMUNE THERAPY IN A PREGNANT WOMAN WITH AIDS.
Ash, Stephen, Karpas A*, Sheppard K, Lynn W. The Paseur Suite, Ealing Hospital Trust,
London, England; *MRC (Haematology) Unit, Cambridge, England.
Objective: Passive hyperimmune therapy [PHIT] with plasma containing high levels of neutralising antibody to HIV might reduce the risk of vertical transmission of HIV Part of the
U.K. PHIT programme is to offer pregnant HIV seropositive women PHIT during their pregnancy and labour.The first such volunteer is presented here as testimony to the feasibility
and safety of such a potential therapy as part of an on-going programme to investigate
whether the administration of HIV- neutralising antibodies may reduce the exposure of the
foetus to infective HIV virions during pregnancy and delivery
Methods: Plasma from HIV seropositive, asymptomatic donors with CD4 lymphocyte
counts>500/ul and high neutralising antibody levels against HIV, was collected at monthly
plasmapheresis sessions.The plasma was sterilised with beta-proprionolactone and pooled
with other donations of the same blood group. 500ml aliquots were given by monthly IV
infusions to recipients.The first pregnant HIV positive volunteer received hyperimmune plasma monthly from her eighth week of pregnancy and through delivery. At that time she was
aged 24, had a CD4 count of 150/ul, and had had oesophageal candidlasis 8 months previously She had acquired HIV through a contaminated blood transfusion some six years
before. She had been taking both AZT and ddl, and these were both continued throughout
pregnancy.
Results: The patient experienced a normal pregnancy except for the development of
cholestasis of pregnancy for four weeks prior to delivery PHIT (and AZT and ddl) were
well tolerated.The baby was born by spontaneous vaginal delivery was clinically normal, and
had normal laboratory parameters.The mother's CD4 counts fell gradually during pregnancy to I O/ul, but recovered post-partum to 160/ul.Titres of antibodies, measured by ELISA,
fell gradually in the baby over the first 18 months of life. Since the age of 20 months, the
baby has repeatedly tested HIV-antibody negative.
Conclusion: PHIT with HIV-hyperimmune plasma may present a safe therapy to reduce the
vertical transmission of HIV
SA Ash,The Pasteur Suite, Ealing Hospital, Uxbridge Road, London UBI 3HW, UK
Telephone: [phone] I; Fax: [phone]
Tu.C.2587
PHYSICAL BREAKS IN THE PLACENTAL TROPHOBLASTIC SURFACE:
SIGNIFICANCE IN VERTICAL TRANSMISSION OF HIV.
Burton GJ*, O'Shea S**, Rostron T**, Mullen JET*, Aiyer S**, Smith R+, Banatvala E**
*Dept of Anatomy University of Cambridge, UK, **Dept of Virology United Medical and
Dental Schools of Guys and StThomas's Hospitals, London, UK, +Dept of Obstetrics and
Gynaecology Ninewells Hospital and Medical School, Dundee, UK.
Objective:Transmission of HIV from mother to infant may occur in utero, during delivery
or postnatally as a result of breast feeding, but the relative proportion and efficiency of
transmission at these times is unclearThe placenta provides a barrier between the maternal
and fetal circulations but only limited attention has been focused on its possible role in
transmission of HIV. Breaks in the continuity of the syncytiotrophoblastic surface of the placenta could provide a potential mechanism for infection of the placenta with subsequent
transmission to the fetus.The significance of physical breaches of the trophoblast in transmission of HIV from mother to child has therefore been evaluated.
Methods: Samples of peripheral blood were obtained from 17 HIV infected women during
pregnancy and at delivery and HIV RNA quantified by Nucleic Acid Sequence Based Assay
(NASBA).At delivery a small sample of placental tissue was obtained and placed into fixative and placental damage quantified and evaluated using transmission electron microscopy.
Infants were followed up at regular intervals and their HIV infection status determined using
PCR, virus isolation, detection of p24 antigen ard measurement of class specific antibodies.
Results: Maternal viral load was generally low (mean 8,237 RNA copies/mI of plasma, range
230-37,233 copies/ml).There was evidence of breaks in the trophoblastic surface to the
depth of the basement membrane in all 17 placentas and perivillous fibrinoid deposits were
observed to a varying degree in all samples.Virus particles, consistent morphologically with
HIV, were seen in a cytotrophoblast cell of one placenta and in the trophoblast basement
membrane of one other Budding of viral particles from the surface of leucocytes within the
fetal capillaries was also observed in one case. None of the 13 infants available to follow-up
had evidence of infection with HIV.
Conclusions: Superfcial damage to the trophoblastic surface of the placenta, with exposure
of the basement membrane and potentia exposure of CD4 expressing cells, is unlikely to
be a significant factor in the transmission of HIV from mother to snfant during pregnancy.

Profesor JE Banatvala, Dept of Virology, United Medical and Dental Schools of Guys and St
Thomas's Hospitals. London SEI 7EH,UK.Tele 0171 928 9292 x2405, Fax [phone]

O
a)
0
U
uC
re
c0
(.)
U
a)
a)
C
0
U
cC
0
me
C
a)
C
X
366