Abstract Book Vol. 2 [International Conference on AIDS (14th: 2002: Barcelona, Spain)]

66 Abstracts WePeB5944-WePeB5948

XIV International AIDS Conference

WePeB5944I Determinants of an undetectable viral load at
delivery in a cohort of HIV+ pregnant women
K. Bunge, D. Virgo, J. Keller, T. Weikel, C. Holcroft, B. Ross, G. Hanna,
J. Anderson. Johns Hopkins University, Baltimore, MD, United States
Background: Maternal HIV-RNA level (VL) at the time of delivery is an established risk factor for perinatal transmission.
Methods: Retrospective cohort study of 64 HIV+ women treated with antiretrovirals (ARV) in pregnancy to identify predictors of undetectable VL (<500 c/ml).
Summary statistics and logistic regression used for analysis.
Results: Mean age: 28 y. Race: Black 76%, White 22%, Hispanic 2%. HIV risk:
sex 67%, sex/IDU 11%, IDU 11%. Current drug use in pregnancy 33%. ARV
regimen: HAART 61%, non-HAART (generally ZDV only) 39%. Average time on
treatment 22 wk. 54% exposed to ARV prior to pregnancy. Median VL beginning
pregnancy: 7485 c/ml, median CD4 377; median delivery VL 476 c/ml, median
CD4 453. 52% achieved undetectable VL. Logistic regression model for predictors of undetectable VL: ARV naive prior to pregnancy OR 5.7, p 0.04; log beginning VL OR 0.68, p 0.71; delivery CD4 OR 1.006, p 0.005; and HAART OR 102,
p <.001.
Conclusion: Predictors of undetectable VL at delivery include no prior ARV use
before pregnancy, lower VL at beginning of pregnancy, higher CD4 at delivery,
and HAART. These predictors are similar to those predicting optimal VL response
in nonpregnant individuals.
Presenting author: Katherine Bunge, 1917 Bank St. #2, Baltimore, Maryland,
21231, United States, Tel.: [phone], Fax: [phone], E-mail:
[email protected]
WePeB5945 Rapid HIV testing of women in labor: a comparison
of protocols in two hospitals in the U.S.
B.W.C. Forsyth1, S.R. Barringer1, M.L. Landry', D. Ferguson', T Tinghitella2,
M. Unfricht 2, E. Luchansky2, T. Walls1, U. Magriples'. 'Department of
Pediatrics, Yale University School of Medicine, CT United States; 2Bridgeport
Hospital, Bridgeport, CT United States
Background: Conducting rapid HIV testing in labor may have an important role
for preventing perinatal HIV transmission when testing has not been done earlier
in pregnancy. Studies have shown that if a woman is found to be positive, initiation
of antiretroviral treatment even after delivery might decrease the risk of transmission, although this might be beneficial only when treatment is started within 12
hours of birth. In the U.S., however, little attention has been paid to the development of protocols for instituting testing in labor. The purpose of this study was
to compare different protocols for HIV testing of women in labor, and determine
what proportion of women had results back early enough to intervene if results
had been positive.
Methods: An observational study was conducted of protocols used in two hospitals: Hospital A initially ran ELISAs each day (N=64) and then changed to using
rapid tests (SUDS) (N=36). Hospital B used only SUDS (N=56). Records were
reviewed to determine the timing of critical events relating to labor and delivery.
Results: With the use of SUDS in Hospital A, results were reported significantly
quicker than with the ELISA protocol in the same hospital (P<0.0001) or the
SUDS protocol in Hospital B (P<0.05). The more rapid turnaround for SUDS
testing in hospital A compared to Hospital B meant that results were more often
available before critical events; for example, in the two hospitals respectively, 67%
and 43% of results were back before the child's birth (P<0.05), and 97% and 78%
were back within 12 hours after birth (P<0.05).
Conclusions: If HIV testing in labor is to have its maximum effect on decreasing
the risk of perinatal transmission, rapid HIV tests need to be available, but also
attention needs to be paid to the protocols developed for ensuring results are
reported as quickly as possible.
Presenting author: Brian Forsyth, Department of Pediatrics, Yale University
School of Medicine, 333 Cedar St., New Haven, CT 06520-8064, United States,
Tel.: [phone], Fax: [phone], E-mail: brian.forsyth @ yale.edu
WePeB5946 Mother to Child transmission (MTCT) of HIV in an
inner city population in New York (USA)
S. Bakshi1, M. Purswani2, E. Stuard2, V. Kalakata2, M. Crane3, S. Duggirala2,
J. McGowan2. Bronx-Lebanon Hospital Center, Albert Einstein college of
Medicine, Pediatric Infectious Disease, Bronx Lebanon Hospital center, 1650
Selwyn Ave, 2C, Bronx, NY10457, United States; 2Bronx-Lebanon Hospital
Center, Albert Einstein college of Medicine, Bronx, NY United States;
3Bronx-Lebanon Hospital Center, Bronx, NY United States
Background: Despite decline in MTCT, failures continue to occur. We reviewed
trends in MTCT since 1998 in an inner-city hospital in Bronx where HIV seroprevalance (SP) in women at delivery is high, and evaluated factors contributing
to MTCT.

Methods: HIV infection (inf) was diagnosed in women by Ab test in pregnancy,
newborn Ab screen (NS), and if no documented test, by expedited test (Exp) in
labor, on cord blood or newborn. Maternal characteristics comparing women who
transmitted inf to their babies and those who did not, included age, length of HIV
inf, CD4+ count, VL, delivery type and length of ruptured membranes. Infant characteristics in infected and uninfected infants included prematurity, birth weight and
neonatal complications.

Results: Since 1998, 176 of 189 HIV+ women delivered live infants. 12 women
transmitted inf to their babies (TRs),164 did not (NonTR). 5/12 TRs had HIV inf
for >1year [chronic (ch) inf], 1/12 was diagnosed at the time her baby developed
01 at age 3 months. In 6/12 TRs diagnosis of HIV was made on NS (5),or Exp (1)
vs in 9/164 nonTR (NS 8 or Exp 1), (p<0.0000001). In 5/12 TR with chr HIV inf,
3 received no intrapartum (IP) ZDV. Additionally, 3 had poor adherence to ART
in pregnancy, but infants of all 5 with chr inf received neonatal ZDV prophylaxis.
1/12 TRs mom presumably transmitted by breast feeding (baby had -ve HIV DNA
PCR at 4 wks, but +ve at 8 wks). There was no difference in obstetric/ infant characteristics between TR and NonTR. From timing of +ve PCR, 3/5 infants of chr inf
moms acquired infection in utero.
Year 1998 1999 2000 2001
Total Deliveries 2577 2636 2532 2508
# HIV+ve deliveries (%SP) 55(2.13) 49(1.86) 40(1.58) 45(1.5)
# HIV+ve Babies 6 4 0 2
%MCTC 11 8 0 4
Conclusions: SP of HIV in women at delivery in an inner city NY has shown modest decline since 1998. MCTC is linked to lack of HIV diagnosis in pregnancy.
Mandatory HIV testing during pregnancy from mid 1999 has led to decrease
in MTCT. The potential for transmission remains in women who are diagnosed
late, cannot be engaged in care, do not receive adequate treatment, or continue
breast-feeding.
Presenting author: Saroj Bakshi, Pediatric Infectious Disease, Bronx Lebanon
Hospital center, 1650 Selwyn Ave, 2C, Bronx, NY10457, United States, Tel.:
[phone], E-mail: sarojbakshi@prodigynet
WePeB5947 Pasteurisation of expressed breast milk: The
experience of HIV infected women
W. Kitinyal, A. Kesesa2, A. Jorgensen3. 'University of Tanzania, Brandekilde
Vade vej 36, Dk-5250 Odense SV, Denmark, Tanzania; 2Amref, Tanzania, Dares
Salaam, Tanzania; 3Solarlnnovation, Odense, Denmark
Issues: There has been a growing concern over transmission of HIV through
breastfeeding. UNAIDS has estimated that 5-700 children are infected with HIV
every day through breast milk and it is estimated that 4000 infants die every day
because they were not adequately breastfed. For HIV infected women, who must
decide how to feed their infants, this is a cruel dilemma.
A device for home-pasteurisation of breast milk was developed and tested as previously reported. The device can be powered by grid electricity or by solar energy
to make it easy to use everywhere. The device heats the milk for 30 min. at 60
degrees C. This treatment kills the virus effectively but preserves the content of
milk like immunoglobulins quite well.
Description: Women in Dar Es Salaam, Tanzania were after counselling, testing and anti viral treatment during pregnancy counselled on how to feed their
infants. The women that chose to express their breast milk and to pasteurise it
were followed-up and had a session every Saturday for the six months they were
breastfeeding.
Lessons learned: from case stories: J is single and lives with her brother. Had
no problem to tell what she was doing but felt that if she moves she wouldn't like
the neighbours to know. The expressing takes about 15 min. and she does it 4
times a day
M has twins. Is living with her mother and others in the house. Does not have
any problem with secrecy. Spend a lot of time to express milk enough for the two
infant and has to supplement with formula milk.
F was married one year before but her husband died soon after. After she had
found out that her husband had been taking some medicine secretly, she was
tested. She is staying alone and expressing milk two times a day Has severe
problems because her husband's relatives are claiming the house.
Recommendations: The method can be an option for HIV infected mothers but
in poor economic settings we think follow-up and support is necessary
Presenting author: Anders Fjendbo Jorgensen, Brandekilde Vade vej 36, Dk5250 Odense SV, Denmark, Denmark, Tel.: [phone], Fax: [phone],
E-mail: [email protected]
WePeB5948 In vivo human placental transfer of nevirapine and
protease inihibitors and its detection in amniotic
fluid
A. Ginqelmaier1, M. Kurowski2, R. Kaestner1, S. Kriegel1, A. Knobbe1,
B.H. Belohradsky3, M. Stauber1, T.A. Grubert 1 '/. Frauenklinik der Universitaet
Muenchen, I. Erauenklinik der Universitaet Muenchen, Maistr. 11, 80337
Muenchen, Germany; 2Auguste-Viktoria-Krankenhaus, Berlin, Germany; 3Dr
Von Haunersches Kinderspital, Muenchen, Germany
Background: In Europe the number of women getting pregnant under a current

HAART is increasing. However, little is known about potential adverse effects for
the babies. In a pilot study we tried to learn more about the in vivo pharmacokinetics of these drugs. We investigated the placental transfer of nevirapine and
protease inhibitors (PIs) and its concentration in the amniotic fluid depending on
the drug level in maternal blood during birth.
Methods: This study represents a prospective analysis of 16 mothers and their
newborn babies. All babies were deliverd by planned cesearean section. We mea