Final Program and Oral Abstracts [International Conference on AIDS (8th: 1992: Amsterdam, Netherlands)]

ThC 1578-ThC 1582

TRACK C: EPIDEMIOLOGY

ThC 1 578 VIRUS ISOLATION, PCR AND NEURO-DEVELOPMENTAL DELAY IN CHILDREN
S 1578 WHO ARE HIV SEROREVERTERS (P-3). Wiznia, Andrew*; Conroy, J**;
Liu,HK***; Nozyce, M*; *Bronx Lebanon Hospital Center, Bronx,
N.Y., **NYS Wadsworth Center for Laboratories and Research,
Albany, N.Y., ***Memorial Sloan Ketterring Laboratory, N.Y.
Objectives: To compare results of HIV virus isolation, HIV-1 PCR amplification,
laboratory and clinical studies in a population of HIV-seroreverter children (CDC
Classification P-3).
Methods: 102 HIV-seronegative children known to be HIV-seropositive at birth (P3) underwent the following laboratory evaluations: HIV antibody determination by
ELISA, HIV culture from peripheral blood mononuclear cells; HIV gag and env
amplification via the polymerase chain reaction; lymphocyte subset analysis; CBC,
and immunoglobulins. Clinical evaluations included a routine physical examination
and an age appropriate neuropsychological assessment.
Results: No child exhibited persistent symptoms consistent with HIV infection.
All children were HIV-seronegative and had age appropriate levels of
immunoglobulins. 148/149 HIV cultures and 189/190 PCR analysis performed in these
children were negative for HIV. Both children with positive results were negative
on two repeat samples. Mean CD4 counts for age groups were: 2714 (6-12 months of
age), 2365 (12-18 mos.), 1919 (18-24 mos.), 1717 (24-36 mos.) and 1315 (>36
mos.). 28/69 (41%) children scored within the normal range on neurodevelopmental
testing. On the MDI of the Bayley Scales 19/69 (30%) scored 1 S.D. below the
norm, 17 (25%) scored 2 S.D. below the norm. On the PDI, 14 (20%) scored 1 S.D.
below the norm, 11 (16%) scored 2 S.D. below the norm.
Conclusions: We were not able to consistently detect immunosilent HIV-infection
in a large cohort of HIV seroreverters (P-3) using culture and PCR technology.
CD4+ lymphocyte cell counts and immunoglobulin levels were normal for age. Many
infants and children born to HIV-infected mothers in our patient population have
significant delay on neuropsychological assessments.
Andrew Wiznia, M.D. Bronx Lebanon Hospital Center 1650
Grand Concourse Bronx, N.Y. 10457 U.S.A. [phone]
[phone] (fax)
ThC 1580 PCR AND HIV-IGA FOR EARLY INFANT DIAGNOSIS OF PERINATAL HIV
INFECTION. Shaffer Nathan; Ou C-Y; Abrams EJ; Krasinski K; Parekh B;
Thomas P; Bamji M; Moore J; Kilbourne B; George R; Rogers M; and the NYC
Perinatal HIV Transmission Collaborative Study. CDC, Atlanta, GA; NY, NY, USA.
Objective: To evaluate the polymerase chain reaction (PCR) and HIV-specific IgA (HIV-IgA) as early
diagnostic tests for infection in children born to HIV-1 infected mothers.
Methods: HIV-exposed children were enrolled at birth, beginning in 1986, and followed prospectively at 2-3
month intervals for at least 15 months to establish definitive infection status. PCR was performed on
previously frozen peripheral blood mononuclear cells, using 2 primer pairs (SK 38/39 and SK 68/69) on the
first 2 or more available infant samples. PCR tests performed before July, 1989 used "P radioactive
oligonucleotide probes; subsequent tests were performed using chemoluminescent probes (Gen-Probe). HIVIgA WB testing, using GammaBind G (Genex), was performed on stored sera from a subset of these children.
Results were evaluated by the age interval of the collected specimen (1 specimen/child/interval).
Results: The sensitivity for each test (combining PCR test methods), by age interval, was:
Test <1 Week 1-3 Weeks 1-3 Months 4-6 Months >6 Months
PCR 14/25(56%) 6/11(55%) 18/19(95%) 20/21(95%) 19/19(100%)
HIV-IgA 2/10(20%) 3/5(60%) 4/12(33%) 13/23(57%) 18/25(72%)
Among the 36 neonatal (< 1 month) samples tested by PCR from different infected children, 5/15 (33%) birth
samples and 5/16 (31%) overall were positive by the "P probe, compared with 9/10 (90%) birth samplesand
15/20 (75%) overall by the chemoluminescent probe. Across intervals, infected infants showed various HIVIgA patterns, including: early and late onset positives, persistently positive, persistently negative, and
intermittently positive. The specificity (# neg/# uninfected tested) of PCR was 100% (66/66 < 1 month,
151/151 > 1 month), and of IgA was 98% (2/4 < 1 month, 106/106 > 1 month).
Conclusions: Using chemoluminescent probes, the sensitivity of PCR to detect infant infection at or near
birth is significantly higher than previously reported; 3/4 of infected infants were detected in the neonatal
period, and >95% thereafter. HIV-IgA is a promising new test, but lacks sensitivity in the early months.
Both tests are highly specific, although neonatal IgA false-positives were noted.
Shaffer Nathan, Centers for Disease Control, MS E-45, Atlanta, GA, 30333, USA,
Telephone: [phone], FAX: [phone]
ThC 1 82 HIV-1-SPECIRC URINARY ANTIBODIES EXCRETED
S5 BY INFANTS OF HIV-1-INFECTED WOMEN. Bauer Gerhard; Johnson, J.P.; Lewis, G.K.;
Tuskan, R.; Hines, S.E.; Nair, P.; Umovitz, H.B.*; Cole, G.A. School of Medicine, University of
Maryland, Baltimore, MD and *Calypte Biomedical Corporation, Berkeley, CA; USA
Objectives: Because, in adults, IgG reactive specifically with gp160 is detectable in urine during HIV infection,
(Cao, Y.Z. et al., Lancet 1988; i:831-32) we are assaying urine samples from infants of HIV-infected women for
such antibodies to determine whether their presence is diagnostic of perinatal HIV transmission.
Methods: A microplate ELISA (Calypte Biomedical) was used to detect HIV-1 envelope-specific IgG in single
undiluted urine samples from each of 37 infants and in paired samples from each of 13 infants. Al infants were
between 2 weeks and 1 year of age at the time(s) of sample collection.
Results: Of 39 infants classified as having indeterminate infection (CDC classification PO), seven (18%) have
been found positive for envelope-specific urinary antibodies at ages averaging 2 months. Eleven infants are
known to be HIV-infected (P1 and P2), 10 of whom (91%) are urinary antibody-positive. Of 6 infected infants
who are still asymptomatic (P1), five were excreting antibodies at ages between 2 weeks and 2 months. All of
the 5 symptomatic (P2) infants were excreting antibodies when sampled between 4-8 months of age.
Concordant results were obtained for each pair of urine samples.
Conclusions: These preliminary findings suggest that the excretion of envelope-specific urinary antibodies by
infants of HIV-infected women may be an indicator of vertically transmitted infection and, possibly, one of its
initial manifestations. The diagnostic value of these antibodies will depend on whether their detection can be
shown to: 1) regularly precede the appearance of clinical disease, and 2) coincide with the direct
demonstration of HIV in patient material. Efforts towards achieving the latter are now in progress.
Bauer, Gerhard, University of Maryland School of Medicine, 655 West Baltimore St.,
Baltimore, MD 21201, USA; Telephone: [phone]; FAX: [phone]

ThC 1579 ACID DISSOCIATION OF IMMUNE COMPLEXES IMPROVES THE DIAGNOSTIC
UTILITY OF P24 ANTIGEN DETECTION IN PERINATALLY ACQUIRED HIV-1
INFECTION. Ouinn. T.C; Kline, R.; Moss, M.; Livingston, RA.; Hutton, N. NIAID,
Bethesda, MD and Johns Hopkins Univ., Balto., MD, U.SA.
Objective: Detection of p24 antigen in perinatally infected children as a diagnostic assay has been limited
due to immune complex formation with passively transferred maternal anti-p24 antibody. Since acid
treatment of serum is known to disrupt immune complexes, we examined the diagnostic utility of the p24
antigen assay following acid treatment of serum samples from children born to HIV-infected women.
Methods: 345 serum samples from 158 children born to HIV-1 infected women followed at The Johns
Hopkins Hospital were available for serologic testing. Infection status of the infants was confirmed by
HIV IgG antibody in children after 15 months of age. Infants were clinically staged according to the
pediatric clinical criteria of the CDC. Samples were analyzed for p24 antigen both before and after acid
dissociation on the Coulter HIV-1 p24 antigen assay.
Results: Although the p24 antigen assay after acid treatment was negative in 9 HIV-infected children less
than one week of age, antigen was detectable at high levels in all 30 samples obtained from infected
children between 1 and 9 months of age. The mean level of p24 antigen at 1-3 months of age was 491
pg/ml which declined to 139 pg/ml by 12 months. Overall, antigenemia was detected in 145 (sensitivity
89.5%) of 162 samples from 47 HIV-1 infected children one month of age or older. In contrast, the
sensitivity of the p24 antigen assay without acid dissociation was only 18% (p < 0.001). Among 76
uninfected children, 132 (specificity 99.2%) of 133 samples were p24 antigen negative following acid
dissociation. The positive and predictive values excluding children less than one week of age were 99.3%
and 88.6%, respectively.
Conclusion: The high sensitivity, specificity and predictive values of the p24 antigen assay following acid
treatment demonstrate its utility for the diagnosis of perinatally acquired HIV-1 infection after one month
of age. Early diagnosis with this assay in combination with other tests should aid in the early detection
of perinatally acquired HIV-1 infection and lead to early initiation of specific therapy in infected children.
Thomas C. Quinn, M.D., Div. of Infectious Diseases, Blalock 1111, Johns
Hopkins Hospital, 600 N. Wolfe St., Baltimore, MD 21205. Phone: [phone]; fax [phone].
ThC 1 581 HIV INFECTION IN NEWBORNS: EVIDENCE FOR VIRAL
REPLICATION WITHIN THE FIRST WEEKS OF LIFE.
Krivine Anne*; Firtion, G**; Cao, L*; FrancoualC*; HenionR**; LebonP*.
*Hopital Saint-Vincent-de-Paul**Clinique Universitaire Port-Royal, Paris, France
Objectives: we have conducted a prospective longitudinal study to evaluate the HIV
status of 50 infants born to HIV seropositive mothers within the first weeks of life.
Methods: blood samples were obtained at birth, 4-9 weeks and 5-9 months and were
tested for HIV-1 by polymerase chain reaction (PCR), viral culture and p24 antigenemia.
In addition, 13 infants were tested retrospectively for the presence of genomic HIV-1
RNA in plasma by PCR.
Results: among the 50 children, 16 were identified as HIV-infected by the age of 4-9
weeks by PCR and culture; in contrast, infection could be detected in only 5 of them at
birth, 3 of whom having positive p24 antigenemia. No changes in the HIV status were
observed between 4-9 weeks and 5-9 months in the 44 children who could be retested.
HIV-1 RNA was detected at 4-9 weeks in the plasma of 7 HIV-infected infants, versus
only 1 at birth.
Conclusion: perinatal HIV-1 infection can be diagnosed within the first two months of
life, whether PCR or viral culture is used. The fact that we failed to detect HIV-1
infection at birth in almost 70% of the babies subsequently found infected is indicative of
an active replication of HIV during the first weeks of life. These findings are consistent
with the detection of HIV-1 RNA in the plasma of 7 infants by the age of 4-9 weeks. Our
results could be in favour of the frequency of late transmission of HIV-1, either at the end
of pregnancy or at delivery.
Anne Krivine Laboratoire de Virologie, Hopital Saint-Vincent-dePaul,75014 Paris, France. Tel: [phone] FAX: [phone]

NOTES

Th82