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

Track C: Epidemiology & Public Health
aliquoted. One aliquot of saliva and the blood sample were tested by
radio immune precipitation assay (RIPA) and for IgG levels.

Mo.C.1529 - Mo.C.1533

Results:
Device
EIA
HIV Result
HIV+
HIVTotal

HIV Results for Saliva Specimens by Collection

OmniSal
Biochem Detect HIV 1,2
172
171
343

OraSure
Organon Teknika HIV I
118
174
291

There was 100% concordance between serum and OF results for all
devices. All 5 OF specimens collected from each patient performed
(signal/COV CV< 15%) for the EIA recommended for use with that d
manufacturer.
Conclusions: There was excellent correlation between HIV results obtaiw
and conventional serology. Performance of each collection device was co
different specimens collected from the same patient. Results of further c
the blood and OF specimens by western blot, RIPA and lgG levels will bi
next few months.The establishment of this reference panel of saliva spec
comparative studies of both collection devices and HIV antibody test kits
Margaret Fearon, HIV Laboratory, 8 I Resources Rd.,Toronto, Ontario M
Telephone: [phone] Facsimile: [phone]
Mo.C.1529
CORRELATION OF HIV PLASMA VIREMIA WITH CLINICAL STATU
COUNT
Touloumi Giota*, Katsoulidou A*, SipsaV*, Lazanas M**, Saroglou G**,
Mandalaki T***, PaparizosV****, Stavrianeas N****, Stratigos I***s,Velc
Papademitriou T*****, Hatzakis A*. *Med Sch; **Evangelismos Hosp;:
****A. Syngros Hosp; *****HCC of AIDS and STD's, Athens, Greece
Objective: To examine the relation between the quantity of human imrr
type I RNA (HIV-RNA) in plasma and clinical status and CD4 count.
Methods: Three hundred and eighty HIV- I positive individuals have bee
ongoing cohort study for assessment of predictive value of HIV RNA inr
of HIV infection. Clinical status, CD4 counts, HIV-Ag, neopterin, f32 micri
samples are obtained every 6 months. HIV-RNA in plasma is quantitate(
DNA (b-DNA) signal amplification method.The baseline data of 359 H
als with measured CD4 counts and HIV-RNA at entry were analysed cr
Results: Plasma HIV-RNA was detected in 73% of the specimens (n=3_
cy of detection according to CD4 levels was as follows: 47%, CD4>500
200-499 (n= 152); 94%, CD4<200 (n= 147).The virus load levels were 1
AIDS patients [Geometric mean (GM) = 50.5 x 103, 95% CI (30.5 x I
medium in symptomatic [GM = 28.6 x 103, 95% Cl (19. I x 103, 42.9 x
asymptomatic individuals [GM =11.8 x 10 3, 95% Cl (8.7 x 0I 3, 16. 1
being statistically significant (P<0.001). A significant linear relationship be
count and log 10 HIV-RNA levels was found (r=-0.47, P<0.0 1), holding I
retroviral treatment (r=-0.45, n=237, P<0.0 I) as well as for the untreat
n=106, P<0.0 I).
Conclusions: We confirmed that there is a strong relation between HIN
disease stage or CD4 levels. However, the ability of viral load to predict
independently of the immunological markers, needs further assessment
G.Touloumi, Athens University Medical School, M. Asias 75, I 1527, Athe
Tel. [phone], Fax(301I) 7704225 email: [email protected].
Mo.C.1530
THE NEED FOR RAPID ON-SITE HIV TESTING IN AFRICA
Wilkinson D, Wilkinson N. Hlabisa Hospital and South African Medical
Hlabisa, South Africa
Issue: Voluntary counselling and testing (VCT) for HIV infection is impo
and as part for risk reduction strategies. In our rural district hospital in
become aware that many patients who are tested for HIV infection foll
selling and after giving informed consent do not return for their results.
precious resources and defeats the object of testing.
Project: We audited the counselling service at a district hospital in Sout
mine the number of people tested for HIV infection at the hospital, the
positive, and the proportion receiving post test counselling. HIV testing
policy (double ELISA in Durban, 300km away).
Results: In October and November 1995, 305 people were tested for
results were never returned; 142 (59%) of those returned were HIV pc
(17%) people returned for their results and therefore were counselled.
receiving results was 2 1.3 days (range 9-43).
Lessons learned: In response to these findings the HIV team in the hos
project to test the accuracy, feasibility, cost-effectiveness and impact on
rates of on-site testing for HIV infection using "rapid" HIV diagnostic tes
encouraging reports from other centres the Capillus HIV- I/HIV-2 (Canr
been chosen for this project.There is an urgent need to develop cost-e
on-site testing for HIV infection in rural African settings.
Dr David Wilkinson, Box 658, Hlabisa 3937, South Africa.Tel. +27318 3
Fax [phone] E-mail [email protected]

EIA, western blot, Mo.C. I 53 I
COMPARISON OF FIVE COMMERCIAL HIV-I ANTIBODY SCREENING ASSAYS IN
THAI MALES
1 Devic e Chanbancherd, Penprapa., Limpairojn N*, Mason CJ**, Jugsudee A*, Michael RA**, de
S Souza MS**. *Army Institute of Pathology Bangkok,Thailand;** AFRIMS, Bangkok,Thailand
Salivette Objective: To compare the reproducibility of a positive 3rd-generation HIV enzyme-linked
Murex GACELISA immunosorbent assay (ELISA) result with other HIV ELISAs and a particle agglutination
assay in Thai sera; and to determine the proportion of antigenemic subjects in those posi84 tive for the reference ELISA.
1 37 Methods: Seventy Western blot negative or Indeterminate samples were selected which
22 I reacted to a designated reference 3rd-generation ELISA (Abbott) and matched with 70
HIV antibody negative samples. All samples were analysed by an additional 3rd-generation
three collection [LISA (Vironostika), two 2nd-generation peptide ELISAs (Novapath-HIV- 1/2, Detect-HIV).
onsistenatly PA (Serodia-HIV) and HIV p24 antigen assay (Coulter).Ten 3rd-generation Abblot ELISA
ievice by the positive and Western blot positive sera were used to test the sensitivity of the 4 additional assays.
ned by testing OF Results: The 3 additional ELISAs and the agglutination test gave varying positive results with
nsistent for the 5 the Abbott ELISA positive samples - 29% for Vironostika, 13% for Detect I I% for Novapath
haracterization of and 18% for Serodia.The 5 assays were 100% concordant for 6% of the samples. Free HIV
e available within the p24 antigen was detected in 4 (6%) of the Abbott positive sera and I (I%) of the:imens will facilitate Vironostika positive sera.The Abbott antibody negative sera remained negative by all assays
s for saliva testing. except for the Detect peptide assay, which showed 1/70 sera positive. Of the 10 positive
controls, there were 100% concordant results for all antibody assays, with no samples show19P 3T I, Canada ing free H1IV p24 antigen.
Conclusions: The Abbott 3rd-generation ELISA demonstrated the most frequent rate of
false HIV antibody positive results relative to the other assays. Howerver the results of the
p24 antigen assay demonstrated that the Abbott test detected early seroconversion in 5%
JS AND CD4 of cases which would have been missed by the other assays.The two 3rd-generation HIV
antibody [LISAs are more sensitive than the peptide and agglutination assays at indicating
Karafoulidou A***, early HIV infection, but their antibody specificities were markedly different from each other:
onaki A*****,
*Laikon Hosp; ILPC Penprapa Chanbancherd, Army Institute of Pathology, 315 Rajvithi Road, Bangkok
I0400,Thailand
nunodeficiency virus Mo.C.532
Mo.C.1532
n enrolled in an MOLECULAR INVESTIGATION CONFIRMING AN OUTBREAK OF HIV IN A
the natural history SCOTTISH PRISON
oglobulin and plasma D L Yirrell*, DLGoldber,**, P Robertson*, J McMenamin**, S Cameron **, A J Leigh
d with the branched Brown*. *Centre for HIV Studies, University of Edinburgh. **Scottish Centre for
IV positive individu- Infection & Environmental Health, Ruchill Hospital, Glasgow. *** Dept of Virology, Ruchill
ros-sectionally. Hospital. Glasgow.
59) asnd the frequen- Aim:To provide conclusive evidence that an outbreak of HIV occurred within Glenochil
S(n60); 63%, CD4 prison, through phylogenetic analysis of virus obtained from infected inmates.
found highest in Background and Methods: Between April and June 1993, 8 cases of acute Hepatitis B infec03, 83.4 x 103)], tion were observed among inmates of the prison.This coupled with reports of random
S03)] and lowest in sharing of needles and syringes led to a public health initiative involving counselling and HIV
I 03)], the difference testing. All 14 irinsates found to be HIV positive were from Glasgow and Edinburgh, and had
tween the CD4 injected in the prison. On the basis of the date of admission to Glenochil, sequential HIV
for those under test results showing seroconversion profiles, P24 antigen data and clinical observations (pried group (r--0.63, mary HIV infection), there was definitive evidence for 6 transmissions having occurred
inside Glenochil.To establish that remaining cases had become infected there, PBMCs were
V-RNA levels and obtained frlom 13 of the 14 HIV positive inmates. cDNA, generated through reverse trant clinical disease, scription on viral RNA, was used along with pro-viral DNA in nested PCR amplifications
involving the V3/V4 region of the ENV gene and p17 of the GAG gene.
Results: Maximum likelihood phylogenetic trees were established comparing data from the
ns, Greece, 14 cases with that from historical controls (haemophiliacs and other drug users from
gr Edinburgh and Glasgow). For each gene region, remarkable similarity in sequence composition was seen for 13 of 14 cases.
Conclusion: Since HIV is highly susceptible to mutation in response to pressure from the
host's immune system, the absence of appreciable sequence variation in this cohort proResearch Council, vides compelling evidence that the inmates acquired their infection in Glenochil from a single individual directly or along a chain over a short period.The application of molecular
)rtant both clinically techniques in this manner can provide crucial epidemiological information, thus enabling the
South Aftica we have demands for public health interventions (in this case in the prison setting) to be based on
owing pre-test coun- firm evidence.
This is a waste of
Dr David Goldberg, Scottish Centre for Infection and Environmental Health,
th Africa, to deter- Ruchill Hospital, Glasgow G20 9NB.Tel:[phone] I 20. Fax:014 I -946-0860.
proportion HIV email: [email protected].
followed provincial Mo.C.533
Mo.C. 1533
HIV. 65 (2 1.3%) DETECTION OF ANTIBODIES TO HUMAN T-LYMPHOTROPIC VIRUS TYPE I AND
asitive, and only 52 TYPE II IN SAMPLES FROM VARIOUS POPULATIONS
Tse aveiage delay in DeweyN P. Chety, C., Horvath, B., Scruggs P, Witt, D. Organon Teknika Corporation (OTC),
Dusham, NC USA
pital has started a Objective: to evaluate an enzyme-linked inmunosorbent assay ([LISA) consisting of both
post test counselling HTLV-l and HTLV-II antigens on the solid phase for the detection of antibodies so HTLV-l
sts. Following recent and HTI V-Il in serum and plasma samples and to correlate the [LISA results with the
bridge Biotech) has insults 5 F Westesrn Blot (WB) and polymerase chain reaction (PCR).
sffectis'e, high quality Methods: The [I ISA solid phase was coated with a purified HTLV-I viral lysate, a purified
I-!-[tV-II Viial lysaste. and a recombinant p21IF. Immune complexes formed by antibodies in
the samples and the solid phase antigens were detected by anti-human IgG-HRP conjugate
81 155 and visualized by Tetramethylbenzidine (TMB) substrate. Samples for evaluation were
de ved f,-om FTLV-I associated diseases group, high risk injecting drug user (IDU) populations, donor populations, and several well characterized panels. A combination ofWB, PCR,
and a pe1,tide assay was used to discriminate between HTLVl and HTLV-II antibodies.
Results: in ihe HTLV-I associated diseases group, both ELISA and WB were reactive for all
of the samples from patients with Adult T-Cell leukemia (ATL) (n- I 8) and Tropical Spastic
Parapare_. (TSP) (n-- 15). All samples contained high levels of antibody and demonstrated
typical antibody profiles to the gag and the en gene products in the WB assay In an IDU

Va)
0
(U
so:3
V.).)
<C
Ca)
C)
U
C)
C)
x
}4
14