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

Mo.B.1345 - Mo.B.1349

Monday, July 8, 1996

pneumonia (PCP) was evident among women, with an adjusted relative rate of I1.4 (95% CI
1.01-1.9, p=0.04). When stratified according to year of registration, the following rate ratios
were obtained:

Year 1988
RR (95% Cl) 1.3 (0.8-2.2)
P 0.12

1989 1990
2. I (I1.2-3.9) 0.9 (0.4-1.8)
0.01 0.4

RR relative rate, Cl confidence intervals.
Few PCP events were recorded in this group after I199 I1
The risk of bacterial pneumonia was comparable overall, however, there was a higher risk
among women in 1988 (RR 2.3, 95% CI. I -4.8, p=0.03). No survival differences were evident.
Conclusions: There was an increased risk of PCP among women during the period prophylactic regimens were introduced in Switzerland.The introduction of PCP prophylaxis may
have been more rapid for men than for women, possibly because men were more likely to
be enrolled in clinical trials.
M. Battegay Kantonspital Basel, Medizinische Poliklinik, Petersgraben 4, 4031 Basel
Telephone [phone]
Mo.B. 1345
A PROSPECTIVE STUDY OF IMMUNOLOGICAL AND VIROLOGICAL PROGRESSION
MARKERS IN HIV DISEASE
Karin S Froebel, C D'Alessandro, G Raab*, E McKenziev, and R P Brettle"m*. Centre for HIV
Research, University of Edinburgh, *Dept of Mathematics, Napier University; fHDept of
Infectious Diseases, City Hospital, Edinburgh, UK.
Objective: I.To determine immunological and/or virological markers which predict clinical
progression to AIDS, additional to the CD4 count. 2.To validate (or riot) reports of lymphocyte proliferation and virus phenotype as predictors of disease progression.
Methods: 212 HIV+ patients (I 22 men, 90 women) at WHO disease stage I - III, with a
known date of sero-conversion, and who attend an out-patient clinic regularly were
enrolled into a pilot study f-rom August 1993-July 1994, and again into a 30 month study
from September I1995.The majority (83%) were infected by intravenous drug injection.
Blood samples are taken 3-monthly and analysed for expression of CD38/CD45RO and
CD28 markers on CD3/CD8 cells, lymphocyte proliferation to anti-CD3+anti-CD28 and to
pokeweed mitogen (PWM), and for the presence of syncytium inducing (SI) virus.
Results: In the 2 years from the start of the pilot study 29 patients progressed to WHO
disease stage IV (AIDS), and 33 patients died, some from non-AIDS causes. Preliminary
cross sectional analysis of the pilot study results showed an increase in the proportion of
patients unresponsive to stimulation with anti-CD3+anti-CD28, or PWM, with advancing
disease stage, and an increase in mean levels of CD8/CD38/CD45RO and CD3/CD8/CD28
subpopulations in patients at WHO disease stages Ill and IV. Patients with low CD4 counts
(<200) can have either normal or increased levels of CD8 subpopulations.There was no
difference in the mean levels of CD8 subpopulations in patients with SI as opposed to nonsyncytium inducing (NSI) virus.
Conclusions: Considerable numbers of patients in the Edinburgh cohort are now progressing to AIDS.The cohort therefore provides an opportunity to study laboratory markers of
disease progression. Preliminary results suggest that proliferative responsiveness and lymphocyte surface markers, other than the CD4 marker, will segregate groups of patients.
Further longitudinal analysis is required, and will be performed to determine whether the
markers predict clinical progression.
K S Froebel, Centre for HIV Research, University of Edinburgh, Kings Buildings, Edinburgh,
EH93JN, UK.Te: ([phone], fax: [phone], email: [email protected]
Mo.B. 1346
CORRELATION OF IN-HOSPITAL MORTALITY WITH CD4 COUNT AND ACUTE
PHYSIOLOGY SCORE IN HOSPITALIZED PATIENTS WITH HIV INFECTION
Afessa, Bekele. University Medical Center, University of Florida Health Science Center,
Jacksonville, Florida
Objective: The Acute Physiology and Chronic Health Evaluation (APACHE) II system is
used to predict the outcome of patients admitted to intensive care units.The acute physiology score (APS) is a component of the APACHE II and is based on vital signs, Glasgow
coma score and selected laboratory values. We undertook this study to determine the correlation of in-hospital mortality with CD4 lymphocyte count and APS of hospitalized, HIV
infected patients.
Methods: Two hundred and twenty-three patients with HIV infection hospitalized in the 9 -month period between I April and 31 December, 1995 were included in this prospective,
observational study Data collected included age, sex, race, risk factor for HIV infection, the
presence of AIDS and the CD4 count during or within 6 months prior to hospitalization.
The APS was calculated from the worst vital sign, Glasgow coma score and selected laboratory values obtained within 24 hours of hospital admission. In hospital mortality was noted.
Student's t, Mann-Whitney U, chi square and Fisher's exact tests were used for statistical
comparisons between groups. P values <.05 were considered significant.
Results: Sixty-eight percent (I 51/223) of the subjects were male; 83% (I185) were African
American, 17% (37) were Caucasian and I was Hispanic.The main known risk factors for
HIV infection were injection drug use (92) and homosexuality (39). Subjects' mean (~SD)
age was 37.7~9.5 years.Two hundred of the subjects had known AIDS indicator conditions.
The in-hospital mortality of the study subjects was 9% (20/223).There was no significant
difference in age, race and sex between survivors and non-survivors.The median CD4
count of survivors was 40/pL compared vs 20/pL of non-survivors (p--.3179).The median
APS of survivors was 9 compared to 20.5 of non-survivors (p<.0001).
Conclusion: Although the CD4 count is a strong predictor of long term survival, the acute
physiology score correlates better with in-hospital mortality of hospitalized patients with
HIV infection.
Bekele Afessa, MD, 655 W 8th street, Jacksonville, FL 32209, USA Telephone: [phone]
Fax: [phone]

Mo.B. 1347
IMMUNE-COMPLEX DISSOCIATED (ICD) P24 ANTIGEN (AG) IN HIV-INFECTED
CHILDREN FROM THE NICHD IVIG CLINICAL TRIAL: DESCRIPTIVE ANALYSIS AND
CORRELATION WITH MORTALITY AND RNA COPY NUMBER
Rich Kennethh,SMofensc,,on, Korelitz Jt, Meyer W~, Moye Jt, Pahwa S**, Bethel) J., Nugent
Rt. *U of Illinois. Chicaio, IL; tNICHD-NIH, Bethesda, MD: tWestat, Rockville, MD; ~MD
Medical MetPath, Baltimore, MD;**North Shore Hosp, Manhasset, NY
Objective: To evaluate ICD p24 Ag levels in HIV-infected children with mild-moderate HIV
disease prospectively followed in the NICHD IVIG Clinical Trial.
Methods: Blood was collected for central repository storage at entry and 3 month intervals
during the trial. Frozen specimens from children with >3 samples available were retrieved
and tested by ICD p24 Ag assay (HIVAG~, Abbott).
Results: 184 children met the requirements for inclusion.The majority (92%) acquired infection perinatally; mean age at entry was 43 months, mean entry CD4 count 1107, and mean
follow-up was 14.3 months. 946 samples were available for testing (mean. 5 samples/child).
At entry mean ICD p24 Ag level was I 202 pg/ml and median was 366 pg/ml, with values
ranging from non-detectable to I 5,640 pg/ml; entry ICD p24 Ag was 100 pg/ml in 28.3%
and <1000 pg/ml in 65.8% of samples; only I. I% of samples were> 10,000 pg/ml. 16 of the
184 children died during the course of the study (8.7%). Among the 31 children with entry
ICD p24 Ag levels 50 pg/ml, there were no deaths, while 15 of I 6 deaths occurred among
children with entry CD p24 Ag level > 100 pg/ml: using entry ICD p24 Ag cutpoints of 0 -10; I 1-50; 5 1-100; 101-1000 and > 1000 pg/m, the number of deaths were 0/8 (0%); 0/23
(0%); 1/21 (4.8%); 8/69 (I 11.6%): and 7/63 (I I. 1%), respectively (X2 test for trend, p=0.056).
A moderate correlation was seen between entry levels of RNA copy number (NASBA
assay Organon Technika) and ICD p24 Ag (Pearson correlation coefficient [r] -0.36): this
correlation was strongest among children with entry CD4 > 500 (r = 0.4 I) than those
with CD4 0- 199 (r - 0.20) and 200-499 (r - 0.16).
Conclusions: Higher entry ICD p24 Ag levels, particularly > 100 pg/nl, showed a correlation with mortality risk. ICD p24 Ag levels show only a moderate correlation with RNA
copy number, and correlation is strongest among children with higher CD4 count.
Kenneth C. Rich, MD, University of Illinois at Chicago, Department of Pediatrics. M/C 856,
840 S. Wood St., Chicago, IL 60612-7324.Tel [phone]. Fax (3 I 2) 413-8694.
Mo.B. 1348
EVALUATION AND CORRELATION OF VIROLOGICAL AND IMMUNOLOGICAL
PARAMETERS FOR THE FOLLOW UP OF HIV INFECTED AND TREATED PATIENTS
V. Calvez, G. Carcelain, R. Tubiana, N. Ktorza, C. Katlama, B. Autran, JM Huraux, H. Agut.
Laboratoire de Virologie, CNRS EP 57, H6pital Pitie-Salpetriere, 83 Bd de l'H6pital, 75013
Paris, France.
Objectives: To value virological and immunological parameters for the follow up of asymptomatic HIV I infected and treated patients with CD4 cells counts (41- I 504 cells/mm3).
Methods: Fourty subjects were included.Twenty one were treatment naive and had been
treated by ZDV. Nineteen were pretreated by ZDV and had been treated by 3TC or d4T
Fourteen subjects had received an additional antiretroviral drug at different times of the
study Virological and immunological evaluations were performed at entry just before starting therapy (Day 0) and at different times (M I, M4, M8).The following virological procedures was performed: virus isolation, quantitative cell viremia by culture, quantitative plasma
viremia by culture, p24 antigen in association with some immunological measurements such
as CD4+ and CD4+CD25+ cells count.
Results: For the baselines at DO, the CD4 cells count and the cellular viral load were higher
in the naive patients tha in the pretreated patients (p<0.05 in the two cases). Cell viremia
and plasma titers exhibited significant decreases only in the naive patients group at M I. M4
and M8 (P<0.05 in the three cases). A significant difference of the cell viremia between DO
and M8 was observed in the patients who had received an additional antiretroviral therapy
but not in the monotherapy patients group (p<0.05) and no significant difference was
observed in this case with the quantitative plasma viremia. No significant difference was
observed between DO and M I, M4 and M8 for the P24 antigen. Changes of both cell
viremia and plasma viremia were significantly correlated (p<0.5) but they were not correlated with the changes of P24 antigen. Some significant differences was observed in the
CD4+CD25+ cells count but not with the CD4 cells count.We are performing HIV RNA
plasma NASBA to study the correlations to the other mesurements.
Conclusions: Cell culture viremia by culture is a good marker for the follow up of HIV I
treated patients but plasma culture viremia by culture is less efficient for the monitoring of
therapy changes.
Dr Vincent CALVEZ, MD, Laboratoire de Virologie - CERVI - GH Pitie-Salpetriere - 83 Bd
de I'H6pital - 75651 PARIS Cedex 13-FRANCE Telephone: [phone]
Fax: [phone]. II
Mo.B. 1349
HEPATITIS BVIRUS MARKERS AS SURROGATES FOR HIGH RISK SEXUAL
BEHAVIOUR ASSESSMENT IN HIV INFECTION IN THE TROPICS
Mingle ulius A.A./, Sagoe K.W.C. I, Blessie V.2. I Ghana Medical School, Accra, Ghana;
2Public Health Ref Laboratory Accra, Ghana.
Objective:To determine whether Hepatitis B virus markers can be used as surrogates in
place of self assessment of high risk sexual behaviours in HIV epidemiology in a tropical
country like Ghana.
Methods Twenty HIV positive and 15 HIV negative samples were randomly selected from
an ongoing studies on dual infections. Hepatitis B markers namely HBs Ag, HBs Ab. HBc Ab
were determined using beads coated with anti HBs, HBs Ag and HBc Ag respectively in the
sandwich enzyme immunoassay procedure. All the tests were performed as directed by the
manufactures (General Biological Corp.Taiwan).
Results: Seven (7) of the 20 HIV positive (7/20). and six of HIV negative (6/I 5) were reactive in the HBs Ag determination. 13 of HIV positive (I 5/20) and 4 of HIV negative (5/I 5)
were reactive for HB Ag and 3 positive and 2 negative HIV were reactive for HBc Ab.
Analysis of the results in terms of army Hepatitis B marker showed higher refrequency in HIV
positives than HIV negative samples i.e. 16 (I 6/20) for HIV positive and (8/I 5) for HIV negatives. Statistically the results showed significant results of HBs Ab. and also using any HB
marker but not significant for HBs Ag marken

115