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

XIV International AIDS Conference

Abstracts WePeC6236-WePeC6239 139

to 28 Wks. Viraemia (VL), cell-associated HIV DNA and mRNA, T cell subsets
and CMV markers (CMV IgM, IgG and DNA) were determined.
Results: Of 125 pts studied, 79 were CMV IgM or DNA +ve and 46 were CMV IgM
-ve (38/46 IgM -ve/IgG +ve and 8/46 IgM -ve/IgG -ve). CMV +ve pts at BL had
higher median [IOR] CD8 and CD38+ counts (1279 [776,1799] and 353 [169,649]
cells/mm3, respectively) than CMV -ve pts (850 [484,1127] and 583 [562,1058],
cells/mm3, respectively; p<0.005 and p<0.05 [MW]), whereas CD4 counts were
similar (528 [394,667] vs 472 [419,676], respectively). CD8 and CD38 at BL appeared related to the degree of CMV infection, levels being higher in the 38 IgM
-ve/IgG +ve pts than the 8 IgM -ve/IgG -ve pts (871 vs 694 for CD8; 374 vs 169
for CD38).
Time to VL<50 c/mL tended to be longer in the IgM +ve (or DNA +ve) vs IgM
-ve pts (Cox); hazard ratio (95% CI) = 0.69 (0.46,1.02); p=0.065, adjusted for BL
VL. Median log HIV DNA (IQR) at Wk 28 was higher in CMV IgM +ve/DNA +ve
vs IgM -ve pts [1.74 (1.51,2.11) vs 1.38 (0.90,1.81), respectively; p<0.05, MW].
Median log mRNA (IQR) changes from BL were less marked in the CMV IgM
+ve/DNA +ve group than the IgM -ve group (-2.23 [-3.08,-1.65] vs -3.42 [-4.19,2.14]; p<0.05, MW). CD4 counts were similar in the +ve and -ve groups at 28
Wks (676 [532,860] vs (706 [477, 881]).
Conclusions: CMV co-infection during HIV seroconversion is associated with
higher pre-treatment CD8 and CD38 activation that may reflect a cellular
anamnestic or primary CMV response. CMV reactivation also appears to delay
HAART response.
Presenting author: Li-Ean Goh, GlaxoSmithKline Research and Development,
Greenford, Middlesex, United Kingdom, Tel.: [phone], Fax: +44 208
966 4529, E-mail: [email protected]
WePeC6236 Effect of pregnancy on HIV disease progression
and survival from seroconversion among women
in rural Uganda
L.V.F. Van der Paal, R. Muwonge, B. Mayanja, J.A.G. Whitworth. Medical
Research Council - Uganda Virus Research Institute, PO Box 49, Entebbe,
Uganda
Background: To assess the effect of pregnancy on HIV disease progression in
Ugandan women.
Methods: A clinical cohort was established in 1990 in rural SW Uganda: 82 HIV
incident cases are women in reproductive age (15-49). We compared progression to WHO stage 4, to death, and to CD4 count < 200cells/ml among women
reporting one or more pregnancies after seroconversion and those reporting no
pregnancy.
Results: Rate of progression to WHO stage 4 among non pregnant women was
2.8 compared with 6.0/100 person years of observation (pyo) among those reporting pregnancy. Hazard Ratio (HR) for all pregnant women adjusted for age at
seroconversion was 1.9 (p=0.4) and 4.5 (p=0.08) among those with only one pregnancy and 1.5 (p=0.6) with >1 pregnancy. Median time to death was 8.8 years
for non-pregnant women and 7.6 years for those reporting pregnancy. Those with
only one pregnancy had a lower median survival of 6.1 years compared with 9.7
years among those who had >1 pregnancy (p=0.05). The corresponding age adjusted HR were 3.2 (p=0.06) and 0.7 (p=0.4). HR for CD4 count below 200 cells/ml
was 2.7 (p=0.11) for women reporting pregnancy after seroconversion compared
with the non-pregnant women. No significant difference was observed between
women with one and women with more than one pregnancy
Conclusion: Pregnancy in HIV infected women seems to be associated with
faster progression, and this effect is mainly observed in women with one pregnancy. Apparent slower progression to AIDS and death among women with more
than one pregnancy compared with women with only one pregnancy could be
due to healthy pregnant women effect.
Presenting author: Lieve Van der Paal, PO Box 49, Entebbe, Uganda, E-mail:
[email protected]
WePeC6237 Effect of pregnancy on maternal HIV disease
progression in the Women and Infants
Transmission Study (WITS)
R.C. Hershow1, H. Minkoff2, R. Tuomala3, H. Watts4, M. Frederick5, C. Zorrilla6,
H. Hammill7, J. Pitt8. 1University of Illinois at Chicago, School of Public Health,
Division of Epidemiology and Biostatistics, School of Public Health, Division of
Epidemiology and Biostatistics, University of Illinois at Chicago, 1630 W Taylor
Street, Rm # 987, Chicago, Illinois 60612, United States; 2University of New
York, Health Sciences Center at Brooklyn, New York, United States; 3Brigham
and Woman's Hospital Boston, United States; 4National Institute of Child Health
and Human Development, Bethesda, United States; 5Clinical Trials and Survey
Corporation, Baltimore, United States; 6 University of Puerto Rico, School of
Medicine, San Juan, United States Territory; 7Bay/or College of Medicine,
Houston, United States; 8Columbia University College of Physicians and
Surgeons, New York, United States

Background: It is uncertain whether pregnancy hastens HIV disease progression. We examined the effect of a second pregnancy on disease progression
among North American women enrolled while pregnant in the Women and Infants Transmission Study.
Methods: Disease progression (changes in CD4 cell per cents, plasma HIV RNA,
time to 1st CDC class C diagnosis [excluding CD4 cell count< 200], and death

from any cause) was compared between women who had only one pregnancy
during study enrollment (OP:953) and women who had a second pregnancy
(SP:329) during follow-up. For CD4 and HIV RNA data, growth curve models with
random beta coefficients were used to describe mean trajectories over time. Cox
proportional hazard models were used to assess time to first class C event and
death.
Results: At baseline, the SP group was significantly younger (24.7 vs 27.8 yrs),
less likely to receive anti-retroviral therapy (ART) during the enrollment pregnancy
(62% vs 70%), and had a higher CD4% (29.8 vs 27.6%). After adjusting for baseline CD4% and other important covariates, SP was not significantly associated
with higher mean CD4% over time (+0.65%, p=0.063) or significant differences in
log HIV RNA (- 0.16, p=0.13). In addition, rate of CD4% change over time was
almost identical for the OP and SP groups and no significant differences in HIV
RNA patterns were observed between the groups. During follow-up, 182 women
(SP:44, OP:138) progressed to a first class C diagnosis and 91 (SP:10,OP:81)
died. After adjusting for multiple variables including age and ART, progression to
1st class C event was not significantly greater for the SP group (RH 1.18, 95%
CI, 0.67-2.05). SP was significantly protective against mortality (RH 0.17, 95%
CI, 0.05- 0.56).
Conclusion: These finding suggest that a second pregnancy does not accelerate
HIV disease progression as assessed immunologically, virologically or clinically.
Presenting author: Ronald C. Hershow, School of Public Health, Division of
Epidemiology and Biostatistics, University of Illinois at Chicago, 1630 W. Taylor
Street, Rm # 987, Chicago, Illinois 60612, United States, Tel.: [phone],
Fax: [phone], E-mail: [email protected]
WePeC6238 Gastrointestinal infections in aids patients
D.R Pandit, S. Raina, S. Sangle, R. Bharadwaj. Sassoon General Hospitals,
a-2, vishnu sadashiv campus, opp. zilla parishad, pune-411011, maharashtra,
India
Introduction: Gastrointestinal infections are very common in AIDS patients and
diarrhoea is a common clinical presentation.It has a varied aetiology and therapy
needs to be modified accordingly The present study was undertaken to asses
the relative roles of differrent pathogens in causation of diarrhoea.
Materials and methods: Stool samples of 50 AIDS patients with diarrhoea admitted to Sassoon General Hospital, Pune were collected & examined for bacteria,
parasites, fungi and mycobacteria by standard techniques.
Results: Bacterial pathogens were detected in 30/50 (60%) of patients, parasites
in 14 (28%), fungi in 5 (10%) and mycobacteria in 1 (2%) of patients. E.coli was
the most common of bacteria (40%), 83% being Enteroinvasive E.coli (E.I.E.C.).
Cryptosporidium was the most frequent parasite (12%). Candida albicans was the
only fungal isolate. Weight loss was present in 100% patients with parasitic fungal
and mycobacterial infections but in only 87% patients with bacterial infection. Dehydration was marked and common in mycobacterial (100%), parasitic (90.2%)
and bacterial (90%) infection but in only 40% of patients with fungal infection.
Fever was predominant with mycobacteria (100%) and bacteria (90%) while less
frequent with parasites (55%) and fungi (40%). Patients with parasitic infection
had a higher frequency of motions ("d 5-10 motions per day). In other pathogens,
it was less than 5 motions per day. Higher parasitic load was associated with
increased mortality
Conclusions: AIDS patients can be affected by conventional as well as unconventional pathogens. Hence an E. coli isolate from stool of HIV positive patient
should not be ignored. Typical clinical features can give a clue to the pathogen.
Early identification of etiologic agent of diarrhoea in an AIDS patient and institution of therapy accordingly is very useful for reduction of morbidity and mortality
and improving the quality of life.
Presenting author: dakshayani pandit, a-2,vishnu sadashiv campus, opp. zilla
parishad, pune-411011, maharashtra, India, Tel.: [phone], E-mail: [email protected]
WePeC6239 Effect of HSV1 and 2 on HIV viral load after HIV
seroconversion
A.M. Rompalo1, J. Astemborksi2, J.M. Zenilman 1, T.C. Quinn3, D. Vlahav2.
1Johns Hopkins School of Medicine, 1830 East Monument Street, Baltimore,
Maryland 21287-0003, United States; 2Johns Hopkins School of Public Health,
Baltimore, Maryland, United States; 3Johns Hopkins School of Medicine,
Baltimore, MD, United States
Background: Herpes simplex virus (HSV) has been associated in vitro with increased HIV replication. Prevalent HSV infection has been associated with increased HIV transmission and acquisition, but not with time to AIDS. We examined the effect of prevalent HSV type 1 and 2 on first HIV viral load after documented seroconversion among injection drug users enrolled in the ALIVE study.
Methods: HIV seroconverters were defined as participants with documented negative HIV serology followed by reactive HIV serology within one year. Sera prior to

first reactive HIV test were analyzed for HSV1 and 2 antibodies using a gG-based
type specific assay. HIV-1 viral load levels were quantified by RT PCR (Roche
Amplicor Monitor Assay).
Results: Of the 213 seroconverters who were tested for HSV-1 (166 men, 47
women), 186 were positive (87%). Of the 208 tested for HSV-2 (161 men, 47
women) 137 (66%) were positive. HSV seroprevalence among men was 88% for
HSV-1 and 58% for HSV-2. Among women, 85% were HSV-1 infected and 94%