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

486 Abstracts ThPeC7608-ThPeC7611

XIV International AIDS Conference

two infections interact is important for their control. We conducted several studies
aimed at investigating the presence of a potential association between HIV and
malaria.
Methods: During the period 1996-2000, we conducted both epidemiological and
immunological studies at Lacor Hospital, North Uganda: an HIV sero-survey
among 699 adult patients admitted to the medical ward; an unmatched hospitalbased case-control study involving 36 clinical malaria cases and 134 controls
among whom the exposure to HIV was investigated; and a immunophenotype
analysis among 23 HIV- Africans and Italians living in North Uganda and 30 HIVAfricans and Italians living in Italy
Results: The sero-survey shows a very high HIV prevalence among patients admitted for malaria (48.8%, 95% CI: 32.9%-64.9%) when compared to that estimated for the general population living in the hospital's catchment area (17.8% in
1996-1997), suggesting the presence of an association between HIV and malaria.
This hypothesis was tested through the case control study, which showed a significant association between HIV and clinical malaria (OR=3.61, 95% Cl: 1.04 -12.52). Finally, immunological results showed a similar pattern of immune activation in healthy African residents, both Italians and Ugandans. This immune
activation disappears in Ugandans living in Italy, suggesting that environmental
factors, such as parasitic infection (e.g., malaria), influence the abnormal immune
activation that has repeatedly been postulated to be involved in the pathogenesis
of African HIV infection.
Conclusions: The combined results suggest an association between HIV and
malaria that could be very important for public health in sub-Saharan Africa. The
role of other concurrent infections highly diffused in this area should also be investigated.
Presenting author: Silvia Declich, Lab. Epidemiology and Biostatistics, Istituto Superiore di Sanit&, Viale regina Elena 299, 00161 - Rome, Italy, Tel.:
[phone], Fax: [phone], E-mail: [email protected]
ThPeC7608 Impact of CCR5 A32/+ mutation on herpes zoster
among HIV-1 seropositive homosexual men
R.PRM. Lensen', A. Krol', J. Veenstra', R.A. Coutinho', H. Schuitemaker2.
'Municipal Health Service, Postbox 2200, 100CE Amsterdam, The
Netherlands; 2Red Cross Transfusion centre, Amsterdam, The Netherlands
Background: CCR5 is a receptor for chemokines of the CC family Heterozygosity for CCR5 A32, a 32-basepair deletion in the gene encoding for CCR5, is
present in 10-15% of Caucasians and delays HIV-1 disease progression among
homosexual men. A study among 157 normal blood donors demonstrated that of
CCR5 A32 heterozygotes are 9.2 times more likely to be seronegative for varicella zoster virus than CCR5 wild type (WT) individuals. We hypothesise that the
presence of CCR5 A32 may affect the incidence of herpes zoster among HIV-1
seropositive homosexual men.
Methods: 306 HIV-1 seropositive homosexual participants in the Amsterdam Cohort Study were tested for their CCR5 A32 genotype. Herpes zoster was defined
by its clinical presentation. Incidence rates and rate ratios were calculated using
Poisson regression (total follow time: 1444 years). We adjusted for age and CD4+
cell counts.
Results: Median follow up time was 4.2 years. Of the 306 participants, 65 (21%)
were CCR5 A32 heterozygotes and 15 (23%) of these experienced a first episode
of herpes zoster. Of the 241 CCR5 WT individuals, 57 (24%) experienced a first
episode of herpes zoster. The incidence of herpes zoster among CCR5 A32 heterozygotes was 3.85%/year and 4.97%/year among CCR5 WT individuals (rate
ratio: 0.78, 95% CI 0.4-1.4). After adjustment for age and CD4+ cell count, the
rate ratio changed to 0.84 (95% Cl 0.5-1.5). At the first episode of herpes zoster
the mean CD4+ cell count for CCR5 A32 heterozygotes was 437x106/I and for
CCR5 WT individuals 357x1 06/I (p=0.1).
Conclusions: In HIV-1 seropositive homosexual men, a CCR5 A32 heterozygous
genotype has a minor not significant effect on the incidence of herpes zoster. We
think that this result has to be confirmed by larger studies.
Presenting author: Ruud Lensen, Postbox 2200, 10000E Amsterdam, The
Netherlands, Tel.: [phone], Fax: [phone], E-mail: rlensen @gggd.
amsterdam.nl
ThPeC7609 Outcome and viral excretion of measles in
HIV-infected children in Uganda
R.T. Perry1, I. Lubega2, M. Nanyunja3, R. Downing4, J. Mermin4,
M. Grabowsky5, R. Harpaz5, G. Mukasa2. 'Centers for Disease Control and
Prevention (CDC), Atlanta, United States; 2 Dept of Paediatrics and Child Health,
Makerere University Medical School, CDC, 1600 Clifton Rd, NE MS E-61,
Atlanta, GA 30333, Uganda; 3Uganda Virus Research Institute, Entebbe,
Uganda; 4 CDC, Entebbe, Uganda; 5 CDC, Atlanta, United States
Background: In Sub-Saharan Africa both measles and HIV are major causes of
childhood morbidity and mortality. This study examines the outcomes of measles

and the duration of measles viral excretion in children by HIV status.
Methods: Two groups of children 6 months to 5 years presenting to Mulago Hospital in Kampala, Uganda were studied. Children <10 days from the onset of
a rash typical of measles (acute measles) were enrolled and followed prospectively to determine measles outcomes and duration of excretion of measles virus.
Also, children with a history of measles in the past 6 months (post measles) were
enrolled and studied cross-sectionally for excretion of measles virus. At enrol
ment and follow-up, all children were examined, blood drawn for HIV and measles
serologies, and urine and nasopharyngeal aspirate samples taken for measles
culture. HIV infection in children aged <18 months was confirmed using HIV RTPCR.
Results: From September 2000 to August 2001, 668 children were enrolled and
329 were included in the analysis. HIV was found in 13 (6%) of 212 acute cases
and in 13 (11%) of 117 post cases. HIV infected children with acute measles had
a higher case fatality rate, 31% (4/13), compared to 10% (21/199) in HIV uninfected children (p=0.05). Measles complication rates were similar by HIV status.
Measles cultures were positive in 108 of 196 children (55%) enrolled 0-7 days, in
5 of 36 (14%) enrolled 8-30 days, and in none of 97 (95% Cl: 0 - 3.7%) enrolled
>30 days after rash onset, with no significant differences by HIV status. Two children, one HIV infected and one HIV uninfected, had positive measles cultures
both in the first week and when seen 1-2 months after rash onset, then negative
cultures 2-3 months after rash.
Conclusions: In this preliminary analysis, measles infection is more likely to be
fatal in children with HIV compared to those without. We detected no excretion of
measles in children >2 months after rash regardless of HIV status.
Presenting author: Robert Perry, CDC, 1600 Clifton Rd, NE MS E-61, Atlanta,
GA 30333, United States, Tel.: [phone], Fax: [phone], Email: [email protected]
ThPeC7610 Vaginal infections in HIV+ and HIV- women, Rakai,
Uganda
M.P. Meehan', S. Aluma2, D. Serwadda3, R.H. Gray4, M.J. Wawerl,
K. Kasule. Columbia University New York, NY, United States; 2Rakai Project,
Kalisizo, Uganda; 3Makerere University Kampala, Uganda; 4Johns Hopkins
University, Baltimore, United States; 5Rakai Project, Entebbe, Uganda
Vaginal Infections in HIV+ and HIV- Women, Rakai, Uganda
Objective: Human immunodeficiency virus-type 1 (HIV-1) has been associated
with bacterial vaginosis (BV), although little is known about the relationship between the two diseases. We are conducting a prospective study of BV natural
history in HIV+ and HIV- women.
Study design: A cohort of 312 women 13-39 in a rural Ugandan population provide self-collected vaginal swabs for BV assessment at weekly intervals. At baseline, blood was collected for HIV and syphilis testing and self administered vaginal
swabs were collected for trichomonas testing, lactobacillus culture and BV evaluation using the Nugent score. In-depth interviews obtained demographic, sexual
and hygiene practices data, and sources of water. Baseline results are reported
here.
Results: HIV seroprevalence was 11.43% and BV prevalence was 45%. In
HIV+ women BV prevalence was 53% vs. 44% in HIV-(Relative risk (RR)=1.671;
CI=0.75-3.73). HIV+ women were more likely to have severe BV (Nugent score
9-10) than HIV- women (6.25% vs. 5%; RR=1.24; CI=
0.33-4.71). However, Trichomonas vaginalis was higher among HIV- women
(7.05%) than HIV+ (3.13%; RR=1.35; CI=0.97-1.87). Syphilis was significantly
higher in the HIV+ than HIV- (12.5% vs. 1.76%; RR=5.18; CI=2.38-11.27). HIV
positive women were somewhat more likely to be in polygamous marriages in
this cohort (25% vs. 16%; RR=1.09; CI=0.51-2.31),and significantly more likely to
drink alcoholic beverages (34% vs. 12%; RR=2.52; CI=1.28-4.95).
Conclusions: BV is more common in HIV+, whereas Trichomonas is more common in HIV- women. BV may be a consequence of HIV infection.
Presenting author: Mary patricia Meehan, P.RO. box 76, Entebbe, Uganda, Tel.:
[phone], Fax: [phone], E-mail: mmeehan @ rakaiproject.or.ug
ThPeC76l11 Seroprevalence of HIV infection in bad obstetrical
history and its correlation with TORCH and VDRL
M.S. Mathur, M.C. Rele, D. Turbadkar. Department of Microbiology L TM M C
and L TM G H, 25/102, Varun Coop Hsg Soc, Behind Shantivan, New Link Road
Extension, Oshiwara, Andheri( West), Mumbai: 400053, Maharashatra, India
Observation: Determining the prevalence of HIV infection in bad obstetrical history (BOH) and its correlation with TORCH and VDRL.
Toxoplasma, Rubella, Cytomegalovirus, Herpes simplex II (TORCH) group of organisms can cause intrauterine infections in foetus and are responsible for repeated abortions, stillbirths, congenital defects and neonatal deaths.
Method: Over a period of two years (Jan 2000 to Dec 2001) pregnant women with
BOH were screened for TORCH IgM and IgG antibodies using ELISA technique,
HIV antibodies as per NACO guidelines and VDRL test.
Result: Of total 1463 serum samples screened, IgM seropositivity for Rubella was
13.8%(202), Toxoplasma 9.6%(141), Cytomegalovirus 4.4%(65), Herpes simplex type II 4.0% (59); IgG seropositivity for Cytomegalovirus was 63.36%(927),
Rubella 47.98%(702), Toxoplasma 28.5%(417), Herpes simplex type II 32.05%
(469); 31(2.1%) were HIV seropositive and 84(5.7%) were VDRL reactive.
Of the 202 Rubella IgM seropositives, 62(31%) were mixed infections with Toxo
plasma 32(52%), CMV 15(24%), HSV-2 12(19%), one each with Toxoplasma plus
HSV-ll, Toxoplasma plus CMV and Toxoplasma, CMV plus HSV 11.
Of the 31 HIV seropositive cases, 4(12.9%) were also IgM seropositive for CMV,
1(3.2%) each IgM seropositive for HSV II and Toxoplasma, 2(6.5 %) IgM seropositive for CMV plus Rubella and 3(9.67%) were VDRL reactive.
Of the 84 VDRL reactive cases, 9(11%) were also seropositive for Rubella IgM,
4(5%)for HSV II IgM and 3 (3.5%) for Toxoplasma IgM antibodies.