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

Mo.C.334 - Mo.C.342

Monday, July 8, 1996

positive patients.Two methods of duplicate removal were evaluated: 1) a comrputer match
by name/initials, gender and date of birth to a previous positive test and 2) esuts of a pre
vious test reported on the requisition by the health care worker.
Results: Removal of duplicates reduced the number of seropositives to 691 Luphicates
identifed by computer matching and health care worker report could be validated in I100%
and 95% of cases respectively Duplications varied from 13 percent in the first quarter to 27
percent in the third (20 percent overall). No single risk group accounted for the missing
data. Risk information is as follows [raw data/enhanced data]: men having sex with noren
[I 70 (20%)/203 (29%)],; injection drug use (IDU) [253 (29%)/264 (38%) ]; heterosexual [62
(7%)/88 (13%)]; other [68 (8%)/87 (13%)]; not given [315 (36%)/52 (7%)]. Females were
more likely than males to have IDU (48%/38%) or heterosexual contact (27%/ 1I0%) as the
risk factor. Females were also more likely than males to be Native Indian (23%/10%). Peak
age distribution for males was 30 to 39 years and 20 to 29 years for females.
Conclusions: The enhanced data will be suitable for input to a national database of I IV
infection, will provide information for advocacy and commurnity groups to target programs
and for epidemologic surveillance of the epidemic.
Darrel Cook, B. C. Centre for Disease Control, 828 West 10th Avenrrue,Vancouver; BC.,V5Z
I L8, Canada.Telephone [phone] Fax [phone] email: [email protected]
Mo.C.334
OVERVIEW OF THE HIV EPIDEMIC AMONG INJECTING DRUG USERS IN
ENGLAND AND WALES
Madden Peter B, Bennett D, Lamagni T, Brady T, Hughes G, Mortimer J, Kaye P Public Health
Laboratory Service, Communicable Disease Surveillance Centre, London, U.K.
Objective: To assess the state of the HIV epidemic in Injecting Drug Users (IDUs) i
England and Wales.
Methods: Data were obtained from: routine reporting of AIDS cases and HItV infections; the
voluntary unlinked anonymous ItlV and hepatiis B core antibody prevalence monitor ing
programme (Saliva survey); the PHLS Collaborative Laboratory Study o f IIV tests; and laboratory reports of acute hepatitis B.
Results: To the end of December 1995, 4% (438) of AIDS case reports in England and
Wales have been ascribed to injecting drug use. Numbers of AIDS reports in IDUs have
risen steeply since 1990. Approximately 8% (1756/22631) of all HIV infection reports iden
tifed by voluntary confidential testing have been ascribed to injecting drug use. Since 1988,
reports have remained constant at approximately 200 each year Annual HIlV prevalence
rates from the Saliva survey for 1990 through to September 1995 ranged between 0.6% to
1.9% among women, and 0.9% to 2.1 % among men, with no clear temporal trend. Armonsg
I 293 new injectors (clients who began to inject in the last 2 years) there have been only 2
HIV I antibody positive specimens, while hepatitis B core antibody prevalence ranged
between 3% and 19%.The Saliva survey has consistently indicated that approxirmately 20%
of current injectors share needles or syringes, with women and young users being significantly more likely to share. Between 1987 and 1994, 12981 IDUs were tested in the PHLS5
Collaborative laboratory Study of HIV tests. HIV prevalence ranged frorm 1.8% to 3.5%,
with no temporal trend. Laboratory reports of acute hepatitis B in which injecting drug use
was the only identified risk have declined from 527 in 1985 to 67 in 1989 leveling off afther
that with 99 cases reported in 1994.
Conclusion The data suggests relatively low and unchanging HIV prevalence with little evidence of recent transmission. However, the high prevalence of hepatitis B core antibody in
new injectors, the fact that approximately 20% of current injectors report sharing
needles/syringes, and continuing hepatitis B transmissions attributed to injecting, indicates that
IDUs remain behaviourally vulnerable to acquisition of HIV and other blood- borne infections.
PB. Madden, PHLS, CDSC, 6 I Colindale Avenue, London NW 9 Tele: 4 18 1 200 6868
Ext: 4930 Fax: [phone]
Mo.C.335
CASE-CONTROL STUDY OF HIV SEROCONVERSION IN HEALTH CARE WORKERS
AFTER PERCUTANEOUS EXPOSURE TO HIV-INFECTED BLOOD; IMPLICATIONS
FOR POSTEXPOSURE MANAGEMENT
Cardo D, Srvastav P*, Cie sielski C, Culver D, Marcus R, McKibben P, Abiteboul I 1,
Lot F*++, Heptonstall J'^"**, Bell David. CDC, Atlanta, GA, USA; * INRS GERES, Paris,
*** Reseau National de Sante Publique, St. Maurice, France;**' Public He, lth Laboratory
Service, London, UK.
Objective: To ass s risk factors for HIV seroconverson in health care workers (HCWs)
after percutaneous exposure (PE) to HIV-infected blood.
Methods: Cases were HCWs who seroconverted to I-tlV after a PE to HIV-infected blood,
reported from national surveillance in the U.S., France, and United Kingdor. Controls were
HCWs in a prospecte CDC surveillance project who did not seroconvert after PE to
HIV-infected blood. Analysis included variables related to the HCW (age, race, sex, occupation, whether postexposure zidovudine [ZDV] was offered, whether ZDV was used), PE
(needle gauge, glove use, depth of PE, visible blood on device, procedure, ermergency hospital location), and source patient (stage of disease and use of antiviral drugs)
Results: Multivariate analysis of data from 31 cases and 679 controls identifed the following
risk factors: deep injury (odds ratio [OR] 16. I, 95% confidence interval [CI] 6. I-44.6), visi
ble blood on device (OR-5.2; 95% Cl, I.8 17.7), procedure involving needle placed d rectly
in a vein or artery (OR= 5. I; 95% CI, 1.9 14.8), terrinal illness in source patent (OR=6.4;
95% CI, 2.2-18.9). Cases were signifcantly less likely to use ZDV than controls (OR- 0.2;
95% CI, 0. I-0.6). No differences were detected in the rates at which cases and controls
were offered ZDV.
Conclusion: These data suggest that a larger inoculum of blood and terminl illness in the
source patient are risk factors for HIV infection after PE to HIV infected blood. In addition,
the use of ZDV postexposure may be protective for HCWs. Potential limitations of the
study should be considered when interpreting these results.The U.S. Public Health Service
is assessing whether chemoprophylaxis should be recommended after occupational HIV
exposure and if so, how to address practical issues; e.g., need to begin prophylaxis promptly
and increased need to know the source patient's HIV status.An update regarding this
assessment will be presented.
David M. Bell, M.D., 1600 Clifton Road, M/S E 68, Atlanta, GA 30333
Telephone: [phone] Fax: [phone] Email: [email protected]. GOV

Mo.C.340
ASYMPTOMATIC GENITAL TRACT INFECTIONS IN A RURAL DISTRICT OF UGANDA
Paxton L n A lS.r., - no N2,Wawer MJ GrayP R3, Aabwire-Mangen F2. Serwadda
D, McNarirn I), k I JI e j2, Li C3, Kiwanuka N.
Columbia tliuv, NY t BAA Makerere Univ, Kamrpala, Uganda 3Johns Hopktns Univ, Baltimore,
USA, 4Rakai Pr oject, Jganda.
Introduction: Previous research suggests that genital tract infections (GTIs) may be impor
tant in the transmission of HIV. As part of a study to determine if mass treatment of G TIs
will reduce HIV transmission, we documented the prevalence of asymptomatic infections in
a representative sample of the rural population of Rakai District, Utganda.
Methods: All adults aged 15 to 59 residing in 58 communities were eligible for enrollment.
Participants gave blood samples for HIV-1, syphilis (RPR wth TPHA confirmation), and H
dttcreyi serology; urine was tested for C. trochorrrotois and N ponorrhe using ligase chain reac
tion. Self administered vaginal swabs were used to detect T vcogpnis and bacteril vaginosis
(BV). All participants were asked about symptoms of GTI n the prior- six months.
Results: From November 1994 to June 1995, 12823 adults were enrolled (9 I% of the per
manent population). Infection prevalence was high: among men, 16.4% were infected with
HIV, 31% had evidence of exposure to H. ducreyi and 12% had active syphilis.Two percent
of miales aged 15 to 29 were infected with C truchomoatis Testing for N. gonorrr in
progress. Among women, 2 1% had HIV 18% had evidence of exposure to HI duic it, 12.5%
had active syphilis, 54% had BV and 24% had T voagiinas. Four percent of I 5 to 29 year olds
had chlamydia infection. Although almost 65% of women had a vaginal infection at the tine
of interview only 8% reported discharge. Six percent of males and I,% of fermales reported genital ulcer in the prior 6 months.The sensitivity of reported ulcer for serologic syphi
or chancroid was only 46% for men and 39% for women.
Conclusions: In this rn ural cohort with high rates of HIV, most adults had evidence of one
or more potentially treatable GTI yet the correlation between infection and self- reported
symnptoms was poor In populations with high pre valen ce of asymptonmatic G Is ha ty
of infected persons will not be reached by a standard clinic-based approah. To both determine the role of GTIs in HIV transmission and to reduce their prevalence, innovative public
health interventions that take into account asymptomatic infections are essentia.
Lynn A. Paxton, PO. Box 209, Entebbe, Uganda lel: 256 42-20276 Fax: [phone]
e-mail: [email protected]
Mo.C.34 I
IS ASYMPTOMATIC URETHRITIS IN MEN AN OBSTACLE TO EFFECTIVE STD
CONTROL? A COMMUNITY STUDY IN MWANZA,TANZANIA.
Buve A, Mosha F.2'3, Watson Jones D4, Mugeye K3, West B 3J, Gabone R2, Gavyole A
Todd J,4, Hayes R4, Grosskurth H4, Mabey D4, Laga M 1, Mayaud Ph I. Institute of
Tropical Medicine, Antwerp; Belgium; 2. National Institute o Medical Resear ch; 3. AMREF
Mwanza,TTanzania; 4. London School of Hygiene & Tropical Medicine, London, UK; (5)
Regional Medical Officer
Objectives: A community based study carried out in Mwanza Region in 199 I-i 992 found
that at least 6.6% of men in the general population were irfected with N gonorrhoea
and/or C trachomatis and that about 90% of these men were asymptomatic.The objectives
of the present study were to confirm the above findings and (I) to estimate the prevalence
of infections with N gonorrhoea, C trachomatis and T vaginalis in men in the general popu
lation; (2) to assess the proportion of infected men who are asymptomatic and are unikely
to seek care for their urethritis.
Methods: All I,100 men aged 15 to 54 years, living within a radius of one health entre,
were included in the study After giving informed consent the study subjects were inter
viewed on signs and symptoms of STD's, care sought for ST D's and sexual behaviourThey
then submitted a urine sample and were examined, including swabbing of the u rethra
Infection with N gonorrhoea was ascertained by gram stair and culture of the urethral
swab; chlamydial infection was tested for by PCR on the ur ne sample; infection w vith T vag
nalis was established by direct examination of the urine sediment and by culture. All men
were requested to report back to the health centre 2 weeks later, during which visit they
were again interviewed on signs and symptoms of urethritis and they were told their test
results and gen treatment if needed.
Results: By the end of January 996 about 400 men were interviewed and examined and
in March the following data will be available for presentaion: (I) prievalence of in ecton
with N gonorrhoea, C trachomatis andT vaginal,s; (2) % of infected rer who are,symptomat c; (3 of the asymptomatic infected men, the % who took an incomplete treatment and
the % who are still asymptomatic at follow-up 2 weeks late.
Conclusions:The effectiveness of passive case detection and management Or the control of
STD's depends, arnong others, on the awareness of symptoms and the health care seekin
behaviour of infected subjects. A high prevalence of asymptomatic inftions is in obstacle
to effective case detection and management of STD's.
Anne Buve, Institute of iropical Medicine, Nationalestraat I55, B 2000 Antwerp Belg ium
Tel [phone] Fax [phone] e-nmail ABUVE(@MICROBIOI.I TG.BE
Mo.C.342
COST-EFFECTIVENESS OF MANAGEMENT STRATEGIES OF ACUTE URETHRITIS
de Clerk M., Crabbe F.**,Vuylsteke B,**, Laga M.* d * Mddecins Sans Frontees (Doors
Without Borders), Brussels, Belgium; ** Institute ofTropical ledicine, Antwerp, Belyiun
Objective:The most effective antibiotics (3rd generation cephalosporines) for the tiea rr
of gonorrhoea are in most developing countries unaffordable.The obectve o this study was
to evaluate alternative treatment strategies (using a cheaper: less effectve antibiotic irst, or
using Gram stain to confirm the presence of N. gonoarrhoee) for their cost effeciventess
Methods:A computer model has been developed to assess the cost and effectiteness of 3
urethritis management strategies for 1000 male patients: strategy (I) all patients treated
with ceftriaxone; strategy (2) all patients treated with cotrinoxazole or kanartcin, ceftrax
one for patients not responding to the Crst treatment; strategy (3) patients treated with cef
triaxone only if Gram stain is positive. Concurrent treatment with doxycycline (for non
gonococcal urethritis) was included in the 3 strategies. Clinical outcome considered in
model included epididymitis in men, cervicitis and complicateons in female partners. Stratey
effectiveness was proportional to complications prevented in men and women. Strate y
costs included all consultations, lab diagnosis (where applicable) and all treatments.

37