Bridging the Gap: Conference Record [Abstract book, International Conference on AIDS (12th: 1998: Geneva, Switzerland)]

300 Abstracts 22188-22191

12th World AIDS Conference

22188 Validation of a multisystem organ failure (MSOF) score
for pneumocystis carinii pneumonia (PCP) and acute
respiratory failure (ARF)
David Forrest1, O. Djurdjev2, C. Zala3, J. Russel', J.S.G. Montaner2. 1Rm 667,
St Paul's Hospital, 1081 Burrard Street Vancouver, British Columbia; 2Canadian
HIV Trials Network Vancouver BC, Canada; 3Unibersidad de Buenos Aires,
Buenos Aires, Argentina
Objectives: To validate a modified MSOF score previously developed at our institution with and without addition of lactate dehydrogenase (LDH) as a predictor
of survival to hospital discharge in patients with AIDS-related PCP and ARF.
Methods: A retrospective chart review was conducted of patients with PCP and
ARF admitted to ICU between April 1, 1991 and September 30, 1996. Age, gender, Acute Physiology And Chronic Health Evaluation (APACHE II) score, Acute
Lung Injury (ALI) score, AIDS score, and LDH on admission were recorded.
Patients were included who had confirmed and presumptive PCP and who were
and were not mechanically ventilated. ARF was defined as PaO2/FiO2 <150.
Measurements were compared between survivors and non survivors. Patients
were stratified by LDH < or >2000 U/L. Receiver operating characteristic (ROC)
curves were constructed for each score ~ LDH. The MSOF score ~ LDH was
assessed further as predictive of outcome by logistic regression.
Results: There were 40 cases of ARF with mortality of 53%. 98% were male
and 63% were ventilated. Although AIDS and ALI scores were similar between
original and validation populations, the patient groups studied were different in
other respects (ventilation rate, use of adjunctive steroids, distribution of APACHE
II and MSOF scores). APACHE II and MSOF scores were predictive of mortality,
while other scores and LDH were not predictive of outcome. Addition of LDH
enhanced predictability of all scores. Results were unchanged when analysis was
restricted to patients who were ventilated and those in whom the diagnosis of
PCP was confirmed. ROC curves showed the MSOF and APACHE II scores were
superior to other scores and LDH, but there was no significant difference between
the areas under the ROC curves. The area under all ROC curves was enhanced
by addition of LDH. With the MSOF score ~ LDH included in a multivariate
analysis, no other score contributed to prediction of outcome.
Conclusions: Our results successfully validate the modified MSOF score as an
effective predictor of survival to hospital discharge of patients with AIDS-related
PCP and acute respiratory failure. We have also confirmed that performance
of the score is enhanced by the addition of LDH. The consistent performance,
simplicity and clinical relevance of the modified MSOF score supports its use in
prognostication for such patients.
22189 Randomized, double-blind, placebo-controlled trial of
erythromycin versus ciprofloxacin for the treatment of
chancroid
Stephen Moses', I.M. Malonza2, M. Tyndall3, J. Perriens4, F.A. Plummer5,
A.R. Ronald5, J.O. Ndinya-Achola2. 1Dept. of Medical Microbiology, University of
Manitoba, 730 William Ave., Winnipeg, MB, R3E OW3, Canada; 2University of
Nairobi, Nairobi, Kenya; 3McMaster University Hamilton, ON, Canada; 4U.S.
Joint Program on AIDS, Geneva, Switzerland; 5University of Manitoba,
Winnipeg, MB, Canada
Background and Objectives: Limited data are available on effective treatment
for chancroid, particularly the efficacy of single-dose regimens and response to
treatment in HIV-positive individuals. Effective chancroid treatment is particularly
important in view of its facilitating effect on HIV transmission. We examined the
efficacy and safety of two WHO-recommended treatment regimens in HIV infected
and uninfected individuals.
Methods: 245 men and women with clinically suspected chancroid presenting
to an STD clinic in Nairobi, Kenya, were randomized to receive either oral erythromycin, 500 mg. tid for 7 days (E), or single-dose oral ciprofloxacin, 500 mg.
(C). Drug administration was double-blind and placebo-controlled. Each patient
was evaluated microbiologically, and was followed weekly until cured (clinical
cure defined as a fully healed ulcer).
Results: The mean patient age was 28.2 years, 85% were male and 35%
were HIV-1 seropositive. From culture, serology and PCR results, etiological
diagnoses were: any H. ducreyi (HD) 45%; any T pallidum (TP) 34%; and any
H. simplex (HS) 20%. There were 13% mixed infections, and 16% with no discernible etiology. After initial treatment, 32 (13%) were lost to follow-up. Of 212
evaluable patients, 179 (84.4%) were either cured or improved at last follow-up.
The cure rate (including those improved at last follow-up) in the C group was
80.8% (84/104) and in the E group 88.0% (95/108), P = 0.19. Most treatment
failures occurred in patients with TP or HS infection. In 91 patients with confirmed
HD infection, the cure rates were 91.2% (44/48) in the C group and 90.7% (39/43)
in the E group, P = 1.0, and 3 of the 4 failures in the C group were co-infections
with TP. Among 23 HIV-1 seropositive patients with confirmed HD infection, the
cure rates were 80.0% (8/10) in the C group and 100% (13/13) in the E group, P
= 0.17. Both failures in the C group were co-infections with TP.
Conclusion: Both the ciprofloxacin and erythromycin regimens appear to be
effective for the treatment of chancroid in both HIV infected and uninfected in
dividuals. T pallidum and H. simplex were etiologic agents in large numbers of
clinically suspected chancroid ulcers, at times in combination with H. ducreyi. The
presence of these agents accounted for the majority of ulcer treatment failures,
reinforcing the importance of the syndromic management of genital ulcer disease,
even by experienced clinicians.

S22190 Declines in syphilis prevalence in pregnancy and in STD
syndromes among clinic attenders at five primary health
care centres in Nairobi, Kenya, 1993 to 1997
Stephen Moses', E.N. Ngugi3, D.J. Jackson3, A. Gikuni3, L. Gathingu3,
R.C. Brunham2, F.A. Plummer2. 1Dept. of Medical Microbiology, University of
Manitoba, 730 William Ave., Winnipeg, MB; 2University of Manitoba, Winnepeg,
MB, Canada; 3University of Nairobi, Nairobi, Kenya
Background and Objectives: A decentralized STD/HIV prevention and control
program was developed in 1991/92 in Nairobi, Kenya. By 1993, staff at 5 municipal health centres were trained to diagnose and treat STDs using a syndromic
approach, provide STD/HIV counselling, and screen for and treat syphilis in pregnancy on site. Antenatal clinic attenders were screened by RPR test, treated at
the same visit if positive, and advised to refer their partners for treatment (overall,
66% of partners were treated). We monitored program outcomes.
Methods: Two indicators of program effectiveness were examined: number of
patients presenting with defined STD syndromes over time; and syphilis prevalence over time among antenatal clinic attenders.
Results:
1993 1994 1995 1996 1997 % Reduction
SIP 8.3% (654) 7.2% (533) 6.2% (656) 5.3% (743) 4.1% (414) 51%
GUF 1237 1429 1094 771 761 38%
GUM 2139 1673 1417 953 678 68%
PID 1625 1598 1420 1171 1074 34%
UD 6329 5346 3927 2438 1673 74%
VD 7695 7302 6643 5660 4493 42%
(SIP = syphilis in pregnancy, percent positive (number positive); GUF = genital ulcer in females;
GUM = genital ulcer in males; PID = pelvic inflammatory disease; UD = urethral discharge; VD
= vaginal discharge)
Conclusion: The program of improved STD management, STD/HIV counselling, and screening for and treating syphilis in pregnancy, was paralleled by
declines in syphilis prevalence in pregnancy (P < 0.001, x2 test for trend), and in
clinic attendance for STD syndromes. Changes in health care seeking behaviour
or effects of other unrelated interventions could also account for these trends,
but there is no evidence to suggest that either was the case. These findings
demonstrate that sustained changes in STD rates are achievable in the Kenyan
context.
122191 Prevalence and clinical expression of herpes simplex
virus infections among women enrolled in the Women's
Interagency HIV Study (WIHS), United States
Ruth Greenblatt', R. Ashley2, N. Ameli3, M. Augenbraun4, M. Young5,
D. Burns6, M. Cohen7. 1 UCSF Box 1352 Dept. Infectious Dis., San Francisco,
California 94143; 2Univ. Washington, Seattle, WA; 3Univ. California, San
Francisco, San Francisco, CA; 4State Univ. New York Brooklyn, Brooklyn, NY;
5Georgetown University, Washington DC; 6Nat. Instit. Child Health Human
Develop, Bethesda, MD; 7Cook County Hospital, Chicago, IL, USA
Objectives: To determine the prevalence of infection with herpes simplex viruses
(HSV) type 1 and 2 and self-reported genital herpes among women with HIV
infection, or those at high risk for HIV.
Design: Longitudinal cohort study based in 5 metropolitan areas in the United
States.
Methods: 2,058 HIV-infected and 568 uninfected WIHS participants completed
medical and gynecological histories, physical examination and phlebotomy. Sera
were tested for HSV type-specific antibodies using gG-1 and gG-2 EIAs. Sera
with high negative or equivocal results in either EIA were tested by HSV Western
blot. We report on data from 615 HIV-infected participants with HSV serological
results in this abstract. The remainder of the cohort is currently being tested.
Results: Antibodies to HSV-1 and HSV-2 were detected in 526 (86%) and 482
(78%) of HIV-Infected participants respectively. Of the 482 HSV-2 seropositive
women, 28% reported they had genital herpes; and 23% reported having a genital
sore during the preceding 6 months. In multivariate logistic regression analysis,
HSV-2 antibody was associated with greater age (OR = 1.47, 95% confidence
intervals = 1.09-2.00), 25-100 lifetime sexual partners (OR = 3.90, 1.23-12.44),
>100 sexual partners (OR = 7.7, 1.9-31.01), African-American ethnicity (OR =
2.80, 1.83-4.29), less than 12 years education (1.68, 1.07-2.65), and income
under $18,000 (OR = 1.95, 1.22-3.11), but not with recent number of sexual
partners, history of genital herpes, or history of genital sore. HSV-1 antibody was
associated with maternal education attainment of less than 12 years (OR = 3.1,
1.62-5.95) and participant's income of less than $18,000 (OR = 2.17, 1.19-3.96),
but not participant's educational attainment, number of sexual partners, ethnicity
or age.
Conclusion: Herpes simplex virus infections are highly prevalent among
women living with HIV infection. Infection with HSV-2 infection is influenced
by long-term sexual behavior, socio-economic status, and age. The majority
of HSV-2 infected women appear to be unaware of this infection, despite the
potential benefits of recognition and treatment.