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

306 Abstracts 22217-22222

12th World AIDS Conference

or AmB. The impact of HAAT with PI on candidemia incidence requires further
investigation.
122217 Histoplasmosis survival in French Guiana I
Pierre Couppie1, R. Pradinaud2, M. Sobesky1, E. Ciyti2, D. Sainte Marie2,
C. Azenar3, B. Benoit4. 1 Hopital De Jour CISIH CHG De Cayenne, BP 6006
97306, Cayenne; 2Service De Dermatologie CHG, Cayenne; 3Laboratorie De
Mycologie, CHG, Cayenne; 4CTS CHG De Cayenne, USA
Objectives: To analyse the prognostic factors of histoplasmosis in patients at
AIDS clinical stage in French Guiana.
Design: Prospective study.
Methods: 23 HIV patients with a diagnosis of disseminated histoplasmosis
and receiving antifungical therapies were included in the study; demographic,
clinical, and biological variables at diagnosis of histoplasmosis were gathered.
Kaplan-Meier survival analysis and Logrank-Test were performed to analyse the
prognostic factors of survival, hazard ratios (RH) are presented with their 95%
confidence interval (CI).
Results: 87% of the patients included were men. At initial diagnosis of histoplasmosis mean age was 38 years old and the mean CD4+ cell count was
40.106/1. 65.2% of patients died during follow-up and the median survival was
5.3 months. Survival was shorter in patients with no adenopathy [RH = 8.9; Cl
= 2-40.8] and dyspnea [RH = 4.5; Cl = 1.5-13.9]. High lactico-deshydrogenase
level (>1000 UI/1) was associated with poorer survival [RH = 8.2; Cl = 2.1-32],
patients with low platelets count (<150.000.106/1) had also a increased risk to
die [RH = 8.4; Cl = 2.2-32.5].
Conclusion: Patients with no adenopathy or dyspnea and biological results
such as lactico-deshydrogenase level and platelet count seem to be significant
prognostic factors in histoplasmosis survival. This is important to consider before
initialize an antifungical therapy adapted to the prognostic of histoplasmosis.
22218 Treatment of HIV-related oral candidosis when
fluconazole and itraconazole solution fail
Jonathan David Cartledge1, J. Midgley2, B. Gazzard2. 1Mortimer Market Centre,
Off Capper St., London; 2Chelsea & Westminster Hospital, London, UK
Objectives: To describe the efficacy of different treatment strategies for fluconazole resistant oral candidosis and describe the characteristics of patients with this
condition.
Subjects: A cohort of 100 consecutive HIV-positive patients with fluconazole
unresponsive candidosis.
Methods: Patients provided mouth rinse specimens for Candida isolation, culture and in vitro susceptibility testing. Treatment strategies were prospectively
evaluated and considered successful if no signs of candidosis persisted after 7
days' therapy.
Results: All patients with positive Candida cultures (95 cases) were infected
with isolates resistant to fluconazole in vitro. Median time since first AIDS diagnosis was 12 months, median CD4 count was 9 cells/mm3 and median survival
was 5 months. Response rates to oral azoles as initial treatment for fluconazole
resistant candidosis were poor overall (<30%), except to itraconazole solution
(86%), which was significantly more likely to clear candidosis than all other subsequent treatment strategies except intravenous amphotericin B, unless hepatic
enzyme inducers. In those failing itraconazole solution, treatment was difficult,
though some responded to intravenous amphotericin B (8/14), high dose (>800
mg/day) fluconazole (4/8) or clotrimazole troches (2/2).
Conclusions: The most successful treatment strategy for this cohort of patients with HIV-related fluconazole resistant candidosis was itraconazole solution.
When itraconazole solution failed, amphotericin or high dose fluconazole (>800
mg/day) were each effective in approximately 50% of cases.
122219 Correlation of serum fluconazole (FLU) levels with
susceptibility testing and clinical response to treatment
of oral thrush
Janine Maenza1, W.G. Merz2, M.J. Rom Agnoli2, J.C. Keruly2, R.D. Moore2,
J.E. Gallant2. ' 1830 E. Monument Street #8074 Baltimore, MD; 2Johns Hopkins
University, Baltimore MD, USA
Background: The objective of this study was to correlate clinical response to
FLU treatment of oral thrush with in vitro susceptibility testing and with serum
FLU levels. (Data on susceptibility testing were previously reported. These data
describe serum drug levels.)
Methods: Oral fungal cultures were obtained from 105 HIV-infected patients
who presented with thrush. Cultures were repeated during and/or following treatment at 2 and 4 weeks after baseline culture. Serum azole levels were obtained
at each timepoint from patients treated with an oral azole.
Results: Among the 14 patients who began a new course of FLU during the
study, the median prescribed dose was 100 mg/d (range 100-200 mg/d) and

baseline cultures showed C. albicans with a median FLU MIC of 0.5 pg/ml.
Follow-up of these patients during treatment showed: the median FLU level was 5
/pg/ml; 10/14 patients had resolution of thrush (7 with negative cultures and 3 with
clearance of C. albicans, but with persistent or new non-C. albicans species); 3/14
had persistent thrush with cultures showing organisms with FLU MICs higher than
their serum FLU level; and 1/14 had an exam clinically consistent with thrush,
but with a negative culture (and FLU level of 8.1 /pg/ml). There were 16 patients

taking FLU at study entry (i.e. patients with thrush while on FLU therapy). In this
group the median FLU dose was 100 mg/d (range 100-800 mg/d); cultures all
showed C. albicans, with a median FLU MIC of 64 pg/ml; and the median serum
FLU level was 6.9 pg/ml.
Conclusions: Among patients beginning a course of FLU most initial cultures
showed a low FLU MIC, the patients responded to treatment, and the serum
FLU levels at follow-up were well above initial FLU MICs. Serum FLU levels are
not high enough on standard doses of FLU (100-200 mg/d) to manage disease
caused by organisms with decreased in vitro susceptibility to this agent.
22220 Coccidioidomycosis: An environmental risk for HIV
patients living in an endemic region?
Naomi E. Aronson1'2, L.B. McNicol3. 18011 Kentbury DR, Bethesda MD 20814;
2 Walter Reed Army Medical Center, Washington, DC; 3 William Beaumont Army
Medical Center, El Paso, TX, USA
Background: Active coccidioidomycosis is recognized as an opportunistic infection in HIV patients living in coccidioidal endemic areas. Using a military population
with significant outdoor occupational exposure in the southwest United States, we
plan to determine the incidence of active coccidioidal infection, and to assess the
routine diagnostic modalities of serial chest radiograph, coccidioidal serologies,
and spherulin skin testing in HIV patients.
Methods: This is a prospective cohort study of 78 HIV infected persons with
periodic evaluations every 6-12 months which included an interview, examination, T cell subset analysis, coccidioidal serologies, spherulin 1:100 skin test, and
chest radiograph.
Results: 68 males, 10 females were followed with a median time in study of
47 months (range 1-120). The racial distribution is 39% black, 34% caucasian,
23% hispanic, 3% native american. The median time living in an endemic area
is 60 months (range 1-612). 22% of this cohort died during the study. No active
coccidioidomycosis was identified. Median CD4 count at entry is 403 (range
28-1678), median latest CD4 count is 263 (range 2-1163). 18% had a positive
coccidioidal skin test on entry, 25% of the remainder converted their spherulin
skin test during the study. 23% demonstrated anergy to non-coccidioidal skin test
antigens. 5% had reactive immunodiffusion tube precipitin serologies and 5% had
coccidioidal complement fixation titers of 1:2 to 1:4.
Conclusion: 40% of this HIV cohort had positive spherulin skin tests suggesting
infection with Coccidioides immitis. 37% had a CD4 lymphocyte count less than
200, consistent with an immunocompromised state, yet no patient has developed
active coccidioidal infection. This result differs from other reported experience.
S22221| Introduction of proteinase inhibitors results in
decreasing incidence of oropharyngeal candidiasis in an
HIV patient cohort study
Jorge Garbino, L. Kaiser, A. Auckentaler, V. Geneva, B. Hirschel. University of
Geneva 24 R Michaeli de Crest, Infectious Disease Division, Switzerland
Objective: To determine the effect of the introduction of proteinase inhibitors
(HAART) on the incidence of oropharyngeal candidiasis in HIV patients.
Methods: Retrospective survey of the incidence of oropharyngoesophageal
candidiasis (OPEC) before and after starting HAART as part of combination
antiretroviral therapy in 9946 clinical records entered in the Swiss cohort HIV
study. Comparison of number of episodes of OPEC before and after introduction
of HAART therapy. Observation and comparison of changes in CD4 cell count
30 days before and 30 days after OPEC, and 30 days before and after HAART
administration.
Results: Out of the 9449 patients in the cohort study, 4449 had a diagnosis of
OPEC. There were 3398 patients with OPEC diagnosis not treated with HAART
and 1051 treated with HAART. Out of these 1051 patients, 88 developed OPEC
after initiation of HAART treatment, but only 69 had a CD4 count and met the
inclusion criteria for this study. The mean CD4 cell count for that group was 106.
In the group of patients with OPEC diagnosis treated without HAART only 1793
met the criteria to be included in the study. Their mean CD4 cell count was 167.
Conclusion: This cohort study confirms that HAART introduction was accompanied by a statistically significant reduction in the incidence of OPEC. The
episodes of OPEC in HAART-treated patients occurred in those with higher levels
of immunosuppression.
22222 | National attitudes and antifungal utilization practices of
US physicians providing care to the HIV population
Jose Vazquez1 2, L. McMillen2, J.D. Sobel2. 14160 John R Suite 2140 Detroit
Michigan 48201; 2 Wayne State University Detroit MI, USA
Objective: Characterization of the attitudes and antifungal utilization practices of
US physicians providing care to the HIV population.
Design: A survey of antifungal utilization in the HIV population for initial and
refractory oropharyngeal (OPC) and esophageal (EC) candidiasis was mailed

between March-June 1996 to physicians (P) randomly selected from the IDSA
Directory.
Results: 566 surveys were returned for a response rate of 18%. 42% of
responders were university, 41% were community based. More than 50% in
both groups cared for >50 HIV/AIDS pts. The drugs of choice for 1st episodes
of OPC were clotrimazole (Cz) 53% and fluconazole (Fz) 30%. Cz failure was
treated with Fz in -75%, followed by AMB, ketoconazole (Kz) and itraconazole