Memoria CD.indd - ISHAM

7-05Evaluation of an antigen-capture ELISA to detect Histoplasma capsulatum antigenuria inimmunocompromised patientsC.M. Scheel 1 , B. Samayoa 2 L. Benjamin 1 , P. Riley 1 , M. Lindsley 1 , A. Herrera 2 , G. Raxcacoj 2 , S. Lima 2 , R. Miramontes 1 ,T.M. Chiller 1 , M. Brandt 1 , E. Arathoon 2 , B.L.Gómez 11Mycotic Diseases Branch, Centers for Disease Control and Prevention, Atlanta, GA. 2 Clinica Familiar Luis Angel Garcia (CFLAG), HospitalGeneral San Juan de Dios, Guatemala City, Guatemala. e-mail: BGomez@cdc.govBackground: Histoplasma capsulatum infection causes significant morbidity and mortality in HIV-infectedindividuals, particularly those in developing countries without access to sophisticated diagnostics or highly-activeantiretroviral therapy. Furthermore, symptoms of histoplasmosis are non-specific, yet most deaths occur in thefirst 2 weeks after diagnosis. Current diagnostic methods can be complex, expensive and slow. A simple, rapidmethod to detect H. capsulatum infection would dramatically decrease time to diagnosis and treatment and reducemorbidity and mortality. We evaluated an antigen-capture ELISA to detect antigen in urine of HIV patients withhistoplasmosis in Guatemala.Methods: Urine samples were collected prior to treatment in HIV patients with culture-confirmed histoplasmosis(n = 48) or non-histoplasmosis fungal and non-fungal diseases (n = 113). Urine from healthy controls (n = 83)were used to define specificity. An antigen-capture ELISA was developed which utilizes polyclonal rabbit anti-H.capsulatum antibody as both capture and detection reagent. A standard curve was included in each assay plateto insure inter-assay reproducibility. Urine specimens were run twice on separate days and repeated a third timeif the coefficient of variance (CV) was greater than 20%.Results: The H. capsulatum antigen-capture ELISA demonstrates a sensitivity of 81% for confirmedhistoplasmosis and a specificity of 96% (all other disease controls, 95.0%; healthy controls, 97%) when testedagainst baseline urine specimens in these patient cohorts. Thirteen of the patients with follow-up urine specimensshowed decreased antigenuria during the course of antifungal chemotherapy.Conclusions: In this analysis, the antigen ELISA assay shows high sensitivity and specificity as a simple rapiddiagnostic test for histoplasmosis in HIV-infected individuals. This assay has the potential to be easily adaptedto laboratories across the world. A simple test using urine would allow for the rapid diagnosis and initiation oftreatment. Longitudinal analysis of H. capsulatum antigenuria in serial specimens during therapeutic interventionmay prove useful for monitoring patient recovery in a clinical setting.7-06An analysis of histoplasmosis in an endemic, resource poor area with a high HIV prevalencerate - Guatemala, 2007R. Miramontes 1 , B. Samayoa 2 , A. Herrera 2 , C. Scheel 1 , B. L. Gómez 1 , E. Arathoon 2 T. Chiller 1 .1Mycotic Diseases Branch, Centers for Disease Control and Prevention; Clinica Familiar Luis Angel Garcia (CFLAG), 2 Hospital GeneralSan Juan de Dios, Guatemala City, Guatemala. e-mail: TChiller@cdc.govBackground: Disseminated histoplasmosis is a serious opportunistic infection in AIDS, often representingthe first manifestation of the syndrome in endemic regions. Central America is a known endemic area forhistoplasmosis; however little is known about the true prevalence in persons with AIDS. In Guatemala, barriers topatient care such as lack of medication and extended travel to health facilities are common and rapid diagnosticsare not available.Methods: A prospective cohort study of hospital patients was conducted from February 2005 - December2007 in large public hospital in Guatemala City. Study criteria required that a patient be HIV-infected andhave three out of five of the following: Fever, pancytopenia, weight loss, radiological evidence consistent withhistoplasmosis, or skin/mucosal lesions suspicious for histoplasmosis. A histoplasmosis case was defined asa positive Histoplasmosis capsulatum culture from a clinical specimen, or positive tissue sample suggestive ofhistoplasmosis.Results: Of the 279 patients that met surveillance criteria, 217 (78%) were enrolled in the study. A total of63 (29%) of 217 patients met the case definition for histoplasmosis. There were 29 (46%) deaths among the 63case patients. The median time to death was 17 days. A total of 11 (17%) case patients were co-infected withtuberculosis and 4 of those were among reported deaths. The median CD4 cell count among case patients ondate of diagnoses was 25 (1-193).Conclusion: The incidence of histoplasmosis in this cohort of patients with HIV in Guatemala is high.Mortality occurred rapidly after admission and in more than a third of the patients. Patient care is complicated inthis setting by the lack of availability of a rapid diagnostic test. Rapid diagnosis is critical in this population. Thisstudy highlights the importance of histoplasmosis as an opportunistic infection in persons with AIDS in Guatemala.More work is needed to better define the burden of this disease in patients with HIV and its incidence among thosepatients co-infected with TB in order to provide guidance for better diagnosis and more rapid treatment.217