Delayed HIV treatment, barriers in access to care and mortality in tuberculosis/HIV co-infected patients in Cali, Colombia / Retraso en el tratamiento para VIH, barreras en el acceso a la atención en salud y mortalidad en personas coinfectadas con tuberculosis y VIH en Cali, Colombia

Abstract Objective: To determine factors associated with mortality in tuberculosis/HIV co-infected patients in Cali, Colombia Methods: This retrospective cohort design included tuberculosis/HIV co-infected persons. Kaplan-Meier and Cox regression were used to estimate survival and risk factors associated with mortality. Results: Of the 279 tuberculosis/HIV co-infected participants, 27.2% died during the study. Participants mainly were adults and males. CD4 count information was available for 41.6% (the median count was 83 cells/mm3), and half were subject to tuberculosis susceptibility testing. The median time between HIV diagnosis and antiretroviral therapy initiation was 372 days. HIV was identified prior to tuberculosis in 53% and concurrent HIV-tuberculosis were diagnosed in 37% of patients. 44.8% had tuberculosis treatment success. Body mass index above 18 kg/m2, initiation of tuberculosis treatment within two weeks, having any health insurance coverage and CD4 count information conferred a survival advantage. Conclusions: Delays in treatment initiation and factors associated with limited health care access or utilization were associated with mortality. As HIV and tuberculosis are both reportable conditions in Colombia, strategies should be focused on optimizing treatment outcomes within both tuberculosis and HIV programs, particularly improving early HIV diagnosis, early antiretroviral therapy treatment initiation, and adherence to tuberculosis treatment.

Resumen Objetivo: Determinar factores asociados con mortalidad en personas con co-infeccion Tuberculosis/VIH en Cali, Colombia. Métodos: Este diseño de cohorte retrospectiva incluyó personas co-infectadas con tuberculosis /VIH. Se utilizó Kaplan Meier y regresion de Cox para estimar supervivencia y factores de riesgo asociados con mortalidad. Resultados: De los 279 participantes coinfectados con tuberculosis/VIH, el 27.2% falleció durante el estudio. Los participantes fueron principalmente adultos y hombres. Se dispuso de información de recuento de CD4 en el 41.6% (la mediana del recuento fue 83 células/mm3), y en la mitad se realizaron pruebas de susceptibilidad para tuberculosis. La mediana de tiempo entre el diagnóstico de VIH e inicio de terapia antirretroviral fue 372 días. Se identificó VIH previo a tuberculosis en un 53%, e infección concurrente tuberculosis-VIH en el 37% de los pacientes. El 44.8% presentó éxito en el tratamiento para tuberculosis. Un índice de masa corporal superior a 18 kg/m2, inicio del tratamiento para TB dentro de las primeras dos semanas, contar con aseguramiento en salud y con recuento de CD4 se asociaron con mayor supervivencia. Conclusiones: Retraso en el inicio de tratamiento y factores relacionados con brechas en el acceso a atención en salud se asociaron con mortalidad. Dado que VIH y tuberculosis son enfermedades de notificación obligatoria en Colombia, las estrategias deben centrarse en optimizar los desenlaces del tratamiento dentro de ambos programas, en particular mejorar el diagnóstico temprano de VIH, el inicio temprano de la terapia antirretroviral y fomentar la adherencia al tratamiento para tuberculosis.

Delayed HIV treatment, barriers in access to care and mortality in tuberculosis/HIV co-infected patients in Cali, Colombia.

Objective: To determine factors associated with mortality in tuberculosis/HIV co-infected patients in Cali, Colombia. Methods: This retrospective cohort design included tuberculosis/HIV co-infected persons. Kaplan-Meier and Cox regression were used to estimate survival and risk factors associated with mortality. Results: Of the 279 tuberculosis/HIV co-infected participants, 27.2% died during the study. Participants mainly were adults and males. CD4 count information was available for 41.6% (the median count was 83 cells/mm3), and half were subject to tuberculosis susceptibility testing. The median time between HIV diagnosis and antiretroviral therapy initiation was 372 days. HIV was identified prior to tuberculosis in 53% and concurrent HIV-tuberculosis were diagnosed in 37% of patients. 44.8% had tuberculosis treatment success. Body mass index above 18 kg/m2, initiation of tuberculosis treatment within two weeks, having any health insurance coverage and CD4 count information conferred a survival advantage. Conclusions: Delays in treatment initiation and factors associated with limited health care access or utilization were associated with mortality. As HIV and tuberculosis are both reportable conditions in Colombia, strategies should be focused on optimizing treatment outcomes within both tuberculosis and HIV programs, particularly improving early HIV diagnosis, early antiretroviral therapy treatment initiation, and adherence to tuberculosis treatment.

Objetivo: Determinar factores asociados con mortalidad en personas con co-infeccion Tuberculosis/VIH en Cali, Colombia. Métodos: Este diseño de cohorte retrospectiva incluyó personas co-infectadas con tuberculosis /VIH. Se utilizó Kaplan Meier y regresion de Cox para estimar supervivencia y factores de riesgo asociados con mortalidad. Resultados: De los 279 participantes coinfectados con tuberculosis/VIH, el 27.2% falleció durante el estudio. Los participantes fueron principalmente adultos y hombres. Se dispuso de información de recuento de CD4 en el 41.6% (la mediana del recuento fue 83 células/mm3), y en la mitad se realizaron pruebas de susceptibilidad para tuberculosis. La mediana de tiempo entre el diagnóstico de VIH e inicio de terapia antirretroviral fue 372 días. Se identificó VIH previo a tuberculosis en un 53%, e infección concurrente tuberculosis-VIH en el 37% de los pacientes. El 44.8% presentó éxito en el tratamiento para tuberculosis. Un índice de masa corporal superior a 18 kg/m2, inicio del tratamiento para TB dentro de las primeras dos semanas, contar con aseguramiento en salud y con recuento de CD4 se asociaron con mayor supervivencia. Conclusiones: Retraso en el inicio de tratamiento y factores relacionados con brechas en el acceso a atención en salud se asociaron con mortalidad. Dado que VIH y tuberculosis son enfermedades de notificación obligatoria en Colombia, las estrategias deben centrarse en optimizar los desenlaces del tratamiento dentro de ambos programas, en particular mejorar el diagnóstico temprano de VIH, el inicio temprano de la terapia antirretroviral y fomentar la adherencia al tratamiento para tuberculosis.

Sample adequacy controls for infectious disease diagnosis by oral swabbing.

Oral swabs are emerging as a non-invasive sample type for diagnosing infectious diseases including Ebola, tuberculosis (TB), and COVID-19. To assure proper sample collection, sample adequacy controls (SACs) are needed that detect substances indicative of samples collected within the oral cavity. This study evaluated two candidate SACs for this purpose. One detected representative oral microbiota (Streptococcus species DNA) and the other, human cells (human mitochondrial DNA, mtDNA). Quantitative PCR (qPCR) assays for the two target cell types were applied to buccal swabs (representing samples collected within the oral cavity) and hand swabs (representing improperly collected samples) obtained from 51 healthy U.S. volunteers. Quantification cycle (Cq) cutoffs that maximized Youden's index were established for each assay. The streptococcal target at a Cq cutoff of ≤34.9 had 99.0% sensitivity and specificity for oral swab samples, whereas human mtDNA perfectly distinguished between hand and mouth swabs with a Cq cutoff of 31.3. The human mtDNA test was then applied to buccal, tongue, and gum swabs that had previously been collected from TB patients and controls in South Africa, along with "air swabs" collected as negative controls (total N = 292 swabs from 71 subjects). Of these swabs, 287/292 (98%) exhibited the expected Cq values. In a paired analysis the three oral sites yielded indistinguishable amounts of human mtDNA, however PurFlockTM swabs collected slightly more human mtDNA than did OmniSwabsTM (p = 0.012). The results indicate that quantification of human mtDNA cannot distinguish swabs collected from different sites within the mouth. However, it can reliably distinguish oral swabs from swabs that were not used orally, which makes it a useful SAC for oral swab-based diagnosis.