Comparing diffusing capacity reference equation performance in evaluating abnormal dlco in post-pneumonia ugandan adults

SESSION TITLE: Pulmonary Physiology Posters SESSION TYPE: Original Investigation Posters PRESENTED ON: 10/10/2023 12:00 pm - 12:45 pm PURPOSE: The ability to measure DLco in sub-Saharan Africa is limited, and existing reference equations do not include people from this region in their derivation sets. Current ATS/ERS guidelines for the interpretation of DLco recommend the use of the Global Lung Initiative (GLI) reference equation, which was derived from a Caucasian sample. The fit of GLI and other commonly used DLco reference equations in a sub-Saharan African population is not presently known. Here, we compared DLco estimates from a population of Ugandan adults using three reference equations. METHODS: This is a cross-sectional analysis of all ATS/ERS compliant DLco measurements obtained in the Inflammation, Aging, Microbes, Obstructive Lung Disease, and Diffusion Abnormalities (I AM OLD-DA) post-pneumonia cohort of Ugandan adults with and without HIV followed for lung function change over time after acute pneumonia. The first passing DLco was used for all participants. The percent predicted and the lower limit of normal (LLN) were estimated using GLI, Neas, and Crapo DLco reference equations. We compared mean DLco %predicted estimates using ANOVA and the proportion of DLco RESULTS: We included DLco measurements from 141 participants (51% female) with a mean age of 32 years. Among them, 60% had HIV, 72% had prior TB, 18% were ever smokers, and 71% reported biomass fuel exposure at home. Data were normally distributed against all equations. Mean DLco %predicted was 97.7% with GLI, 82.9% with Neas-White, 90.1% with Neas-African American, and 75.5% with Crapo; these estimates differed significantly from each other (p<0.001). Proportions of individuals with DLco CONCLUSIONS: There were significant differences in the mean DLco %predicted and the proportion of individuals with DLco CLINICAL IMPLICATIONS: DLco interpretation in this sample of Ugandan adults varied significantly depending on the reference equation used. As capacity to measure DLco increases in sub-Saharan Africa, there is a growing need for the development of region-specific reference sets and better representation of Ugandans in universal reference equations. DISCLOSURES: No relevant relationships by Rebecca Abelman No relevant relationships by Nirav Bhakta No relevant relationships by Jake Branchini No relevant relationships by Katerina Byanova No relevant relationships by Patrick Byanyima No relevant relationships by Jessica Fitzpatrick No relevant relationships by Laurence Huang donation of drug for NIH-sponsored trial relationship with Merck Please note: since 2022 Added 04/13/2023 by Peter Hunt, source=Web Response, value=Consulting fee honorarium and research grant relationship with Gilead Please note: 2019-present Added 04/13/2023 by Peter Hunt, source=Web Response, value=also honorarium for lecture honorarium relationship with Viiv Please note: 2021 to present Added 04/13/2023 by Peter Hunt, source=Web Response, value=honorarium Advisory Committee Member relationship with Biotron Please note: 2019 to present Added 04/13/2023 by Peter Hunt, source=Web Response, value=Consulting fee No relevant relationships by Jue Lin Employee relationship with Infectious Diseases Research Collaboration Please note: 2010 Added 04/04/2023 by Ingvar Sanyu, source=Web Response, value=Salary No relevant relationships by Abdulwahab Sessolo No relevant relationships by William Worodria No disclosure on file for Josephine Zawedde No relevant relationships by Michelle Zhang