Feasibility and inter-reporter variability of submaximal outcomes derived from cardiopulmonary exercise testing in people with advanced cystic fibrosis lung disease

BACKGROUND: Cardiopulmonary exercise testing (CPET) provides prognostic information in people with advanced cystic fibrosis lung disease (pwACFLD). This project aimed to ascertain feasibility and inter-reporter variability in the identification of submaximal CPET outcomes for pwACFLD as potential predictors of prognosis where no peak exercise data are available. METHODS: We utilised data from an international retrospective multicentre study involving pwACFLD, for whom raw CPET data were available. Two experienced operators independently reviewed and analysed CPET tests with a focus on three pre-defined measures: oxygen uptake (V̇O$_{2}$) at the anaerobic threshold (AT), the breathing reserve index at the AT (BRIAT), and the slope of the minute ventilation to carbon dioxide production ratio (V̇$_{E}$/V̇CO$_{2}$-slope). We calculated intra-class correlation coefficients (ICC) with their 95% confidence intervals (CI), and limits of agreement using the Bland-Altman method. RESULTS: The original cohort included 174 pwACFLD. Among those, raw CPET data were available for 101 individuals, of which 89 tests were of sufficient technical quality for submaximal analysis. In 72 out of 89 technically-acceptable tests (81%), the AT could be confidently identified by both operators. Furthermore, ICCs indicated good-to-excellent inter-reporter agreement for V̇O$_{2}$at the AT [ICC (95% CI) 0.79 (0.62 to 0.88)], the V̇$_{E}$/V̇CO$_{2}$-slope [0.95, 0.93 to 0.97)], and BRIAT [0.76, 0.63 to 0.85]. CONCLUSIONS: Submaximal CPET data can be reliably obtained in most pwACFLD by trained CPET operators. Future studies may ascertain the prognostic value of submaximal CPET outcomes in pwACFLD.