Association of Pulmonary Involvement at Baseline with Exercise Intolerance and Worse Physical Functioning 8 Months...

Objectives: We aimed to describe the cardiopulmonary function during exercise and the health-related quality of life (HRQoL) in patients with a history of COVID-19 pneumonia, stratified by chest computed tomography (CT) findings at baseline. Methods: Among 77 consecutive patients with COVID-19 who were discharged from the Pulmonology Ward between March 2020 and April 2021, 28 (mean age 54.3 ± 8.6 years, 8 females) agreed to participate to the current study. The participants were analyzed in two groups based on pulmonary involvement (PI) at baseline chest CT applying a threshold of 25%. A consequent artificial intelligence (AI)-guided total opacity score (TOS) was calculated in a subgroup of 22 patients. A cardiopulmonary exercise test (CPET) was conducted on average 8.4 (±1.9) months after discharge from the hospital. HRQoL was defined using the short-form (SF-36) questionnaire. The primary outcome was exercise intolerance that was defined as a peak oxygen uptake (V'O_2peak) < 80% predicted. Secondary outcomes were ventilatory limitation, defined as breathing reserve < 15%, circulatory limitation, defined as oxygen pulse predicted below 80%, and deconditioning, defined as exercise intolerance in the absence of ventilatory and circulatory limitations. Other secondary outcomes included the SF-36 domains. Results: In all, 15 patients had at least 25% PI (53.6%) at baseline chest CT. Exercise intolerance was observed in ten patients (35.7%), six due to deconditioning and four due to circulatory limitation; none had ventilatory limitation. AI-guided TOS was 30.1 ± 24.4% vs. 6.1 ± 4.8% (p < 0.001) at baseline, and 1.7 ± 3.0% vs. 0.2 ± 0.7% (nonsignificant) at follow-up in high and low PI groups, respectively. The physical functioning (PF) domain score of the SF-36 questionnaire was 66.3 ± 19.4 vs. 85.0 ± 13.1 in high and low PI groups, respectively (p = 0.007). Other SF-36 domains did not differ significantly between the groups. A high PI at baseline was inversely correlated with V'O_2peak (standardized β coefficient = -0.436; 95% CI -26.1; -0.7; p = 0.040) and with PF scores (standardized β coefficient -0.654; 95% CI -41.3; -7.6; p = 0.006) adjusted for age, sex, body mass index and lung diffusion capacity. Conclusions: One-third of participants experienced exercise intolerance eight months after COVID-19 pneumonia. A higher PI at baseline was significantly associated with exercise intolerance and PF. Notwithstanding, the radiological PI was resolved, and the exercise intolerance was mainly explained not by ventilatory limitation but by circulatory limitation and deconditioning.