Potential Diaphragm Muscle Weakness-related Dyspnea Persists 2 Years after COVID-19 and Could Be Improved by Inspiratory Muscle Training: Results of an Observational and an Interventional Clinical Trial.

Rationale: Diaphragm muscle weakness might underlie persistent exertional dyspnea, despite normal lung and cardiac function in individuals who were previously hospitalized for acute coronavirus disease (COVID-19) illness. Objectives: The authors sought, first, to determine the persistence and pathophysiological nature of diaphragm muscle weakness and its association with exertional dyspnea 2 years after hospitalization for COVID-19 and, second, to investigate the impact of inspiratory muscle training (IMT) on diaphragm and inspiratory muscle weakness and exertional dyspnea in individuals with long COVID. Methods: Approximately 2 years after hospitalization for COVID-19, 30 individuals (11 women, 19 men; median age, 58 years; interquartile range [IQR] = 51-63) underwent comprehensive (invasive) respiratory muscle assessment and evaluation of dyspnea. Eighteen with persistent diaphragm muscle weakness and exertional dyspnea were randomized to 6 weeks of IMT or sham training; assessments were repeated immediately after and 6 weeks after IMT completion. The primary endpoint was change in inspiratory muscle fatiguability immediately after IMT. Measurements and Main Results: At a median of 31 months (IQR = 23-32) after hospitalization, 21 of 30 individuals reported relevant persistent exertional dyspnea. Diaphragm muscle weakness on exertion and reduced diaphragm cortical activation were potentially related to exertional dyspnea. Compared with sham control, IMT improved diaphragm and inspiratory muscle function (sniff transdiaphragmatic pressure, 83 cm H2O [IQR = 75-91] vs. 100 cm H2O [IQR = 81-113], P = 0.02), inspiratory muscle fatiguability (time to task failure, 365 s [IQR = 284-701] vs. 983 s [IQR = 551-1,494], P = 0.05), diaphragm voluntary activation index (79% [IQR = 63-92] vs. 89% [IQR = 75-94], P = 0.03), and dyspnea (Borg score, 7 [IQR = 5.5-8] vs. 6 [IQR = 4-7], P = 0.03). Improvements persisted for 6 weeks after IMT completion. Conclusions: To the best of the authors' knowledge, this study is the first to identify a potential treatment for persisting exertional dyspnea in long COVID and provide a possible pathophysiological explanation for the treatment benefit. Clinical trial registered with www.clinicaltrials.gov (NCT04854863, NCT05582642).