Rationale Our group recently demonstrated diaphragm muscle weakness as a potential correlate for persistent dyspnea in patients 15 months after COVID-19 in whom lung and cardiac function are normal.

Objective The aim of this study was to determine the impact of inspiratory muscle training (IMT) on diaphragm muscle weakness and exertional dyspnea in these patients.

Methods 18 patients (7 female, age 62±15 years) with diaphragm muscle weakness and dyspnea underwent a 6 weeks randomized sham-controlled IMT trial. Both arms trained 30 breaths twice daily. Before, immediately after IMT and 6 weeks after terminating IMT a comprehensive respiratory muscle assessment including twitch transdiaphragmatic pressure following cervical magnetic stimulation of the phrenic nerve roots, a diaphragm fatigue protocol and evaluation of dyspnea was performed.

Main results Following IMT diaphragm muscle function (treatment arm: sniff pDI 83±11 vs 100±40 cmH2O; p-value =0.02), diaphragm fatigueability (treatment arm: time to task failure 472±247 vs 1219±979 seconds; p-value =0.05), diaphragm voluntary activation index (treatment arm: 77±16 vs 85±31 %; p-value =0.03) and dyspnea (treatment arm: borg scale 6.89±1.83 vs 5.67±1.50; p=0.03) all improved in the treatment but not in the sham control arm. These effects persisted even 6 weeks after the end of the IMT.

Conclusions IMT improves dyspnea in patients with persistent diaphragm muscle weakness related dyspnea following a hospitalization for COVID-19. The current study therefore is the first to identify a potential treatment therapy of persisting COVID-associated dyspnea.