Post-COVID changes in lung function 6 months after veno-venous extracorporeal membrane oxygenation: a prospective observational clinical trial

Abstract

Background: Severe coronavirus disease 2019 (COVID-19) may require veno-venous extracorporeal membrane oxygenation (V-V ECMO). While V-V ECMO is offered in severe lung injury to COVID-19, long-term respiratory follow-up in these patients is missing. Therefore, we aimed at providing comprehensive data on the long-term respiratory effects of COVID-19 requiring V-V ECMO support during the acute phase of infection.

Methods: In prospective observational cohort study design, patients with severe COVID-19 receiving invasive mechanical ventilation and V-V ECMO (COVID group, n = 9) and healthy matched controls (n = 9) were evaluated 6 months after hospital discharge. Respiratory system resistance at 5 and 19 Hz (R₅, R₁₉), and the area under the reactance curve (AX₅) was evaluated using oscillometry characterizing total and central airway resistances, and tissue elasticity, respectively. R₅ and R₁₉ difference (R₅-R₁₉) reflecting small airway function was also calculated. Forced expired volume in seconds (FEV₁), forced expiratory vital capacity (FVC), functional residual capacity (FRC), carbon monoxide diffusion capacity (DLCO) and transfer coefficient (KCO) were measured.

Results: The COVID group had a higher AX₅ and R₅-R₁₉ than the healthy matched control group. However, there was no significant difference in terms of R₅ or R₁₉. The COVID group had a lower FEV₁ and FVC on spirometry than the healthy matched control group. Further, the COVID group had a lower FRC on plethysmography than the healthy matched control group. Meanwhile, the COVID group had a lower DLCO than healthy matched control group. Nevertheless, its KCO was within the normal range.

Conclusion: Severe acute COVID-19 requiring V-V ECMO persistently impairs small airway function and reduces respiratory tissue elasticity, primarily attributed to lung restriction. These findings also suggest that even severe pulmonary pathologies of acute COVID-19 can manifest in a moderate but still persistent lung function impairment 6 months after hospital discharge.

Trial registration: NCT05812196.

Keywords: ECMO; plethysmography; post-COVID-19; respiratory failure; respiratory oscillometry; spirometry.

FIGURE 1

Patient flow chart. Altogether 55 patients were assessed for eligibility and were divided into the two study groups. We allocated 18 patients with severe acute COVID-19, all of whom received V-V ECMO support at our tertiary center. Nine patients in the COVID group were excluded as they did not survive, leaving nine post-COVID patients for the final analyses. Control group of patients were recruited from an ongoing study applying the same methodology as for healthy adults. Exclusion criteria included a history of smoking, chronic respiratory disease, or COVID-19-induced pneumonia requiring hospitalization. We selected 9 control subjects using propensity score matching, based on demographic characteristics relevant to lung function outcomes such as sex, age, height, and weight.

FIGURE 2

Mechanical parameters obtained via respiratory oscillometry characterizing airflow resistance at oscillation frequencies of 5 and 19 Hz (R₅, R₁₉), and their difference (R₅–R₁₉) reflecting the frequency dependence of the real part of the respiratory impedance spectra. Data were obtained at 6 months after hospital discharge in patients requiring veno-venous extracorporeal membrane oxygenation in the acute phase of coronavirus disease 2019 (COVID, gray shading) and in healthy matched controls (H, white shading). Data were reported as absolute values (left panels) and percent predicted (right panel), where the latter is available. Different symbols represent parameter values obtained in the individual patients. *p < 0.05 between the COVID-19 and healthy matched control groups.

FIGURE 3

Respiratory tissue mechanical parameters obtained via respiratory oscillometry representing the area under the reactance curve at 5 Hz, and the resonant frequency (AX₅) and the resonant frequency (f_res). Data were obtained at 6 months after hospital discharge in patients requiring veno-venous extracorporeal membrane oxygenation in the acute phase of coronavirus disease 2019 (COVID, gray shading) and in healthy matched controls (H, white shading). Data were reported as absolute values (left panels) and percent predicted (right panels). *p < 0.05 between the COVID-19 and healthy matched control groups.