One-year pulmonary impairment after severe COVID-19: a prospective, multicenter follow-up study

The Coronavirus disease 2019 (COVID-19) pandemic, caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and initiated in December 2019, expanded dramatically throughout the world [1]. Pneumonia and acute respiratory distress syndrome (ARDS), frequent manifestations of COVID-19, may cause pulmonary sequelae including pulmonary fibrosis [2, 3]. Short-term pulmonary sequelae have been described in cohorts of patients followed up between 3 and 6 months after discharge and range from mild respiratory impairment, with moderately reduced DLCO in asymptomatic patients, to more significant restrictive ventilatory dysfunction in patients suffering persistent pulmonary symptoms, mainly exertional dyspnea [4, 5]. The severity of respiratory failure and the need of higher respiratory support (endotracheal intubation and invasive mechanical ventilation (IMV)) during pneumonia together with the extension of the radiological involvement were identified as factors associated to the development of pulmonary sequelae both functional and radiological [4‐6].

A short-term follow-up may not be adequate to evaluate the long-term prognosis of respiratory impairment, therefore studies with longer follow-ups are warranted. Preliminary data from Chinese cohorts showed that up to 47% of patients showed residual abnormalities on pulmonary Computed Tomography (CT) scan performed at 1 year from the pneumonia, with ground glass attenuation and reticular abnormalities as the major radiologic patterns [7]. Furthermore, when comparing 6 to 12-month follow-up radiological exams, fibrotic interstitial lung abnormalities (ILA) and traction bronchiectasis remained stable, while non fibrotic ILA were completely or partially resolved [8]. The largest observational 1-year follow-up study available till now and performed in Wuhan, China, reported a prevalence of lung diffusion impairment up to 54% in critically ill patients and a significant burden of symptoms with 30% of patients still complaining of dyspnea [9]. However, large observational studies on long-term pulmonary sequeale in European cohorts are still missing.

Moreover, prior experience with the SARS due to SARS-CoV-1 reported the presence of pulmonary sequealae, although of mild entity, even years after the infection [10, 11].

This study aims to identify and characterize the pulmonary sequelae, in patients hospitalized for SARS-CoV-2 pneumonia, at 12 months follow-up after hospital discharge, and to evaluate their association with the maximum ventilatory support received during hospitalization.

In our cohort of 287 patients at 12-month follow-up from hospitalization due to SARS-CoV-2 pneumonia fibrotic sequelae at HRCT scans were found in a strict minority of patients (3, 1% of the cohort). Mild non-fibrotic radiological abnormalities were observed in the majority of cases (66% of the entire cohort) with interstitial lung involvement, particularly GGO and RA, as subpleural curvilinear lines, as the main radiologic pattern. However, the anatomical extension of these abnormalities was limited, with a mean lobar involvement that ranges between 13 and 17% of each entire single lobe. Similar results were observed in a Chinese cohort of 41 patients where GGO and RA were the most common HRCT finding, although only 47% of the cohort showed residual radiological aberration [7]. Only a narrow minority of patients developed fibrotic sequelae (honeycombing was observed in 1% of cases) and irreverisible abormalities such as bronchiectasis, however we cannot predict the evolution of the more common non-fibrotic sequelae (mainly GGO and RA) and studies with longer follow-ups are required. Furthermore, it is difficult to identify whether the lung damage is entirely due to the viral action or is at least partially secondary to baro- and volotrauma during IMV. In fact, while we did not observe cystic alterations and pneumatoceles in our cohort, fibrotic sequelae might also be favored by barotrauma [19, 20].

In our cohort older patients with more severe pneumonia were at higher risk of developing radiological sequelae, which nicely fit with the results from a Chinese cohort by Chen et al. [7].

Almost 40% of patients showed DLCO impairment of mild entity in the majority of cases, and an even smaller percentage showed a restrictive pattern (between 7 and 11.5% according to the definition used). DLCO impairment was more common in patients in the “oxygen only” and IMV group. However, the number of patients in the “oxygen only” group was limited and the proportion of patients lost to follow-up was slightly higher compared to the other groups, which may have led to the selection of the most severe cases. Also interesting to note for the primary end-point, the DLCO values improved between the 6- and 12-month evaluation. Despite the mild entity of the functional sequelae, a consistent proportion of patients at 1-year from SARS-CoV-2 pneumonia still report exertional dyspnea (35%) with a worsening trend compared to the 6-month visit. Similar results were reported by the largest 1-year follow-up cohort described to date (1276 patients): Huang et al. observed a general improvement in functional and radiological lung sequelae between 6- and 12-month follow-up visits, however the number of patients with exertional dyspnea slightly increased between the two time-points reaching 30% of the entire cohort [9]. We did not collect other non-respiratory symptoms in our patients, however Huang et al. reported a slight increase also in anxiety and depression between 6 and 12 months after hospitalization for pneumonia. Furthermore, a recent study on the main symptoms reported by patients 1-year after COVID-19 hospitalization found that fatigue, anxiety and myalgia were among the most common [21]. All these debilitating symptoms were recently included in the definition of “long COVID” syndrome [22]. In our cohort the worsening of exertional dyspnea was similar among the three groups (oxygen only, CPAP and IMV) suggesting that the mechanisms causing breathlessness may be at least partially independent of the severity of pneumonia. These data also suggest the importance of psychological follow-up and rehabilitation programs in patients with a persistent burden of symptoms months after COVID-19 recovery.

In our cohort of patients with COVID-19, we observed a three times higher risk of DLCO impairment in patients with asthma compared to those without this comorbidity. The impact of asthma on COVID-19 remains largely unknown. The available literature suggests that asthma is not associated with worse COVID-19 short-term outcomes, including mortality [23, 24]. However, data on long-term outcomes after COVID-19 in large cohorts of patients with asthma is not yet available. The DLCO impairment we observed in our patients with asthma might be related both to COVID-19 sequelae and asthma itself, although none of the patients presented a disease exacerbation at the time of the follow-up visit. However, in our study, asthmatic patients were only 16 and results should be interpreted with care. When we excluded asthma from the models (due to the low number of asthmatic patients) results were consistent.

Our results and those of the available literature suggest that elderly patients with more severe pneumonia (IMV group) may require a more standardized follow-up including complete PFTs and chest HRCT to better evaluate the presence of long-term pulmonary sequelae. However, even patients with less severe pneumonia (“oxygen only” group) and less functional (DLCO) and radiological involvement, still showed a slight worsening of dyspnea and physical performance at 6MWT between 6- and 12-month follow-up, suggesting that, in case of persistence or appearance of new respiratory symptoms, a personalized follow-up may be required. In this scenario, HRCT scan and DLCO appear as the most sensitive tools to identify pulmonary sequelae.

Among the main strengths of our study we acknowledge (1) the multicentric design, which included both university and non-university hospitals, that increased the generalizability of the results; (2) the selection criteria excluded patients with pre-existing structural lung diseases that may have hampered the possibility of identifying sequelae of SARS-CoV-2 pneumonia.

Our study presents also some limitations: (1) the study visits were conducted during the third pandemic wave and this may have contributed to the lost to follow-up of some patients, however the distribution of age and gender was similar among all patients recruited and patients actually visited; (2) data on the severity of radiological involvement during hospitalization, that may have had an impact on the development of pulmonary sequelae, were not collected; (3) we did not collect any pre-COVID-19 CT scan, therefore it is impossible to evaluate if minor interstitial lung abnormalities, emphysema or bronchiectasis were subclinical pre-existing alterations; (4) We decided to use predefined cut-offs retrieved from the literature to define the presence and degree of PFTs alterations instead of the Global Lung Function Initiative (GLI) guidelines although this choice may have lead to an under- or overestimation of the proportion of patients with impaired PFTs.

In conclusion, we observed in the majority of patients at 12 months from SARS-CoV-2 pneumonia hospitalization minor non-fibrotic chest HRCT abnormalities. DLCO was the most sensitive functional parameter to identify lung sequelae and we observed its improvement between 6- and 12-month follow-up. Nevertheless, a considerable and increasing proportion of patients still reports exertional dyspnea regardless of the initial severity of the disease. Older patients and those who required IMV are at higher risk of developing pulmonary sequelae, however also patients with persistent or worsening respiratory symptoms may require a personalized follow-up. Further studies with longer follow up (2–3 years) are required to evaluate the possible progression of non-fibrotic interstitial lung abnormalities.