key: cord-0683923-ye6k2yx3
authors: Ferioli, Martina; Prediletto, Irene; Bensai, Serena; Betti, Sara; Daniele, Federico; Di Scioscio, Valerio; Modolon, Cecilia; Rimondi, Maria Rita; De Molo, Chiara; Serra, Carla; Nava, Stefano; Fasano, Luca
title: Spontaneous Evolution of COVID-19 Lung Sequelae: Results from a Double-Step Follow-Up
date: 2022-01-18
journal: Respiration
DOI: 10.1159/000521316
sha: 74a0a1fdd9c44173feefeb906cf1c50b3aa34a2a
doc_id: 683923
cord_uid: ye6k2yx3

BACKGROUND: Few studies have reported a double-step follow-up of patients after hospitalization for COVID-19. OBJECTIVES: We designed an observational double-step follow-up study with a clinical, functional, and radiological evaluation at 2 and 6 months after COVID-19. The primary outcome was to describe symptoms, spirometry, and 6-minute walking test (6MWT) at 2 and 6 months. Secondary outcomes were to identify if the lowest PaO2/FiO2 during hospitalization is related with functional and radiological evolution and to assess the correlation between radiological and functional abnormalities at 6 months. METHODS: Symptoms, spirometry, and 6MWT were assessed at 2 and 6 months; arterial blood gas, chest x-ray, and lung ultrasound were performed at 2 months; body plethysmography, diffusing capacity for carbon monoxide (DLCO), and CT scan were performed at 6 months. RESULTS: Sixty-four per cent and 42% of patients reported at least one symptom at 2 and 6 months, respectively. The most common 6-month functional alteration was DLCO impairment (57% of patients). An improvement of FEV1, FVC, and 6MWT was observed between 2 and 6 months (p < 0.001). Patients with PaO2/FiO2 <200 during hospitalization performed worse at 6MWT at 2 and 6 months (p < 0.05) and reported more extended radiological abnormalities at 6 months (p < 0.001) compared with patients with PaO2/FiO2>200. At 6 months, more extended radiological abnormalities were related with worse 6MWT, DLCO, and total lung capacity (p < 0.05). DISCUSSION: DLCO and 6MWT impairment seem to be the functional hallmark of COVID-19 and are related with the severity of acute pneumonia. At 6 months, radiological abnormalities were related to functional impairment.

months; body plethysmography, diffusing capacity for carbon monoxide (DLCO), and CT scan were performed at 6 months. Results: Sixty-four per cent and 42% of patients reported at least one symptom at 2 and 6 months, respectively. The most common 6-month functional alteration was DLCO impairment (57% of patients). An improvement of FEV1, FVC, and 6MWT was observed between 2 and 6 months (p < 0.001). Patients with PaO 2 /FiO 2 <200 during hospitalization performed worse at 6MWT at 2 and 6 months (p < 0.05) and reported more extended radiological abnormalities at 6 months (p < 0.001) compared with patients with PaO 2 /FiO 2 >200. At 6 months, more extended radiological abnormalities were related with worse 6MWT, DLCO, and total lung capacity (p < 0.05). Discussion: DLCO and 6MWT impairment seem to be the functional hallmark of COVID-19 and are related with the severity of acute pneumonia. At 6 months, radiological abnormalities were related to functional impairment.

Few studies have addressed the problem of clinical and functional aftermath in patients after hospitalization for COVID-19 Pneumonia [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] . Most of these studies have assessed only one aspect of lung sequelae, such as symptoms [1] , pulmonary function tests (PFTs) [2] [3] [4] , and high-resolution computed tomography (HRCT) [5, 6] . Other studies performed a comprehensive evaluation, reporting both clinical and functional/radiological aspects [7] [8] [9] [10] . Recently, a comprehensive large Chinese monocentric cohort study showed that 6 months after the acute disease 76% of patients reported at least one symptom (especially fatigue, muscle weakness, and sleep difficulties), up to 56% of patients had diffusion dysfunction and that the most common HRCT pattern was pulmonary interstitial changes (ground-glass opacities [GGOs] and irregular lines) [7] . In a monocentric French study [8] about previously hospitalized patients, 51% of 478 subjects reported during a telephone interview at least one symptom (especially fatigue, cognitive impairment, or dyspnoea) that was not present before COVID-19; new onset dyspnoea was reported in 16% of patients. One hundred seventy-seven patients were assessed in the outpatient clinic 4 months after discharge: HRCT abnormalities and diffusing capacity for carbon monoxide (DLCO) impairment were more frequent in patient who underwent intubation during hospitalization. In the same series, left ventricular ejection fraction was >50% in 90% of the patients, and in none it was <40%. All the subjects with an ejection fraction <50% were former intensive care unit patients [8] . An Italian study reported that 58% and 44% of patients showed alterations of DLCO and chest X-ray (CXR), respectively [10] .

Very few studies assessed the natural history of the disease by a double-step follow-up after withdrawing any therapy of the acute phase. An observational study about corticosteroid treatment was performed in England in patients with persistent HRCT abnormalities and PFTs dysfunction 4 weeks after discharge and an improvement of symptoms, radiological signs, and lung function was shown after treatment [11] . Li et al. [12] reported spirometry at 2 and approximately 4 weeks from discharge, showing a progressive functional improvement in vital capacity, forced vital capacity (FVC), and forced expiratory volume in 1 s (FEV1) in 18 patients. Liu et al. [13] described the respiratory functional effects of a 6-week respiratory rehabilitation training in 72 patients by 2 evaluations: the first almost at 6 months from the onset of COVID-19 and the second after 6 weeks. An improve-ment of spirometry, DLCO, and 6-Minute Walking Distance (6MWD) was observed [13] . Indeed, only a guidance by British Thoracic Society (BTS) suggesting a different follow-up according to the severity of the acute pneumonia had already been published when this study was designed [14, 15] . We therefore planned a prospective observational monocentric follow-up study to assess the clinical, functional, and imaging evolution of the disease (2 months and 6 months after discharge) without any pharmacologic therapy and out of any rehabilitation program. Symptoms, spirometry, and 6MWT were assessed at 2 and 6 months; arterial blood gas (ABG), CXR, and lung ultrasound (LUS) were performed at 2 months; body plethysmography, DLCO, and HRCT were performed at 6 months. The primary outcome was to describe symptoms, spirometry, and exercise capacity (6MWT) at 2 and 6 months.

Secondary outcomes were: • To evaluate if the patients with more severe acute respiratory failure defined by impairment of gas exchange (lowest value of the ratio of arterial oxygen partial pressure [PaO 2 in mmHg] to fractional inspired oxygen [FiO 2 ], PaO 2 /FiO 2 nadir) had worse functional (lung function impairment and 6MWT) and radiological sequelae (HRCT severity score) at 6 months [16] . • To assess the correlation between radiological abnormalities (HRCT severity score) and functional impairment at 6 months.

Inclusion criteria were a previous molecular diagnosis of CO-VID-19 (by reverse transcriptase-polymerase chain reaction on naso-pharyngeal swab at admission to hospital) and an age older than 18 years. Patients signed a written informed consent. No exclusion criteria were present. Patients included in the study were hospitalized during the first Italian epidemic wave (March 1, 2020-May 31, 2020) and were followed up at the Pulmonary and Respiratory Intensive Care Unit of Sant'Orsola-Malpighi hospital until November 30, 2020. Our 2 steps follow-up included: an "early" appointment within 2 months from discharge and a second one, within 6 months from discharge. All patients were seen face to face by 2 trained physicians. Figure 1 shows the flow chart of the study, with the number of patients that were present at each evaluation. The data regarding the acute phase (the interval between symptoms onset to discharge from hospital) were taken retrospectively from the hospital medical records and they are shown on Table 1 .

At 2 months we recorded PFTs (spirometry), 6MWT, ABG, CXR, LUS, and symptoms. To do that, we have pre-identified 12 symptoms on a checklist, and the patients reported or not their presence (cough, dyspnoea, chest pain, fatigue, fever, muscle weakness, tachycardia, throat pain, nausea, diarrhoea, balance disorder, and stomach ache). Indeed, they were free to report any other symptoms. At 6 months, we recorded symptoms, PFTs (spirometry, body plethysmography, and DLCO), HRCT, and 6MWT. ABG was performed only at 2 months as we expected a normalization of PaO 2 /FiO 2 since its value at discharge was already above 300 mm Hg.

6MWT was performed according to actual guidelines, including the minimum number of tests [17] ; we recorded 6MWD, the percent predicted value (calculated according to Gibbons et al. [18] ), the lowest saturation during 6MWT (lowest oxygen saturation, SpO 2 %) and the highest heart rate during 6MWT (highest HR), the difference between the initial SpO 2 value and the lowest SpO 2 value during the 6MWT (6MWT desaturation %, e.g., from 98 to 95% desaturation is 3%), and the distance saturation product (DSP: 6MWD × lowest SaO 2 during the 6MWT [19] ).

Concerning PFTs, we defined obstructive pattern by a FEV1/ FVC ratio less than 70, we defined possible restrictive pattern by a FVC less than 80% of the predicted value (since it was impossible to perform body plethysmography at 2 months), and we defined a possible mixed pattern if the FEV1/FVC ratio was less than 70 and FVC was less than 80% predicted value, according to Johnson et al. [20] .

CXR was performed at 2 months in 74 patients as 26 refused another radiological study as many were performed in hospital during COVID-19 acute phase, and a HRCT was already planned within 4 months. CXR was classified from a team of 3 expert radiologists as negative, presence of opacities, interstitial thickening, or peripheral fibrotic linear images. A semi-quantitative scoring system called Brixia CXR score, that assigns a score (from 0 to 3) to each of 6 zones on frontal chest projection, was used to perform a radiological quantification of lung abnormalities by 3 radiologists [21, 22] .

The LUS examinations were all performed with a My Lab TM Eight machine (Esaote, Genova, Italy) by 2 experienced operators. The convex probe (3-5 MHz) was used, with the widest acoustic window and maximum depth of 10 cm, with the focus on the pleural line. Mechanical Index was <0.9. The examination was performed in the sitting, lateral, or supine position. Several scoring systems are reported in literature to evaluate COVID-19 pneumonia during acute phase [23, 24] . According to the systematic protocol of scanning already described in literature for each patient 12 areas have been explored and reported in the LUS score (LUSs) [23] . The areas have been registered as right and left lung areas (R and L) and divided by using the anterior and posterior axillary lines resulting in 3 areas per hemithorax (anterior, lateral, and posterior), each of them splits in superior and inferior. Each area was examined in the sagittal and axial views. All the lung areas were explored, and for each one the score was defined with the semiquantitative assessment of pulmonary aeration loss: 0: normal lung (A lines); 1: non-coalescent B lines (B lines occupying less than 50% of the intercostal space in the transversal plane); 2: coalescent B lines (B lines occupying more than 50% of the intercostal space); 3: consolidation >1 cm. The sum of the 12 different lung areas represents the LUSs and it can variate between 0 and 36. The presence of irregular thickening of the pleural line was reported as "irregular pleural line pattern (IPP)" and could be observed both alone and together with B lines. We specifically reported IPP because pleural line irregularity is one of the most common sonographic findings of COVID-19 patients [25] . The whole examination lasted approximately 10 min. As suggested by the international evidence-based recommendations for point-of-care LUS, a positive finding of in- terstitial pulmonary syndrome is the presence of 2 or more positive regions bilaterally, with a LUSs cut-off ≥2 [26] . HRCT severity score was visually and independently calculated by 3 different radiologists according to Lessmann et al. [16] . A score 0-5 was assigned to each lobe: 0 = no lobe involvement; 1 = involvement <5%; score 2 = involvement >5% and <25%; score 3 = involvement >25% and <50%; score 4 = involvement >50% and <75% and score 5 = involvement >75%; so that the range of total score was 0-25. Discrepancies were solved by mutual agreement of the 3 radiologists after reviewing HRCT. We excluded from the radiological analysis patients with prior interstitial lung disease (ILD) because it could be difficult to differentiate between prior ILD (maybe worsened after COVID-19) and COVID-19 sequelae.

At 2 months steroids were withdrawn in patients still on treatment for organizing pneumonia-like lesions as the pattern was felt to be stabilized because the triggering viral infection had cleared since at least 30 days and because of the clinical improvement with no ongoing driver of inflammation. Standard inhalation therapy was given to patients with a diagnosis of chronic obstructive pulmonary disease (COPD). No patient underwent other respiratory treatments or a formal rehabilitation program.

Statistical analysis was made using IBM SPSS Statistics version 21.0. Differences between variables were analysed using one-way analysis of variance, when appropriate. Relationship between variables was studied using Spearman's correlation, p < 0.05 was considered statistically significant.

One hundred consecutive patients with COVID-19 were studied, including the full range of pneumonia severity. 10 patients underwent invasive mechanical venti- Table 1 . The mean age was 60 years; 37% of patients were former smokers (24 pack/ years of average), 50% had a cardiovascular comorbidity (hypertension 44%, chronic heart disease 11%, and diabetes 14%) and 22% had a respiratory comorbidity (COPD 9%, asthma 10%, ILD 3%); 40% were obese (body mass index >29) and 40% overweight (25< body mass index <30). Cardiovascular comorbidities were significantly more prevalent in the group with higher ventilatory support (p < 0.001). PaO 2 /FiO 2 nadir and HRCT extension score were worse in the group of patients with higher ventilatory support (p < 0.001). Lactate dehydrogenase at hospital admission was higher in the group with higher ventilatory support (p < 0.001). More patients of the higher ventilatory support group received therapies (hydroxychloroquine, azithromycin, steroids, antivirals, and tocilizumab) compared to the lower ventilatory support group. Clinical, functional, and radiological characteristics of patients at 2 and 6 months are shown in Table 2 .

Symptoms reported are shown in Table 2 . At 2 months, 64% of patients were still symptomatic. Exertional dyspnoea (44%), fatigue (13%), and cough (10%) were the most prevalent symptoms, whilst other symptoms were reported by less than 10% of patients. At 6 months, the total rate of patients reporting at least one symptom decreased (42%, −22%). The variation of dyspnoea assessed by Modified British Medical Research Council Questionnaire (mMRC) from 2 to 6 months was statistically significant (Table 2 ; Fig. 2a) . Table 3 shows a significant correlation between symptoms at 6 months and time to hospital admission (p < 0.05).

Mean results of spirometry, body plethysmography, and DLCO are shown in Table 2 . At 2 months, spirometry was performed in 99 patients and showed a normal pattern in 76% of cases versus 86% at 6 months, when it was performed in 87 subjects. A statistically significant improvement was observed between 2 and 6 months for FEV1 and FVC (p < 0.001). Patients with a normal pattern significantly increase from 2 to 6 months (+10%, p < 0.05). At 6 months 43% of the patients had a normal value of DLCO, 36% showed a mild reduction, and 21% a moderate/severe impairment (shown in Fig. 2b) . The mean PaO 2 /FiO 2 value at 2 months was 444 mm Hg. Six-Minute Walking Test.6MWT distance and variables are reported in Table 2 . 83 patients performed the 6MWT at 2 months and 75 at 6 months. Patients who did not perform the test had orthopaedic abnormalities, were unable to walk unassisted after COVID-19 hospitalization or refused it. Only the patients who performed 6MWT at 2 months were asked to perform it again at 6 months. The mean distance (6MWD) was 418 m (63% predicted) at 2 months and 489 m (73% predicted) at 6 months (p < 0.001) with 84% of patients walking <80% of predicted at 2 months and 64% at 6 months (shown in Fig. 2c , d) (p < 0.05, Table 2 ). At 2 and 6 months, 45% and 34% experienced a desaturation larger than 3% (p < 0.001), respectively. DSP was 395 at 2 months versus 466 at 6 months (p < 0.001). Table 3 shows a significant correlation between 6MWT desaturation and DSP and hospital length of stay (HLOS), between 6MWD and DSP and PaO 2 /FiO 2 at admission, and between 6MWT desaturation and HRCT extension score during hospitalization.

Data concerning CXR, LUS, and HRCT are shown in Table 2 .

Chest X-ray. Two months after hospital discharge, we performed CXR in 74 patients: 46% of them were normal; areas of parenchymal opacities were present in 16% of patients, interstitial thickening in 34%, and mostly peripheral fibrotic linear images in 27% of cases. The lowest Brixia Score 2 months after acute phase of COVID-19 was 0 (18 patients) and the highest score was 8 (4 patients), with a mean score of 3 (±3).

Lung Ultrasound. LUS was performed 2 months after discharge in 53 patients. The lowest LUS score 2 months after acute phase of COVID-19 was 0 (20 patients) and the highest score was 25 (1 patients), with a mean score of 7 (±6). A small pleural effusion was found in 12 patients (22%), often localized in posterior-basal regions (R6-L6). Irregular pleural line pattern was reported in 50 patients (94%), whilst irregular pleural pattern in absence of B lines was found in 20 patients (38%). Interstitial pulmonary syndrome (the presence of 2 or more positive regions bilaterally, with a LUSs cut off ≥2) was diagnosed at 2 months in 31 patients (59%).

High-Resolution Computed Tomography. Eighty-four patients performed HRCT during the acute phase (hospitalization) and 76 patients performed HRCT after 6 months. HRCT at 6 months was compared to HRCT during hospitalization using the HRCT severity score: there was a reduction in extension of the disease (mean score decreased from Table 3 .) with a shift from consolidation and crazy paving during hospitalization to fibrotic lines and normal lung at 6 months (shown in Fig. 2e, f) . Furthermore, the density of GGOs was decreasing at 6 months as if they were fading away. Table 4 shows the clinical and radiological features of the patients in the acute phase, at 2 and at 6 months divided according to the lowest PaO 2 /FiO 2 (PaO 2 /FiO 2 nadir). Patients with PaO 2 /FiO 2 <200 during hospitalization performed significantly worse at 6MWT (p < 0.05 for all the recorded parameters) 2 months after discharge. At 6 months, there is no longer a significant difference in 6MWD and DSP, whilst a significant difference was still present for lowest SpO 2 and 6MWT desaturation %. A significant difference between these 2 groups in HRCT severity score was present both during hospitalization and at 6 months.

As shown in Table 3 , several correlations were found between different physiological and radiological variables. DLCO at 6 months was positively correlated to PaO 2 /FiO 2 nadir during hospitalization and inversely to HLOS. DSP during 6MWT was inversely correlated to HLOS and CRP. 6MWT desaturation % was inversely related to HRCT severity score during hospitalization (shown in Fig. 2g) . The overall symptoms were not correlated with any parameters except for the time lag between the onset of symptoms and hospitalization (p < 0.05). Bold p values are significant. SD, standard deviation; CRP, C-reactive protein at the time of hospital admission (first available value); IL-6, interleukin 6; Max, highest value during hospitalization; SpO 2 , oxygen saturation. Table 5 shows some significant correlations between 6 months HRCT severity score and 6MWT, DLCO and body plethysmography at 6 months. There was a significant inverse correlation between SpO 2 measured at clinical evaluation and HRCT severity score at 6 months (p < 0.05). Concerning 6MWT, a higher HRCT severity score is associated with a shorter distance (6MWD), a lower SpO 2 during 6MWT, and consequently a lower DSP (p < 0.05). There was a significant inverse correlation between DLCO and total lung capacity (TLC) and HRCT severity score at 6 months (p < 0.05).

This is a real-life assessment of the clinical, functional, and radiological trajectory of recovery in patients hospitalized for COVID-19 pneumonia. We planned a doublestep follow-up: the interim 2 months evaluation included symptoms, ABG and imaging (CXR and LUS) to rule out complications (e.g., new pulmonary infiltrates and pneumothorax) and a simple functional evaluation (spirometry and 6MWT) according to the staff availability in our centre at that time; the 6-month visit was a comprehensive assessment of functional and radiological lung evaluation.

In our series, patients who received a higher respiratory support during acute phase had a higher prevalence of cardiovascular comorbidities, a lower PaO 2 /FiO 2 nadir, a higher HRCT extension score, and a higher lactate dehydrogenase at admission (p < 0.001) ( Table 1 ). We also found that more patients of the high ventilatory support group received therapies (hydroxychloroquine, azithromycin, steroids, antivirals, and tocilizumab) compared to the low ventilatory support group, probably due to the severity of pneumonia. Anyway, these results may be misunderstood as treatment was not standardized (patient received different dosages, different duration, and at different times of the disease).

From a clinical point of view, our study reports that 64 and 42% of patients were still symptomatic at 2 and 6 months, respectively. At 2 months Carfì et al. [1] reported that 87% of patients complained persistence of at least one symptom; at 4 months; 51% of patients declared at least one symptom in a French series [8] ; 69% of patients complained at least 1 symptom at 5 months in another Italian study [9] , whilst 76% of patients reported at least 1 symptom at 6 months in a large Chinese series [7] . The percentage of symptomatic patients in our study is lower than in the others; this is probably due to the absence of psychological or cognitive items in our checklist (e.g., sleep disturbances, anxiety or depression, memory difficulties, and concentration problems). In our series, exertional dyspnoea (44%), fatigue (13%), and cough (10%) were the most prevalent symptoms, and this is consistent with other studies [1, 8, 9] .

Our series shows a progressive improvement of spirometry from 2 to 6 months after discharge. Li et al. [12] showed a similar improvement soon after discharge (2 and 4 weeks); therefore, this initial functional gain seems to last, even in the absence of any therapy, for several months.

Impairment of diffusion capacity represents the most common abnormality of lung function in discharged CO-VID-19 survivors (51% of patients) according to Mo et al. [2] , with increasing impairment as the acute pulmonary involvement increased in severity. Our data show that at 6-month DLCO is reduced in 57% of patients and that DLCO was significantly worse in patients with lower PaO 2 /FiO 2 nadir during hospitalization (p < 0.05, Table 3). These results are consistent with Huang et al. [7] , Morin et al. [8] and an Italian series [10] . Therefore, DLCO impairment seems to be the functional hallmark of COVID-19 and it is worse the worst acute pneumonia was, whatever is the way to categorize the severity of acute pneumonia (PaO 2 /FiO 2 or highest ventilatory support).

Six-Minute Walking Test. Our results show a significant improvement of 6MWD from 2 (418 m) to 6 months (489 m, p < 0.001). 6MWD at 6 months is consistent with the distances reported by 2 other Italian series (Anastasio et al. [9] 500-520 m and Faverio et al. [10] 450-485 m). Another Italian study proved the efficacy of formal rehabilitation in increasing 6MWD in patients with severe COVID-19 pneumonia soon after discharge from the intensive care unit [4] . Six months after the acute phase, Liu et al. [13] performed a randomized-controlled trial (6- week respiratory rehabilitation training vs. no rehabilitation) and showed a statistically significant improvement of 6MWD only in the intervention group. These data suggest that 6MWD improves even out of any formal rehabilitation program within 6 months from discharge, but as rehabilitation has proved its efficacy, this window of time might be an opportunity for active training.

Correlating exercise performance during follow-up and severity of the acute phase, our results show that patients with a more severe acute pneumonia (PaO 2 /FiO 2 nadir ≤200) has a significantly lower 6MWD and 6MWD % predicted at 2 but not at 6 months (Table 4 ). This result is consistent with an Italian series that found no difference in 6MWD at 6 months between groups divided according to the highest support during hospitalization [10] .

Although distance is the most measured variable during 6MWT as it represents the most reliable measure of exercise capacity and has an important prognostic value, we also reported other variables related to 6MWT, such as lowest SpO 2 , desaturation % and DSP [17, 19] . Our results show that patient with a more severe acute pneumonia (PaO 2 /FiO 2 nadir ≤200) has a lower minimum SpO 2 and a higher desaturation % during 6MWT at both 2 and 6 months (Table 4 ). This result is consistent with Anastasio et al. [9] who found that lowest SpO 2 during 6MWT at 6 months is still worse in patients with more severe acute pneumonia. Taken together, these data suggest that at 6 months the patients with more severe acute pneumonia seem to have annulled the 6MWD gap with those with milder acute disease while gas exchange is still impaired. Causes of gas exchange impairment might be persistent interstitial pathology or ventilation/perfusion (V/Q) mismatch caused by chronic thromboembolism. At 6 months, we found an inverse correlation between the extent of HRCT involvement and 6MWT parameters (6MWD, lowest SpO 2 , and DSP) ( Table 5 ), suggesting that a worse exercise performance might be linked to more severe parenchymal abnormalities. The same table shows the correlation between HRCT extension score and PFTs, demonstrating a significant inverse correlation with TLC and DLCO (p < 0.05); a better correlation was found with TLC. Taken together the significant relation between HRCT extension score, PFTs (above all TLC), and 6MWT may suggest a prevalent role of postinfectious ILD in the impairment of exercise performance, whilst long-term vascular involvement may play a minor role. This hypothesis is quite speculative, but the absence of significant long-term cardiovascular alterations has been also suggested in 2 larger post-COVID series [7, 8] . Obviously, other studies are needed to confirm this assumption.

Chest-x-Ray. Since CT scans were overloaded during the first wave of COVID-19, even if CXR has limited sensitivity and specificity for COVID-19, several patients were monitored with CXR to assess acute complications (e.g., pneumothorax) [27] . An Italian group validated a semi-quantitative assessment of pulmonary involvement of CXR in hospitalized patients with COVID-19, called Brixia Score [21, 22] . The median Brixia CXR score reported in literature during acute phase is 7 [21] , while our data report a mean 2-month Brixia score of 3 (±3), suggesting an improvement of radiological abnormalities. In our series, we include CXR at 2-month evaluation to explore the possibility of medium-term complication (e.g., lung abscesses and pleural effusions) and to assess the presence of parenchymal opacities, interstitial thickening, or fibrotic lines.

Lung Ultrasound. Even if the role of LUS in the acute phase of COVID-19 is globally accepted, the utility of LUS in long-term follow-up of COVID- 19 has not yet been defined. Our results show that at 2 months from acute phase 59% of patients have an interstitial pulmonary syndrome. These results are consistent with an Italian study which found interstitial pulmonary syndrome in 63% of 38 patients at 3 months after the acute phase of COVID-19 [28] .

High-Resolution Computed Tomography. The radiological evolution of COVID-19 pneumonia starts from a prevalence of GGO and reticulation with a predominantly peripheral, middle-lower, and posterior distribution; then it goes through an increase of GGO plus reticular pattern and consolidation pattern with the appearance of subpleural lines and bronchus distortion; after about 14 days from the onset of symptoms the lung involvement 12 DOI: 10.1159/000521316 starts to decrease [5] . One month after discharge, the abnormalities in the lungs (GGO, consolidation, interlobular septal thickening, and irregular lines) seem to gradually reduce [6] , but data at 4 months reveal the persistence of abnormalities in most patients (58% of non-intubated and in 75% of intubated patients) [8] . Our results show that lung abnormalities at 6 months decrease, with a significant reduction in the severity score from 13 (acute phase) to 6 (at 6 months) (p < 0.001). Furthermore, a higher score at 6 months was found in patients with PaO 2 / FIO 2 nadir <200 (p < 0.001). These data at 6 months are consistent with the results of a large Chinese series that show a trend to larger extension of HRCT impairment in patients with more severe acute pneumonia [7] . At 6 months, the most common HRCT patterns reported by Huang et al. [7] were GGO (≥40%) and irregular lines (≥10%). Our series shows similar results, with GGO in 57% and lines in 18% of patients.

Our investigation has some limitations. Firstly, it is an uncontrolled study, including only COVID-19 patients without a non-COVID-19 comparison group. Then we have no PFTs and HRCT before the disease and at discharge, making any firm conclusion about the fulfilment and the time of recovery quite speculative. Very few patients in this series had chronic respiratory symptoms or pre-existing lung diseases (ILD and COPD) before CO-VID-19, so it is impossible to have any baseline lung function test or HRCT. In addition, pre-existing lung diseases may affect the ability to distinguish the extent of post-COVID alterations from functional impairment. During the first wave of COVID-19 in Italy, there was a shortage of beds, which induced precocious discharges from hospitals; furthermore, the staff was almost completely dedicated to the care of acute hospitalized patients and the guidelines for a safe performance of lung function tests were not yet available so lung function studies before discharge were not a priority at that time. The same reason justifies the absence of DLCO at 2 months. In addition, unluckily, some radiological data are missing (I.e. not every patient underwent all the imaging) but this was a "real life" study, begun when the first pandemic wave was not yet over, so it was not always possible to match the patients' requirements with the machines' availability. The lack of LUS and CXR in certain patients makes impossible to perform a full-correlation analysis on these tests. Finally, we do not include therapies administered during hospitalization in the follow-up correlation analysis because our population did not receive homogeneous treat-ment (e.g., type of steroid used, number of treatment days, and type of tapering), and therefore, any speculation regarding the influence of the treatment on the recovery trajectory would be biased.

This study suggests that COVID-19 patients improve between 2 and 6 months after discharge. A mild defect in gas exchange and of exercise capacity seems to be the functional hallmark of aftermath. According to the results of our study, severity of acute pneumonia is related to the severity of functional and radiological alterations (Tables 3, 4 ). This correlation is consistent with 2 longterm follow-up studies [7, 8] and endorses BTS guidance to pay more attention at follow-up of patients with more severe pneumonia [14, 15] .

Gemelli Against COVID-19 Post-Acute Care Study Group. Persistent symptoms in patients after acute COVID-19

Abnormal pulmonary function in CO-VID-19 patients at time of hospital discharge

Abnormal carbon monoxide diffusion capacity in COVID-19 patients at time of hospital discharge

Pulmonary rehabilitation in patients recovering from CO-VID-19

Imaging features and evolution on CT in 100 COVID-19 pneumonia patients in Wuhan, China

Chest computed tomography and clinical follow-up of discharged patients with CO-VID-19 in Wenzhou City

6 months consequences of COVID19 in patients discharged from hospital: a cohort study

Writing Committee for the COMEBAC Study Group

Four-month clinical status of a cohort of patients after hospitalization for COVID-19

Medium-term impact of COVID-19 on pulmonary function, functional capacity and quality of life

Six-month pulmonary impairment after severe COVID-19: a prospective, multicentre follow-up study. Respiration

Persistent post-COVID-19 inflammatory interstitial lung disease: an observational study of corticosteroid treatment

Lung ventilation function characteristics of survivors from severe COVID-19: a prospective study

Respiratory rehabilitation in elderly patients with COVID-19: a randomized controlled study

BTS guidance on respiratory follow up of patients with a clinico-radiological diagnosis of COVID-19 pneumonia

Respiratory followup of patients with COVID-19 pneumonia

Automated assessment of COVID-19 reporting and data system and chest CT severity scores in patients suspected of having COVID-19 using artificial intelligence

An official European Respiratory Society/American Thoracic Society technical standard: field walking tests in chronic respiratory disease

Reference values for a multiple repetition 6-minute walk test in healthy adults older than 20 years

An official systematic review of the European Respiratory Society/American Thoracic Society: measurement properties of field walking tests in chronic respiratory disease

A stepwise approach to the interpretation of pulmonary function tests

COVID-19 outbreak in Italy: experimental chest X-ray scoring system for quantifying and monitoring disease progression

Radiographic severity index in COVID-19 pneumonia: relationship to age and sex in 783 Italian patients

Lung ultrasound in the ICU: from diagnostic instrument to respiratory monitoring tool

Lung ultrasound in the emergency department for early identification of COVID-19 pneumonia. Respiration

Lung sonographic findings in COVID-19 patients

International Liaison Committee on Lung Ultrasound (ILC-LUS) for International Consensus Conference on Lung Ultrasound (ICC-LUS). International evidence-based recommendations for point-of-care lung ultrasound

Multimodality imaging of COVID-19 pneumonia: from diagnosis to follow-up

Lung ultrasonography for long-term follow-up of COVID-19 survivors compared to chest CT scan

The research was conducted ethically in accordance with the World Medical Association Declaration of Helsinki. The study was approved by the local Ethical Committee (CE-AVEC, Comitato Etico Area Vasta Emilia Centro) (n. 714/2020). Written informed consent was obtained from patients to participate in the study.

The authors have no conflicts of interest to declare.

No funding was received. 

All data generated or analysed during this study are included in this article. Further enquiries can be directed to the corresponding author.