key: cord-0861822-6tkd9y81 authors: Noel-Savina, Elise; Viatgé, Thibault; Faviez, Guillaume; Lepage, Benoît; Mhanna, Laurent; Pontier, Sandrine; Dupuis, Marion; Collot, Samia; Thomas, Pascal.; Lacasia, Jon Idoate; Crognier, Laure; Bouharaoua, Sihem; Sifontes, Stein Silva; Mazières, Julien; Prévot, Grégoire; Didier, Alain title: Severe COVID-19 pneumonia: clinical, functional, and imaging outcomes at 4 months date: 2021-04-28 journal: Respir Med Res DOI: 10.1016/j.resmer.2021.100822 sha: be26566dd27a6b12e95df6fc60941c4e8aa070b5 doc_id: 861822 cord_uid: 6tkd9y81 Introduction: Given the pathophysiology of coronavirus disease 19 (COVID-19), persistent pulmonary abnormalities are likely. Methods:We conducted a prospective cohort study in severe COVID-19 patients who had oxygen saturation <94% and were primarily admitted to hospital. We aimed to describe persistent gas exchange abnormalities at 4 months, defined as decreased diffusing capacity of the lungs for carbon monoxide (DLco) and/or desaturation on the 6-minute walk test (6MWT), along with associated mechanisms and risk factors Results: Of the 72 patients included, 76.1% required admission to the intensive care unit (ICU), while 68.5% required invasive mechanical ventilation (MV). 39.1% developed venous thromboembolism (VTE). At 4 months, 61.4% were still symptomatic. Functionally, 39.1% had abnormal carbon monoxide test results and/or desaturation on 6MWT; high-flow oxygen, MV, and VTE during the acute phase were significantly associated. Restrictive lung disease was observed in 23.6% of cases, obstructive lung disease in 16.7%, and respiratory muscle dysfunction in 18.1%. A severe initial presentation with admission to ICU (p=.0181), and VTE occurrence during the acute phase (p=.0089) were associated with these abnormalities. 41% had interstitial lung disease in computed tomography (CT) of the chest. Four patients (5.5%) displayed residual defects on lung scintigraphy, only one of whom had developed VTE during the acute phase (5.5%). The main functional respiratory abnormality (31.9%) was reduced capillary volume (Vc<70%). Conclusion: Among patients with severe COVID-19 pneumonia who were admitted to hospital, 61% were still symptomatic, 39% of patients had persistent functional abnormalities and 41% radiological abnormalities after 4 months. Embolic sequelae were rare but the main functional respiratory abnormality was reduced capillary volume. A respiratory check-up after severe COVID-19 pneumonia may be relevant to improve future management of these patients. ARDS damages the alveoli by mechanisms which depend on the aetiology, and increases the risk of progression to pulmonary fibrosis. Meduri et al. [4] described fibrotic lesions in the lungs in half of patients who died from ARDS. It has been reported that post-ARDS fibrotic lesions lead to decreased diffusing capacity of the lungs for carbon monoxide (DLco) 3 months after extubation; and that 80% of patients recovered normal respiratory function within a year [5] [6] [7] . J o u r n a l P r e -p r o o f 8 ARDS damages the alveoli. Its mechanisms may differ from one aetiology to the next, pulmonary thromboembolism. A severe hypercoagulable state has been demonstrated in patients admitted to ICU, which makes the case for uncontrolled hyper-inflammatory phase as mentioned above [8] . In a recent cohort of 184 critical COVID-19 patients admitted to ICU, 31% developed venous or arterial thrombotic complications, 80% of which involved pulmonary embolism [9, 10] . The long-term impact of such thromboembolic complications is uncertain. Besides parenchymal pulmonary lesions, COVID patients are also at higher risk of pulmonary thromboembolism. A severe hypercoagulable state has been demonstrated in those in intensive care, lending further credence to the suspected significant, uncontrolled hyper-inflammatory phase mentioned above [12] . In a recent cohort of 184 critical COVID patients being treated in intensive care, 31% developed venous or arterial thrombotic complications, 80% of which involved pulmonary embolism [13, 14] . The long-term repercussions of such thromboembolic complications are uncertain. There are suggested follow-up protocols for patients after COVID-19 pneumonia, particularly for respiratory function [11] . The respiratory repercussions of such pneumonia on lung parenchyma and vascularisation are unclear. Georges et al. [11] suggested screening for interstitial and pulmonary vascular disease, particularly if embolism occurred during the acute phase. This study aims to describe the characteristics of persistent gas exchange abnormalities 4 months after severe COVID-19 pneumonia, in patients without prior cardiopulmonary disease. There are suggested protocols for following patients after COVID-19 pneumonia, particularly as regards respiratory function, but no official recommendations [19] . The This prospective cohort study was conducted in Toulouse University Hospital, France, between April and September 2020. Its epidemiological aim was to determine the prevalence of persistent pulmonary abnormalities 4 months after hypoxemic pneumonia. The inclusion criteria were COVID-19 pneumonia diagnosed by positive polymerase chain reaction (PCR) assay using a respiratory sample from the saliva, nasopharynx, bronchi, tracheal aspirate or bronchoalveolar lavage. The patient's condition had to require hospitalization, with clinical and radiological lung involvement: saturation below 94% without respiratory support and computed tomography (CT) of the chest, presenting common features of COVID-19 pneumonia. Patients had to be 18 years old or more, and give a written informed consent. The inclusion criteria were pneumonia of definite COVID-19 origin based on a positive polymerase chain reaction (PCR) assay using a respiratory specimen from the saliva, nasopharynx, bronchi, tracheal aspirate or bronchoalveolar lavage. The pneumonia had to require admission to hospital. The lung insult had to be clinical and radiological: saturation had to below 94% at room temperature on diagnosis, and computed tomography (CT) of the chest had to yield consistent findings. Patients had to be aged 18 or more and had to give written consent to participate in the study. Our primary aim was to describe persistent gas exchange abnormalities at 4 months. Our study was approved by the institutional review board of Brest University Hospital The primary objective was to describe persistent gas exchange abnormalities at 4 months, defined by abnormal pulmonary diffusing capacity and oxygen desaturation during a 6MWT. The secondary objectives were to identify the mechanisms of persistent gas exchange abnormalities and their severity; determine the prevalence of persistent respiratory symptoms; determine the nature of persistent bronchial and ventilatory abnormalities; describe the persistent pulmonary abnormalities on high-resolution computed tomography (HRCT) and lung scintigraphy; compare symptomatic and asymptomatic patients after 4 months to identify the pathophysiological mechanisms of persistent symptoms; and identify risk factors for persistent respiratory abnormalities in clinical, laboratory and imaging assessments or related to initial patient management. To determine the mechanism of gas exchange abnormalities, several parameters were restrictive lung disorderdisease (total lung capacity < 80% of predicted value) impairment of respiratory muscles consistent with diaphragmatic dysfunction (maximum inspiratory pressure and/or sniff nasal inspiratory pressure < 60% of predicted value, or decrease a fall in vital capacity when lying down compared with sitting > 15%) -eosinophilic bronchial inflammation (increased fractional exhaled NO) readings suggestive of proximal or distal airway obstruction with a decrease in impedance measurements on the forced oscillation technique [12] .on the forced oscillation technique [22] . The check-up was completed with laboratory tests: ionogram, urea, creatinine, creatinine clearance, liver function test, Nt-proBNP, D-Dimers, cardiolipid antibodies, anti-Beta-2 GP1 antibodies, lupus anticoagulant, complete blood count, CRP, ferritin, CPK, LDH, coagulant test, fibrinogen, SARS-CoV-2 serology, and Beta-HCG in women of childbearing age. The second-line assessment was left to investigator's judgement depending on respiratory abnormalities: bronchoalveolar lavage for interstitial disease; thoracic CT angiography, echocardiography, right catheterism, cardiopulmonary exercise testing (CPET) for vascular abnormalities; methacholine testing, diaphragmatic echography in the event of respiratory muscle impairment; hyperventilation testing, CPET and echocardiography for dyspnoea of undetermined cause. Characteristics of symptomatic patients were compared to asymptomatic. Symptoms were correlated to findings from additional examinations. The characteristics of symptomatic patients were compared against those of asymptomatic ones. Symptoms were correlated with the findings of the additional examinations conducted to determine their cause or causes. The required sample size was defined by the expected accuracy in estimating the prevalence of these abnormalities: with at least 60 patients and a prevalence of 50%, the 95% confidence interval (CI) was ± 13% (37%, 63%). The required sample size was calculated as a function of the expected precision in estimating the prevalence of these abnormalities at 4 months. With at least 60 patients and a prevalence of 50%, the 95% confidence interval (CI) was ±13% (37%, 63%). Exchange abnormalities at 4 months were described by estimating their prevalence with a 95% CI. Regarding second endpoints, prevalence of symptoms and pulmonary complications were estimated with a 95% CI. Our primary composite endpoint was an adjusted decreased diffusing capacity of the lungs for carbon monoxide (DLco) <70% of predicted value on the CO test and/or decrease of 4% or more in oxygen saturation on the 6MWT 4 months after COVID-19 pneumonia. (table 1) 93 patients were selected, 72 of whom were retained for analysis (figure 1). Some 93 patients were selected for the study, of whom 72 were retained for analysis (figure 1). Their respiratory function was assessed a median of 4.3 months after PCR diagnosis. The population was 76.4% men, and the mean age was 60.5 years (±12.8). Their comorbidities were excess weightoverweight (42.6%), obesity (40%), hypertension (38.5%), diabetes (23%), coronary artery disease (8%) and cancer (12%). Some 76.1% were admitted to intensive careto ICU for a mean of 21. The Figure 2 illustrates our respiratory function test results. Only four patients (5.5%) displayed residual defects on ventilation/perfusion lung scintigraphy at 4 months and none at 6 months, one of whom had developed VTE during the acute phase and was still being treated with anticoagulants when reexamined. . All four had normal scintigraphy after three months of anticoagulation. 31% of HRCT were normal, 46% presented ground-glass opacities, 25% reticular opacities, 6% alveolar opacities, 7% micronodules, 23% bronchiectasis and no honeycombing. After a multidisciplinary meeting, it was concluded that 44.4% of patients had persistent ILD at 4 months, 56% of them with architectural distortion Admission to intensive careICU, invasive MV and its duration, and prone positioning were associated with abnormal gas exchange at 4 months, as were complications like such as VTE, ARDS and tetraparesis. Patients presenting abnormal gas exchange tended to have received corticosteroids later, specifically 15.3 days versus 10 days after PCR diagnosis. This effect was adjusted for initial severity of the disease. Initial CT severity was also associated with abnormal gas exchange (p=0.01). Those who displayed abnormal gas exchange tended to have received corticosteroids later, specifically 15.3 days versus 10 days after PCR diagnosis. Admission to intensive careICU was generally associated with an increased risk of (p=0.001), as were MV, HFO and prone positioning. Most patients with ILD at 4 months were admitted to ICU (6.3% vs. 93.8%, p=0.001) and treated with MV, HFO and prone position. They also had significantly more complications including VTE, ARDS and tetraparesis. It seems that persistence of ILD was influenced by time delays from diagnosis to commencement of treatment with steroids. In patients presenting ILD, steroids were started 15.3 days after PCR diagnosis, but 8.6 days in those who did not (nonsignificant result, p = 0.0726), regardless of the severity of the initial disease. However, patients with ILD were not more symptomatic nor dyspnoeic than others. Patients with ILD at 4 months were more likely to have been admitted to intensive care (6.3% vs. 93.8%, p=0.001) and to be treated with MV, HFO and prone positioning. They also had significantly more complications like VTE, ARDS and tetraparesis. ILD patients with distortion seemed to have developed more severe forms than those without, and required MV and muscle relaxants more often. They also developed VTE more often. Additionally, 50% of ILD patients who had distortion received corticosteroids as against 57.1% of those who did not. Persistent symptoms at 4 months did not significantly correlate with any of the risk factors assessed. Four months after a moderate to severe SARS-CoV-2 infection, less than half the patients had persistent diffusion abnormalities. The main predictive factors for this respiratory impairment were disease severity during initial presentation and VTE occurrence during the acute phase. abnormalities. The main predictive factors for these were the disease severity on initial presentation with admission to intensive care and VTE occurrence during the acute phase. Nearly half of patients displayed ILD on CT examination at 4 months. Bronchial distortion was frequently observed, even in those who were not treated with MV. Such pulmonary lesions may partly explain why gas exchange abnormalities persisted at 4 months. Some studies suggest using corticosteroids to reduce the fibroblast proliferation and collagen deposition provoked by the release of pro-inflammatory cytokines [7] . In our study, corticosteroids were mostly initiated in severe cases sometime after initial diagnosis. Patients who received corticosteroids earlier tended to present fewer functional and radiographic sequelae at 4 months. Guidance on the role of corticosteroids in treatment has been provided by the World Health Organization [17] although dedicated studies might shed light on when such therapy should be initiated and for how long. Sonnweber et al. [17] described in a prospective study, 3 months after mild to severe SARS-Cov-2 infection: 36% dyspnoea, 21% restrictive ventilatory disorder, 11% impaired DLco, with ILD pattern in 63% of 145 individuals, 50% of whom required oxygen supply and 22% of whom were admitted to ICU. Arnold et al. [18] described at 3 months: of 110 COVID-19 survivors, 39% had dyspnoea, 11% restrictive ventilatory disorder, and 14% desaturation in a 6-minute walk test (6MWT). Shah et al. [19] described, 3 months after SARS-CoV-2 infection, without specifying severity in their population: 20% dyspnoea, 7% SpO2 ≤ 88% at the end of the 6MWT, 52% DLCO < 80% with 45% concurrent restrictive ventilatory deficit, and 11% airflow obstruction. In the HRCT assessment, 83% of patients had ground glass, 65% reticulation and only 12% with neither imaging abnormality. Chest CT sequelae and DLco impairment were significantly associated with numerous days on oxygen supplementation, used as a proxy for acute phase severity. A study of respiratory function 3 and 6 months after a SARS infection [23] revealed that 15% of survivors had reduced surface area for gas exchange, while DLco was significantly lower at 6 months in the subgroup of patients admitted to intensive care. Those abnormalities were only partly due to the lung insult, since other factors like deconditioning on exertion and respiratory muscle dysfunction were also at play, as in our study. Prospective studies would therefore be useful to determine the risk of pulmonary [20, 21, 30] . However, none of the factors investigated were predictive, not even initial severity and management, implying that multiple factors may be at work. Our study aimed to investigate respiratory sequelae in patients who developed severe lung disease and to describe the natural course of this new disease using non-invasive, clinically relevant additional examinations such as DLco and 6MWT desaturation. A prospective observational cohort was therefore suited to achieving these aims. Lung disease is unusually common in SARS-CoV-2 infections and unusually severe. There is little risk of developing chronic thromboembolic pulmonary disease with or without pulmonary hypertension whatever the initial presentation, so routine screening is unwarranted. That said, a basic respiratory workup 3 to 4 months after infection is necessary given the genuine risk of developing extensive pulmonary sequelae. This workup should include a physical examination, a walking test, a diffusion capacity test and chest CT. Early and more prolonged corticosteroid therapy may lower the prevalence of such sequelae. Dedicated studies will be needed. Among patients with severe COVID-19 pneumonia who were admitted to hospital, 61% were still symptomatic after 4 months but no predicting factor was found. Moreover, 39% of patients had persistent functional abnormalities and 41% radiological abnormalities. Embolic sequelae were rare but the main functional respiratory abnormality was reduced capillary volume, probably as a result of capillary inflammation. A respiratory check-up after severe COVID-19 pneumonia may be useful to improve future management of patients: physical examination, 6 MWT, diffusion capacity test and chest CT. Conflict of interest: The authors have no conflicts of interest to disclose. Support statement: The study was supported by the institution: Toulouse University Hospital Pulmonary embolism Clinical Characteristics of Coronavirus Disease 2019 in China Dexamethasone in Hospitalized Patients with Covid-19 -Preliminary Report Late adult respiratory distress syndrome Open lung biopsy does not correlate with pulmonary function after the adult respiratory distress syndrome Recovery of function in survivors of the acute respiratory distress syndrome COVID-19-Related Severe Hypercoagulability in Patients Admitted to Intensive Care Unit for Acute Respiratory Failure Incidence of thrombotic complications in critically ill ICU patients with COVID-19 Prevalence of venous thromboembolism in patients with severe novel coronavirus pneumonia Respiratory follow-up of patients with COVID-19 pneumonia ERS Task Force on Respiratory Impedance Measurements. The forced oscillation technique in clinical practice: methodology, recommendations and future developments and by The ERS Executive Committee Imaging of interstitial lung disease Impact of severe acute respiratory syndrome (SARS) on pulmonary function, functional capacity and quality of life in a cohort of survivors Long-term outcomes of pandemic 2009 influenza A(H1N1)-associated severe ARDS Patient outcomes after hospitalisation with COVID-19 and implications for follow-up: results from a prospective UK cohort A prospective study of 12-week respiratory outcomes in COVID-19-related hospitalisations Early CT features and temporal lung changes in COVID-19 pneumonia in Wuhan, China Survivors: Predictive Factors and Risk Reduction Strategies What's next after the clot? Residual pulmonary vascular obstruction after pulmonary embolism: From imaging finding to clinical consequences Incomplete Recovery of Lung Perfusion After 3 Months in Patients with Acute Pulmonary Embolism Treated with Antithrombotic Agents SARS-CoV-2 is an appropriate name for the new coronavirus