A Nomogram-Based Model to Predict Respiratory Dysfunction at 6 Months in Non-Critical COVID-19 Survivors.

Objective: To assess the prevalence of respiratory sequelae of Coronavirus disease 2019 (COVID-19) survivors at 6 months after hospital discharge and develop a model to identify at-risk patients. Patients and Methods: In this prospective cohort study, hospitalized, non-critical COVID-19 patients evaluated at 6-month follow-up between 26 August, 2020 and 16 December, 2020 were included. Primary outcome was respiratory dysfunction at 6 months, defined as at least one among tachypnea at rest, percent predicted 6-min walking distance at 6-min walking test (6MWT) ≤ 70%, pre-post 6MWT difference in Borg score ≥ 1 or a difference between pre- and post-6MWT oxygen saturation ≥ 5%. A nomogram-based multivariable logistic regression model was built to predict primary outcome. Validation relied on 2000-resample bootstrap. The model was compared to one based uniquely on degree of hypoxemia at admission. Results: Overall, 316 patients were included, of whom 118 (37.3%) showed respiratory dysfunction at 6 months. The nomogram relied on sex, obesity, chronic obstructive pulmonary disease, degree of hypoxemia at admission, and non-invasive ventilation. It was 73.0% (95% confidence interval 67.3-78.4%) accurate in predicting primary outcome and exhibited minimal departure from ideal prediction. Compared to the model including only hypoxemia at admission, the nomogram showed higher accuracy (73.0 vs 59.1%, P < 0.001) and greater net-benefit in decision curve analyses. When the model included also respiratory data at 1 month, it yielded better accuracy (78.2 vs. 73.2%) and more favorable net-benefit than the original model. Conclusion: The newly developed nomograms accurately identify patients at risk of persistent respiratory dysfunction and may help inform clinical priorities.

Prevalence of Long COVID-19 Symptoms After Hospital Discharge in Frail and Robust Patients.

Background: A motley postacute symptomatology may develop after COVID-19, irrespective of the acute disease severity, age, and comorbidities. Frail individuals have reduced physiological reserves and manifested a worse COVID-19 course, during the acute setting. However, it is still unknown, whether frailty may subtend some long COVID-19 manifestations. We explored the prevalence of long COVID-19 disturbs in COVID-19 survivals. Methods: This was an observational study. Patients aged 65 years or older were followed-up 1, 3, and 6 months after hospitalization for COVID-19 pneumonia. Results: A total of 382 patients were enrolled. Frail patients were more malnourished (median Mini Nutritional Assessment Short Form score 8 vs. 9, p = 0.001), at higher risk of sarcopenia [median Strength, Assistance with walking, Rising from a chair, Climbing stairs, and Falls (SARC-F) score 3 vs. 1.5, p = 0.003], and manifested a worse physical performance [median Short Physical Performance Battery (SPPB) score 10 vs. 11, p = 0.0007] than robust individuals, after hospital discharge following severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pneumonia. Frailty was significantly associated with: (i) confusion, as a presenting symptom of COVID-19 [odds ratio (OR) 77.84, 95% CI 4.23-1432.49, p = 0.003]; (ii) malnutrition (MNA-SF: adjusted B -5.63, 95% CI -8.39 to -2.87, p < 0.001), risk of sarcopenia (SARC-F: adjusted B 9.11, 95% CI 3.10-15.13, p = 0.003), impaired muscle performance (SPPB: B -3.47, 95% CI -6.33 to -0.61, p = 0.02), complaints in mobility (adjusted OR 1674200.27, 95% CI 4.52-619924741831.25, p = 0.03), in self-care (adjusted OR 553305.56, 95% CI 376.37-813413358.35, p < 0.001), and in performing usual activities of daily living (OR 71.57, 95% CI 2.87-1782.53, p = 0.009) at 1-month follow-up; (iii) dyspnea [modified Medical Research Council (mMRC): B 4.83, 95% CI 1.32-8.33, p = 0.007] and risk of sarcopenia (SARC-F: B 7.12, 95% CI 2.17-12.07, p = 0.005) at 3-month follow-up; and (iv) difficulties in self-care (OR 2746.89, 95% CI 6.44-1172310.83, p = 0.01) at the 6-month follow-up. In a subgroup of patients (78 individuals), the prevalence of frailty increased at the 1-month follow-up compared to baseline (p = 0.009). Conclusion: The precocious identification of frail COVID-19 survivors, who manifest more motor and respiratory complaints during the follow-up, could improve the long-term management of these COVID-19 sequelae.

Prevalence of Long COVID-19 Symptoms After Hospital Discharge in Frail and Robust Patients

Background A motley postacute symptomatology may develop after COVID-19, irrespective of the acute disease severity, age, and comorbidities. Frail individuals have reduced physiological reserves and manifested a worse COVID-19 course, during the acute setting. However, it is still unknown, whether frailty may subtend some long COVID-19 manifestations. We explored the prevalence of long COVID-19 disturbs in COVID-19 survivals. Methods This was an observational study. Patients aged 65 years or older were followed-up 1, 3, and 6 months after hospitalization for COVID-19 pneumonia. Results A total of 382 patients were enrolled. Frail patients were more malnourished (median Mini Nutritional Assessment Short Form score 8 vs. 9, p = 0.001), at higher risk of sarcopenia [median Strength, Assistance with walking, Rising from a chair, Climbing stairs, and Falls (SARC-F) score 3 vs. 1.5, p = 0.003], and manifested a worse physical performance [median Short Physical Performance Battery (SPPB) score 10 vs. 11, p = 0.0007] than robust individuals, after hospital discharge following severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) pneumonia. Frailty was significantly associated with: (i) confusion, as a presenting symptom of COVID-19 [odds ratio (OR) 77.84, 95% CI 4.23–1432.49, p = 0.003];(ii) malnutrition (MNA-SF: adjusted B –5.63, 95% CI –8.39 to –2.87, p < 0.001), risk of sarcopenia (SARC-F: adjusted B 9.11, 95% CI 3.10–15.13, p = 0.003), impaired muscle performance (SPPB: B –3.47, 95% CI –6.33 to –0.61, p = 0.02), complaints in mobility (adjusted OR 1674200.27, 95% CI 4.52–619924741831.25, p = 0.03), in self-care (adjusted OR 553305.56, 95% CI 376.37–813413358.35, p < 0.001), and in performing usual activities of daily living (OR 71.57, 95% CI 2.87–1782.53, p = 0.009) at 1-month follow-up;(iii) dyspnea [modified Medical Research Council (mMRC): B 4.83, 95% CI 1.32–8.33, p = 0.007] and risk of sarcopenia (SARC-F: B 7.12, 95% CI 2.17–12.07, p = 0.005) at 3-month follow-up;and (iv) difficulties in self-care (OR 2746.89, 95% CI 6.44–1172310.83, p = 0.01) at the 6-month follow-up. In a subgroup of patients (78 individuals), the prevalence of frailty increased at the 1-month follow-up compared to baseline (p = 0.009). Conclusion The precocious identification of frail COVID-19 survivors, who manifest more motor and respiratory complaints during the follow-up, could improve the long-term management of these COVID-19 sequelae.

A Pilot Study of the Efficacy and Economical Sustainability of Acute Coronavirus Disease 2019 Patient Management in an Outpatient Setting.

Objective: To report a preliminary experience of outpatient management of patients with Coronavirus disease 2019 (COVID-19) through an innovative approach of healthcare delivery. Patients and Methods: Patients evaluated at the Mild-to-Moderate COVID-19 Outpatient clinics (MMCOs) of San Raffaele University Hospital and Luigi Sacco University Hospital in Milan, Italy, from 1 October 2020 to 31 October 2021 were included. Patients were referred by general practitioners (GPs), Emergency Department (ED) physicians or hospital specialists (HS) in case of moderate COVID-19. A classification and regression tree (CART) model predicting ED referral by MMCO physicians was developed to aid GPs identify those deserving immediate ED admission. Cost-effectiveness analysis was also performed. Results: A total of 660 patients were included. The majority (70%) was referred by GPs, 21% by the ED and 9% by HS. Patients referred by GPs had more severe disease as assessed by peripheral oxygen saturation (SpO2), ratio of arterial oxygen partial pressure to fractional inspired oxygen (PaO2/FiO2), C-reactive protein (CRP) levels and interstitial involvement at lung ultrasound. Among them, 18% were addressed to the ED following MMCO assessment. CART analysis identified three independent predictors, namely home-measured SpO2, age and body mass index (BMI), that robustly divide patients into risk groups of COVID-19 severity. Home-measured SpO2 < 95% and BMI ≥ 33 Kg/m2 defined the high-risk group. The model yielded an accuracy (95% CI) of 83 (77-88)%. Outpatient management of COVID-19 patients allowed the national healthcare system to spare 1,490,422.05 € when compared with inpatient care. Conclusion: Mild-to-moderate COVID-19 outpatient clinics were effective and sustainable in managing COVID-19 patients and allowed to alleviate pressure on EDs and hospital wards, favoring effort redirection toward non-COVID-19 patients.

Myosteatosis Significantly Predicts Persistent Dyspnea and Mobility Problems in COVID-19 Survivors.

Background: Persistent symptoms including dyspnea and functional impairment are common in COVID-19 survivors. Poor muscle quality (myosteatosis) associates with poor short-term outcomes in COVID-19 patients. The aim of this observational study was to assess the relationship between myosteatosis diagnosed during acute COVID-19 and patient-reported outcomes at 6 months after discharge. Methods: Myosteatosis was diagnosed based on CT-derived skeletal muscle radiation attenuation (SM-RA) measured during hospitalization in 97 COVID-19 survivors who had available anthropometric and clinical data upon admission and at the 6-month follow-up after discharge. Dyspnea in daily activities was assessed using the modified Medical Research Council (mMRC) scale for dyspnea. Health-related quality of life was measured using the European quality of life questionnaire three-level version (EQ-5D-3L). Results: Characteristics of patients with (lowest sex- and age-specific tertile of SM-RA) or without myosteatosis during acute COVID-19 were similar. At 6 months, patients with myosteatosis had greater rates of obesity (48.4 vs. 27.7%, p = 0.046), abdominal obesity (80.0 vs. 47.6%, p = 0.003), dyspnea (32.3 vs. 12.5%, p = 0.021) and mobility problems (32.3 vs. 12.5%, p = 0.004). Myosteatosis diagnosed during acute COVID-19 was the only significant predictor of persistent dyspnea (OR 3.19 [95% C.I. 1.04; 9.87], p = 0.043) and mobility problems (OR 3.70 [95% C.I. 1.25; 10.95], p = 0.018) at 6 months at logistic regression adjusted for sex, age, and BMI. Conclusion: Myosteatosis diagnosed during acute COVID-19 significantly predicts persistent dyspnea and mobility problems at 6 months after hospital discharge independent of age, sex, and body mass. Clinical Trial Registration: [www.ClinicalTrials.gov], identifier [NCT04318366].

A Nomogram-Based Model to Predict Respiratory Dysfunction at 6 Months in Non-Critical COVID-19 Survivors

Objective To assess the prevalence of respiratory sequelae of Coronavirus disease 2019 (COVID-19) survivors at 6 months after hospital discharge and develop a model to identify at-risk patients. Patients and Methods In this prospective cohort study, hospitalized, non-critical COVID-19 patients evaluated at 6-month follow-up between 26 August, 2020 and 16 December, 2020 were included. Primary outcome was respiratory dysfunction at 6 months, defined as at least one among tachypnea at rest, percent predicted 6-min walking distance at 6-min walking test (6MWT) ≤ 70%, pre-post 6MWT difference in Borg score ≥ 1 or a difference between pre- and post-6MWT oxygen saturation ≥ 5%. A nomogram-based multivariable logistic regression model was built to predict primary outcome. Validation relied on 2000-resample bootstrap. The model was compared to one based uniquely on degree of hypoxemia at admission. Results Overall, 316 patients were included, of whom 118 (37.3%) showed respiratory dysfunction at 6 months. The nomogram relied on sex, obesity, chronic obstructive pulmonary disease, degree of hypoxemia at admission, and non-invasive ventilation. It was 73.0% (95% confidence interval 67.3–78.4%) accurate in predicting primary outcome and exhibited minimal departure from ideal prediction. Compared to the model including only hypoxemia at admission, the nomogram showed higher accuracy (73.0 vs 59.1%, P < 0.001) and greater net-benefit in decision curve analyses. When the model included also respiratory data at 1 month, it yielded better accuracy (78.2 vs. 73.2%) and more favorable net-benefit than the original model. Conclusion The newly developed nomograms accurately identify patients at risk of persistent respiratory dysfunction and may help inform clinical priorities.

Evaluation of Muscle Mass and Stiffness with Limb Ultrasound in COVID-19 Survivors.

BACKGROUND: acute illnesses, like COVID-19, can act as a catabolic stimulus on muscles. So far, no study has evaluated muscle mass and quality through limb ultrasound in post-COVID-19 patients. METHODS: cross sectional observational study, including patients seen one month after hospital discharge for SARS-CoV-2 pneumonia. The patients underwent a multidimensional evaluation. Moreover, we performed dominant medial gastrocnemius ultrasound (US) to characterize their muscle mass and quality. RESULTS: two hundred fifty-nine individuals (median age 67, 59.8% males) were included in the study. COVID-19 survivors with reduced muscle strength had a lower muscle US thickness (1.6 versus 1.73 cm, p =0.02) and a higher muscle stiffness (87 versus 76.3, p = 0.004) compared to patients with normal muscle strength. Also, patients with reduced Short Physical Performance Battery (SPPB) scores had a lower muscle US thickness (1.3 versus 1.71 cm, p = 0.01) and a higher muscle stiffness (104.9 versus 81.07, p = 0.04) compared to individuals with normal SPPB scores. The finding of increased muscle stiffness was also confirmed in patients with a pathological value (≥ 4) at the sarcopenia screening tool SARC-F (103.0 versus 79.55, p < 0.001). Muscle stiffness emerged as a significant predictor of probable sarcopenia (adjusted OR 1.02, 95% C.I. 1.002 - 1.04, p = 0.03). The optimal ultrasound cut-offs for probable sarcopenia were 1.51 cm for muscle thickness (p= 0.017) and 73.95 for muscle stiffness (p = 0.004). DISCUSSION: we described muscle ultrasound characteristics in post COVID-19 patients. Muscle ultrasound could be an innovative tool to assess muscle mass and quality in this population. Our preliminary findings need to be confirmed by future studies comparing muscle ultrasound with already validated techniques for measuring muscle mass and quality.

Six-month respiratory outcomes and exercise capacity of COVID-19 acute respiratory failure patients treated with continuous positive airway pressure.

BACKGROUND: COVID-19 long-term sequelae are ill-defined since only a few studies have explored the long-term consequences of this disease so far. AIMS: To evaluate the 6-month respiratory outcome and exercise capacity of COVID-19 acute respiratory failure (ARF) patients treated with continuous positive airway pressure (CPAP) during the first wave of the ongoing COVID-19 pandemic. METHODS: A retrospective observational study included COVID-19 patients with ARF. Interventions included CPAP during hospitalisation and 6-month follow up. Frailty assessment was carried out through frailty index (FI), pO2 /FiO2 during hospitalisation and at follow up, respiratory parameters, 6-min walking test (6MWT) and the modified British Medical Research Council (mMRC) and Borg scale at follow up. RESULTS: More than half of the patients had no dyspnoea according to the mMRC scale. Lower in-hospital pO2 /FiO2 correlated with higher Borg scale levels after 6MWT (ρ 0.27; P 0.04) at the follow-up visit. FI was positively correlated with length of hospitalisation (ρ 0.3; P 0.03) and negatively with the 6MWT distance walked (ρ -0.36; P 0.004). CONCLUSIONS: Robust and frail patients with COVID-19 ARF treated with CPAP outside the intensive care unit setting had good respiratory parameters and exercise capacity at 6-month follow up, although more severe patients had slightly poorer respiratory performance compared with patients with higher PaO2 /FiO2 and lower FI.

Physical and psychological sequelae at three months after acute illness in COVID-19 survivors.

BACKGROUND: Coronavirus disease 2019 (COVID-19) may leave behind an altered health status early after recovery. We evaluated the clinical status of COVID-19 survivors at three months after hospital discharge. METHODS: In this prospective observational cohort study, hospitalized patients aged ≥18 years, evaluated at one (M1) and three (M3) months post-discharge were enrolled. 251 patients (71.3% males, median [IQR] age 61.8 [53.5-70.7] years) were included. Median (IQR) time from discharge to M3 was 89 (79.5-101) days. Primary outcome was residual respiratory dysfunction (RRD), defined by tachypnea, moderate to very severe dyspnea, or peripheral oxygen saturation ≤95% on room air at M3. RESULTS: RRD was found in 30.4% of patients, with no significant difference compared with M1. Chronic obstructive pulmonary disease and length of stay were independent predictors of RRD at multivariable logistic regression (odds ratio, OR, [95% confidence interval, CI] 4.13 [1.17-16.88], p 0.033; OR [95% CI] 1.02 [1.00-1.04], p 0.047, respectively). Obesity and C-reactive protein levels upon admission were additional predictors at regression tree analysis. Impaired quality of life (QoL) was reported by 53.2% of patients. Anxiety and insomnia were each present in 25.5% of patients, and PTSD in 22.4%. No difference was found between M1 and M3 in QoL, anxiety or PTSD. Insomnia decreased at M3. Current major psychiatric disorder as well as anxiety, insomnia and PSTD at M1 independently predicted PTSD at M3. CONCLUSIONS: Clinical damage may persist at three months after discharge in COVID-19 survivors. Post-recovery follow-up is an essential component of patient management.

Weight trajectories and abdominal adiposity in COVID-19 survivors with overweight/obesity.

BACKGROUND: COVID-19 is associated with unintentional weight loss. Little is known on whether and how patients regain the lost weight. We assessed changes in weight and abdominal adiposity over a three-month follow-up after discharge in COVID-19 survivors. METHODS: In this sub-study of a large prospective observational investigation, we collected data from individuals who had been hospitalized for COVID-19 and re-evaluated at one (V1) and three (V2) months after discharge. Patient characteristics upon admission and anthropometrics, waist circumference and hunger levels assessed during follow-up were analyzed across BMI categories. RESULTS: One-hundred-eighty-five COVID-19 survivors (71% male, median age 62.1 [54.3; 72.1] years, 80% with overweight/obesity) were included. Median BMI did not change from admission to V1 in normal weight subjects (-0.5 [-1.2; 0.6] kg/m2, p = 0.08), but significantly decreased in subjects with overweight (-0.8 [-1.8; 0.3] kg/m2, p < 0.001) or obesity (-1.38 [-3.4; -0.3] kg/m2, p < 0.001; p < 0.05 vs. normal weight or obesity). Median BMI did not change from V1 to V2 in normal weight individuals (+0.26 [-0.34; 1.15] kg/m2, p = 0.12), but significantly increased in subjects with overweight (+0.4 [0.0; 1.0] kg/m2, p < 0.001) or obesity (+0.89 [0.0; 1.6] kg/m2, p < 0.001; p = 0.01 vs. normal weight). Waist circumference significantly increased from V1 to V2 in the whole group (p < 0.001), driven by the groups with overweight or obesity. At multivariable regression analyses, male sex, hunger at V1 and initial weight loss predicted weight gain at V2. CONCLUSIONS: Patients with overweight or obesity hospitalized for COVID-19 exhibit rapid, wide weight fluctuations that may worsen body composition (abdominal adiposity). CLINICALTRIALS. GOV REGISTRATION: NCT04318366.

Antibody response to multiple antigens of SARS-CoV-2 in patients with diabetes: an observational cohort study.

AIMS/HYPOTHESIS: The aim of the study was to characterise the humoral response against severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in patients with diabetes. Demonstrating the ability to mount an appropriate antibody response in the presence of hyperglycaemia is relevant for the comprehension of mechanisms related to the observed worse clinical outcome of coronavirus disease 2019 (COVID-19) pneumonia in patients with diabetes and for the development of any future vaccination campaign to prevent SARS-CoV-2 infection. METHODS: Using a highly specific and sensitive measurement of antibodies by fluid-phase luciferase immunoprecipitation assays, we characterised the IgG, IgM and IgA response against multiple antigens of SARS-CoV-2 in a cohort of 509 patients with documented diagnosis of COVID-19, prospectively followed at our institution. We analysed clinical outcomes and antibody titres according to the presence of hyperglycaemia, i.e., either diagnosed or undiagnosed diabetes, at the time of, or during, hospitalisation. RESULTS: Among patients with confirmed COVID-19, 139 (27.3%) had diabetes: 90 (17.7%) had diabetes diagnosed prior to the hospital admission (comorbid diabetes) while 49 (9.6%) had diabetes diagnosed at the time of admission (newly diagnosed). Diabetes was associated with increased levels of inflammatory biomarkers and hypercoagulopathy, as well as leucocytosis and neutrophilia. Diabetes was independently associated with risk of death (HR 2.32 [95% CI 1.44, 3.75], p = 0.001), even after adjustment for age, sex and other relevant comorbidities. Moreover, a strong association between higher glucose levels and risk of death was documented irrespective of diabetes diagnosis (HR 1.14 × 1.1 mmol/l [95% CI 1.08, 1.21], p < 0.001). The humoral response against SARS-CoV-2 in patients with diabetes was present and superimposable, as for timing and antibody titres, to that of non-diabetic patients, with marginal differences, and was not influenced by glucose levels. Of the measured antibody responses, positivity for IgG against the SARS-CoV-2 spike receptor-binding domain (RBD) was predictive of survival rate, both in the presence or absence of diabetes. CONCLUSIONS/INTERPRETATION: The observed increased severity and mortality risk of COVID-19 pneumonia in patients with hyperglycaemia was not the result of an impaired humoral response against SARS-CoV-2. RBD IgG positivity was associated with a remarkable protective effect, allowing for a cautious optimism about the efficacy of future vaccines against SARs-COV-2 in people with diabetes. Graphical abstract.

Post-COVID-19 follow-up clinic: depicting chronicity of a new disease.

BACKGROUND AND AIM OF THE WORK: The coronavirus disease-19 (COVID-19) outbreak is posing considerable challenges to healthcare systems and societies worldwide. While the knowledge on the acute phase of the disease has rapidly expanded, little is known on the consequences of COVID-19 following clinical remission. We set up a multidisciplinary COVID-19 follow-up outpatient clinic to identify and address the clinical needs of COVID-19 survivors. Here we describe the features of our follow-up programme. METHODS: The multidisciplinary assessment comprises a complete physical examination, respiratory evaluation (peripheral oxygen saturation, respiratory rate, dyspnoea assessment, lung ultrasound and pulmonary function), cardiovascular assessment (electrocardiography, echocardiography), nutritional assessment (anthropometrics, mini Nutritional Assessment screening tool), neurological examination including cognitive tests, and mental health assessment. All data are prospectively collected, and blood is sampled for biobanking. RESULTS: Since 7 April to 5 June, 2020, 453 out of the 1388 COVID-19 survivors managed at our University Hospital have been evaluated at the Outpatient COVID-19 Follow-up Clinic. The characteristics of the follow-up cohort are similar to those of the whole cohort of COVID-19 in terms of demographics, comorbidities, and COVID-19 severity upon ED presentation, indicating that the follow-up cohort is representative of the whole cohort. CONCLUSIONS: Continuous patient monitoring might give an answer to the numerous unsolved questions about what comes next in this pandemic and beyond. This will help physicians and researchers establish strategies to face future pandemics and develop preventative and therapeutic strategies for similar hyperinflammatory conditions.

Interleukin-1 blockade with high-dose anakinra in patients with COVID-19, acute respiratory distress syndrome, and hyperinflammation: a retrospective cohort study.

BACKGROUND: Mortality of patients with coronavirus disease 2019 (COVID-19), acute respiratory distress syndrome (ARDS), and systemic inflammation is high. In areas of pandemic outbreak, the number of patients can exceed maximum capacity of intensive care units (ICUs), and, thus, these individuals often receive non-invasive ventilation outside of the ICU. Effective treatments for this population are needed urgently. Anakinra is a recombinant interleukin-1 receptor antagonist that might be beneficial in this patient population. METHODS: We conducted a retrospective cohort study at the San Raffaele Hospital in Milan, Italy. We included consecutive patients (aged ≥18 years) with COVID-19, moderate-to-severe ARDS, and hyperinflammation (defined as serum C-reactive protein ≥100 mg/L, ferritin ≥900 ng/mL, or both) who were managed with non-invasive ventilation outside of the ICU and who received standard treatment of 200 mg hydroxychloroquine twice a day orally and 400 mg lopinavir with 100 mg ritonavir twice a day orally. We compared survival, mechanical ventilation-free survival, changes in C-reactive protein, respiratory function, and clinical status in a cohort of patients who received additional treatment with anakinra (either 5 mg/kg twice a day intravenously [high dose] or 100 mg twice a day subcutaneously [low dose]) with a retrospective cohort of patients who did not receive anakinra (referred to as the standard treatment group). All outcomes were assessed at 21 days. This study is part of the COVID-19 Biobank study, which is registered with ClinicalTrials.gov, NCT04318366. FINDINGS: Between March 17 and March 27, 2020, 29 patients received high-dose intravenous anakinra, non-invasive ventilation, and standard treatment. Between March 10 and March 17, 2020, 16 patients received non-invasive ventilation and standard treatment only and comprised the comparison group for this study. A further seven patients received low-dose subcutaneous anakinra in addition to non-invasive ventilation and standard treatment; however, anakinra treatment was interrupted after 7 days because of a paucity of effects on serum C-reactive protein and clinical status. At 21 days, treatment with high-dose anakinra was associated with reductions in serum C-reactive protein and progressive improvements in respiratory function in 21 (72%) of 29 patients; five (17%) patients were on mechanical ventilation and three (10%) died. In the standard treatment group, eight (50%) of 16 patients showed respiratory improvement at 21 days; one (6%) patient was on mechanical ventilation and seven (44%) died. At 21 days, survival was 90% in the high-dose anakinra group and 56% in the standard treatment group (p=0·009). Mechanical ventilation-free survival was 72% in the anakinra group versus 50% in the standard treatment group (p=0·15). Bacteraemia occurred in four (14%) of 29 patients receiving high-dose anakinra and two (13%) of 16 patients receiving standard treatment. Discontinuation of anakinra was not followed by inflammatory relapses. INTERPRETATION: In this retrospective cohort study of patients with COVID-19 and ARDS managed with non-invasive ventilation outside of the ICU, treatment with high-dose anakinra was safe and associated with clinical improvement in 72% of patients. Confirmation of efficacy will require controlled trials. FUNDING: None.