Antibiotic definitive treatment in ventilator associated pneumonia caused by AmpC-producing Enterobacterales in critically ill patients: a prospective multicenter observational study.

BACKGROUND: Ventilator associated pneumonia (VAP) due to wild-type AmpC-producing Enterobacterales (wtAE) is frequent in intensive care unit (ICU) patients. Despite a low level of evidence, definitive antimicrobial therapy (AMT) with third generation cephalosporins (3GCs) or piperacillin is discouraged. METHODS: Observational prospective study including consecutive wtAE VAP patients in 20 French ICUs. The primary objective was to assess the association of the choice of definitive AMT, i.e. piperacillin ± tazobactam (PTZ), 3GCs or other molecule (4GCs, carbapenems, quinolones, cotrimoxazole; control group), with treatment success at day-7. Recurrence of infection was collected as a secondary outcome, and analyzed accounting for the competing risk of death. RESULTS: From February 2021 to June 2022, 274 patients were included. Enterobacter cloacae was the most prevalent specie (31%). Seventy-eight patients (28%) had PTZ as definitive AMT while 44 (16%) had 3GCs and 152 (56%) were classified in the control group. Day-7 success rate was similar between the 3 groups (74% vs. 73% vs. 68% respectively, p = 0.814). Recurrence probability at day-28 was 31% (95% CI 21-42), 40% (95% CI 26-55) and 21% (95% CI 15-28) for PTZ, 3GCs and control groups (p = 0.020). In multivariable analysis, choice of definitive AMT was not associated with clinical success, but definitive AMT with 3GCs was associated with recurrence at day-28 [csHR(95%CI) 10.9 (1.92-61.91)]. CONCLUSION: Choice of definitive antimicrobial therapy was not associated with treatment success at day 7. However, recurrence of pneumonia at day-28 was higher in patients treated with third generation cephalosporins with no differences in mortality or mechanical ventilation duration.

Risk factors and prognosis of orotracheal intubation in aquaporin-4-IgG neuromyelitis optica spectrum disorder attacks.

BACKGROUND: Aquaporin-4 immunoglobulin G Neuro Myelitis Optica spectrum disorders attacks (NMOSD-AQP4-IgG+ attacks) can cause respiratory failure requiring orotracheal intubation (OTI), but the risk factors and outcomes of OTI during attacks remain unclear. Our primary objective was to identify the clinical and radiological risk factors for OTI in NMOSD-AQP4-IgG+ attacks. As a secondary objective, we aimed to evaluate the prognosis of OTI-attacks. METHODS: We retrospectively analyzed NMOSD-AQP4-IgG+ attacks at the Pitié-Salpêtrière Hospital (Jan 2010-Jan 2021), excluding isolated optic neuritis. The primary outcome was the need for OTI due to neurological dysfunction an attack (OTI-attack). The secondary outcome was attack's poor recovery after 12 months, defined as a modified Rankin score (mRS) > 2 in patients with an initial mRS ≤ 2, or an increase ≥ 1 point in mRS in other patients. Analyses were performed using a binomial generalized linear mixed model, with a random intercept for the patient ID to account for within-patient correlations. RESULTS: Seventy-three attacks in 44 patients NMOSD-AQP4-IgG+ were analyzed. Of 73 attacks, 8 (11%) required OTI during the attack, related to acute restrictive respiratory failure (n = 7) and/or severe swallowing disorder (n = 2). None of the OTI-attacks occurred in patients previously treated with active disease-modifying treatment (DMT), while 36 (55.4%) of the non-OTI-attacks occurred in patients who were already on active DMT. On admission, OTI-attacks were more likely to have upper limbs motor paresis of (75.0% versus 29.2%, p = 0.366) and dyspnea (3 [50.0%] versus 4 [6.6%], p = 0.002) compared to non-OTI-attacks. MRI analysis showed that OTI-attacks had edematous lesions in the cervical spinal cord, mainly at levels C1 (75% versus 0% in non-OTI-attacks), C2 (75% versus 1.9%), C3 (62.5% versus 1.9%), and C4 and C5 levels (50% versus to 3.9%). One OTI-attack resulted in the death of one patient. Five patients with OTI-attack had mRS ≤ 2 one year after OTI-attack. Two (25%) OTI-attacks had poor recovery compared to 15 (24.2%) non-OTI-attacks (p = 0.468). CONCLUSION: OTI-attacks occurred in untreated NMOSD-AQP4-IgG+ patients and were associated with edematous upper cervical lesions. The prognosis of these attacks may be favorable, and warrant maximal medical and supportive treatment. Trial registration This was a retrospective observational monocentric cohort study nested in the NOMADMUS cohort (ClinicalTrials.gov Identifier: NCT02850705).

Blood-brain barrier dysfunction in intensive care unit.

The central nervous system is characterized by a peculiar vascularization termed blood-brain barrier (BBB), which regulates the exchange of cells and molecules between the cerebral tissue and the whole body. BBB dysfunction is a life-threatening condition since its presence corresponds to a marker of severity in most diseases encountered in the intensive care unit (ICU). During critical illness, inflammatory response, cytokine release, and other phenomena activating the brain endothelium contribute to alterations in the BBB and increase its permeability to solutes, cells, nutrients, and xenobiotics. Moreover, patients in the ICU are often old, with underlying acute or chronic diseases, and overly medicated due to their critical condition; these factors could also contribute to the development of BBB dysfunction. An accurate diagnostic approach is critical for the identification of the mechanisms underlying BBB alterations, which should be rapidly managed by intensivists. Several methods were developed to investigate the BBB and assess its permeability. Nevertheless, in humans, exploration of the BBB requires the use of indirect methods. Imaging and biochemical methods can be used to study the abnormal passage of molecules through the BBB. In this review, we describe the structural and functional characteristics of the BBB, present tools and methods for probing this interface, and provide examples of the main diseases managed in the ICU that are related to BBB dysfunction.

FX06 to rescue SARS-CoV-2-induced acute respiratory distress syndrome: a randomized clinical trial.

BACKGROUND: Vascular leakage is a major feature of acute respiratory distress syndrome (ARDS). We aimed to evaluate the efficacy of FX06, a drug under development that stabilizes interendothelial cell junctions, at reducing vascular leakage during SARS-CoV-2-induced ARDS. METHODS: This multicenter, double-blinded, randomized trial included adults with COVID-19-associated ARDS who had received invasive mechanical ventilation for < 5 days and were randomized to receive either intravenous FX06 (400 mg/d, for 5 days) or its vehicle as placebo. The primary endpoint was the lowering-from day 1 to day 7-of the transpulmonary thermodilution-derived extravascular lung-water index (EVLWi). RESULTS: Twenty-five patients were randomized to receive FX06 and 24 the placebo. Although EVLWi was elevated at baseline (median [IQR] 15.6 mL/kg [13.5; 18.5]), its declines from day 1 to day 7 were comparable for FX06 recipients and controls (respectively, - 1.9 [- 3.3; - 0.5] vs. - 0.8 [- 5.5; - 1.1] mL/kg; estimated effect - 0.8 [- 3.1; + 2.4], p = 0.51). Cardiac indexes, pulmonary vascular permeability indexes, and fluid balances were also comparable, as were PaO2/FiO2 ratios and durations of mechanical ventilation. Adverse event rates were similar for the 2 groups, although more FX06 recipients developed ventilator-associated pneumonia (16/25 (64%) vs. 6/24 (24%), p = 0.009). CONCLUSIONS: In this unique-dosing-regimen study, FX06 did not lower SARS-CoV-2-induced pulmonary vascular leakage. Future investigations will need to evaluate its efficacy at earlier times during the disease or using other regimens. Trial registration NCT04618042. Registered 5 November 2020.

One-Year Outcomes in Patients With Acute Stroke Requiring Mechanical Ventilation.

BACKGROUND: Long-term outcomes of patients with severe stroke remain poorly documented. We aimed to characterize one-year outcomes of patients with stroke requiring mechanical ventilation in the intensive care unit (ICU). METHODS: We conducted a prospective multicenter cohort study in 33 ICUs in France (2017-2019) on patients with consecutive strokes requiring mechanical ventilation for at least 24 hours. Outcomes were collected via telephone interviews by an independent research assistant. The primary end point was poor functional outcome, defined by a modified Rankin Scale score of 4 to 6 at 1 year. Multivariable mixed models investigated variables associated with the primary end point. Secondary end points included quality of life, activities of daily living, and anxiety and depression in 1-year survivors. RESULTS: Among the 364 patients included, 244 patients (66.5% [95% CI, 61.7%-71.3%]) had a poor functional outcome, including 190 deaths (52.2%). After adjustment for non-neurological organ failure, age ≥70 years (odds ratio [OR], 2.38 [95% CI, 1.26-4.49]), Charlson comorbidity index ≥2 (OR, 2.01 [95% CI, 1.16-3.49]), a score on the Glasgow Coma Scale <8 at ICU admission (OR, 3.43 [95% CI, 1.98-5.96]), stroke subtype (intracerebral hemorrhage: OR, 2.44 [95% CI, 1.29-4.63] versus ischemic stroke: OR, 2.06 [95% CI, 1.06-4.00] versus subarachnoid hemorrhage: reference) remained independently associated with poor functional outcome. In contrast, a time between stroke diagnosis and initiation of mechanical ventilation >1 day was protective (OR, 0.56 [95% CI, 0.33-0.94]). A sensitivity analysis conducted after exclusion of patients with early decisions of withholding/withdrawal of care yielded similar results. We observed persistent physical and psychological problems at 1 year in >50% of survivors. CONCLUSIONS: In patients with severe stroke requiring mechanical ventilation, several ICU admission variables may inform caregivers, patients, and their families on post-ICU trajectories and functional outcomes. The burden of persistent sequelae at 1 year reinforces the need for a personalized, multi-disciplinary, prolonged follow-up of these patients after ICU discharge. REGISTRATION: URL: https://www. CLINICALTRIALS: gov; Unique identifier: NCT03335995.

Clinical features, etiologies, and outcomes in adult patients with meningoencephalitis requiring intensive care (EURECA): an international prospective multicenter cohort study.

PURPOSE: We aimed to characterize the outcomes of patients with severe meningoencephalitis requiring intensive care. METHODS: We conducted a prospective multicenter international cohort study (2017-2020) in 68 centers across 7 countries. Eligible patients were adults admitted to the intensive care unit (ICU) with meningoencephalitis, defined by an acute onset of encephalopathy (Glasgow coma scale (GCS) score [Formula: see text] 13), a cerebrospinal fluid pleocytosis [Formula: see text] 5 cells/mm3, and at least two of the following criteria: fever, seizures, focal neurological deficit, abnormal neuroimaging, and/or electroencephalogram. The primary endpoint was poor functional outcome at 3 months, defined by a score of three to six on the modified Rankin scale. Multivariable analyses stratified on centers investigated ICU admission variables associated with the primary endpoint. RESULTS: Among 599 patients enrolled, 589 (98.3%) completed the 3-month follow-up and were included. Overall, 591 etiologies were identified in those patients which were categorized into five groups: acute bacterial meningitis (n = 247, 41.9%); infectious encephalitis of viral, subacute bacterial, or fungal/parasitic origin (n = 140, 23.7%); autoimmune encephalitis (n = 38, 6.4%); neoplastic/toxic encephalitis (n = 11, 1.9%); and encephalitis of unknown origin (n = 155, 26.2%). Overall, 298 patients (50.5%, 95% CI 46.6-54.6%) had a poor functional outcome, including 152 deaths (25.8%). Variables independently associated with a poor functional outcome were age > 60 years (OR 1.75, 95% CI 1.22-2.51), immunodepression (OR 1.98, 95% CI 1.27-3.08), time between hospital and ICU admission > 1 day (OR 2.02, 95% CI 1.44-2.99), a motor component on the GCS [Formula: see text] 3 (OR 2.23, 95% CI 1.49-3.45), hemiparesis/hemiplegia (OR 2.48, 95% CI 1.47-4.18), respiratory failure (OR 1.76, 95% CI 1.05-2.94), and cardiovascular failure (OR 1.72, 95% CI 1.07-2.75). In contrast, administration of a third-generation cephalosporin (OR 0.54, 95% CI 0.37-0.78) and acyclovir (OR 0.55, 95% CI 0.38-0.80) on ICU admission were protective. CONCLUSION: Meningoencephalitis is a severe neurologic syndrome associated with high mortality and disability rates at 3 months. Actionable factors for which improvement could be made include time from hospital to ICU admission, early antimicrobial therapy, and detection of respiratory and cardiovascular complications at admission.

Description and Outcome of Severe Hypoglycemic Encephalopathy in the Intensive Care Unit.

BACKGROUND: Disorders of consciousness due to severe hypoglycemia are rare but challenging to treat. The aim of this retrospective cohort study was to describe our multimodal neurological assessment of patients with hypoglycemic encephalopathy hospitalized in the intensive care unit and their neurological outcomes. METHODS: Consecutive patients with disorders of consciousness related to hypoglycemia admitted for neuroprognostication from 2010 to 2020 were included. Multimodal neurological assessment included electroencephalography, somatosensory and cognitive event-related potentials, and morphological and quantitative magnetic resonance imaging (MRI) with quantification of fractional anisotropy. Neurological outcomes at 28 days, 3 months, 6 months, 1 year, and 2 years after hypoglycemia were retrieved. RESULTS: Twenty patients were included. After 2 years, 75% of patients had died, 5% remained in a permanent vegetative state, 10% were in a minimally conscious state, and 10% were conscious but with severe disabilities (Glasgow Outcome Scale-Extended scores 3 and 4). All patients showed pathologic electroencephalography findings with heterogenous patterns. Morphological brain MRI revealed abnormalities in 95% of patients, with various localizations including cortical atrophy in 65% of patients. When performed, quantitative MRI showed decreased fractional anisotropy affecting widespread white matter tracts in all patients. CONCLUSIONS: The overall prognosis of patients with severe hypoglycemic encephalopathy was poor, with only a small fraction of patients who slowly improved after intensive care unit discharge. Of note, patients who did not improve during the first 6 months did not recover consciousness. This study suggests that a multimodal approach capitalizing on advanced brain imaging and bedside electrophysiology techniques could improve diagnostic and prognostic performance in severe hypoglycemic encephalopathy.

Outcomes of Patients With Active Cancer and COVID-19 in the Intensive-Care Unit: A Multicenter Ambispective Study.

Background: Several studies report an increased susceptibility to SARS-CoV-2 infection in cancer patients. However, data in the intensive care unit (ICU) are scarce. Research Question: We aimed to investigate the association between active cancer and mortality among patients requiring organ support in the ICU. Study Design and Methods: In this ambispective study encompassing 17 hospitals in France, we included all adult active cancer patients with SARS-CoV-2 infection requiring organ support and admitted in ICU. For each cancer patient, we included 3 non cancer patients as controls. Patients were matched at the same ratio using the inverse probability weighting approach based on a propensity score assessing the probability of cancer at admission. Mortality at day 60 after ICU admission was compared between cancer patients and non-cancer patients using primary logistic regression analysis and secondary multivariable analyses. Results: Between March 12, 2020 and March 8, 2021, 2608 patients were admitted with SARS-CoV-2 infection in our study, accounting for 2.8% of the total population of patients with SARS-CoV-2 admitted in all French ICUs within the same period. Among them, 105 (n=4%) presented with cancer (51 patients had hematological malignancy and 54 patients had solid tumors). 409 of 420 patients were included in the propensity score matching process, of whom 307 patients in the non-cancer group and 102 patients in the cancer group. 145 patients (35%) died in the ICU at day 60, 59 (56%) with cancer and 86 (27%) without cancer. In the primary logistic regression analysis, the odds ratio for death associated to cancer was 2.3 (95%CI 1.24 - 4.28, p=0.0082) higher for cancer patients than for a non-cancer patient at ICU admission. Exploratory multivariable analyses showed that solid tumor (OR: 2.344 (0.87-6.31), p=0.062) and hematological malignancies (OR: 4.144 (1.24-13.83), p=0.062) were independently associated with mortality. Interpretation: Patients with cancer and requiring ICU admission for SARS-CoV-2 infection had an increased mortality, hematological malignancy harboring the higher risk in comparison to solid tumors.

Brain Biopsy for Neurological Diseases of Unknown Etiology in Critically Ill Patients: Feasibility, Safety, and Diagnostic Yield.

OBJECTIVES: Brain biopsy is a useful surgical procedure in the management of patients with suspected neoplastic lesions. Its role in neurologic diseases of unknown etiology remains controversial, especially in ICU patients. This study was undertaken to determine the feasibility, safety, and the diagnostic yield of brain biopsy in critically ill patients with neurologic diseases of unknown etiology. We also aimed to compare these endpoints to those of non-ICU patients who underwent a brain biopsy in the same clinical context. DESIGN: Monocenter, retrospective, observational cohort study. SETTING: A French tertiary center. PATIENTS: All adult patients with neurologic diseases of unknown etiology under mechanical ventilation undergoing in-ICU brain biopsy between January 2008 and October 2020 were compared with a cohort of non-ICU patients. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Among the 2,207 brain-biopsied patients during the study period, 234 biopsies were performed for neurologic diseases of unknown etiology, including 29 who were mechanically ventilated and 205 who were not ICU patients. Specific histological diagnosis and final diagnosis rates were 62.1% and 75.9%, respectively, leading to therapeutic management modification in 62.1% of cases. Meningitis on prebiopsy cerebrospinal fluid analysis was the sole predictor of obtaining a final diagnosis (2.3 [1.4-3.8]; p = 0.02). ICU patients who experienced therapeutic management modification after the biopsy had longer survival (p = 0.03). The grade 1 to 4 (mild to severe) complication rates were: 24.1%, 3.5%, 0%, and 6.9%, respectively. Biopsy-related mortality was significantly higher in ICU patients compared with non-ICU patients (6.9% vs 0%; p = 0.02). Hematological malignancy was associated with biopsy-related mortality (1.5 [1.01-2.6]; p = 0.04). CONCLUSIONS: Brain biopsy in critically ill patients with neurologic disease of unknown etiology is associated with high diagnostic yield, therapeutic modifications and postbiopsy survival advantage. Safety profile seems acceptable in most patients. The benefit/risk ratio of brain biopsy in this population should be carefully weighted.

Endothelial cell biomarkers in critically ill COVID-19 patients with encephalitis.

COVID-19 is associated with encephalitis in critically ill patients and endothelial dysfunction seems to contribute to this life-threatening complication. Our objective was to determine the hallmark of endothelial activation in COVID-19-related encephalitis. In an observational study in intensive care unit (ICU), we compared vascular biomarkers of critically ill COVID-19 patients with or without encephalitis. To be classified in the encephalitis group, patients had to have new onset of central neurologic symptom, and pathological findings on either brain magnetic resonance imaging (MRI) and/or electroencephalogram (EEG). Among the 32 critically ill COVID-19 consecutive patients, 21 were categorized in the control group and 11 in the encephalitis group. Encephalitis patients had a longer ICU stay than control patients (median length [25th-75th percentile] of 52 [16-79] vs. 20.5 [11-44] days, respectively, p = 0.04). Nine-month overall follow-up mortality reached 21% (7/32 patients), with mortality rates in the encephalitis group and the control group of 27% and 19%, respectively. Encephalitis was associated with significant higher release of soluble endothelial activation markers (sE-selectin, tumor necrosis factor-α (TNF-α), interleukin 6, placental growth factor, and thrombomodulin), but these increases were correlated with TNF-α plasmatic levels. The hypoxia-inducible protein angiopoietin-like 4 (ANGPTL4) was at significantly higher levels in encephalitis patients compared to control patients (p = 0.0099), and in contrary to the other increased factors, was not correlated with TNF-α levels (r = 0.2832, p = 0.1163). Our findings suggest that COVID-19-related encephalitis is a cytokine-associated acute brain dysfunction. ANGPTL4 was the only elevated marker found in encephalitis patients, which was not correlated with systemic inflammation, suggesting that ANGPTL4 might be a relevant factor to predict encephalitis in critically ill COVID-19 patients.

Electrical Impedance Tomography Monitoring of Bronchoalveolar Lavage in Patients With Acute Respiratory Distress Syndrome.

OBJECTIVES: The impact of bronchoalveolar lavage on regional ventilation in mechanically ventilated patients with acute respiratory distress syndrome has rarely been described. Our objectives were use electrical impedance tomography to describe lung impedance variation post bronchoalveolar lavage and identify morphologic patterns according to respiratory failure severity. DESIGN: Monocenter physiologic study on mechanically ventilated patients. SETTING: University medical ICU. INTERVENTIONS: After a recruitment maneuver, tidal impedance variation distributions (a surrogate for impact of bronchoalveolar lavage on tidal volume distribution), end-expiratory lung impedance (correlated with end-expiratory lung volume and used to quantify postbronchoalveolar lavage derecruitment), respiratory mechanics, and blood gases were recorded before and over 6 hours post bronchoalveolar lavage with Pao2 to the Fio2 ratio. Patients were grouped according to their prebronchoalveolar lavage, that is, Pao2 to the Fio2 ratio less than 200 or greater than or equal to 200. RESULTS: Twenty-one patients (median [interquartile range] age 55 yr [50-58 yr]; 13 males), 13 with Pao2 to the Fio2 ratio less than 200, were included. Unlike that latter group, bronchoalveolar lavage significantly impacted tidal impedance variation distribution in patients with Pao2 to the Fio2 ratio greater than or equal to 200, with a ventilation shift to the contralateral lung (from 54% to 42% in the bronchoalveolar lavage side), which persisted up to 6 hours post bronchoalveolar lavage. Similarly, end-expiratory lung impedance was less distributed in the bronchoalveolar lavage side post procedure of patients with Pao2 to the Fio2 ratio greater than or equal to 200, but the difference did not reach statistical significance (p = 0.09). As reported for tidal impedance variation, end-expiratory lung impedance distribution in patients with severe or moderate acute respiratory distress syndrome did not change significantly during the 6 hours post bronchoalveolar lavage. Although bronchoalveolar lavage effects on gas exchanges were minor in all patients, static compliance in patients with Pao2 to the Fio2 ratio greater than or equal to 200 was significantly lower post bronchoalveolar lavage (p = 0.04). CONCLUSIONS: The negative impact of bronchoalveolar lavage on regional ventilation, which persisted at least 6 hours, appeared to be more profound in patients with normal lung function or mild acute respiratory distress syndrome. In contrast, regional ventilation, lung recruitment, respiratory mechanics, and gas exchanges were modestly impacted by the bronchoalveolar lavage in patients with severe or moderate acute respiratory distress syndrome. That finding is reassuring and supports not summarily proscribing bronchoalveolar lavage for the most severely ill with acute respiratory distress syndrome.

Relationship between ventilator-associated pneumonia and mortality in COVID-19 patients: a planned ancillary analysis of the coVAPid cohort.

BACKGROUND: Patients with SARS-CoV-2 infection are at higher risk for ventilator-associated pneumonia (VAP). No study has evaluated the relationship between VAP and mortality in this population, or compared this relationship between SARS-CoV-2 patients and other populations. The main objective of our study was to determine the relationship between VAP and mortality in SARS-CoV-2 patients. METHODS: Planned ancillary analysis of a multicenter retrospective European cohort. VAP was diagnosed using clinical, radiological and quantitative microbiological criteria. Univariable and multivariable marginal Cox's regression models, with cause-specific hazard for duration of mechanical ventilation and ICU stay, were used to compare outcomes between study groups. Extubation, and ICU discharge alive were considered as events of interest, and mortality as competing event. FINDINGS: Of 1576 included patients, 568 were SARS-CoV-2 pneumonia, 482 influenza pneumonia, and 526 no evidence of viral infection at ICU admission. VAP was associated with significantly higher risk for 28-day mortality in SARS-CoV-2 (adjusted HR 1.70 (95% CI 1.16-2.47), p = 0.006), and influenza groups (1.75 (1.03-3.02), p = 0.045), but not in the no viral infection group (1.07 (0.64-1.78), p = 0.79). VAP was associated with significantly longer duration of mechanical ventilation in the SARS-CoV-2 group, but not in the influenza or no viral infection groups. VAP was associated with significantly longer duration of ICU stay in the 3 study groups. No significant difference was found in heterogeneity of outcomes related to VAP between the 3 groups, suggesting that the impact of VAP on mortality was not different between study groups. INTERPRETATION: VAP was associated with significantly increased 28-day mortality rate in SARS-CoV-2 patients. However, SARS-CoV-2 pneumonia, as compared to influenza pneumonia or no viral infection, did not significantly modify the relationship between VAP and 28-day mortality. CLINICAL TRIAL REGISTRATION: The study was registered at ClinicalTrials.gov, number NCT04359693.

Distinct cytokine profiles associated with COVID-19 severity and mortality.

BACKGROUND: Markedly elevated levels of proinflammatory cytokines and defective type-I interferon responses were reported in patients with coronavirus disease 2019 (COVID-19). OBJECTIVE: We sought to determine whether particular cytokine profiles are associated with COVID-19 severity and mortality. METHODS: Cytokine concentrations and severe acute respiratory syndrome coronavirus 2 antigen were measured at hospital admission in serum of symptomatic patients with COVID-19 (N = 115), classified at hospitalization into 3 respiratory severity groups: no need for mechanical ventilatory support (No-MVS), intermediate severity requiring mechanical ventilatory support (MVS), and critical severity requiring extracorporeal membrane oxygenation (ECMO). Principal-component analysis was used to characterize cytokine profiles associated with severity and mortality. The results were thereafter confirmed in an independent validation cohort (N = 86). RESULTS: At time of hospitalization, ECMO patients presented a dominant proinflammatory response with elevated levels of TNF-α, IL-6, IL-8, and IL-10. In contrast, an elevated type-I interferon response involving IFN-α and IFN-ß was characteristic of No-MVS patients, whereas MVS patients exhibited both profiles. Mortality at 1 month was associated with higher levels of proinflammatory cytokines in ECMO patients, higher levels of type-I interferons in No-MVS patients, and their combination in MVS patients, resulting in a combined mortality prediction accuracy of 88.5% (risk ratio, 24.3; P < .0001). Severe acute respiratory syndrome coronavirus 2 antigen levels correlated with type-I interferon levels and were associated with mortality, but not with proinflammatory response or severity. CONCLUSIONS: Distinct cytokine profiles are observed in association with COVID-19 severity and are differentially predictive of mortality according to oxygen support modalities. These results warrant personalized treatment of COVID-19 patients based on cytokine profiling.

The cerebral network of COVID-19-related encephalopathy: a longitudinal voxel-based 18F-FDG-PET study.

PURPOSE: Little is known about the neuronal substrates of neuropsychiatric symptoms associated with COVID-19 and their evolution during the course of the disease. We aimed at describing the longitudinal brain metabolic pattern in COVID-19-related encephalopathy using 18F-FDG-PET/CT. METHODS: Seven patients with variable clinical presentations of COVID-19-related encephalopathy were explored thrice with brain 18F-FDG-PET/CT, once in the acute phase, 1 month later and 6 months after COVID-19 onset. PET images were analysed with voxel-wise and regions-of-interest approaches in comparison with 32 healthy controls. RESULTS: Patients' neurological manifestations during acute encephalopathy were heterogeneous. However, all of them presented with predominant cognitive and behavioural frontal disorders. SARS-CoV-2 RT-PCR in the CSF was negative for all patients. MRI revealed no specific abnormalities for most of the subjects. All patients had a consistent pattern of hypometabolism in a widespread cerebral network including the frontal cortex, anterior cingulate, insula and caudate nucleus. Six months after COVID-19 onset, the majority of patients clinically had improved but cognitive and emotional disorders of varying severity remained with attention/executive disabilities and anxio-depressive symptoms, and lasting prefrontal, insular and subcortical 18F-FDG-PET/CT abnormalities. CONCLUSION: The implication of this widespread network could be the neural substrate of clinical features observed in patients with COVID-19, such as frontal lobe syndrome, emotional disturbances and deregulation of respiratory failure perception. This study suggests that this network remains mildly to severely impaired 6 months after disease onset.