key: cord-0891477-mdfzpyzj
authors: Fournier, Morgane; Faille, Dorothée; Dossier, Antoine; Mageau, Arthur; Roland, Pascale Nicaise; Ajzenberg, Nadine; Borie, Raphael; Bouadma, Lila; Bunel, Vincent; Castier, Yves; Choquet, Christophe; Crestani, Bruno; Daugas, Eric; Deconinck, Laurene; Descamps, Diane; Descamps, Vincent; Dieudé, Philippe; Ducrocq, Gregory; Faucher, Nathalie; Goulenok, Tiphaine; Guidoux, Céline; Khalil, Antoine; Lavallée, Philippa; Lescure, François Xavier; Lortat-Jacob, Brice; Mal, Hervé; Mutuon, Pierre; Pellenc, Quentin; Steg, Philippe Gabriel; Taille, Camille; Timsit, Jean Francois; Yazdanpanah, Yazdan; Papo, Thomas; Sacré, Karim
title: Arterial Thrombotic Events in Adult Inpatients With COVID-19
date: 2020-11-27
journal: Mayo Clin Proc
DOI: 10.1016/j.mayocp.2020.11.018
sha: bfe30b741f88791e5a40890c5cac57733525cc2d
doc_id: 891477
cord_uid: mdfzpyzj

Objective Coronavirus disease 2019 (COVID-19) infection triggers thrombotic events. We aimed to evaluate the clinical course and risk factors for arterial thrombotic events (AT) in adult inpatients with COVID-19. Patients and Methods All consecutive adult patients admitted for COVID-19 infection in a referral center in France and discharged from the hospital between April 1 and April 30, 2020 were included. All AT that occurred through discharge were considered for analysis. Epidemiological, demographic, clinical, laboratory, treatment, and outcome data were extracted from electronic medical records using a standardized data collection form. Results Overall, 531 COVID-19+ patients were analyzed. Among them, 30 (5.6%) experienced AT events. AT in the setting of COVID-19 infection happened at a median of 11 [5, 20] days after the first symptoms of infection, occurred in high risk patients according to traditional cardiovascular risk factors, had an atypical pattern such as thrombosis of the aorta, upper limb or renal arteries, or cerebral microvasculopathy in 7 (23.3%) cases and were associated with an in-hospital mortality rate of 40%. AT increased the risk of death by 3 folds in COVID-19+ patients (HR 2.96 95%CI [1.4-4.7], P=.002). Subdistribution survival hazard model showed that a concentration of D-dimer >1250 ng/mL increased by >7 (sdHR 7.68 95%CI [2.9-20.6], P<.001) the risk of AT in COVID-19 + patients. Conclusion A dramatically high rate of in-hospital death was observed in patients who suffered AT in the setting of COVID-19 infection. A D-dimer level >1250 ng/mL at entry may identify COVID-19+ patients at risk for AT.

Given the rapidly growing pandemic, there is an urgent need to measure the rate of thrombotic manifestations associated with COVID-19 and analyze the impact on survival.

Most studies in the field have reported on venous thrombosis and focused on patients hospitalized in intensive care units. The primary goal of this study was to assess the incidence of, and risk factors for, arterial thrombotic events in all hospitalized patients with COVID-19 in a referral center in France over 1 month.

All consecutive adult patients admitted for COVID-19 infection at Bichat Hospital, Paris, and discharged from the hospital between April 1 and April 30, 2020 were retrospectively reviewed. Bichat Hospital, Paris, is a referral center in France for patients with suspected or confirmed emerging infectious diseases 2 . Inclusion criteria for COVID 19 inpatients were: age > 18 years old, diagnosis of COVID-19 according to WHO interim guidance, and positive SARS-CoV-2 RT-PCR testing on a respiratory sample (nasopharyngeal swab or invasive respiratory sample). COVID-19 infection was considered despite a negative SARS-CoV-2 RT-PCR testing at time of admission in patients who had 1-close contact with a confirmed COVID-19 case prior to onset of symptoms, 2-symptoms suggestive of COVID-19 infection (cough, fever, shortness of breath, sudden onset of anosmia, ageusia or dysgeusia), 3-typical CT scan appearance ("crazy paving" pattern) and 4-no alternative diagnosis. Patients not admitted because of mild clinical symptoms or who were transferred to another hospital or in whom COVID-19 infection were not confirmed after review were excluded. International Classification of Disease code (ICD-10) for COVID 19 infection (U071) was used to screen J o u r n a l P r e -p r o o f patients. Data were extracted from the French Diagnosis Related Groups (DRG)-based information system (PMSI) database. Information regarding eligibility conditions, exclusion criteria, epidemiological, demographic, clinical, laboratory, treatment, and outcome data were extracted from electronic medical records using a standardized data collection form.

Routine blood tests included at least complete blood count, electrolytes, creatinine, liver tests, lactate dehydrogenase (LDH), C reactive protein (CRP), D-Dimer and ferritin.

Laboratory confirmation for SARS-CoV-2 was obtained at the Bichat Hospital Virology Department, Paris, France. RT-PCR assays were performed in accordance with the protocol The severity of COVID-19 was analyzed at the time of admission based on the extent (<10%, between 10 and 25%, between 25 and 50% and >50%) of lung involvement on CT-scan.

All consecutive adult COVID-19 negative patients (Supplemental Table) admitted for arterial thrombotic events at Bichat Hospital, Paris, and discharged from the hospital between March 1 and March 31, 2020 (Supplemental Figure 1 ) were used as controls

International Classification of Disease codes (ICD-10) for arterial thrombotic events (AT) were used to screen patients and controls including I21.x, I236, I240, I513, I63.x, I65.x, I66., I676, I74.x I788, H348, H349. All arterial thrombotic events that occurred through discharge were considered for analysis. Extensive screening for conventional cardiovascular risk factors (age, sex, smoking status, body-mass index, dyslipidemia, past cardiovascular events (CVE), high blood pressure, diabetes mellitus) was undertaken in all cases. 

Our study is a human non-interventional study where 1-subjects were assigned to a diagnosis strategy within current practice, 2-study involved products with a marketing authorization that are prescribed in the usual manner and used in accordance with French agencies authorizations, 3-epidemiological methods were used to analyze the data, and 4-information used in the study were collected for clinical care. According to the Public Health French Law 

Thirty COVID-19+ patients (5.6%) experienced arterial thrombotic events (AT) that occurred 2 [0,11] days after admission and 11 [5, 20] days after first symptoms of COVID-19 infection.

AT was the first symptom of COVID infection in 1 (3.3%) patient and the main reason for admission in 12 (40%), respectively. Arterial thrombotic events consisted of myocardial infarction (n=9, 30%), stroke (n=8, 26.7%) and acute/subacute limb ischemia (n=6, 20%).

Interestingly, age, sex, and high blood pressure, smoking, diabetes, dyslipidemia, past CVE and BMI were not statistically different between COVID-19+ and COVID-19-control inhospital patients with AT (Supplemental Table) . However, despite a similar risk for CVE, COVID-19+ patients displayed atypical AT (n=7, 23.3%), including massive aortic thrombosis (n=2), splenic infarct (n=1), renal artery thrombosis (n=1), diffuse cerebral microvasculopathy (n=1) and upper extremity arterial thrombosis (n=1), that were not observed in the control group (Supplemental Table) . In-hospital death was dramatically higher in COVID-19+ patients with AT as compared to COVID-19-AT controls (40% vs 

Considering its poor prognosis, we aimed to identify risk factors for AT in COVID-19+ patients. By comparing the characteristics of COVID-19+ patients with and without AT, we observed that an history of past CVE and a higher plasma level of D-dimer were both associated with AT (Table 1 ). In the multivariable analysis, only D-dimer level was associated with AT in COVID-19+ patients ( Table 2) . 

Prior studies varied regarding the incidence of thrombosis in patients with COVID-19 but all suggested a heightened risk, particularly for deep vein thrombosis and pulmonary embolism, 3, 4 . Few studies however specifically focused on arterial thrombotic events in COVID-19+ patients. A previous published study reported an incidence of 11.1% of AT among 3334 consecutive hospitalized COVID-19 patients in 4 hospitals in New York City 5 . In our referral center in France we found a lower AT rate of 5.6%, more consistent with other published studies in Spain and Italy 6, 7 . D-dimer level at admission was high and consistent with an early distinct coagulation disorder associated with COVID-19. Evidence for abnormal coagulation parameters associated with COVID-19 appeared in early reports from China 8 . Disseminated intravascular coagulation could be ruled out since D-dimer levels were increased far out of proportion with any abnormalities in PT, APTT, fibrinogen level, or platelets count (Table 1) Whether the elevated D-dimer levels and thrombotic manifestations are caused by a specific COVID-19 coagulopathy or merely related to coagulation activation associated with severe inflammation is unknown. In our study, contrasting with previous reports 13, 14 , arterial thrombotic events were not associated with inflammatory markers such as CRP.

Elevated D-dimer levels at admission, are known to be associated with increased mortality in COVID-19 1, 5, 13, 15, 16 . In our study, D-dimer level was also independently and strongly associated with arterial thrombotic events. According to our findings, D-dimer levels at admission may also identify patients at high risk for arterial thrombotic events. Low molecular weight heparin was associated with lower mortality in patients with markedly elevated d-dimer 17 . D-dimer has also been used to guide anticoagulation 18 . Noteworthy in our series arterial thrombotic events occurred although most patients were receiving anticoagulation according to current recommendations 19 .

Both the fact that numerous arterial events were unusual -such as thrombosis of the aorta, upper limb or renal arteries, or cerebral microvasculopathy-and the strikingly high risk of death, as compared with AT control patients with no COVID infection, are suggestive of a COVID-19 specific arterial vasculopathy. COVID-19 infection causes cytokine storm, hypoxic injury, hypercoagulability, increased platelet activity and endothelial damage 1, 20-23 .

In our study, COVID-19+AT+ patients shared the same cardiovascular risk burden than the general population. COVID-19 infection may act as an acute second hit in patients at otherwise high risk for cardiovascular events according to traditional risk factors.

Our study has several limitations. First, it is a retrospective monocentric observational study.

Second, around 6% of patients have highly probable but not microbiologically confirmed COVID-19 infection. In clinical practice however, many patients presented with a negative RT-PCR test for COVID-19 even though they have typical lung manifestations and a highly suspected contact history. In our study, patients with a negative RT-PCR had recent exposure, [1] [2] [3] [4] [5] [6] [7] [8] [9] [10] [11] [12] [13] [14] days before AT onset in AT+ patients) and available for analysis in 318 patients. The reference concentration of D-dimer at our institution is < 234 ng/mL AT, arterial thrombotic events

COVID-19 and its implications for thrombosis and anticoagulation

Challenges and issues about organizing a hospital to respond to the COVID-19 outbreak: experience from a French reference centre

Prevalence of venous thromboembolism in patients with severe novel coronavirus pneumonia

Incidence of thrombotic complications in critically ill ICU patients with COVID-19

Thrombosis in hospitalized patients with COVID-19 in a New York City Health System

Incidence and consequences of systemic arterial thrombotic events in COVID-19 patients

Venous and arterial thromboembolic complications in COVID-19 patients admitted to an academic hospital in

Epidemiological and clinical characteristics of 99 cases of 2019 novel coronavirus pneumonia in Wuhan, China: a descriptive study

SSC on Disseminated Intravascular Coagulation of the ISTHJ. The scoring system of the Scientific and Standardisation Committee on Disseminated Intravascular Coagulation of the International Society on Thrombosis and Haemostasis: a 5-year overview

The perils of D-dimer in the medical intensive care unit

D-dimer correlates with proinflammatory cytokine levels and outcomes in critically ill patients

COVID-19 and coagulation: bleeding and thrombotic manifestations of SARS-CoV-2 infection

C-reactive protein and risk of venous thromboembolism: results from a population-based case-crossover study

Clinical Characteristics of Coronavirus Disease 2019 in China

Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Anticoagulant treatment is associated with decreased mortality in severe coronavirus disease 2019 patients with coagulopathy

Thromboembolic risk and anticoagulant therapy in COVID-19 patients: emerging evidence and call for action

Scientific and Standardization Committee on DIC of the International Society on Thrombosis and Haemostasis. The progression from coagulopathy to disseminated intravascular coagulation in representative underlying diseases

Subcommittee on Disseminated Intravascular Coagulation

Coagulation abnormalities and thrombosis in patients with COVID-19

Deep Vein Thrombosis in Hospitalized Patients With COVID-19 in Wuhan, China: Prevalence, Risk Factors, and Outcome

59 Follow up Length of hospitalisation, days

AT, arterial thrombotic events; BMI, body mass index, calculated as weight in kilograms divided by height in meters squared

ESRD, endstage renal disease; IQR, interquartile range Critical illness defined as mechanical ventilation. In-hospital death defined as death at discharge. Inflammatory disorders were considered when requiring steroid use and included asthma (n=13) and chronic autoimmune/inflammatory systemic diseases (n=25)

colic (n=7) cancers, low grade non-Hodgkin lymphoma (n=7), skin cancer (n=5), myelodysplastic syndrome (n=4), lung (n=2), endometrial/ovarian (n=2), liver cancer (n=1), pancreatic (n=1), and bladder (n=1) cancers. Solid-Organ Transplantation included kidney (n=19)

Anti-IL1 treatment was anakinra; Anti-IL6 treatment was tocilizumab; Antiviral drugs included lopinavir/ritonavir (n=155) and remdesivir (n=7)

Anticoagulant treatment was given at prophylactic or therapeutic dose in 217 (81.3%) and 50 (18.7%) patients, respectively. The reference concentration of D-dimer at our institution is < 234 ng/mL