key: cord-1038184-7k49jyaw
authors: Jimenez-Guiu, Xavier; Huici-Sánchez, Malka; Romera-Villegas, Antonio; Izquierdo-Miranda, Alexandre; Sancho-Cerro, Ana; Vila-Coll, Ramon
title: Deep vein thrombosis in non-critically ill patients with COVID-19 pneumonia. DVT in non-ICU patients.
date: 2020-09-07
journal: J Vasc Surg Venous Lymphat Disord
DOI: 10.1016/j.jvsv.2020.08.028
sha: c2139ee90967948f773f7825c4271b6edcaad02d
doc_id: 1038184
cord_uid: 7k49jyaw

BACKGROUND / OBJECTIVES: Venous thromboembolic events are one of the main causes of mortality among hospitalized patients with COVID-19 pneumonia. The aim of our study is to describe the prevalence of deep vein thrombosis in non-critical patients with COVID-19 pneumonia and correlate such observations with the thromboprophylaxis received. PATIENTS/METHODS: Prospective cohort study of 68 patients admitted to the hospital for COVID-19 pneumonia. Diagnosis was confirmed with PCR testing of nasopharyngeal specimens. Deep veins were examined using compression with the transducer on B-mode. Patients were separated into two groups for statistical analysis: those undergoing LMWH prophylaxis and those receiving intermediate or complete anticoagulation treatment. Risk analysis and logistic regression were performed. RESULTS: We included 57 patients, on whom a CDUS was performed. Of these patients, 49.1% were women, with a mean age of 71.3 years old. Six patients were diagnosed with DVT; the prevalence of in-hospital DVT in patients with COVID-19 pneumonia was 10.5%. All patients who presented with a DVT were on LMWH prophylaxis. Patients undergoing prophylactic anticoagulation treatment had a higher risk of DVT by 16.21% [95% CI 0.04-0.28 (p-value 0.056)], when compared to those treated with intermediate or complete anticoagulation treatment. We also found a protective factor for DVT in the intermediate or complete anticoagulation treatment group [OR of 0.19 (95% CI 0.08-0.46) p-value <0.05]. CONCLUSIONS: Non-critically ill, hospitalized patients with COVID-19 pneumonia are at high risk of deep vein thrombosis despite correct, standard thromboprophylaxis.

This prospective single-center study found a prevalence of 10.5% 12 DVT in patients admitted to our hospital due for COVID-19 pneumonia despite LMWH 13 prophylaxis, and found that intermediate or complete anticoagulation treatment was a protective 14 factor for DVT. The study suggest that non-critically ill patients with COVID-10 pneumoniae are 15 at high risk for DVT. 16

After the outbreak of SARS-CoV-2, healthcare systems from around the world have unified 2 resources and knowledge to contain the spread of the virus. Although coronaviruses have been 3 well studied, SARS-CoV-2 has been reported as having particular characteristics that distinguish 4 it from other viruses of the same family. When compared with SARS-CoV, SARS-CoV-2 has a 5 lower mortality rate; however its ability to infect human hosts appears higher and the incubation 6 period longer [1] . During infection, plasma levels of proinflammatory and prothrombotic 7

cytokines such as C-RP, IL-6, IL-8 and D-dimer may have a role in the physiopathology of 8 COVID-19-induced SARS. Some studies have reported high levels of those cytokines, especially 9

in patients with more serious conditions [2] . 10

One main complication that has been described to arise in hospitalized patients with COVID-19 11 is venous thromboembolic events. In general, venous thromboembolic events are one of the 12 main causes of mortality among hospitalized patients [2] . Diagnostic algorithms rely upon 13 clinical history, physical examinations and prior diseases to assess clinical probability scales; 14 compression duplex ultrasound (CDUS) and CTA are gold-standard diagnostic tools for deep 15 vein thrombosis (DVT) and pulmonary embolism (PE) [3, 4] We therefore aim to describe the prevalence of DVT among non-critically ill, hospitalized 5 patients with COVID-19 pneumonia and correlate such findings with the thromboprophylaxis 6 received. 7 8

We performed a prospective cohort study of 68 patients admitted to our hospital due for COVID-10 19 pneumonia during April 2020. Diagnosis was confirmed with PCR testing of nasopharyngeal 11

All patients were clinically assessed by a vascular surgeon and Wells score was calculated. 13

Inclusion criteria were as follows: patients with COVID-19 pneumonia who were admitted to 14 the emergency department and required hospitalization. Exclusion criteria were as follows: (1) 15 palliative treatment, (2) pregnancy, (3) thromboembolic event diagnosis prior to hospital 16 admission, (4) required intensive care treatment and (5) refusal to participate in this study. 17

Demographic data included cardiovascular risk factors (dyslipidemia, high blood pressure, 18 diabetes mellitus, smoking), chronic pulmonary or cardiac diseases (chronic pulmonary 19 obstructive disease and chronic ischemia heart disease or atrial fibrillation), need for oxygen 20 therapy and need for corticotherapy (1mg/kg of methylprednisolone administered every 24 hours 21 intravenously). Categorical variables were presented as the number of cases and percentages; continuous 7 variables were presented as the mean and standard deviation when there was normal distribution; 8 and median and interquartile range (IQR) when there was not normal distribution. 9

Patients were divided into two groups for statistical analysis: those undergoing LMWH 10 prophylaxis and those receiving intermediate or complete anticoagulation treatment. 11

We performed a risk analysis and logistic regression test. 12

Differences between groups were assessed by the chi-squared test (Yates correction) and T-13 student when appropriate. 14 Statistical significance was considered to have been achieved if P value was less than 0.05. 15

The investigation project was approved by the Investigation Ethics Committee of Bellvitge 16

Our study population included 67 patients admitted to the hospital with unilateral or bilateral 5 pneumonia caused by COVID-19 during April 2020. 6

A total of 10 patients were excluded from the study in accordance with the criteria set: 3, refusal 7

to participate in the study; 3, in palliative treatment; and 4, thromboembolic event diagnosis prior 8 to hospital admission (3 presented with DVT and PE; 1, with popliteal DVT). Fifty-seven 9 patients were finally included, comprising 28 (49.1%) women and 29 (50.9%) men. The mean 10 age of this series of patients was 71.3 (SD 12.7) years. All included patients underwent CDUS 11 within a median time of 9 days of hospitalization (IQR 6-15). Baseline characteristics of our 12 sample are described in Table I . 13

Clinical pretest probability of venous thromboembolism as determined by the Wells score was 14 unlikely in 98.2% of patients and likely in 1.8% of patients. Six patients were diagnosed with 15

DVT. The prevalence of in-hospital DVT in patients with COVID-19 pneumonia was 10.5%. 16

Of patients with DVT, five had an unlikely Wells score and one patient had a likely Wells score.

All of our patients experienced asymptomatic DVT except for one individual who mentioned 1 pain in the calf region and had a swollen leg in the prior 24 hours. The most common area 2 affected was distal (83.3%) and one patient presented with femoropopliteal DVT (16.7%). Corticosteroid treatment was used in 25 (43%) patients. The incidence of DVT was 12% in 18 patients with corticosteroid treatment and 9.3% in patients without corticosteroid treatment. 19

There was no statistically significant difference between both groups (p-value 0.74). 20

Sepsis was present in 19 (33%) patients. There was a tendency of more cases of DVT in patients 21 with sepsis (21%) than in those without sepsis (5.2%) (p-value of 0.067). 22 1 DVT and 576ng/mL (SD 570) in patients with DVT (p-value 0.64). 2

Patients with DVT had a mean IL-6 level of 12.8pg/mL (SD 11.8) and patients without DVT had 3 a mean IL-6 level of 15.2pg/mL (SD13.9) (p-value 0.69). 4

Bleeding complications were found in one patient belonging to the intermediate or complete 5 anticoagulation group (5%). No differences were observed between both groups (p-value 0.22). Venous thromboembolic disease is reported as one of the major complications arising patients 11

with COVID-19 [9] . Data reported from initial autopsies shows that non-suspected PE or DVT 12 may account for 58% of deaths and 33% as the main cause of death [16] . Nonetheless, it is not 13 clearly known whether venous thromboembolic events were the main cause of clinical worsening 14 and death among hospitalized patients, or complications induced by severe pneumonia and 15 inflammatory responses. For that reason, we hold the belief that our efforts should also be to 16 focus on a non-critically ill, hospitalized population to prevent thromboembolic complications 17 that are the most common events. 18

Our study is indeed one of the first to analyze asymptomatic DVT in a non-ICU setting. Even 19 with correct, standard prophylaxis with LMWH, a prevalence of DVT of 10.5% was observed. 20

While there was one case of a femoropopliteal DVT, the remaining cases were infrapopliteal 21 DVT. Management of infrapopliteal DVT remains under discussion; however, previously 22

reported data of a standard population demonstrated a proximal progression of 15.5% [17] . In the 23 opinion of our group, infrapopliteal DVT may be treated, as there is an unstudied procoagulant 1 state in patients with COVID-19 that could lead to higher progression rates. 2

The high proportion of patients with DVT may be explained by dysregulation of the coagulation 3 system, which has been observed in patients with COVID-19 pneumonia. A recent study by 4

Panigada M et al investigated the inflammatory response secondary to COVID-19 by measuring 5 blood from ICU-admitted patients with thromboelastography. These authors' findings 6 highlighted that, more than an acute disseminated intravascular coagulation (with low platelet 7 count and fibrinogen clotting activity, and high fibrin degradation products), the inflammatory 8 response leads to a hypercoagulable state with high levels of fibrinogen, D-dimer, C-Reactive 9

Protein, VIII factor, Von Willebrand factor and C-reactive protein; low levels of antithrombin; 10 and normal levels of platelets, prothrombin time and activated partial thromboplastin time [18] . 11

A high procoagulant state, alongside sepsis and medical treatment (e.g. corticotherapy) may be 12 the reason as to why patients with COVID-19 pneumonia are at a higher risk of thromboembolic 13 events when compared to a standard hospitalized population. 19 may be different from that in patients without COVID-19 pneumonia. Investigators suggested 8 that in situ immunothrombosis was the main cause. However, despite sample size, the main 9 limitation of the study was that patients did not undergo a lower limb CDUS scan [21] . 10

Our study group believes that physiopathology notwithstanding, selected patients may benefit 11 from intermediate anticoagulation dosages and avoid a thromboembolic event. Despite standard 12

prophylaxis, this was the case in six DVT. 13

These findings and results are noteworthy, as we are asked to wonder whether standard 14

prophylaxis treatments suffice and if the scientific community should indeed be more aggressive 15 in DVT prevention in patients with COVID-19 pneumonia. Clinically randomized trials would 16 best address these questions to improve clinical practices and provide better patient care. J o u r n a l P r e -p r o o f

Non-critically ill patients admitted to hospital for COVID-19 pneumonia are at high risk of deep 2 vein thrombosis despite correct

of Novel Coronavirus-Infected Pneumonia

Clinical course and risk factors for mortality of adult 5 inpatients with COVID-19 in Wuhan, China: a retrospective cohort study

Diagnostic methods for deep vein thrombosis

CTPA as the gold standard for the diagnosis 10 of pulmonary embolism

Risk assessment of venous 13 thromboembolism and bleeding in COVID-19 patients. Pulmonology

Der Nigoghossian C 15 et al. COVID-19 and Thrombotic or Thromboembolic Disease: Implications for Prevention

All authors contributed substantially to the study design, acquisition, analysis, and interpretation 6 of the data. X.J. Guiu, M. H. Sanchez and A.R. Villegas drafted the first version of the 7 manuscript. All authors revised and approved the final version to be published. 8

We would like to thank Eva Cuenca Velasco and Gloria Plana Meler for their help during the 11

We would also like to thank Anthony Armenta for providing medical editing assistance for the 13 article at hand. 14