key: cord-0949065-fju6f33n
authors: Di Minno, Alessandro; Ambrosino, Pasquale; Calcaterra, Ilenia; Di Minno, Matteo Nicola Dario
title: COVID-19 and Venous Thromboembolism: A Meta-analysis of Literature Studies
date: 2020-09-03
journal: Semin Thromb Hemost
DOI: 10.1055/s-0040-1715456
sha: 814af6370e3cece1daa70d391d50e82fac343bbd
doc_id: 949065
cord_uid: fju6f33n

Coronavirus disease 2019 (COVID-19) may have a wide spectrum of clinical presentations, leading in some cases to a critical condition with poor long-term outcomes and residual disability requiring post-acute rehabilitation. A major concern in severe COVID-19 is represented by a concomitant prothrombotic state. However, contrasting data are available about the prevalence of venous thromboembolism (VTE), including deep vein thrombosis (DVT) and/or pulmonary embolism (PE). A detailed search on the association of COVID-19 with thromboembolic complications was conducted in the main electronic databases (PubMed, Web of Science, and Scopus) according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. The weighted mean prevalence (WMP) with 95% confidence interval (95% CI) was calculated with the random-effects model. Twenty studies enrolling 1,988 COVID-19 patients were included. The WMP of VTE was 31.3% (95% CI: 24.3–39.2%). The WMP of DVT was 19.8% (95% CI: 10.5–34.0%), whereas the WMP of PE was 18.9% (95% CI: 14.4–24.3%). Similar results were obtained when specifically analyzing studies on patients admitted to intensive care units and those on patients under antithrombotic prophylaxis. Regression models showed that an increasing age was associated with a higher prevalence of VTE (Z-score: 3.11, p = 0.001), DVT (Z-score: 2.33, p = 0.002), and PE (Z-score: 3.03, p = 0.002), while an increasing body mass index was associated with an increasing prevalence of PE (Z-score = 2.01, p = 0.04). Male sex did not impact the evaluated outcomes. The rate of thromboembolic complications in COVID-19 patients is definitely high. Considering the risk of fatal and disabling complications, adequate screening procedures and antithrombotic strategies should be implemented.

of clinical presentations. 4,5 COVID-19 may range from a mild disease with flulike symptoms to a critical care respiratory condition requiring specialized management at intensive care units (ICUs), [6] [7] [8] with poor long-term outcomes and residual chronic disability. 9, 10 Besides the respiratory manifestations with severe disabling complications, another major concern is represented by evidence of consistent hemostatic changes in patients with severe or critical COVID-19, likely related to a prothrombotic switch. 11, 12 Among COVID-19 patients, it is reasonable to assume that those with a very severe disease could exhibit high risk of venous thromboembolism (VTE), including deep vein thrombosis (DVT) and/or pulmonary embolism (PE). Critically ill patients with COVID-19 are bedridden in ICU because of acute infective disease determining respiratory failure, thus resulting in a high VTE risk, as predicted by a Padua score (bedridden: 3 points; infection: 1 point; respiratory failure: 1 point). 13 Contrasting data are currently available about the prevalence of thromboembolic events in COVID-19 patients.

The aim of the present study was to perform a systematic literature search with meta-analysis of studies reporting the prevalence of VTE, DVT, or PE in patients with COVID-19. Moreover, we implemented some meta-regression models to evaluate the effect of demographic and clinical variables on the evaluated outcomes.

We developed a protocol for this systematic review, defining the search strategy, the outcomes, the inclusion and exclusion criteria, the approach for quality assessment, and the statistical methods.

To detect all available studies on the association of COVID-19 with thromboembolic complications, we conducted a systematic literature search in PubMed, Web of Science, and Scopus according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. 14 The last search was performed on May 19, 2020, with no language restriction, by using the following terms in all possible combinations: COVID-19, SARS-CoV-2, VTE, DVT, PE, vein, venous, thrombosis, thrombosis thrombotic, thromboembolism, pulmonary, embolism, embolic, and occlusion. Moreover, the lists of the bibliographic references of all included articles were manually screened. In case of missing data among studies fulfilling the inclusion criteria, the authors were contacted by e-mail to try to claim the original data. Two authors (A.D.M. and P.A.) analyzed each article and separately performed the extraction of data. In case of disagreement, a third investigator was consulted (M.N.D.D.M.). Discrepancies were resolved by consensus. Overall, selection results showed a high interreader agreement (κ ¼ 1.00) and were reported according to PRISMA flow-chart (►Fig. 1).

According to a predetermined protocol, all studies reporting data about the association of COVID-19 with thromboembolic complications were included. Reviews and articles on animal models were excluded. Overall, we included in the analysis all studies providing data about the prevalence of VTE, DVT, or PE among COVID-19 patients, with VTE being defined as the presence of any thromboembolic complication (DVT, PE, or a combination of both). 15 Studies reporting on the autopsy evidence of thromboembolic events were also included. Superficial vein thrombosis was not considered as VTE episode and was excluded from the analysis. In each study, data regarding sample size, mean age of enrolled patients, percentage of male patients, and body mass index (BMI) were extracted. Formal quality score adjudication was not used since most included studies were small cohort studies.

Data synthesis and analyses were performed by using comprehensive meta-analysis (version 2; Biostat). The prevalence of VTE, DVT, and of PE was expressed as weighted mean prevalence (WMP) with 95% confidence intervals (95% CIs). The pooled effect was tested by means of Z-scores, with p < 0.05 being considered statically significant. We evaluated statistical heterogeneity among studies with chi-squared Cochran's Q-test and with I 2 index, which measures the inconsistency among results of studies and defines the proportion of total variation in study estimates that is due to heterogeneity rather than sampling error. In particular, an I 2 value of 25% corresponds to low, 25 to 50% to moderate, and 50% to high heterogeneity. 16 Funnel plots of the logit event rate versus the standard error were used as a graphical representation of publication bias. To detect a potential small-study effect, funnel plots were visually inspected for asymmetry. Moreover, the Egger test was used to assess publication bias over and above any subjective evaluation, with p < 0.10 being considered statistically significant. 17 To be as conservative as possible, the random-effects method was used to take into account the variability among included studies.

We performed separate subgroup analyses on (1) studies specifically including ICU patients; (2) studies specifically including patients receiving antithrombotic prophylaxis; and (3) studies systematically performing VTE screening.

We hypothesized that mean age of the study population, the percentage of males, and BMI may affect the rate of thromboembolic complications observed in different studies. Thus, we planned to perform meta-regression analyses after implementing regression models with the prevalence of VTE, DVT, and PE as dependent variables (y) and the aforementioned covariates as independent variables (x). Comprehensive meta-analysis software (version 2, Biostat) was used for the multivariate approach.

After excluding duplicates, the search detected 222 articles. Of these, 140 were excluded because they were off the topic (after scanning the title and/or the abstract), and another 40 because they were reviews/commentaries or lacked of data of interest. Another 22 studies were excluded after fulllength paper evaluation (►Fig. 

Major characteristics of included studies are shown in ►Table 1. The number of patients varied from 11 to 328, with mean age of 64.3 years (from 51.7 to 80.5 years). Eleven studies 18, 19, 21, 24, 25, 28, 30, [32] [33] [34] 36 specifically enrolled patients admitted to ICU. A total of 66.8% of patients were males (range: 45.7-90.9%) and the mean BMI was 28.2 kg/m 2 (range: 23.6-31.3 kg/m 2 ). Antithrombotic prophylaxis was used by 78.4% of patients (range: 0-100%).

The WMP of VTE in the 1,988 COVID-19 patients was 31.3% (95% CI: 24.3-39.2%, ►Fig. 2). [18] [19] [20] [21] [22] [23] [24] [25] [26] [27] [28] [29] [30] [31] [32] [33] [34] [35] [36] [37] The heterogeneity among studies was significant (I 2 : 89.2%, p < 0.001) and was not reduced by the exclusion of one study at time. Results were similar when specifically analyzing studies on ICU patients (WMP: 32.7%, 95% CI: 21.9-45.7%; I 2 : 89.9%, p < 0.001). 18, 19, 21, 24, 25, 28, 30, [32] [33] [34] 36 When analyzing studies specifically enrolling patients receiving antithrombotic prophylaxis, the WMP of VTE was 23.9% (95% CI: 15.9-34.4%; I 2 : 87.8%, p < 0.001). [18] [19] [20] 24, 25, 28, 30, 33, 34 

Regression models (►Fig. 4) showed that an increasing age was associated with a higher prevalence of VTE (Z-score: 3.11, p ¼ 0.001), DVT (Z-score: 2.33, p ¼ 0.002), and PE (Z-score: 3.03, p ¼ 0.002) in COVID-19 patients. In addition, an increasing BMI was associated with an increasing prevalence of PE (Z-score: 2.01, p ¼ 0.04), but not of VTE (Z-score: À0.20, p ¼ 0.843) or DVT (Z-score: À0.65, p ¼ 0.517). The rate of VTE, DVT, or PE in COVID-19 patients was not impacted by the prevalence of male sex (Z-score: 0.009, p ¼ 0.999; Z-score: À0.99, p ¼ 0.321; and Z-score: 1.03, p ¼ 0.304, respectively).

Results of the present meta-analysis consistently show that COVID-19 can be associated with thromboembolic complications. In particular, we found that the prevalence of VTE is %30% in COVID-19 patients, with DVT being reported for %20% and with PE being reported for %18% of patients. Regression models showed that an increasing age is associated with a higher prevalence of VTE, DVT, or PE, while an increasing BMI is associated with an increasing prevalence of PE. No impact of male sex on the evaluated outcomes was found.

To the best of our knowledge, this is the first meta-analysis summarizing the growing amount of literature data on the prevalence of thromboembolic events in patients with COVID-19, a respiratory transmitted virus that in 6 months afflicted more than 7 million people and killed more than 400,000 people worldwide. Besides the respiratory syndrome with severe disabling complications, the hypercoagulable state in COVID-19 patients may lead to a high thrombotic risk with dramatic impact on prognosis and mortality. 35 It is plausible that the severity of COVID-19 infection could be an important determinant in the onset of VTE as critically ill patients are bedridden in ICU due to an acute infective disease determining respiratory failure, with a Padua prediction score always >5. 13 In keeping with this, it should be considered that obesity and an older age, which may further increase the VTE risk and the final Padua score, are potential risk factors for severe COVID-19. 5 Accordingly, our regression models showed that an increasing age and BMI were associated with an increased prevalence of thromboembolic events.

However, the relationship between SARS-CoV-2 infection and the hemostatic imbalance of COVID-19 patients seems to be more complex and not fully explained by the presence of traditional VTE risk factors. In keeping with this, Clayton et al reported a 4% incidence of VTE in patients with pneumonia secondary to respiratory tract infections. 38 Our results show that, compared with this evidence, the rate of VTE in COVID-19 patients is %7to 8-fold higher as compared with other respiratory infections. Accordingly, based on previous metaanalytical data, 39 the frequency of thromboembolic complications in ICU patients is %12%. Thus, an indirect comparison with results of our subgroup analysis on ICU suggests that the probability of VTE is higher in severe COVID-19 as compared with the general ICU patient.

In our meta-analysis, the fact that most included studies reported ongoing thromboprophylaxis at the time of VTE by using standard or even therapeutic doses of heparin further supports the hypothesis that SARS-CoV-2 itself may be an additional thrombotic stimulus, over and above traditional VTE risk factors. Accordingly, in the frame of subgroup analyses, we confirmed a high prevalence of VTE (%24%), DVT (%10%), and PE (%15%) also when analyzing studies on patients receiving antithrombotic prophylaxis.

The thrombogenicity of COVID-19 has led some authors to wonder whether such an exceptionally high prevalence of thrombotic complications should be interpreted as a consequence of a localized thrombotic process in the lungs rather than an embolic phenomenon. 40 In detail, it has been hypothesized that the inflammatory reaction in the lungs causes endothelial dysfunction, thus leading to the formation of pulmonary microthrombi. 40 In keeping with this, the histological analysis of pulmonary vessels in patients with COVID-19 showed widespread thrombosis and microangiopathy, with alveolar capillary microthrombi being ninefold more prevalent in COVID-19 patients than in patients with H1N1 influenza. 41 This may have relevant implications as the composition of pulmonary emboli and pulmonary microthrombi is different. In fact, while emboli reflect the composition of the distal part of the venous thrombi from which they originate, pulmonary microthrombi contain a large amount of platelets and fibrin. 42 It is reasonable to assume that such different composition may significantly impact both therapeutic and prophylactic strategies. 40 Taken together, the currently available data may suggest that SARS-CoV-2 infection can rapidly evolve into a severe condition with pulmonary and multiorgan involvement, potentially resulting in a consistent hemostatic imbalance. However, the mechanisms behind such hemostatic changes in severe patients with COVID-19 are still unclear.

Some previous articles focused on the association of COVID-19 severity with changes in primary and secondary hemostatic parameters, reporting that severe patients show longer prothrombin time (PT) and higher D-dimer values, with a lower platelet count. 5, 11, 12, [43] [44] [45] Similar results were found in nonsurvivors to COVID-19 as compared with survivors. Among 183 hospitalized patients with COVID-19, it has been documented that 71.4% of nonsurvivors have overt disseminated intravascular coagulation (DIC) as compared with only 0.6% of survivors. 46, 47 The presence of consumption coagulopathy among critically ill patients with SARS-CoV-2 infection may be in line with the observed changes in primary and secondary hemostatic parameters. The pathogenesis of DIC is complex and multifactorial, and the "cytokine storm" described in ICU patients with SARS-CoV-2 infection may somehow contribute to its development. 48 However, many patients with severe COVID-19 may not fulfill the International Society on Thrombosis and Haemostasis (ISTH) criteria for DIC, 49 as coagulopathy seems to have its peculiar features among these patients. For instance, thrombocytopenia is less pronounced in severe COVID-19 patients as compared with sepsis-related DIC, while D-dimer levels are particularly high. 50 Moreover, although hyperfibrinogenemia is a common finding in COVID-19 patients, low fibrinogen levels have also been documented in nonsurvivors. 48 Overall, based on available data, it can be hypothesized that the coagulopathy observed in severe COVID-19 may be a combination of a low-grade DIC with its own features and a localized pulmonary thrombotic microangiopathy. 50 Considering the high thrombotic risk of COVID-19 patients, further specifically designed studies are needed to better address the issue of the hemostatic changes associated with COVID-19 severity. This could allow for a better understanding of the pathogenesis of this potentially fatal complication. Moreover, given the unknown long-term outcomes of COVID-19 and the increasing amount of literature data suggesting the presence of residual chronic disability and rehabilitation needs among survivors, 51-57 it is compelling to identify new biomarkers related to the disease to timely predict its clinical progression and to adequately manage critical patients even after the acute phase (e.g., general ward, rehabilitation centers, and home). 58 In the meantime, in all hospitalized COVID-19 patients, following the WHO and ISTH indications, 59-61 prophylactic low-molecular-weight heparin (LMWH) or subcutaneous unfractionated heparin (UFH) should be considered.

Nevertheless, despite the lack of evidence supporting the prescription of high-dose prophylactic regimens, which may conversely increase the bleeding risk (particularly in patients with liver and renal impairment), the high rate of VTE reported during adequate thromboprophylaxis deserves to be considered and further investigated. 62 The fact that some authors suggested the use of therapeutic doses in all hospitalized patients with COVID-19 even in the absence of randomized trial evidence 25 reflects the belief in the medical community that standard thromboprophylaxis may be inadequate in the presence of such a strong prothrombotic switch.

Some potential limitations of our meta-analysis should be discussed. First, we have to consider that most included studies had a retrospective design and few studies (n ¼ 5) enrolled consecutive COVID-19 patients undergoing imaging tests (Doppler ultrasound and/or computed tomography angiography) for VTE screening. Considering that symptoms of respiratory failure from COVID-19 are difficult to separate from those of PE, the fact that most studies based the execution of imaging tests on signs and symptoms of VTE may contribute to the significantly high heterogeneity affecting our results. However, although we could not definitively establish sources of such heterogeneity, our findings were substantially confirmed by appropriate sensitivity and subgroup analyses. In particular, we extended our results analyzing studies systematically performing thrombosis screening in all enrolled patients, showing a fourfold increase in DVT diagnosis rate. This might support the hypothesis that a strict screening of thromboembolic complications in all COVID-19 patients is needed. Moreover, the impact of clinical and demographic variables on results was evaluated by means of meta-regression models. Finally, we excluded the presence of publication bias by using different methods.

Results of the present meta-analysis show that the rate of VTE in COVID-19 patients is not negligible and claims the attention on the need of adequate screening procedures and antithrombotic strategies to manage VTE. Failure of prevention and interventional strategies for the prothrombotic state of severe COVID-19 patients may dramatically impact their prognosis. While waiting for data from a larger number of studies, the high thromboembolic risk of COVID-19 patients should be considered to timely predict and manage the long-term outcomes of these patients with potentially disabling and fatal complications.

No financial support was received for this study.

China Novel Coronavirus Investigating and Research Team. A novel coronavirus from patients with pneumonia in China

World Health Organization

World Health Organization

A familial cluster of pneumonia associated with the 2019 novel coronavirus indicating person-toperson transmission: a study of a family cluster

China Medical Treatment Expert Group for COVID-19. Clinical characteristics of coronavirus disease 2019 in China

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

Clinical features and treatment of COVID-19 patients in northeast Chongqing

Real-time tentative assessment of the epidemiological characteristics of novel coronavirus infections in Wuhan, China, as at 22

Covid-19 crisis triageoptimizing health outcomes and disability rights

The COVID-19 response must be disability inclusive

Clinical characteristics of refractory COVID-19 pneumonia in Wuhan, China

Risk factors associated with acute respiratory distress syndrome and death in patients with coronavirus disease 2019 pneumonia in Wuhan, China

A risk assessment model for the identification of hospitalized medical patients at risk for venous thromboembolism: the Padua Prediction Score

Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statement

Cardiology patient pages. Pulmonary embolism and deep vein thrombosis

Measuring inconsistency in meta-analyses

Systematic reviews in health care: investigating and dealing with publication and other biases in meta-analysis

Humanitas COVID-19 Task Force. Venous and arterial thromboembolic complications in COVID-19 patients admitted to an academic hospital in

Incidence of venous thromboembolism in hospitalized patients with COVID-19

Pulmonary embolism in patients with Covid-19 pneumonia

Prevalence of venous thromboembolism in patients with severe novel coronavirus pneumonia

Incidence of asymptomatic deep vein thrombosis in patients with COVID-19 pneumonia and elevated D-dimer levels

Acute pulmonary embolism associated with COVID-19 pneumonia detected by pulmonary CT angiography

High risk of thrombosis in patients with severe SARS-CoV-2 infection: a multicenter prospective cohort study

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

Pulmonary arterial thrombosis in COVID-19 with fatal outcome: results from a prospective, singlecenter, clinicopathologic case series

Acute pulmonary embolism in COVID-19 patients on CT angiography and relationship to D-dimer levels

High incidence of venous thromboembolic events in anticoagulated severe COVID-19 patients

Upsurge of deep venous thrombosis in patients affected by COVID-19: preliminary data and possible explanations

Pulmonary embolism in COVID-19 patients: awareness of an increased prevalence

Acute pulmonary embolism and COVID-19

Extremely high incidence of lower extremity deep venous thrombosis in 48 patients with severe COVID-19 in Wuhan

COVID-19-related severe hypercoagulability in patients admitted to intensive care unit for acute respiratory failure

Thrombotic complications of patients admitted to intensive care with COVID-19 at a teaching hospital in the United Kingdom

Autopsy findings and venous thromboembolism in patients with COVID-19

Fibrinolysis shutdown correlation with thromboembolic events in severe COVID-19 infection

Deep vein thrombosis in hospitalized patients with coronavirus disease

China: prevalence, risk factors, and outcome

Recent respiratory infection and risk of venous thromboembolism: case-control study through a general practice database

The impact of deep vein thrombosis in critically ill patients: a meta-analysis of major clinical outcomes

SARS-2 coronavirus-associated hemostatic lung abnormality in COVID-19: Is it pulmonary thrombosis or pulmonary embolism?

Pulmonary vascular endothelialitis, thrombosis, and angiogenesis in Covid-19

The distinctive structure and composition of arterial and venous thrombi and pulmonary emboli

Clinical and immunological features of severe and moderate coronavirus disease 2019

Clinical features of patients infected with 2019 novel coronavirus in Wuhan, China

Changes in blood coagulation in patients with severe coronavirus disease 2019 (COVID-19): a meta-analysis

Clinical course and outcomes of critically ill patients with SARS-CoV-2 pneumonia in Wuhan, China: a single-centered, retrospective, observational study

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

Abnormal coagulation parameters are associated with poor prognosis in patients with novel coronavirus pneumonia

Towards definition, clinical and laboratory criteria, and a scoring system for disseminated intravascular coagulation

Coagulation abnormalities and thrombosis in patients with COVID-19

Critical care rehabilitation -is it the answer for reducing morbidity in ARDS survivors? Regarding "Acute respiratory distress syndrome: a clinical review

Editorial: Covid-19 and physical and rehabilitation medicine

Rehabilitation of COVID-19 patients

Sequelae of acute respiratory distress syndrome: interest of rehabilitation

Canadian Critical Care Trials Group. Functional disability 5 years after acute respiratory distress syndrome

COVID-19 and functional disability: current insights and rehabilitation strategies

Rehabilitation after critical illness in people with COVID-19 Infection

Recommendations for minimal laboratory testing panels in patients with COVID-19: potential for prognostic monitoring

Clinical management of severe acute respiratory infection when novel coronavirus (2019-nCoV) infection is suspected. Interim guidance 28

ISTH interim guidance on recognition and management of coagulopathy in COVID-19

Supported by the ESC Working Group on Pulmonary Circulation and Right Ventricular Function. COVID-19 and thrombotic or thromboembolic disease: implications for prevention, antithrombotic therapy, and follow-up: JACC stateof-the-art review

COVID-19 and its implications for thrombosis and anticoagulation

Seminars in Thrombosis & Hemostasis

None.