key: cord-287781-bvintr10
authors: Sakka, Mehdi; Connors, Jean M; Hékimian, Guillaume; Martin-Toutain, Isabelle; Crichi, Benjamin; Colmegna, Ines; Bonnefont-Rousselot, Dominique; Farge, Dominique; Frere, Corinne
title: Association between D-Dimer levels and mortality in patients with coronavirus disease 2019 (COVID-19): a systematic review and pooled analysis
date: 2020-05-27
journal: J Med Vasc
DOI: 10.1016/j.jdmv.2020.05.003
sha: 
doc_id: 287781
cord_uid: bvintr10

Abstract Background: Several observational studies have reported elevated baseline D-dimer levels in patients hospitalized for moderate to severe coronavirus disease 2019 (COVID-19). These elevated baseline D-dimer levels have been associated with disease severity and mortality in retrospective cohorts. Objectives: To review current available data on the association between D-Dimer levels and mortality in patients admitted to hospital for COVID-19. Methods: We performed a systematic review of published studies using MEDLINE and EMBASE through 13 April 2020. Two authors independently screened all records and extracted the outcomes. A random effects model was used to estimate the standardized mean difference (SMD) with 95% confidence intervals (CI). Results: Six original studies enrolling 1355 hospitalized patients with moderate to critical COVID-19 (391 in the non-survivor group and 964 in the survivor group) were considered for the final pooled analysis. When pooling together the results of these studies, D-Dimer levels were found to be higher in non-survivors than in-survivors. The SMD in D-Dimer levels between non-survivors and survivors was 3.59 µg/L (95% CI 2.79–4.40 µg/L), and the Z-score for overall effect was 8.74 (p < 0.00001), with a high heterogeneity across studies (I²=95%). Conclusions: Despite high heterogeneity across included studies, the present pooled analysis indicates that D-Dimer levels are significantly associated with the risk of mortality in COVID-19 patients. Early integration of D-Dimer testing, which is a rapid, inexpensive, and easily accessible biological test, can be useful to better risk stratification and management of COVID-19 patients.

15-20% of patients progress to severe pneumonia resulting in hypoxia and respiratory failure which necessitate supportive care for critical illness and supplemental nasal oxygen, and 5% of them progress to acute respiratory distress syndrome (ARDS) or multiple organ failure (MOF) requiring admittance to an intensive care unit (ICU) with mechanical ventilation or extracorporeal membrane oxygenation (ECMO). [9] Risk factors for developing ARDS and MOF throughout the course of COVID-19 are not fully understood yet. Due to their limited capacity, health care systems are facing an unprecedented critical crisis worldwide. Early optimized care may improve prognosis in patients at high risk for developing the most severe forms, however early identification of poor prognosis and risk stratification remains challenging. Therefore, reliable biomarkers or risk assessment models allowing to earlier predict poor prognosis are warranted to help physicians in decision making regarding the most appropriate therapeutic approach to further provide the best possible care for each individual patient. D-dimers are degradation products of cross-linked fibrin produced when plasmin cleaves fibrin to break down clots. D-dimer testing is rapid, reproducible, inexpensive, and easily accessible. Early retrospective cohort studies of COVID-19 patients admitted to hospital reported increased baseline Ddimer levels in 36% to 43.6% of cases, [1] [2] [3] [4] particularly in patients admitted to ICU. In a recent pooled analysis of 4 studies including 553 patients, baseline D-Dimer levels were associated with COVID-19 severity.

[10] While a retrospective cohort of 183 severe COVID-19 patients suggested that in the most severe cases high D-dimer levels could be related to disseminated intravascular coagulation (DIC) [11] , two recent prospective studies aiming to characterize the coagulation profile of COVID-19 ARDS patients with standard and viscoelastic coagulation tests reported a pro-coagulant profile without evidence of DIC. [12, 13] In the present study, we reviewed all current available data on the association between D-Dimer levels and mortality in COVID-19 patients in order to assess the potential of D-dimer testing for predicting outcomes in patients admitted to hospital.

We performed a comprehensive literature search of published studies from all languages using MEDLINE and EMBASE through 13 April 2020. The following keywords (MeSH terms) were used:

'D-Dimer' AND ('Coronavirus disease 2019' OR 'COVID-19' OR 'SARS-CoV-2' OR 'novel corona' OR '2019-ncov'). We also screened pre-print articles from arXiv.

Studies were eligible to be included in the meta-analysis if they met the following inclusion criteria: (i) original cohort study, (ii) written in English language, (iii) reporting separately D-Dimers levels in hospitalized COVID-19 patients with fatal outcome and in those without fatal outcome. No restriction was made on the patient disease severity (i.e COVID-19 patients with moderate, severe or critical disease). Comments, letters to the editor, editorial, reviews, case reports and basic science studies were excluded.

Two authors (M.S. and C.F.) independently screened all records identified through database searching for study eligibility based on title and abstracts. The agreement between the reviewers for study selection was assessed using the kappa statistic.

[14] Any discrepancies were resolved by consensus after discussion between the two authors or upon consensus from a third author (D.F.). The Newcastle-Ottawa scale was used to assess the studies quality and risk of bias. [15] Data were independently extracted by two authors [including first author's last name, publication year, study design, country, study population, study period, sample size, sex, age, mortality rate, and means and standard deviations (SD) of D-Dimer in COVID-19 hospitalized patients with fatal outcome and in those without fatal outcome] using a standardized data extraction form.

Continuous data were represented by mean, SD and standardized mean difference (SMD) with 95% confidence intervals (CI). For studies reporting medians and interquartile ranges (IQR), means and SD were estimated according to the algorithm proposed by Hozo et al. [16] Heterogeneity among studies was assessed by the I² statistic, I² >50% representing high degree of heterogeneity. A random-effects model was used in order to take into account within-study and between-study variance. Visualization of funnel plots was used to assess for publication bias. A p value <0.05 was considered as statistically significant. All statistical analyses were performed using the Cochrane's Review Manager software (RevMan, version 5.3, Copenhagen, Denmark).

A total of 26 records were identified through database searching, of which 20 were excluded because they were editorial (n=1), review article (n=1), basic science study (n=1), they were written in foreign (Chinese) language (n=5), they did not assess mortality as an outcome or they did not reported D-dimer levels separately in COVID-19 patients with fatal outcome and in those without fatal outcome (n=12).

Finally, 6 original studies enrolling a total of 1355 patients hospitalized for moderate to critical COVID-19 (391 in the non-survivor group, and 964 in the survivor group) were considered for the final pooled analysis. [5] [6] [7] 11, 17, 18] The agreement between authors for study selection was 100% (kappa statistic

The main characteristics of included studies are summarized in Table 1 . All studies were retrospective cohorts conducted in China between December 2019 and February 2020. [5] [6] [7] 11, 17] The duration of follow-up was unclear except in one study [6] . All studies were at high risk of bias according to the Newcastle-Ottawa scale.

The median age of patients ranged from 51 to 62 years. Most of patients (58.6%) were men. The mortality rate varied from 11.5 to 29.8% (Table 1 ). Of note, most studies excluded patients who were still hospitalized at the end of the study period (i.e. neither recovered nor died). When pooling together the results from these 6 studies, D-Dimer levels were found to be higher in nonsurvivors than in-survivors. The SMD in D-Dimer levels between non-survivors and survivors was 3.59 µg/L (95% CI 2.79-4.40 µg/L) (Figure 1) , and the Z-score for overall effect was 8.74 (p < 0.00001), with a high heterogeneity across studies (I²=95%).

Retrospective cohorts early reported increased baseline D-dimer levels in one third of overall COVID-19 patients admitted to hospital. [1] [2] [3] [4] In two studies, significantly higher levels of D-Dimers were observed in small samples of patients with severe disease (i.e. admitted to ICU) compared to those with mild or moderate disease (i.e. not requiring an admission to ICU). [1, 4] al. [17] reported that D-Dimer levels were significantly associated with 28-days mortality (OR 1.058, 95 % CI 1.028-1.090; p<0.001) in a retrospective cohort of 449 patients with severe COVID-19

To the best of our knowledge, six risk assessment models (RAM) for predicting mortality among patients with suspected or confirmed COVID- 19 have been yet developed. [19] [20] [21] [22] [23] [24] Only three of these RAMs included biological data (C-reactive protein, lactic dehydrogenase, and lymphocyte count). However, none of them included D-dimer levels as a predictor.

More recently, a retrospective study investigated the prognosis and the optimal cut-off values for baseline D-dimer levels to predict mortality in 343 COVID-19 patients. Thirteen deaths occurred during hospitalization. A D-dimer value ≥2.0 µg/mL at admission was found to be significantly associated with mortality (Hazard ratio 51.5, 95% CI 12.9-206.7). This cutoff value of 2.0 µg/mL predicts mortality with a C-index of 0.89, a sensitivity of 92.3% and a specificity of 83.3%. [25] The mechanisms underlying the increase in D-Dimer in COVID-19 patients is not fully understood. In their retrospective cohort, Tang et al. found that 71.6% of non-survivors fulfilled the criteria for sepsisinduced coagulopathy (SIC) compared to 0.6% in survivors. [11] However, 2 recent prospective studies reported a pro-coagulant profile in COVID-19 ARDS patients using standard and viscoelastic coagulation tests without further evidence of DIC, [12, 13] indicating that the increase in D-Dimers levels, which is constantly observed in severe COVID-19 cases, might not be related to DIC. Moreover, in a prospective cohort of 184 patients admitted to ICU for COVID-19 pneumonia, none of the patients developed DIC. [26] In the most severe cases of COVID-19, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection results in uncontrolled inflammatory innate and impaired adaptive immune responses, as reflected by increased serum levels of pro-inflammatory cytokines such as IL-6, IL-1β, IL-2, IL-8, IL-17, G-CSF, GM-CSF, IP10, MCP1, MIP1α and TNF. [27] Importantly, there is a crosstalk between inflammation and blood coagulation, the production of pro-inflammatory cytokines such as TNF-α, IL-1β and IL-6, leading to the upregulation of tissue factor (TF) expression on the endothelial cells and monocytes surfaces, further resulting in a procoagulant activity. [28] In addition, autopsy reports of COVID-19 patients found extensive neutrophil infiltration in pulmonary capillaries, with acute capillaritis and fibrin deposition, indicating that neutrophil extracellular traps (NETs) may contribute to organ damage and promote thrombosis. [29] Hypoxia has also been demonstrated to trigger a procoagulant activity through the upregulation of hypoxia-inducible transcription factors that modulate the expression of several coagulation and fibrinolysis factors such as tissue factor pathway inhibitor, TF, and PAI-1. [30] Accordingly, there is mounting evidence that COVID-19 is associated with a prothrombotic state [12, 13] Potential limitations of the present study should be acknowledged. First, our analysis included only observational retrospective studies conducted in a single country, which may have resulted in biases inherent to such studies. Selected studies were heterogeneous in terms of study design, study population, outcome definitions, and length of follow-up. Second, we converted non-normally distributed statistics (i.e. median and range) to normally distributed statistics (i.e. mean and SD), which may also have introduced some bias in the results. Finally, the amount of included studies and their sample size were limited.

In conclusion, evidence has emerged that D-Dimer testing in combination with clinical factors or other biomarkers might be useful to early stratify patients with COVID-19 admitted to hospital and to further individualize treatment. However, it seems premature to use D-Dimer alone to guide clinical decisionmaking. Further large prospective studies to validate the prognosis performances of D-Dimer and to determine the best cutoff value to be used are warranted. 

The authors declare they have no conflict of interest. 

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

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

Hospitalized Patients With 2019 Novel Coronavirus-Infected Pneumonia in Wuhan, China

Clinical characteristics of 113 deceased patients with coronavirus disease 2019: retrospective study

Risk Factors Associated With Acute Respiratory Distress Syndrome and Death in Patients With Coronavirus Disease

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

Risk factors associated with disease progression in a cohort of patients infected with the 2019 novel coronavirus

Characteristics of and Important Lessons From the Coronavirus Disease 2019 (COVID-19) Outbreak in China: Summary of a Report of 72 314 Cases From the Chinese Center for Disease Control and Prevention

D-dimer is Associated with Severity of Coronavirus Disease 2019: A Pooled Analysis

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

The procoagulant pattern of patients with COVID-19 acute respiratory distress syndrome

Hypercoagulability of COVID-19 patients in Intensive Care Unit. A Report of Thromboelastography Findings and other Parameters of Hemostasis

Evidence-Based Medicine Teaching Tips Working Group. Tips for learners of evidence-based medicine: 3. Measures of observer variability (kappa statistic)

The Newcastle-Ottawa Scale (NOS) for assessing the quality of nonrandomized studies in meta-analyses. Ottawa Health Research Institute

Estimating the mean and variance from the median, range, and the size of a sample

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

Clinicolaboratory study of 25 fatal cases of COVID-19 in Wuhan

Development and external validation of a prognostic multivariable model on admission for hospitalized patients with COVID-19

Estimation of risk factors for COVID-19 mortalitypreliminary results

ACP risk grade: a simple mortality index for patients with confirmed or suspected severe acute respiratory syndrome coronavirus 2 disease (COVID-19) during the early stage of outbreak in Wuhan, China

A machine learning-based model for survival prediction in patients with severe COVID-19 infection

Association of radiologic findings with mortality of patients infected with 2019 novel coronavirus in Wuhan

Host susceptibility to severe COVID-19 and establishment of a host risk score: findings of 487 cases outside Wuhan

D-dimer levels on admission to predict inhospital mortality in patients with Covid-19

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

COVID-19: immunopathology and its implications for therapy

The Role of Gap Junction-Mediated Endothelial Cell-Cell Interaction in the Crosstalk between Inflammation and Blood Coagulation

Targeting potential drivers of COVID-19: Neutrophil extracellular traps

The stimulation of thrombosis by hypoxia

Acute pulmonary embolism and COVID-19 pneumonia: a random association?

COVID-19 Complicated by Acute Pulmonary Embolism

Coagulopathy and Antiphospholipid Antibodies in Patients with Covid-19

Prevalence of venous thromboembolism in patients with severe novel coronavirus pneumonia

Attention should be paid to venous thromboembolism prophylaxis in the management of COVID-19