By Arnaldo Lopez Ruiz, MD, and Bibidh Subedi, PharmD, BCCCP

Dr. Lopez Ruiz is Attending Physician, Division of Critical Care, AdventHealth Medical Group, AdventHealth, Orlando, FL;
Dr. Subedi is Multi-System ICU Pharmacist, AdventHealth, Orlando, FL.

Drs. Lopez Ruiz and Subedi report no financial relationships relevant to this field of study.

SYNOPSIS: In a retrospective study involving 449 patients with severe COVID-19 requiring intensive care unit admission, those patients with a positive sepsis coagulation score or D-dimer greater than 3.0 mcg/mL who received prophylactic doses of low molecular weight heparin exhibited lower 28-day mortality.

SOURCE: Tang N, Bai H, Chen X, et al. Anticoagulant treatment is associated with decreased mortality in severe coronavirus disease 2019 patients with coagulopathy. J Thromb Haemost 2020;18:1094-1099.

The novel coronavirus disease 2019 (COVID-19) pandemic has shown that a large proportion of patients requiring admission to the intensive care unit (ICU) exhibit abnormal coagulation markers suggesting an activation of the coagulation cascade. In fact, patients with elevated D-dimer levels had higher mortality than patients with lower or normal D-dimer levels.1 Tang et al2 reported a retrospective observational study from Jan. 1 to Feb. 13, 2020, involving 449 ICU patients with severe COVID-19 at Tongji Hospital of Huazhong University of Science and Technology in Wuhan, China. Severe COVID-19 was defined as respiratory rate ≥ 30 breaths/minute, arterial oxygen saturation ≤ 93% at rest, and PaO2/FiO2 ≤ 300 mmHg. Anticoagulation therapy was defined as receiving unfractionated heparin (UFH) or low molecular weight heparin (LMWH) for seven days or longer. The authors compared mortality between heparin users vs. nonusers. The authors performed a multivariate analysis to correlate age, gender, use of heparin, prothrombin time (PT), platelet count, and D-dimer levels with 28-day mortality, and they risk-stratified patients based on sepsis-induced coagulopathy (SIC) score and D-dimer levels. Patients were excluded if they had bleeding diathesis, a hospital stay less than seven days, and a lack of information about coagulation parameters.

Out of 449 patients, 99 (22%) patients received anticoagulation for at least seven days, with 94 patients receiving LMWH (40 mg to 60 mg enoxaparin/day) and five receiving UFH (10,000-15,000 U/day). These patients were compared to the remaining 350 patients without anticoagulation or who were anticoagulated for less than seven days (control group). The mean age was 65 ± 12 years, and comorbid conditions were present in 60.6% of the patients, with hypertension (39.4%) and diabetes (20.7%) the most common. There was no difference in 28-day mortality between heparin users vs. nonusers (30.3% vs. 29.7%; P = 0.910). In a multivariate logistic regression model, D-dimer, PT, and age were positively correlated, and platelet count was negatively correlated with 28-day mortality. Furthermore, 97 (21.6%) patients out of the total had a positive SIC score (≥ 4 points). In these patients, anticoagulation use showed a statistically significant decrease in mortality (40% vs. 64.2%; odds ratio [OR], 0.372; 95% confidence interval [CI], 0.15-0.90; P = 0.029). There also was a 20% mortality reduction in heparin users who had D-dimer greater than 3.0 mcg/mL (32.8% vs. 52.4%; OR, 0.442; 95% CI, 0.226-0.865; P = 0.017). The investigators concluded that anticoagulation using LMWH (40-60 mg/day) or UFH (10,000-15,000 UI/day) was associated with lower 28-day mortality in patients with severe COVID-19 who had an SIC score ≥ 4 or D-dimer ≥ 3.0 mcg/mL.


Severe COVID-19 disease has been associated with activation of the coagulation cascade, leading to a procoagulant and proinflammatory state due to diffuse endothelial damage affecting multiple organs.3 About 80% of ICU non-survivors with COVID-19 had elevated D-dimer levels (median 5.2 mcg/mL; interquartile range, 1.5-21.1). A D-dimer greater than 1 mcg/mL has been described as a prognostic factor for increased mortality in severe COVID-19.1 Therefore, the presence of elevated D-dimer levels in patients with severe COVID-19 admitted to the ICU requires in-depth evaluation to determine whether these patients would benefit with a full dose of anticoagulation.

This study by Tang et al is an interesting and hypothesis-generating study with several limitations that need consideration before widespread implementation of anticoagulation therapy in severe COVID-19. This was a retrospective study with unmatched cases (treated with anticoagulation for at least seven days) and control patients (no anticoagulation treatment or treated for less than seven days) admitted to the ICU over a short period of six weeks. The doses of LMWH or UFH reported in the article are approved and recognized as prophylactic dosing for patients with normal renal function. In fact, in the United States or Europe, the use of a prophylactic anticoagulation regimen with LMWH or UFH is considered standard of care for any ICU patient unless specifically contraindicated. However, here only 99 (22%) patients were receiving prolonged prophylactic anticoagulation despite being a high-risk population (e.g., bed-ridden, critically ill) for deep venous thrombosis (DVT) or pulmonary embolism (PE). Therefore, the title and conclusion of the article should be interpreted with caution because the purported benefit was due to only a prophylactic dose of LMWH or UFH.

The authors did not describe details on the care measures implemented during the patients’ ICU stays; specifically, data regarding the use of glucocorticoids or immunomodulators (tocilizumab, hydroxychloroquine, anakinra) are lacking. These medications attenuate the development of SIC, and it raises the question of whether they may have favored the effect of anticoagulants. Furthermore, the authors did not report information on the proportion of patients who developed deterioration in clinical status, septic shock, acute kidney injury, severe (PaO2/FIO< 100) acute respiratory distress syndrome (ARDS), or coinfection. Any of these conditions can predispose to coagulopathy and impact ICU mortality.4 The mortality benefit was not detected in the initial analysis between heparin users vs. nonusers; instead, it was found in a subset of patients with an elevated SIC score or D-dimer level. Whether this was due to statistical chance or actual causation is hard to determine. Implementing an anticoagulation protocol either using full or prophylactic dosing based on SIC score may not be appropriate, since only 21.6% of the patients had an elevated SIC score.

Furthermore, it still is not clear whether a high D-dimer level is revealing an active thrombotic state or whether a high D-dimer is a consequence of overt disseminated intravascular coagulation (DIC) that may increase the chances of bleeding down the road. Therefore, it is important to assess objectively whether these patients with elevated D-dimer levels have a hypercoagulable or hypocoagulable state. Finally, the study showed that anticoagulation therapy in patients with D-dimer below 1 mcg/mL had a strong tendency (non-statistically, but probably clinically relevant) for higher mortality than untreated patients. This finding should be examined further in a larger study.

The fact that viral infections promote a severe inflammatory response resulting in activation of the coagulation cascade is well-known, and it can manifest with bleeding, thrombosis, or both.5,6 However, it is unclear why some non-hemorrhagic viruses cause bleeding (e.g., influenza H5N1), while others are associated with thrombosis (e.g., cytomegalovirus, severe acute respiratory syndrome [SARS]-associated coronavirus [SARS-CoV], influenza H1N1, or parvovirus B19), and yet others show both complications (e.g., varicella-zoster virus [VZV] or herpes simplex virus [HSV]).5,7 The COVID-19 pandemic, like SARS, has revealed many cases of coagulopathy. It has been reported that 71.4% of patients who die due to COVID-19 have coagulopathy and meet criteria for DIC, compared to only 0.6% of patients who survive exhibiting signs of DIC.8 Case series indicate that these patients do not have a bleeding diathesis, but rather a prothrombotic DIC with frequent venous and arterial microthrombotic events causing ischemic limbs, strokes, ventricular thrombosis, pulmonary embolism, central line, dialysis catheter, or hemofilter thromboses that responded to tissue plasminogen activator (tPA).9,10 A large Danish observational study of 184 ICU patients with COVID-19 pneumonia found a cumulative incidence of the composite outcome of symptomatic PE, DVT, ischemic stroke, myocardial infarction, or systemic arterial embolism in 31% (95% CI, 20-41%) of the patients, with PE being the most frequent thrombotic complication (n = 25, 81%).11 Using these data, we have to consider that a significant proportion of patients with severe COVID-19 are at high risk for thrombotic complications and will meet criteria for DIC.

It is important to note that elevated D-dimer levels alone or in combination with abnormal prothrombin time or fibrinogen levels have very low specificity for DVT or PE.12,13 Therefore, additional tests may be helpful to identify hypercoagulability, like thromboelastography (TEG)14,15 or rotational thromboelastometry (ROTEM)14 and measuring procoagulant markers (factor V, factor VIII, Von Willebrand factor),4,15 anticoagulant markers (antithrombin III, activated protein C),4,15 or D-dimer/fibrinogen ratio.12,13 These tests may aid in deciding whether a patient with severe COVID-19 would benefit from full anticoagulation or not.4 At this time, a meta-analysis of 24 trials involving 14,767 patients indicates that full anticoagulation therapy in sepsis is not associated with a reduction in mortality, including in patients with sepsis-induced coagulopathy.16,17 However, despite the increase in bleeding events, a significant reduction in mortality (risk ratio 0.72, 95% CI, 0.62-0.85) was observed in the population with sepsis-induced DIC.16 These data reinforce the goal of correctly identifying patients with severe COVID-19 who may benefit from full anticoagulation.

Solving the anticoagulation dilemma in COVID-19 remains challenging. Increasing anticoagulant doses alone may not change outcomes in patients with severe COVID-19. Novel anti-inflammatory agents, such as IL-6 or IL-1 antagonists, glucocorticoids, and other approaches to attenuate the hyperinflammatory response driving the prothrombotic state should be considered. More studies are needed to define biomarkers or laboratory tests that may help in identifying hypercoagulable patients, along with prospective randomized controlled trials to assess the impact of full-dose anticoagulation in these patients. Until then, institutions should create an anticoagulation strategy based on clinical signs, symptoms, and selected laboratory tests to better identify those patients with a hypercoagulable or hypocoagulable state to guide the degree of anticoagulation.


  1. Zhou F, Yu T, Du R, et al. Clinical course and risk factors for mortality of adult inpatients with COVID-19 in Wuhan, China: A retrospective cohort study. Lancet 2020;395:1054-1062.
  2. Tang N, Bai H, Chen X, et al. Anticoagulant treatment is associated with decreased mortality in severe coronavirus disease 2019 patients with coagulopathy. J Thromb Haemost 2020;18:1094-1099.
  3. Varga Z, Flammer AJ, Steiger P, et al. Endothelial cell infection and endothelitis in COVID-19. Lancet 2020;395:1417-1418.
  4. Simmons J, Pittet JF. The coagulopathy of acute sepsis. Curr Opin Anaesthesiol 2015;28:227-236.
  5. Goeijembier M, van Wissen M, van de Weg C, et al. Review: Viral infections and mechanisms of thrombosis and bleeding. J Med Virol 2012;84:1680-1696.
  6. Levi M, van der Poll T, Schultz M. New insights into pathways that determine the link between infection and thrombosis. Neth J Med 2012;70:114-120.
  7. Squizzato A, Gerdes VE, Buller HR. Effects of human cytomegalovirus infection on the coagulation system. Thromb Haemost 2005;93:403-410.
  8. Tang N, Li D, Wang X, Sun Z. Abnormal coagulation parameters are associated with poor prognosis in patients with novel coronavirus pneumonia. J Thromb Haemost 2020;18:844-847.
  9. Han H, Yang L, Liu R, et al. Prominent changes in blood coagulation of patients with SARS-Cov-2 infection. Clin Chem Lab Med 2020; Mar 16. doi: 10.1515/cclm-2020-0188. [Epub ahead of print].
  10. Wang J, Hajizadeh N, Moore EE, et al. Tissue plasminogen activator (tPA) treatment for COVID-19 associated acute respiratory distress syndrome (ARDS): A case series. J Thromb Haemost 2020; Apr 8. doi: 10.1111/jth.14828. [Epub ahead of print].
  11. Klok FA, Kruip MJHA, van der Meer NJM, et al. Incidence of thrombotic complications in critically ill ICU patients with COVID-19. Thromb Res 2020; Apr 10. doi: 10.1016/j.thromres.2020.04.013. [Epub ahead of print].
  12. Kucher N, Kohler HP, Dornhöfer T, et al. Accuracy of D-dimer/fibrinogen ratio to predict pulmonary embolism: A prospective diagnostic study. J Thromb Haemost 2003;1:708-713.
  13. Wuillemin WA, Korte W, Waser G, Lämmle B. Usefulness of the D-dimer/fibrinogen ratio to predict deep venous thrombosis. J Thromb Haemost 2005;3:385-387.
  14. Drumheller BC, Stein DM, Moore LJ, et al. Thromboelastography and rotational thromboelastometry for the surgical intensivist: A narrative review. J Trauma Acute Care Surg 2019;86:710-721.
  15. Koami H, Sakamoto Y, Ohta M, et al. Can rotational thromboelastometry predict septic disseminated intravascular coagulation? Blood Coagul Fibrinolysis 2015;26:778-783.
  16. Favaloro EJ. Laboratory testing in disseminated intravascular coagulation. Semin Thromb Hemost 2010;36:458-467.
  17. Umemura Y, Yamakawa K, Ogura H, et al. Efficacy and safety of anticoagulant therapy in three specific populations with sepsis: A meta-analysis of randomized controlled trials. J Thromb Haemost 2016;14:518-530.
  18. Iba T, Levy JH, Raj A, Warkentin TE. Advance in the management of sepsis-induced coagulopathy and disseminated intravascular coagulation. J Clin Med 2019;8. doi: 10.3390/jcm8050728.