Thromboelastometry and D-Dimer Elevation in Coronavirus-2019.

To the Editor:

SEVERE elevation of D-dimer is a hallmark of septic shock and a predictor of mortality in coronavirus-2019 (COVID-19) disease. D-dimer reflects the extent of plasmin-mediated degradation of cross-linked fibrin, thereby causing intravascular coagulation. Use of thromboelastometry has gained popularity to assess systemic fibrinolysis in liver transplantation and major trauma, but its utility has not been fully elaborated in the critical care setting. We therefore analyzed the laboratory and thromboelastometry data from 11 critically ill patients receiving mechanical lung ventilation and intensive care support for COVID-19 at the R Adams Cowley Shock Trauma Center over a 2-day period. The Institutional Review Board approved the study. Patients were characterized as follows (data in median [25%-75% quartiles] or percentage); median age 53 years (45.5-65.5 y), body mass index 28.1 (27.1-34.6), 64% male, 54.5% hypertensive, and 45.5% diabetic. Patients were dichotomized into 2 groups on the basis of D-dimer levels 5 times the upper limit of normal (649 ng/mL fibrinogen equivalent unit). Three of 6 patients in the high D-dimer group were on extracorporeal membrane oxygenation support. Despite highly significant C-reactive protein and D-dimer elevations in the latter group, systemic fibrinolysis was not detected either on EXTEM or FIBTEM (maximal lysis 0%). D-dimer has a half-life of about 8 hours and reflects in vivo thrombus formation. On the other hand, thromboelastometry only measures the reserve hemostasis capacity in the collected blood using a high-dose coagulation trigger (eg, tissue factor). Tissue plasminogen activator is an important trigger of fibrinolysis in vivo, but its half-life is normally less than 3 minutes. Circulating plasminogen activator inhibitor-1 levels are increased during Severe Acute Respiratory Syndrome (SARS) corona virus infection. Systemic fibrinolysis thus is unlikely to occur in COVID-19 patients with cytokine storm (Table 1 ).

Table 1

Laboratory Data of Patients with Moderate versus Severe D-Dimer Elevations

	D-Dimer (ng/mL)
	≤3,245	>3,245
Standard laboratory	n = 5	n = 6
CRP (mg/dL)	4.9 (3.8-26.1)	27.5 (13.0-32.7)
D-dimer (ng/mL)	2,410 (1,220-2,800)	15,465 (8,050-19,730)
Fibrinogen (mg/dL)	478 (351-1,057)	734 (567-1,016)
Hematocrit (%)	28.4 (24.4-30.3)	25.9 (22.1-28.7)
Platelet (× 109/mL)	211 (152-269)	144 (104-301)
PT (sec)	14.7 (13-14.7)	15.1 (14.9-15.4)
Thromboelastometry
EXTEM-CT (s)	73 (69-74)	76.5 (73-91.5)
EXTEM-A10 (mm)	63 (60-70)	67 (61.5-68.9)
FIBTEM-A10 (mm)	30 (30-36)	36.5 (32.8-43.4)
EXTEM-ML (%)	0	0

NOTE. Thromboelastometry was performed on the ROTEM Delta (TEM Innovations, Munich, Germany). EXTEM and FIBTEM reagents contain hexadimethrine bromide, that neutralizes heparin. Five patients in the high D-dimer group were on intravenous heparin. Reference ranges: C-reactive protein <1 mg/dL; D-dimer <640 ng/mL fibrinogen equivalent unit; fibrinogen 216-438 mg/dL; hematocrit 37%-50%; platelet 153-367 × 109/mL; prothrombin time 9.6-11.2 sec; EXTEM clotting time 43-82 seconds; EXTEM clot amplitude at 10 minutes 46-67 mm; FIBTEM clot amplitude at 10 minutes 7-24 mm; EXTEM maximal lysis <15%.

Abbreviations: A10, clot amplitude at 10 minutes; CRP, C-reactive protein; CT, clotting time; ML, maximal lysis; PT, prothrombin time.

Raza et al. previously showed that only 5% of trauma patients had fibrinolysis on ROTEM, whereas 57% of patients had moderate fibrinolysis with a median D-dimer level of 38,687 ng/mL. In our patients, a median D-dimer fibrinogen equivalent unit of 15,465 ng/mL and fibrinogen 734 mg/dL showed that only 0.21 % of fibrinogen was converted to D-dimer. In contrast, the data in the study by Raza et al showed that 1.84% of fibrinogen (median 210 mg/dL) was converted to D-dimer. Taken together, critically ill COVID-19 patients demonstrated significant elevations in D-dimer consistent with microvascular thromboses, but only small fractions of fibrin seem to be broken down locally and systemic fibrinolysis is rarely observed.

Conflict of Interest

None.