D-dimer testing in patients with COVID-19

D-Dimer is a product of fibrin degradation (fibrinolysis) that in health circulates in blood plasma at low blood concentration. Since activated blood coagulation and consequent fibrinolysis is associated with increased plasma D-dimer concentration, D-dimer has proven a clinically useful biomarker of thrombotic disease.

The most established utility of the D-dimer test is in the investigation of patients suspected of suffering venous thromboembolism (deep vein thrombosis (DVT) and its sequelae, pulmonary embolism (PE). D-dimer is almost always increased in those with DVT/PE and so a normal result is routinely and widely used to help exclude the diagnosis.

An elevated D-dimer cannot be used to diagnose DVT/PE because of its low specificity; there are many other conditions associated with thrombosis (abnormal increased blood coagulation, coagulopathy) and consequent elevated D-dimer. These include: myocardial infarction, stroke, aortic dissecting aneurysm, disseminated intravascular coagulation, severe infection/sepsis, cancer and acute kidney injury.

COVID-19, the pandemic disease caused by infection with the novel virus, SARS-CoV-2 (severe acute respiratory syndrome coronavirus-2) can now be added to the already extensive list of conditions that may be associated with elevated D-dimer.
The discovery that D-dimer may be elevated in COVID-19 was first reported by physicians in Wuhan, China where the epidemic started. A study of 191 patients with COVID-19, who were hospitalised in Wuhan during January 2020 at the outset of the pandemic, revealed that D-dimer was elevated in many of these patients and the magnitude of the elevation was greatest in those who did not survive.

A number of subsequent studies conducted around the world have confirmed that D-dimer is elevated in those with severe COVID-19 and highest in those who are most critically ill and those who do not survive. Much COVID-19 research over the past months has been directed at understanding the significance of D-dimer elevation and the COVID-19 related coagulopathy that is presumed responsible for the elevation.

This highlighted study provides further evidence and detail of the prevalence and consequences of D-dimer elevation in COVID-19. It provides the most robust evidence to date, not least because of its large, unselected study population.

This is a retrospective observational study of 2377 consecutive COVID-19 positive patients (>18 yrs) admitted to a multi-hospital New York health system during the period March 1^st – April 8^th, 2020 and followed up until study end on May 20^th. D-dimer testing at admission and during admission is routinely applied to COVID 19 patients at this hospital system. The upper limit of normal for the D-dimer assay used is 0.23 μg/ml (230 ng/ml).

At admission, 1823 (76%) study patients had raised D-dimer (> 0.23 μg/ml (230 ng/ml)); 932(39%) had D-dimer in the range 0.23-0.50 μg/ml (230-500 ng/ml), 628 (26%) had D-dimer in the range 0.501-2.0 μg/ml (501-2000 ng/ml), and 263 (11%) had D-dimer > 2.0 μg/ml (2000 ng/ml).

Of the total study population, 899 (37.8%) had critical illness (admitted to intensive care unit), 620 (26.1%) required mechanical ventilation, 410 (17.2%) had a thrombotic event (DVT, PE, myocardial infarction, ischemic stroke) and 871 (36.8%) had acute kidney injury (AKI).

For all these serious adverse clinical outcomes of COVID-19, those with elevated D-dimer (> 0.23 μg/ml (230 ng/ml)) at admission were more likely to be affected than those with normal D-dimer (< 0.23 μg/ml (230 ng/ml)).

So, for example, 43.9% of patients with raised D-dimer at admission compared with 18.5% of those with a normal D-dimer at admission developed a critical illness. Similarly, patients with elevated D-dimer at admission were more likely to require mechanical ventilation (29.9% v 13.9%), more likely to suffer AKI (42.4% v 19.0%) and more likely to suffer a thrombotic event (19.4% v 10.2%).

The magnitude of D-dimer elevation at admission was found to be independently associated with risk of all these serious clinical outcomes. After controlling for age, sex, ethnicity, a defined list of co-morbidities and the use of a defined list of prescribed drugs, critical illness was found to be 1.8 times more likely if D-dimer was in the range 0.23-0.50 μg/ml (230-500 ng/ml) than if it were normal (< 0.23 μg/ml (230 ng/ml)), 3.1 times more likely if D-dimer was in the range 0.5-2.0 μg/ml (500-2000 ng/ml) and 5.6 times more likely if D-dimer at admission was > 2.0 μg/ml (2000 ng/ml).

This pattern was evident for the two other serious clinical outcomes: AKI and thrombosis. Irrespective of age, sex, gender, and pre-existing illness, the higher the admission D-dimer, the greater was the likelihood of AKI and thrombosis during hospital admission.

One of the more unique features of this study, made possible by the comprehensive D-dimer testing policy at the hospitals, was recording the trajectory of D-Dimer elevation during COVID-19 hospital admission. This revealed that D-dimer levels continue to rise after admission, reaching a peak level at around day 5, before gradually plateauing at a level lower than peak but greater than that at admission, for the duration of hospital admission. Median D-dimer for the whole study population at admission was 0.387 μg/ml (387 ng/ml) (25th-75th percentile range: 0.237-0.713 μg/ml (237-713 ng/ml)) compared with peak median 0.767 μg/ml (767 ng/ml) (25^th -75^th percentile range: 0.328-3.372 μg/ml (328-3372 ng/ml)) around 5 days later.

Results of the study demonstrate that just as magnitude of admission D-dimer level is independently associated with risk of adverse clinical outcome so too is magnitude of peak D-dimer level. Of the total study population, 301(12.7%) were found to have a peak D-dimer level > 10.0 μg/ml (10,000 ng/ml). This patient group with highest peak D-dimer levels had highest incidence of serious clinical outcomes during hospital admission: 86% had critical illness requiring admission to intensive care; 71% required mechanical ventilation; 81% had AKI; and 39% had a thrombotic event.

Apart from the association between D-dimer levels and adverse clinical outcomes, the study also sought to investigate the relationship between D-dimer level at admission and the ultimate outcome for COVID-19 patients: death or survival.

At study end date (20^th May 2020), 1652 (69.5%) of the 2377 study subjects had survived their COVID-19 illness and been discharged from hospital, 117(4.9%) remained in hospital, and 608 (25.6%) had died (or been transferred to hospice for terminal illness care). Unadjusted mortality rate was higher in those with elevated D-dimer at admission (> 0.23 μg/ml (230 ng/ml)) than those with normal D-Dimer at admission (D-dimer < 0.23 μg/ml (230 ng/ml) (29.9% v 10.8%).

After controlling for variables as listed above (age, sex, gender etc), it was evident that the magnitude of D-dimer elevation at admission is independently associated with risk of death. Irrespective of age, gender, ethnicity and pre-existing co-morbidity, death was found to be 1.7 times more likely if D-dimer was in the range 0.23-0.50 μg/ml (230-500 ng/ml) at admission, than if it were normal (< 0.23 μg/ml (230 ng/ml)), 2.3 times more likely if D-dimer was in the range 0.5-2.0 μg/ml (500-2000 ng/ml) and 4.2 times more likely if D-dimer was > 2.0 μg/ml (2000 ng/ml).

A similar pattern was seen for peak D-dimer levels: magnitude of peak D-dimer (at day 5 of hospital admission) was shown to be independently associated with risk of death. Of those 301 patients with highest peak D dimer (> 10.0 μg/ml (10,000 ng/ml)), 182 (60.5%) had died or been transferred for terminal care at study end.
In discussion of their study, the authors summarise the results of other similar studies and briefly review current understanding of the pathophysiology of COVID-19 related coagulopathy and associated D-dimer elevation.

This study provides the most robust evidence to date that D-dimer is predictive of serious illness and death due to SARS-CoV-2 infection, and thereby provides further support for the notion that point of care D-dimer testing might be a helpful tool for the early triaging of COVID-19 patients presenting to the hospital emergency department and for monitoring patients during the first week of hospital admission.