COVID19 coagulopathy in Caucasian patients.

Although the pathophysiology underlying severe COVID19 remains poorly understood, accumulating data suggest that a lung-centric coagulopathy may play an important role. Elevated D-dimer levels which correlated inversely with overall survival were recently reported in Chinese cohort studies. Critically however, ethnicity has major effects on thrombotic risk, with a 3-4-fold lower risk in Chinese compared to Caucasians and a significantly higher risk in African-Americans. In this study, we investigated COVID19 coagulopathy in Caucasian patients. Our findings confirm that severe COVID19 infection is associated with a significant coagulopathy that correlates with disease severity. Importantly however, Caucasian COVID19 patients on low molecular weight heparin thromboprophylaxis rarely develop overt disseminated intravascular coagulation (DIC). In rare COVID19 cases where DIC does develop, it tends to be restricted to late-stage disease. Collectively, these data suggest that the diffuse bilateral pulmonary inflammation observed in COVID19 is associated with a novel pulmonary-specific vasculopathy termed pulmonary intravascular coagulopathy (PIC) as distinct to DIC. Given that thrombotic risk is significantly impacted by race, coupled with the accumulating evidence that coagulopathy is important in COVID19 pathogenesis, our findings raise the intriguing possibility that pulmonary vasculopathy may contribute to the unexplained differences that are beginning to emerge highlighting racial susceptibility to COVID19 mortality.

Race and ethnicity have major effects upon thrombotic risk, with significantly lower risk in Chinese compared to Caucasian individuals.

Severe COVID19 infection is associated with a significant coagulopathy in Caucasian patients that correlates with disease severity.

Despite significantly increased D‐dimers, progression to overt disseminated intravascular coagulation (DIC) in Caucasian COVID19 patients maintained on prophylactic dose low molecular weight heparin (LMWH) is rare.

COVID19 infection was first described in December 2019 in Wuhan, China. , Since then, this disease has disseminated through most countries around the world and already caused more than 150 000 fatalities. The causative agent of COVID19, i.e. Severe Acute Respiratory Syndrome‐Coronavirus‐2 (SARS‐CoV‐2), is a novel betacorona virus that shares similarities with SARS and Middle East Respiratory Syndrome (MERS) viruses which were previously responsible for endemics in 2003 and 2012. , Studies have estimated overall COVID19 mortality rates ranging from 4·3% to 14·6%. , , This mortality burden is predominantly attributable to a progressive bilateral pneumonia that can ultimately progress to acute respiratory distress syndrome (ARDS).

Although the underlying pulmonary pathophysiology remains incompletely understood, severe COVID19 infection is associated with a marked alveolar inflammatory cell infiltrate, together with a systemic cytokine storm response. Several studies have also reported evidence of a COVID19‐associated coagulopathy. , , Furthermore, multivariate regression analysis in Chinese COVID19 cohorts reported that elevated plasma levels of fibrin degradation D‐dimers constituted an independent biomarker for poor prognosis in COVID19. Consistent with the hypothesis that coagulation activation may play a role in COVID19 pathogenesis, postmortem studies have highlighted marked pathological changes specifically involving the lung microvasculature, including disseminated micro‐thrombi and significant haemorrhagic necrosis. , Moreover, emerging data suggest that severe COVID19 is also associated with a significantly increased risk for developing deep vein thrombosis and pulmonary embolism. ,

Critically, the COVID19 coagulopathy data published to date have been predominantly derived from studies of Chinese patients. , , , This is important because race and ethnicity have major effects upon thrombotic risk. , In particular, epidemiological studies have shown that the incidence of venous thromboembolism (VTE) is approximately 3–4‐fold lower in Chinese compared to Caucasian individuals. , Conversely, VTE risk is significantly higher in African‐Americans compared to Caucasians. This effect has been consistently observed, even in individuals of different ethnicities living within the same geographical location. Importantly, reduced VTE prevalence also contributes to the fact that thromboprophylaxis is utilised less frequently in Chinese hospitals. Given these data, it is clearly important to determine whether there are differences in coagulopathic features in COVID19‐infected Caucasian compared to Chinese patients. In addition, the utility of D‐dimer levels as a prognostic marker in hospitalised Caucasian patients maintained on low molecular weight heparin (LMWH) thromboprophylaxis also needs to be validated.

Methods

Consecutive adult patients with COVID19 were recruited from St James’s Hospital between 13th March and 10th April. Inclusion criteria were individuals with a positive COVID19 polymerase chain reaction test in patients aged 18 years or older. The study was approved by the St James’s Hospital Research Ethics Committee and informed consent was obtained from all participants. Criteria for hospital admission were defined as those requiring inpatient care as a result of the severity of illness based on laboratory and radiological parameters as well as clinical gestalt. Following admission, all patients received supportive care in line with best international practice, which included the use of supplemental oxygen where indicated. In addition, hospitalised patients with COVID19 received weight‐ and renally‐appropriate doses of LMWH thromboprophylaxis unless contra‐indicated as part of standard of care [enoxaparin 20 mg once daily (OD) if <50 kg; enoxaparin 40 mg OD if 50–100 kg; 40 mg twice daily (BD) if 101–150 kg; 60 mg BD if >150 kg]. Eight patients had renal impairment on admission and were therefore treated with enoxaparin 20 mg OD. Four patients were on therapeutic anticoagulation on admission (two on apixaban, one on edoxaban and one on warfarin). Epidemiological, demographic, treatment and outcome data were derived from the hospital electronic patient records using a standard data collection form. For each subject, samples were collected at time of admission and at timepoints during their subsequent admission.

All haemostasis testing was performed in the National Coagulation Laboratory in St James’ Hospital, Dublin. Assays included prothrombin time (PT), activated partial thromboplastin time (APTT), fibrinogen, and D‐dimer levels. The PT, APTT and fibrinogen assays were measured on the ACL Top 550TM analyser using HemosIL(®) RecombiPlasTin 2G, HemosIL(®) SynthASil and HemosIL(®) Fibrinogen‐C reagents respectively. D‐dimer levels were measured in ng/ml Fibrinogen Equivalent Units (FEU) using the D‐Dimer HS 500 assay on the ACL Top 550 analyser (Instrumentation Laboratory, Bedford, MA, USA). Between patient subgroups, normally and non‐normally distributed quantitative data were compared using the Student’s t‐test and Mann–Whitney U‐test, respectively. Results were tabulated as the mean ± SD, median (interquartile range) or number (%) as appropriate. Data were analysed using GraphPad Prism 8 (GraphPad Software, San Diego, CA, USA), and a P < 0·05 was considered statistically significant.

Results and discussion

A total of 83 patients (55 males and 28 females) were enrolled in the study with a median age of 64 (range 26–92) years (Table I). Sixty‐seven patients (81%) were Caucasian, 10 (12%) were Asian, five (6%) were African and one (1%) patient was of Latino/Hispanic ethnicity. Underlying co‐morbidities were identified in sixty‐seven (80·7%) of the cohort. At time of writing, 50 patients (60·2%) had fully recovered and were discharged from hospital, while 20 (24·1%) remained in hospital and 13 (15·7%) had died. Fifty patients (60·3%) were discharged without requiring intensive care unit (ICU) admission, 23 patients (27·7%) were admitted to ICU during their disease course and 10 patients (12%) required ICU support but were deemed clinically unsuitable for ICU admission.

Table I

Demographics and coagulation parameters of COVID19 patients on admission and Day +4.

Parameters	Normal range	Total (n = 83)	Survivors and non‐ICU n = 50	Non‐survivors and/or ICU n = 33	P value
Age (years)		62 ± 16·3	60·5 ± 17·7	67·9 ± 11·9	0·02*
Sex (male/female)		55/28	33/17	22/11
Underlying co‐morbidities†		67 (80·7%)	38 (76%)	29 (87·9%)
On admission
PT (s)	9·9–13·1	12·9 (12–14·5)	12·6 (11·7–14·5)	12·9 (12·2–14·5)	0·11
APTT (s)	24–36	31 (28·2–31·9)	31.3 (29.3‐33.1)	30.4 (28.2‐32.2)	0.52
Fibrinogen (g/l)	1·9–3·5	4·7 (4·4–6·6)	4·5 (3·7–6·2)	5·6 (4·4–6·6)	0·045*
D‐dimer (nanogram/ml)	0–500	732 (553–1580)	804 (513–1290)	1003 (536·5–1782)	0·018*
Platelets (×109/l)	140–450	196 (162·3–227)	201 (161–251)	196 (153–289)	0·47
C‐reactive protein (mg/l)	0–5	56·1 (12·5–122·5)	37·9 (7·9–92·1)	94·8 (35–158·5)	0·0005***
Day 4 of admission
PT (s)	9·9–13·1	12·8 (12·4–13·8)	12·5 (12·2–14·1)	13·1 (12·5–14·5)	0·007**
APTT (s)	24–36	31·1 (29·8–33·1)	31·3 (29·5–35·3)	30·8 (28·3–33·4)	0·35
Fibrinogen (g/l)	1·9–3·5	4·9 (4·6–7·6)	5 (3·6–6·5)	5 (4·2–7·1)	0·29
D‐dimer (nanogram/ml)	0–500	881 (738·5–3459)	803 (529–1549)	1210 (603·5–3623)	0·003**
Platelets (×109/l)	140–450	212 (153–309)	203 (153–323)	231 (154–328·8)	0·52
C‐reactive protein (mg/l)	0–5	71 (33·3–225·7)	46·1 (21–105·9)	107 (58·1–222·5)	0·001**

P values are for survivors and non‐ICU (intensive care unit) compared to non‐survivors and/or ICU admission. Data are presented as mean ± SD or median (interquartile range) as appropriate (ns; not significant; *, P < 0·05, **, P < 0·01; ***, P < 0·001). PT, prothrombin time; APTT, activated partial thromboplastin time.

Underlying co‐morbidities: cardiovascular conditions including hypertension, ischaemic heart disease, stroke, type 2 diabetes mellitus and hyperlipidemia were most common. Other conditions included asthma, chronic obstructive pulmonary disease, solid organ or haematological cancer, and obesity.

Coagulation testing performed on admission in the total cohort demonstrated normal PT and APTT (Table I). In contrast, plasma D‐dimer levels were significantly increased (median 732; range 200–10 000 ng/ml). D‐dimer levels were above the normal range in 67% of the total cohort on admission (Table I). Importantly however, despite the increased D‐dimers, disseminated intravascular coagulopathy (DIC) was not evident. In particular, platelet counts were within the normal range in 83·1% with a platelet count <100 × 109/l observed in only five patients on admission. Fibrinogen levels were significantly increased (median 4·7; range 3·0–9·9 g/l) on admission, with levels remaining persistently elevated throughout hospitalisation. No patient had a fibrinogen level <2·0 g/l at any time point. This increase in fibrinogen levels is probably due to an acute phase response, as significantly elevated C‐reactive protein levels (median 56 mg/l; normal range 0–5 mg/l) were also observed (Table I). Thus, despite the fact that thrombotic risk is much higher in Caucasian patients and the significantly elevated levels of D‐dimers observed, overt DIC as defined according to the International Society on Thrombosis and Haemostasis Scientific Standardization Committee DIC score was present in none of our COVID19 patients at time of admission. Nevertheless, our data confirm that severe COVID19 infection is associated with a significant coagulopathy in Caucasian patients that appears to be similar in magnitude to that previously reported in the original Chinese cohorts. , ,

To assess whether COVID19 coagulopathy at time of admission was indicative of future clinical course, we divided our cohort into two groups based on the requirement of ICU admission for ventilatory support or death due to COVID19 infection versus those who were discharged without requiring ICU support. Median age of non‐survivors was 75·2 years (range 63·5–92) compared to 60·2 years (range 26·9–89) in survivors. Those whose admission resulted in an ICU stay or death were also more likely to have underlying co‐morbidities compared with those who survived and did not require ICU admission (Table I). In keeping with the previous Chinese data, we observed that abnormal coagulation parameters on admission were also associated with a poor prognosis in Caucasian patients with COVID19 infection. In particular, D‐dimer levels were significantly higher in the subgroup who eventually needed ICU admission (median 1 003 vs. 804 ng/ml; P = 0·018) (Table I). Similarly, fibrinogen and C‐reactive protein (CRP) levels were also both significantly elevated in the poor prognosis group. While no significant difference was observed in PT between the two groups on admission, the PT was higher in the adverse prognostic group by Day +4 (median 13·1 vs. 12.5 s; P = 0·007) (Fig 1). In addition, D‐dimer levels remained significantly higher in the poor prognosis group at Day +4 (Fig 1). Cumulatively, these data support the hypothesis that COVID19‐associated coagulopathy probably contributes to the underlying pulmonary pathogenesis.

Fig 1

Serial coagulation parameters in COVID19 patients following admission. Patient data are graphed as the median and interquartile range. Dotted lines represent the upper limit of the local normal range for prothrombin time (PT), activated partial thromboplastin time (APTT), D‐dimers and fibrinogen levels. Dotted line represents the lower limit of the normal range for platelet count. Survivors and patients not requiring ICU admission are in blue compared to non‐survivors and/or those needing ICU admission in red. As many non‐ICU patients had been discharged, no Day +14 data are presented for that group. [Colour figure can be viewed at wileyonlinelibrary.com]

Limited data are available regarding serial follow‐up of coagulation parameters in COVID19‐positive patients. Tang et al. observed a progressive increase in PT and D‐dimers over the first seven days following admission in Chinese patients who did not survive. In contrast, in COVID19 survivors, both PT and D‐dimer levels remained consistent over this period. Similarly, we observed a progressive increase in D‐dimer levels in our poor prognostic COVID19 group, but not in the subgroup who did not require ICU admission (Fig 1). In contrast to the Chinese data however, we observed no progressive increase in PT in the adverse prognostic group (Fig 1). We postulate that these different results probably reflect the fact that our cohort is predominantly Caucasian in origin, and also that our patients were commenced on LMWH at time of admission. Critically, despite the evidence of a progressive COVID19 coagulopathy over time, none of our cohort maintained on prophylactic LMWH developed systemic DIC. This is evident from the fact that platelet count, APTT and fibrinogen levels do not differ over this time period (Fig 1).

In conclusion, our findings demonstrate that severe COVID19 infection is associated with a significant coagulopathy that correlates with disease severity. The marked increase in D‐dimer levels is consistent with progressive coagulation activation, along with concurrent activation of fibrinolysis within the lungs. Critically however, COVID19 patients on prophylactic dose LMWH do not typically develop overt DIC. In rare COVID19 cases where DIC does develop, it tends to be restricted to late‐stage disease. Collectively, these data suggest that the diffuse bilateral pulmonary inflammation observed in COVID19 is associated with a novel pulmonary‐specific vasculopathy which was recently termed pulmonary intravascular coagulopathy (PIC) as distinct to DIC. , Although the biological mechanisms underlying COVID19 pulmonary vasculopathy remain poorly understood, the ACE‐2 receptor utilised by COVID19 is expressed on both type II pneumocytes and vascular endothelial cells (EC) within the lungs, , raising the interesting possibility that the pathobiology may include direct pulmonary EC infection, activation and/or damage. In addition, the cytokine storm associated with COVID19 infection will have major impacts upon thrombin generation and fibrin deposition within the lung. In the context of this lung‐centric vasculopathy, we hypothesise that the refractory ARDS phenotype observed in severe COVID19 is due to concurrent ‘double‐hit’ pathologies targeting both ventilation (V) and perfusion (Q) within the lungs where alveoli and pulmonary microvasculature exist in close anatomical juxtaposition. Our data suggest that, at least at standard prophylactic doses, LMWH does not significantly impact the progressive increase in D‐dimer levels observed in patients with severe COVID19. Further adequately powered randomised controlled studies will be required to determine whether more intensive anticoagulation and/or targeted anti‐inflammatory therapies may be useful in attenuating PIC in selected patients with severe COVID19. Given that thrombotic risk is significantly impacted by race, coupled with the accumulating evidence that coagulopathy is important in COVID19 pathogenesis, our findings raise the intriguing possibility that pulmonary vasculopathy may contribute to the unexplained differences that are beginning to emerge highlighting racial susceptibility to COVID19 mortality.

Conflicts of interest

JSO’D has served on the speaker’s bureau for Baxter, Bayer, Novo Nordisk, Boehringer Ingelheim, Leo Pharma, Takeda and Octapharma. He has also served on the advisory boards of Baxter, Bayer, Octapharma CSL Behring, Daiichi Sankyo, Boehringer Ingelheim, Takeda and Pfizer. JSO’D has also received research grant funding awards from Baxter, Bayer, Pfizer, Shire, Takeda and Novo Nordisk.

Author contributions

HF, LT, CNC, CB, IML, PB, CLB, RG, AG, MB, KR, NOC, NC, JOS & JOD — conception, patient enrollment, data collection and interpretation. All authors contributed to literature review, final draft writing and critical revision. All the authors have participated sufficiently in this work, take public responsibility for the content and have made substantial contributions to this research.