Thrombomodulin resistance as a novel prothrombotic pathway in COVID-19

Abstract

Hypercoagulability and endothelial dysfunction are strongly involved in the worsening of the clinical condition in COVID-19 patients. In severe cases, the inflammatory process triggers the release of angiopoietin 2, which could decrease circulating thrombomodulin (TM), a major regulatory mechanism in thrombin generation. Although some studies have described an increased TM resistance, further data are needed to obtain robust results. The objective of our study was to evaluate TM resistance in hospitalized COVID-19 patients using the thrombin generation test and its correlation with development of any severe clinical events (SCE). Forty-seven hospitalized COVID-19 patients were included (median age was 59 years (50–75); 42.6% women). Measurement of endogenous thrombin potential (ETP) revealed that 54.8% of patients had a percentage (%) of ETP inhibition < 40%, suggesting TM resistance. 23% (23%) of patients (n = 11/47) presented at least one SCE. Significant resistance to TM was observed in patients with SCE: percentage (%) of ETP inhibition was 24.3% vs. 47.6% (p = 0.019) in the non-SCE group. Furthermore, lower percentage (%) of ETP inhibition significantly correlated with increased clot stiffness (r= -0.372, p = 0.0167). The percentage (%) of ETP inhibition was a strong predictor of SCE, with an AUC of 0.803 (95%CI: 0.670–0.936). To conclude, thrombin generation can be a powerful tool for risk stratification in hospitalized COVID-19 patients. In addition, increased TM resistance, quantified by a lower percentage (%) of ETP inhibition is strongly associated with the development of SCE and shows promise as a powerful and new independent marker of poor prognosis.

Introduction

COVID-19 associated coagulopathy (CAC) is a life-threatening complication of SARS-CoV-2 infection characterized by an excessive release of proinflammatory cytokines (cytokine storm) and vascular endothelial dysfunction, promoting a procoagulant condition^1,2. Indeed, CAC has been described in up to 70% of deceased COVID-19 patients³. Moreover, endothelial cell injury is involved in organ impairment, leading to disruption of normal coagulation balance, with hypofibrinolysis and complement activation⁴. In fact, the endothelial cell injury constitutes a major feature in the pathophysiology of COVID-19 infection, especially during the inflammatory phase⁵.

Thrombomodulin (TM) is a high-affinity thrombin receptor located in the endothelial cell membrane and plays an essential role as an endogenous natural anticoagulant. TM preserves the endothelial microenvironment and, therefore, maintains an anti-inflammatory and anticoagulant state, in cooperation with protein C (PC) and thrombin-activatable fibrinolysis inhibitor (TAFI). In that way, there is an inhibition in the coagulation cascade through its action on activated factor VIII and activated factor V⁶.

A dysregulation in the Tie2-angiopoietin pathway in the endothelium reveals a critical role in the link between inflammation and coagulation in critical illnesses, including COVID-19^7,8. One of the main components involved in this dysregulation is angiopoietin 2 (ANGPT2), an inflammatory cytokine released from Weibel Palade sacs due to endothelial dysfunction. ANGPT2 binds competitively to Tie2 receptors, promoting vascular destabilization⁹. In addition, ANGPT2 binds with high affinity to TM, thereby impacting its anticoagulant capacity^10,11. Thus, the ANGPT2 released in the setting of endothelial damage and its subsequent binding to TM could be contributing to the hypercoagulable state in COVID-19 patients^10,12. Indeed, circulating levels of ANGPT2 have been associated with adverse outcomes in several critical care syndromes such as acute respiratory distress syndrome (ARDS) or sepsis and have recently been described in critically ill COVID-19 patients^7,8,13. Endothelial injury also induces von Willebrand factor (VWF) release, probably associated with acquired non-immune ADAMTS-13 (a disintegrin and metalloproteinase with thrombospondin type 1 motif, member 13) consumption and all these factors together contribute to thrombosis development¹⁴.

Data on thrombin generation (TG) in CAC are scarce. Consistent with endothelial dysfunction, some studies using the thrombin generation test (TGT) have shown that COVID-19 patients have a higher resistance to TM compared to the healthy population¹⁵. This resistance has been defined by the anticoagulant effect of TM, which has been shown to reduce the potential for TG when added to TGT¹⁶. These results were also observed when comparing COVID-19 patients admitted to the intensive care unit (ICU) versus non-ICU patients. The authors suggest that this may be a consequence of PC deficiency (or another defect in the TM-mediated anticoagulation pathway due to endothelial injury), resulting in the increased TG capacity of COVID-19 plasma¹⁷. Increased factor V levels, which may lead to increased TG, either directly or indirectly through a mechanism involving activated PC resistance, could also be involved in this prothrombotic disorder¹⁸. However, other studies did not observe differences in TM resistance between critical and non-critical COVID-19 patients, so its value as a prognostic factor is uncertain. In that way, Benati et al. suggested that the PC system might be a minor player in COVID-19 coagulopathy, observing very similar results between the TG profile with and without TM in COVID-19 patients versus healthy controls¹⁹. However, these authors included only 16 hospitalized COVID-19 patients and 19 healthy controls (potentially limiting statistical power), and they do not specify TG according to clinical severity events. The resistance to TM was not described.

The main aim of the study was to analyze the resistance to TM in hospitalized COVID-19 patients using the TGT and its correlation with clinical events. As a secondary objective, we aimed to evaluate changes in proteins of endothelial origin in severe COVID-19 patients.

Materials and methods

Study design

We designed a prospective observational study including hospitalized COVID-19 patients confirmed by polymerase chain reaction within the first week of diagnosis from May 2021 to December 2021. This study was performed at the General University Hospital Dr. Balmis in Alicante and the Regional University Hospital in Málaga, Spain.

Inclusion criteria

Patients over 18 years old who require hospitalization due to COVID-19 infection were recruited. All patients signed the informed consent prior to inclusion in the study. Hospitalization criteria for patients with COVID-19 were based on disease severity, including the presence of respiratory failure, the need for supplemental oxygen, and the presence of complications. The complications comprised severe pneumonia, pleural effusion, ARDS, secondary bacterial infection and severe coagulopathy. Additionally, factors such as age, the presence of comorbidities, and inadequate response to outpatient treatment were also considered to determine the need for hospitalization.

The study was conducted in accordance with the basic principles of the World Medical Association Declaration of Helsinki and complied with the standards described in the European Union Guidelines for good clinical practice. This study was approved by the local ethic committee at the University General Hospital Dr. Balmis. The ethical approval number is PI2020-089.

Exclusion criteria

Patients with previous history of arterial or venous thrombosis, current pregnancy, postpartum, oral contraceptive intake, hormone replacement treatment, active cancer and individuals receiving anticoagulant or antiplatelet therapy at inclusion were excluded.

Sample collection

Blood samples were obtained by direct vein puncture and anticoagulated with 0.129 M sodium citrate during the first week once diagnosis was confirmed. Platelet-poor plasma (PPP) was immediately obtained by double centrifugation at 2500 g for 15 min at 22º C. The aliquots were stored at -80 °C until processing. Samples were defrosted at 37 °C for 10 min prior to analysis.

All patients received standard thromboprophylaxis and adjusted in case of renal impairment (glomerular filtration < 20 mg/ml), in agreement with the International Society on Thrombosis and Hemostasis (ISTH) and CHEST (American College of Chest Physicians) guidelines recommendations^20,21. Patients who required intensive care, heparin was maintained at prophylaxis doses or adjusted to intermediate doses if several cardiovascular risk factors (CVRF) were present^20,21. Regarding the treatment for COVID-19 during hospitalization, patients were treated with dexamethasone, remdesivir and/or tocilizumab according to local hospital guidelines. Supportive measures tailored to the patient´s condition were also provided.

Clinical variables

The following CVRF were recruited at inclusion: arterial hypertension, diabetes mellitus, coronary heart disease, chronic obstructive pulmonary disease (COPD) and obstructive sleep apnea.

The following severe clinical events (SCE) were considered at inclusion: non-invasive mechanical ventilation, ICU admission, encephalitis, thrombosis or death.

Hematological variables

We measured the activated partial thromboplastin time (APTT), prothrombin time, fibrinogen and D-dimer on an automated coagulation analyzer (STA evolution^®, Stago, Paris, France), following protocols from the manufacturer.

White blood cell count, hemoglobin and platelet count were performed on a fully automatic analyzer (Sysmex XN 550^®, Le Roche, Basel, Switzerland).

TGT was analyzed using the Genesia^® analyzer and the STG^®-Thromboscreen kit (Diagnostica Stago, Paris, France), based on the fluorometric method described by Hemker et al.²². When testing, a new calibration test, three levels of quality control (low, normal, and high TM resistance), and a reference plasma provided by the manufacturer to normalize parameters are assessed. The STG^®-Thromboscreen kit contains a mixture of phospholipids and an intermediate picomolar concentration of human recombinant tissue factor, with and without adding TM, as activator of the coagulation system.TM concentration is that required to obtain a 50% decrease of the endogenous thrombin potential (ETP). The final TM concentration is not disclosed by the manufacturer. TG was measured in PPP samples thawed in a 37ª water bath for 3 min. TGT provides information on the start time of thrombin generation (lag time, LT, minutes), the ETP or total amount of thrombin generated (ETP, nM/min, min: minutes), the peak of thrombin concentration (Max peak, nM), the time to reach the maximum peak of thrombin (Tt Peak, minutes), the time to thrombin neutralization (start tail, ST, minutes) and the velocity of thrombin formation (velocity index, Vel Index, nM/min). We defined TM resistance as an ETP reduction less than 40% after TM addition. This definition was established home-made applying TGT to COVID-19 patients with thrombotic risk (n = 120). We calculated TM resistance applying the following formula to these 120 patients with COVID-19: (1-(ETP with TM/ETP without TM)) X 100, in agreement with the manufacturer´s instructions. Patients with an ETP reduction less than 40% after TM addition had the worst outcome, in terms of thrombotic development or death. All samples were processed and analyzed at the University General Hospital Dr. Balmis.

Global hemostatic analysis in whole blood samples was also measured with the Quantra^® hemostasis analyzer (Qplus^® and QStat^®, HemoSonics, Paris, France). This technique provides information on the clot time (CT, seconds), heparinase clot time (HCT, seconds), clot stiffness (CS, hPa), fibrinogen contribution to clot stiffness (FCS, hPa), platelet contribution to clot stiffness (PCS, hPa) and clot stability to lysis (CLS, %).

TM was measured by Enzyme-Linked Immunosorbent Assay (ELISA, Abcam, Cambridge, UK), designed for the quantitative measurement of TM in serum, plasma buffered solutions and supernatants, following protocols from the manufacturer. Normal range lies between 0.625 ng/ml and 20 ng/ml.

ANGPT2 was measured by Enzyme-Linked Immunosorbent Assay (ELISA, Abcam, Cambridge, UK), designed for the quantitative measurement of human ANGPT2 in serum, plasma buffered solutions and supernatants, following protocols from the manufacturer. Normal range lies between 4.12 pg/ml and 3000 pg/ml.

Quantitative determination of plasminogen activator inhibitor-1 (PAI-1) was assessed by Enzyme-Linked Immunosorbent Assay (ELISA, Asserachrom^®, Diagnostica Stago, Paris, France) in PPP. Normal range lies between 4 and 43 ng/ml.

The TAFI activity was also determined in PPP. Plasma TAFI was activated using thrombin (Stachrom^®, Diagnostica Stago, Paris, France). The activated TAFI partially hydrolyzes a chromogenic substrate, and subsequent hydrolysis by carboxypeptidase A leads to substrate discoloration. The measurement of this discoloration is proportional to the TAFI activity in the sample. Normal range lies between 108 ± 24.5%.

Plasma ADAMTS13 activity and VWF antigen were measured using chemiluminescence (AcuStar, International Laboratory, Bedford, MA, USA) following the protocols from the manufacturer.

Statistical analysis

Descriptive analysis of qualitative variables included percentages. Shapiro–Wilk test was used for testing normality of continuous variables. Normally distributed continuous variables were presented as means ± standard deviations, whereas non-normally distributed variables were presented as median and interquartile range (p25-p75 percentile). Student’s-t tests (parametric) or Mann–Whitney-U tests (non-parametric) were used for group comparison as appropriated.

Association between TG and other continuous laboratory variables was assessed by Pearson’s (parametric) or Spearman’s Rho (non-parametric) correlation tests.

The ROC (Receiver Operating Characteristic) curve was determined to study the probability of outcome. Statistical significance was set at 5% (P < 0.05) and 95% Confidence Intervals (95%CI) were also calculated using bootstrap resampling (1,000 iterations) for non-normally distributed variables. To account for multiple comparisons, we applied the Holm–Bonferroni method, balancing the risk of type I and type II errors. A post-hoc power analysis for the primary outcome (ETP inhibition) was performed using Cohen’s d. Statistical analyses were primarily conducted using IBM SPSS version 26 (IBM Corp., Chicago, IL, USA). Complementary analyses, including bootstrapping and p-value adjustments, were performed using R (version 4.5.1, R Foundation for Statistical Computing, Vienna, Austria) within the RStudio environment (version 2025.05.1, RStudio, PBC, Boston, MA, USA) on Windows 11.

Results

Demographic and clinical characteristics of the patients

A total of 47 COVID-19 hospitalized patients were analyzed. The mean age was 59 years (50–75), and 42.6% of them were women. Almost half of the patients (40.6%) had at least one comorbidity, and a total of 23.4% of patients (n = 11/47) experienced at least one SCE during hospital admission. See Table 1 for demographic and clinical data.

Table 1.

Demographic and clinical characteristics of hospitalized COVID-19 patients.

N = 47 patients	Median (p25-p75) or N (%)
Age (median)	59 years (50–75)
Gender
Male	27 (57.4)
Female	20 (42.6)
Comorbidities
Hypertension	21 (44.7)
Diabetes	16 (34)
Coronary heart disease	6 (12.8)
COPD	4 (8.5)
SCE
Mechanical ventilation	8 (17)
ICU admission	5(10.6)
Death	3 (6.4)
Thrombotic event	3 (6.4)
Encephalitis	1 (2.1)

N: number of patients; p25-p75: 25th – 75th percentile; SCE: severe clinical event; COPD: chronic obstructive pulmonary disease; ICU: intensive care unit.

Increased hypercoagulability in hospitalized COVID-19 patients

Hyperfibrinogenemia was detected in 81.6% of patients (591.5 mg/dL [268–900]). Fibrinogen was higher in SCE-patients compared to non-SCE patients (671 mg/dL [494–900] vs. 554 mg/dL [268–900], p = 0.032). D-dimer was elevated in 67.4% of patients (600.5 mg/dL [131–8731]). SCE-patients also showed higher D-dimer levels than non-SCE patients, although this difference did not reach statistical significance (779 mg/dL [300–2881] vs. 575 mg/dL [131–8731], p = 0.202).

Global hemostatic analysis by Quantra showed a median CS and PCS of 31.25 hPa [15–64] (normal range: 13-33.2) and 25.8 hPa [12–48] (normal range: 11.9–29.8), respectively. An increased CS due to FCS (> 3.7 hPa) was described in 70.2% (33/47) of patients (4.8 hPa [2–15]. SCE-patients showed a higher CS (36.3 hPa [24–64] vs. 26.7 hPa [15–53], p = 0.008), PCS (30.6 [20–48] vs. 22.8 hPa [12–40], p = 0.001) and FCS (6.8 hPa [4–15] vs. 4.3 hPa [2–15], p = 0.011), compared to non-SCE patients.

Hospitalized COVID-19 patients showed a median VWF antigen (VWF: Ag) levels of 250.9% (87–400) and an ADAMTS13 activity of 73.3% [13–114]. The median vWF: Ag/ADAMTS13 ratio in the global cohort was 3.41 (1–17), being higher in the SCE-patients group than in the non-SCE patients group (4.45 [3–7] vs. 3.45 [2–9], p = 0.039).

Increased resistance to thrombomodulin in SCE-patients

The results of the TGT are shown in Table 2.

Table 2.

Thrombin generation parameters after in vitro thrombomodulin addition in patients with and without severe clinical events (SCE).

Variable	Non-SCE patients	SCE patients	Diff (95% CI)	p-value (adjusted)
LT (TM+) (min)	3.1 [2.6–3.5]	2.7 [2.5–2.9]	0.4 (-0.1 to 0.8)	0,490
ETP (TM+) (nM/min)	787 ± 380	1238 ± 381	451 (–729 to − 172)	0,015
ETP inh (%)	47.6 ± 22.6	24.3 ± 10.2	–23.4 (8.3 to 38.4)	0,019
Ttpeak (TM+) (min)	5.0 [4.5–5.7]	4.5 [4.2–4.8]	0.5 (-0.0 to 1.2)	0,251
Max peak (TM+) (nM)	292 ± 119	363 ± 88	71 (–149 to 8)	0,383
ST (TM+) (min)	13.9 [12.8–15.2]	13.8 [13.2–14.4]	0.1 (-0.8 to 1.3)	1
Vel Index (TM+) (nM/min)	138 [72–186]	206 [162–279]	-68 (-184 to -9)	0,031

Data are shown as mean ± SD or median [p25–p75]. Mean or median differences (Diff) with 95% CI are reported. Adjusted p-values (Holm–Bonferroni) account for multiple comparisons between SCE (severe clinical event) and non-SCE groups. LT: Lag time; TM+: samples with added thrombomodulin; min: minutes; ETP: endogenous thrombin potential; inh: inhibition; Ttpeak: Time to maximum thrombin peak (Max peak); ST: Start tail; Vel Index: Velocity Index.

No differences were observed in TGT in the absence of TM in terms of LT, ETP, Tt peak, ST and Vel Index between patients with and without SCE. In contrast, the TGT study with TM revealed a higher ETP (1238 ± 381 nM/min vs. 787 ± 380 nM/min; p = 0.015), max peak (363 ± 88 nM vs. 292 ± 119 nM, p = 0.383), vel index (206 nM/min [162–279] vs. 138 nM/min [72–186], p = 0.031) and a lower Tt peak (4.5 min [4.2–4.8] vs. 5.0 min [4.5–5.7], p = 0.251) than non-SCE patients. These patients also showed greater resistance to TM than those without SCE (24.3 ± 10.2 vs. 47.6 ± 22.6%; p = 0.019) (see Fig. 1). In fact, 100% of SCE-patients showed a % inhibition of ETP with TM < 40% (p < 0.001).

Fig. 1.

Changes in percentage (%) in ETP inhibition between SCE and non-SCE patients.

No correlation between percentage (%) inhibition of ETP and age (r = 0.1822, p = 0.248) was observed. In addition, there was no significant difference in percentage (%) inhibition of ETP between men (mean = 43.6 ± 22.5) and women (mean = 39.8 ± 23.3) (p = 0.598).

Correlation between increased thrombin generation and clot stiffness

Increased CS correlated positively with increased TG. A higher CS significantly correlated with higher max peak thrombin in the presence of TM (r = 0.320, p = 0.042) and a lower percentage (%) inhibition by TM (r=-0.372, P = 0.017). A higher PCS also significantly correlated with a lower percentage (%) inhibition by TM (r=-0.385, P = 0.013).

Similar TGT pattern in hospitalized anticoagulated COVID-19 patients compared to healthy non-COV19 individuals

We evaluated TG in 120 hospitalized patients with COVID-19 infection. These patients received heparin according to ISTH and CHEST guidelines. Normalized data expressed as ratio (in case of LT, Ttpeak and ST) and percentage (%) (in case of ETP and max peak) compared to a reference plasma (i.e., plasma from healthy non-COVID.19 subjects), suggest a similar TG tendency. See Table 3.

Table 3.

Thrombin generation pattern in hospitalized COVID-19 patients and healthy individual.

	Global (N = 120) Absolute results	Global (N = 120) Normalized results*
LT (min)	2.83 [2–17]	1.43 (ratio) [1.24–1.73]
LT (TM+) (min)	2.95 [2–8]	NA
ETP (nM/min)	1,407 [403-3,807]	98.55 (%) [85.37–112.5]
ETP (TM+) (nM/min)	745.8 [63 − 1,953]	NA
ETP inh (%)	47.6 [5–96]	NA
Ttpeak (min)	5.25 [3–24]	1.27 (ratio) [1.11–1.53]
Ttpeak (TM+) (min)	5.13 [3–10]	NA
Max peak (nM)	257.1 [51–597]	106.2 (%) [75.26–120.7]
Max peak (TM+) (nM)	160.1 [10–515]	NA
ST (min)	17.6 [12–50]	1.00 (ratio) [0.86–1.17]
ST (TM+) (min)	14.3 [11–45]	NA
Vel Index (nM/min)	187.4 (9-625)	NA
Vel Index (TM+) (nM/min)	153.8 (11–517)	NA

*Compared to a reference plasma. LT: lag time; TM+: sample with added thrombomodulin; ETP: endogenous thrombin potential; inh: inhibition; Ttpeak: time to peak; Max peak: peak thrombin concentration; ST: start tail; Vel Index: velocity of thrombin formation. NA: Not applicable.

Endogenous thrombin potential as a poor prognosis marker in COVID-19

ETP in the presence of TM proved to be a good predictor of SCE during admission with an AUC (Area Under the Curve) of 0.791 (p < 0.001). The % of ETP inhibition was a strong predictor of SCE, with an AUC of 0.803 (95%CI: 0.670–0.936).

Endothelial origin parameters in SCE and non-SCE COVID-19 patients

A non-significant increase in endothelial origin markers was observed in SCE patients, suggesting a hypercoagulable state. See Table 4.

Table 4.

Changes in endothelial origin parameters in SCE and non-SCE COVID-19 patients.

Variable	Non-SCE patients	SCE patients	Difference (95% CI)	p-value (adjusted)
ANGPT-2 (pg/ml)	643.8 [395.9–1172.5]	355.0 [254.2–599.6]	289 (-74 to 679)	0.361
TM (ng/ml)	3.2 [2.8–4.4]	4.2 [3.6–4.7]	-1.0 (-1.9 to 0.3)	1
TAFI (%)	112.3 ± 32.3	122.9 ± 28.4	10.6 (-32.9 to 11.8)	1
PAI-1 (ng/ml)	33.7 [25.1–44.9]	37.9 [35.6–42.3]	-4.2 (-11.2 to 2.6)	1

Data presentation and statistical methods as in Table 2. ANGPT-2: angiopoietin-2; TM: thrombomodulin; TAFI: thrombin-activatable fibrinolysis inhibitor; PAI-1: plasminogen activator inhibitor-1; SCE: severe clinical event.

ETP: endogenous thrombin potential; SCE: severe clinical event.

Discussion

TGT performed in the presence of TM is increasingly recognized as a valuable tool for assessing hemostatic balance in PPP because of its ability to concomitantly assess the effect of pro- and anticoagulant plasma factors²³.

The majority of inpatients with COVID-19 infection require prophylaxis with heparin and the dose may vary depending on the severity of the disease, considering intermediate or therapeutic doses in critically ill patients²⁴. However, there is controversy in the persistence of the hypercoagulability evaluated by TGT despite anticoagulation. In that way, Campello et al. studied 89 patients (30 in ICU) and determined an increased TG that reached levels of healthy individuals under thromboprophylaxis with heparin¹⁷. In the same way, Cohen et al. did not observe statistical differences from the values obtained from non-anticoagulated healthy controls in 32 COVID-19 patients²⁵. Other studies observed hypercoagulation persistence despite anticoagulation^19,26. The sample sizes in these 2 last studies were small (n = 23 and n = 16 respectively), so these results should be taken cautiously. We found, in agreement with Campello ¹⁷ and Cohen²⁵, no relevant differences between anticoagulated COVID-19 patients and healthy non-anticoagulated individuals. Only a slightly prolonged LT (1.43 times) and Ttpeak (1.27 times) in COVID-19 inpatients, suggesting heparin effect. However, the ETP and max peak were similar (98.55% and 106.2% compared to a reference plasma, respectively). In addition to these findings, we also reported an increase in resistance to TM in patients with SCE (including mechanical ventilation, encephalitis, thrombosis, ICU admission and death) compared to those without SCE and we assessed parameters of endothelial origin such as ANGP2, TM, PAI-1 and TAFI, between those groups. We observed an increasing trend in hypercoagulable markers of endothelial origin in SCE patients, such as TM, PAI-1 and TAFI.

We established TM resistance if percentage (%) of ETP inhibition was < 40%. The ETP inhibition was more remarkable in SCE patients compared to non-SCE patients (24.3 ± 10.2 vs. 47.6 ± 22.6%; p = 0.019), reflecting an accentuated TM resistance in severe COVID-19 patients. In addition, when adding TM, we observed a higher ETP (p = 0.015) and max peak (p = 383) and a shorter Ttpeak (p = 0.251) in SCE patients compared to non-SCE patients. Therefore, SCE patients had an increased TG compared to non-SCE patients and this effect is not dose heparin dependent. All SCE patients received intermediate heparin dose (not prophylactic). Cohen et al. did not find correlation between TG parameters and severity of COVID-19 infection²⁵. However, they assessed a small sample size.

This prothrombotic state correlated with Quantra, which measures global hemostasis as well. In particular, a higher CS and PCS correlated with a lower ETP inhibition and TM resistance. The mean age of our cohort was 59 years old (50–75).

Little data is available about the association between age and resistance to TM. The literature suggests that an increase in procoagulant factors and a reduced protein C activation have been described in older subjects as well as in men and postmenopausal women, and these findings can promote resistance to TM²⁷. In our study, no correlation between percentage (%) of ETP inhibition values and age (r = 0.1822, p = 0.248) was observed. In addition, we did not find a significant difference in percentage (%) of ETP inhibition between men (mean = 43.6 ± 22.5) and women (mean = 39.8 ± 23.3) (p = 0.598).

We observed a higher D-dimer in SCE patients compared to non-SCE patients (although this difference did not reach the statistical significance), and this finding implies a worse clinical outcome, as it has been described in the literature^28,29. In our study, we also described elevated fibrinogen levels as an acute phase reagent in the setting of an acute infection, more remarkable in SCE patients. Zhang et al. predicted mortality in COVID-19 patients with hyperfibrinogenemia³⁰.

The endothelial cell injury plays a key role in CAC, contributing to a hypercoagulable status, with complement activation, hyperfibrinolysis and kinins cascades dysfunction. In addition, the release of pro-inflammatory particles in the setting of the underlying endotheliopathy contributes to the pathogenic mechanisms in lung and other organ injury⁴. To emphasize this endothelial impairment, the median vWF: Ag/ADAMTS13 ratio in the global cohort was higher in the SCE-patients group than in non-SCE patients group (4.45 [3–7] vs. 3.45 [2–9], P = 0.039). This has been previously described by our group¹⁴.

We observed an increasing trend in hypercoagulable markers of other endothelial origin in SCE patients compared to non-SCE patients, in particular, TM, PAI-1 and TAFI. Statistical significance was not reached with any of the endothelial parameters.

The main limitation of this study is the relatively small sample size (n = 47), particularly in the subgroup with severe clinical events (n = 11), which may limit the statistical power to detect smaller differences between groups. Although significant differences were observed in several key parameters, the findings should be interpreted with caution, and validation in larger multicenter cohorts is warranted. Nonetheless, a post-hoc power analysis for the primary outcome (ETP inhibition) showed an estimated power of 86.5% to detect the observed difference between groups (Cohen’s d = 1.14, α = 0.05), indicating moderate-to-high statistical power. These preliminary results encourage us to conduct a multicenter study to confirm our data and strengthen the statistical evidence for clinical application. As for the main strength of our study, to our knowledge, this is the first to include TM resistance assessment in hospitalized COVID-19 patients using the Genesia^®.

To sum up, the TG assay evaluates global hemostasis by concomitantly measuring both pro- and antithrombotic markers and may be a valuable tool in determining hypercoagulable status in COVID-19 patients.

Author contributions

AMR and PMV designed the study. AM and IFB performed the statistical analysis. FLJ and IFB contributed to patients’ recruitment. AMR wrote the draft and PMV supervised the manuscript. All authors approved the final version of the manuscript.

Funding

No funding was received for conducting this study.

Data availability

The datasets generated during the current study are available from the corresponding author on reasonable request.

Competing interests

The authors declare no competing interests.