Pattern of Conventional Coagulation and Thromboelastographic Parameters in Patients with COVID-19 Diseases and Association of COVID-Associated Coagulopathy with Mortality in Intensive Care Unit.

Background: Knowledge of underlying pathophysiology of coagulopathy is evolving and the pattern of coagulation parameters in coronavirus disease 2019 (COVID-19)-associated diseases is still not very clear. Aims: In the present study, we aimed to find out the pattern and distribution of conventional coagulation parameters and thromboelastographic (TEG) parameters in COVID-19-associated coagulopathy (CAC) in survivors and nonsurvivors at 28 days. Setting and Design: The present prospective observational study was conducted at a tertiary care COVID-19 intensive care unit (ICU) facility from March 21, 2020, to July 15, 2021. Materials and
Methods: Admission clinical and laboratory data (conventional coagulation, inflammatory and TEG parameters, and disease severity parameters) of 64 COVID-19 patients admitted to the ICU were collected. Patients were divided into two groups, i.e., survivors and nonsurvivors. Statistical Analysis: Data were compared between two groups, i.e., survivors versus no survivors on 28 days using Student's t-test/Mann-Whitney U-test or Chi-square test/Fisher's exact test.
Results: Admission mean plasma fibrinogen levels (474.82 ± 167.41 mg.dL-1) and D-dimer were elevated (1.78 [0.66, 3.62] mg.mL-1) in the COVID-19 ICU patients. Overall, COVID-19 patients had mean lower normal platelet count (150 ± 50 × 103 cells.mm-3), with marginally elevated prothrombin time (16.25 ± 3.76 s) and activated partial thromboplastin time (38.22 ± 16.72 s). A 65.6% (42/64) TEG profile analysis showed a normal coagulation profile, and the rest 21.9% (14/64) and 12.5% (8/64) had hypercoagulable and hypocoagulable states, respectively. Plasma D-dimer level was markedly elevated in nonsurvivors compared to survivors (P < 0.05), while no other conventional coagulation parameters and TEG profile demonstrated statistically significant between the two groups.
Conclusion: Markedly elevated plasma D-dimer level was observed in nonsurvivors of COVID-19 ICU patients. A large portion of COVID-19 ICU patients had a normal TEG profile. Conventional coagulation parameters and TEG profile were similar between survivors and nonsurvivors. Copyright:

INTRODUCTION

Our understanding of the underlying pathobiology of coronavirus disease 2019 (COVID-19) is limited and evolving with time. Present knowledge points toward complex interaction between viral replication in the lower respiratory tract region with an aberrant immunological response.[1] The dysregulation in coagulation parameters in COVID-19 patients, also known as COVID-19–associated coagulopathy (CAC), is often considered as a part of pathobiology and is alleged as a cause of morbidity and mortality in COVID-19–infected patients.[2] CAC is appreciated as more of thrombotic nature in the literature.[2] The spectrum varies from pulmonary embolism to myocardial infarction and ischemic stroke.[3] Two European autopsy studies in patients with COVID-19 infection revealed thrombosis in the segmental and subsegmental pulmonary arterial vasculature along with diffuse alveolar damage.[45] A study showed thrombotic event in patients who had received prophylactic anticoagulation therapy.[5]

The conventional coagulation tests, viz., prothrombin time, activated thromboplastin time (aPTT), and plasma fibrinogen, often lack the sensitivity to detect coagulation anomalies in trauma patients due to static nature.[6] Moreover, tests are mostly done on plasma which is platelet poor; thus, platelet function, clot dynamics, and fibrinolysis cannot be assessed. Thromboelastography (TEG) is a point-of-care test used to assess clot formation and degradation in whole blood. TEG with its ability to quantitatively measure the dynamics of clot formation is considered to shed lights on CAC.[27] It has been widely used to identify bleeding abnormalities in trauma and guide for blood products' transfusion. COVID-19 is associated with increased risk of a hypercoagulable state characterized by increased clot strength and hypofibrinolysis on viscoelastic tests. However, to date, no much information is available on coagulation parameters and TEG studies in COVID-19 infection, especially in sick intensive care unit (ICU) Indian patients.[7] We conducted a prospective observation study with a primary objective if dynamic coagulation study such as TEG parameters and static conventional coagulation parameters changes in COVID-19 ICU patients and their pattern in survivor and nonsurvivor COVID-infected ICU patients.

MATERIALS AND METHODS

This observational study was conducted at a tertiary care referral hospital with a dedicated COVID care facility. Before initiation of the study, ethical approval was taken from institute's ethical board with a waiver of consent (IEC code: 2020-208-IP-EXP-23, PGI/BE/649/2020).

Samples of consecutive patients admitted to the ICU with a diagnosis of COVID-19 infection from August 1, 2020, to July 15, 2021, were included in the study. The admission clinical and laboratory data were retrieved from the hospital informatics system. Admitted patients were categorized as moderate and severe as per the guidelines from World Health Organization (WHO) for COVID-19 management.[7]

On admission, if patients were on any form of oxygen or on mechanical ventilation, intravenous glucocorticoid therapy (6 mg dexamethasone) once daily for 10 days and prophylactic anticoagulation with either subcutaneous enoxaparin 1 mg.kg−1 daily or 5000 units unfractionated heparin subcutaneous 8 h (depending on clinical situation and the absence of contraindications) were initiated.[8] After blood sample collection for TEG and conventional coagulation study, the treating physician would decide on anticoagulation dose escalation (intermediate or therapeutic dose). Within 24 h of admission, peripheral venous blood for complete hemogram including platelet count (impedance method, Ruby CELL DYN, Abbott) and conventional coagulation tests including prothrombin time and aPTT, plasma fibrinogen level (Clauss method), quantitative D-dimer assay (latex enhanced immunoturbidimetric method), and qualitative fibrinogen degradation products was sent. Stago, Satellite Max hemostatic analyzer was used for conventional coagulation tests. The positive cutoff for qualitative fibrinogen degradation products was taken as >2.5 μg.mL−1 as per the manufacturer's instructions. The samples were processed after quality check using two levels (normal and prolonged) for conventional coagulation parameters and three levels (low, normal, and high) for platelet count. The quality control and methodology were the same for all the samples processed.

Along with the samples for routine coagulation assay, 2.7 mL of blood collected in a vacutainer containing 3.2% buffered trisodium citrate anticoagulant was used for TEG assay. The sample for TEG was processed on Thromboelastograph Hemostasis Analyzer System 5000 (Haemonetics, USA) within 2 h of sample collection. The sample was processed in kaolin cups for patients not on any anticoagulant, while it was processed in heparinase cup along with kaolin as an activator for patients on an anticoagulant. The TEG parameters analyzed were reaction time (R), K time (K), α angle, maximum amplitude (MA), coagulation index (CI), and lysis index (LY) 30, where R: time to initial fibrin formation up to 2 mm; K: time from the end of R until a fixed level of clot strength is reached; α: speed of clot formation; MA: measurement of clot strength and assesses the function of the platelets and to some extent fibrinogen; LY 30%: clot dissolution (percentage decrease in amplitude 30 min post-MA); and CI: an index representative of the entire coagulation process. The normal reference values of TEG parameters provided by the manufacturer for kaolin activated citrated samples were used wherein R = 2–8 min, K = 1–3 min, alpha (α) angle = 55°–78°, MA = 51–69 mm, CI −3 to +3, and LY 30% = 0–8. With respect to CI on TEG, the normal coagulation profile was defined as CI of −3 to +3 while hypercoagulability was defined as CI <−3 and hypercoagulability as CI >+3.

The inflammatory markers such as procalcitonin, fibrinogen, lactate dehydrogenase, and C-reactive protein were also sent at the same time. Variables for disease severity scores such as the partial pressure of oxygen in the blood to fraction of inspired oxygen ratio, the sequential organ failure assessment score, and the International Society on Thrombosis and Haemostasis disseminated intravascular coagulation (DIC) were noted. All the patients were followed up till the end point as survivors at 28 days or mortality in hospital.

Statistical analysis

Continuous variables were presented in mean ± standard deviation if follow normal distribution; otherwise, median (interquartile range) was used when categorical variables were presented in number (%). Independent samples t-test was used to compare the means, Mann–Whitney U-test was used to compare the medians, whereas Chi-square test/Fisher's exact test to compare the proportion between two groups. P < 0.05 was considered statistically significant. Statistical Package for the Social Sciences, version-23 (SPSS-23, IBM, Chicago, USA), and MedCalc Statistical Software version 20.009 (MedCalc-20.009, Ostend, Belgium) were used for data analysis.

RESULTS

A total of 64 COVID-positive ICU patients were included in the study. Out of 64 patients, 57 (89%) were under severe COVID illness and 11% (7 patients) were under moderate COVID category as per the WHO definition. Out of the total patients, 35 (54.7%) patients died. The mean and median age of the patients was comparable (51.91 and 52 years with a range of 24–86 years) between survivors and nonsurvivors. The majority of the study patients were males (n = 46, 71.9%). There was a significant difference in the distribution of age (P = 0.003) between survivors and nonsurvivors. However, other distributions such as male gender, duration of hospital stays, day of sample collection, and the onset of symptom to ICU admission day (each P > 0.05) between survivors and nonsurvivors were comparable. Diabetes (45.3%) and hypertension (43.8) were the most common comorbidities followed by coronary artery disease (12.5%), chronic kidney disease (9.4%), and malignancy and immunological disease (6.3%) such as rheumatoid arthritis, systemic lupus erythematosus, and ANCA-related vasculitis. The patient's baseline profile, severity scores, and inflammatory parameters are shown in Table 1. The conventional coagulation and TEG findings are shown in Table 2. The conventional coagulation and inflammatory parameters, TEG findings, and D-dimer with respect to outcome are shown in Figure 1. The level of D-dimer was elevated significantly (P = 0.006) in the nonsurvivor group. The prothrombin time and partial thromboplastin time were on a marginally higher level in nonsurvivors, but no significant difference was noted between the groups. However, the conventional inflammatory markers were elevated with a more pronounced elevation in nonsurvivors. Statistically significant high values for ferritin and lactate dehydrogenase were noted in nonsurvivors (P < 0.001).

Table 1

The patient’s baseline profile, inflammatory parameters and disease severity scores are shown

Baseline characteristics of COVID-19 patients
Variables	Total (n=64)	Survivor (n=29, 45.3%)	Non-survivor (n=35, 54.7%)	P
Age (years)	51.91±15.91	45.62±12.14	57.11±16.92	0.003
Male sex n(%)	2.5 : 1	20 (69)	26 (74.3)	0.64
Duration of hospital stay (days)	15 (8,33)	14 (8, 33)	16 (8, 23)	0.44
Duration from onset of symptom to ICU admission, days (Median, IQR)	8 (4-16)	8 (5-15)	9 (4-16)	0.49
Co-morbidity n (%) (yes)	55 (85.9)	22 (75.9)	33 (94.3)	0.06
Conventional inflammatory markers
Ferritin (ng/ml)	1098 (514, 017)	812 (213, 1256)	2528 (783, 6760)	<0.001
Procalcitonin (ng/ml)	0.21 (0.09, 1.11)	0.13 (0.06, 0.46)	0.25 (0.10,1.85)	0.03
C -reactive protein (mg/L)	71.94±73.29	56.12±62.65	85.84±79.87	0.11
Lactate dehydrogenase (U/L)	558.34±394.22	345.9±118.51	734.37±453.88	0.001
Disease severity score
P/F ratio	117 (70-230)	130 (82-230)	100 (70-186)	0.08
SOFA	4 (1-8)	4 (1-6)	5 (1-8)	0.44
DIC score	2 (0-3)	2 (0-2)	2 (0-3)	0.56

Data presented in Mean±SD, Independent samples t-test used. #Median (IQR) compared by Mann Whitney U test. $Frequency (%) compared by Chi-square test. P<0.05 is significant. IQR=Interquartile range, SD=Standard deviation, ICU=Intensive care unit, SOFA=Sequential organ failure assessment, DIC=Disseminated intravascular coagulation

Table 2

The conventional coagulation and thromboelastography findings of patients are shown

Coagulation and thromboelastography findings
Conventional coagulation parameters
Variables	Total (n=64)	Survivor (n=29, 45.3%)	Non-survivor (n=35, 54.7%)	P
Platelet count (x103 cells/mm3)	150±50	158 (108, 249)	145 (72, 179)	0.16
Prothrombin time (seconds)	16.25±3.76	15.29±3.92	17.06±3.46	0.06
Activated partial thromboplastin time (seconds)	38.22±16.72	37.81±19.64	38.56±14.09	0.86
Fibrinogen (mg/dl)	474.82±167.41	487.71±144.86	464.21±185.42	0.59
Qualitative Fibrinogen degradation products (positive)	55 (85.9%)	23 (79.3%)	32 (91.4%)	0.16
D-dimer (microg/ml)	1.78 (0.66, 3.62)	1.24 (0.49, 2.08)	2.71 (1.02, 5.70)	0.006
Thromboelastography parameters
R time (minutes)	6.31±2.6	6.39±3.1	6.23±2.15	0.81
K time (minutes)	1.72±0.86	1.79±1.05	1.67±0.69	0.59
Alpha Angle (degree)	65.49±9.79	63.86±11.29	66.84±8.28	0.23
Maximum Amplitude (mm)	67.76±8.53	67.76±8.51	67.76±8.52	0.78
Coagulation index	1.05(-0.9, 2.53)	0.9 (-1, 2.50)	1.10 (-1, 2.90)	0.79
Lysis at (30 minutes)	0 (0, 0.7)	0 (0, 1.43)	0 (0, 0.1)	0.16
G	11.56±4.60	11.59±4.17	11.53±4.97	0.76

Positive cut off for qualitative fibrinogen degradation products >2.5 microg/ml. P<0.05 is significant

Figure 1

Distribution of thromboelastographic variables and D-dimer as per patients’ outcomes due to COVID-19

In TEG analysis, we found a mostly normal coagulation profile of 42/64 (65.6%). The lysis time at 30 min was 0 (0–0.7) while the MA was on the higher side of the normal range (67.76 ± 8.51). The rest of the parameters were within range. The TEG findings of the patients with respect to outcome are presented in Table 3.

Table 3

The thromboelastography findings of patients with respective to outcome is shown

Thromboelastography profile
Variables	Total (n=64)	Survivor (n=29, 45.3%)	Non-survivor (n=35, 54.7%)	P
Normal (%)	42 (65.6)	19 (65.5)	23 (65.7)	0.99
Hypercoagulable (%)	14 (21.9)	6 (20.7)	8 (22.9)	0.83
Hypocoagulable (%)	8 (12.5)	4 (13.8)	4 (11.4)	0.77

DISCUSSION

To the best of our knowledge, this is the first large single-center Indian prospective observational study to see the dysregulation of coagulation in patients with COVID-19 admitted to the ICU based on TEG and conventional coagulation parameters. Finding of hypercoagulable state in COVID-19 was nearly a dictum as per earlier studies.[379] In our study of 64 patients, we noted a 65.6% normal profile in TEG analysis. Noteworthy hypercoagulable state findings, we observed were lower lysis time at 30 min (indicating a reduction or complete shutdown in fibrinolysis) and a higher normal range of MA (indicating high clot strength). Being a referral center, our center received critically ill patients (mostly severe or critical COVID-19 diseases). Most of the patients were in their 2nd week of illness with comorbidities (85.9%), so we expect some involvement of coagulation and fibrinolysis system as disease progress in 2nd week. Most of the patients were on uniform dose prophylactic anticoagulation when the TEG sample was taken. Thus, by this stage of disease, one-time TEG and static coagulation parameters should have shown some coagulation derangement on analysis.

In survival data, we found statistically significant high values for ferritin, procalcitonin, and lactate dehydrogenase in nonsurvivors. This is similar to other studies seen by other authors.[710] The level of D-dimer was elevated significantly (P = 0.006) in nonsurvivor group. The high levels of D-dimer may be due to extensive liberation of plasminogen activators by COVID-19 viral inflammation-mediated endothelial cell dysfunction.[11] Pooled analysis study showed that D-dimer values are significantly higher in severe COVID-19 patients.[12] Raised D-dimer has been found to be of bad prognostic value in several studies.[131415] It has the advantage of being sensitive in detecting coagulation dysregulation, but it lacks specificity.[131516] Higher D-dimer levels along with normal platelet count in a study pointed toward the higher thromboembolic events tendencies as compared to bleeding events in COVID-19 infection.[17] In our study, the platelet count was found to be within range pointing toward thrombotic tendencies among nonsurvivors. Fibrinogen level in our study was elevated. As fibrinogen is a coagulation factor and is an acute-phase reactant, the increase in fibrinogen level can be explained by the marked inflammation associated with COVID-19 infection.[1317] Similar findings were obtained in the studies by Huang et al. and Guan et al.; furthermore, they found severe thrombocytopenia in less than 5% of cases.[1317]

DIC as seen in bacterial sepsis is triggered by pathogen-associated molecular patterns and host-derived damage-associated molecular patterns.[1118] This leads to a thrombocytopenia with an increase in prothrombin time ultimately leading to the fibrinolytic shutdown. The postulated mechanism is pronounced inflammation leading to direct and indirect endothelial dysfunction and release of procoagulant changes.[31415161718] Initially, it is localized to the lung; however, as the disease advances, the hypercoagulability becomes systemic. In CAC, there is a lower degree of thrombocytopenia, while there is a marked elevation in the D-dimer levels. The prothrombin and aPTT remain within normal range (as in our study). Levi and Iba in their study had similar findings and attributed it to accelerated thrombin formation due to the presence of circulating activated clotting factors such as thrombin and factor Xa thrombin.[11] The overall rate of thrombotic events is higher than bleeding events in CAC as compared to classic DIC.[19] On the contrary, some authors have attributed the prolongation in aPTT due to lupus-like anticoagulant which was found in about 91% of cases with severe acute respiratory distress syndrome.[20]

A review of the published literature revealed venous thromboembolism to various sites leading to pulmonary embolism, myocardial infarction, deep vein thrombosis, and ischemic stroke in COVID-19–infected patients.[3] However, no data were available regarding venous thromboembolic events in patients in the present study. The overall rate of thrombotic events is higher than bleeding events in CAC as compared to classic DIC. Meta-analysis by Boonyawat et al. and Fontelo et al. found the incidence of venous thromboembolic events to be as high as 28% in severe COVID-19 infection.[2122] On the other hand, very few articles are available about the incidence of bleeding, and a systemic review was done by Beyrouti et al. which included 12 cohort studies (n = 63,390 patients with COVID-19 infection) and found the incidence of intracranial hemorrhage to be 0.38%.[23] Musoke et al. attributed the increased risk of bleeding with the use of therapeutic anticoagulation.[24]

Limitations

We did not perform serial TEG analysis and conventional coagulation study over a period of time to look into the dynamic nature of coagulation and fibrinolysis process in COVID-19 disease. This is the major limitation of the study. However, we received all the patients as tertiary referral COVID-19 treatment centers from smaller hospitals and clinics and patients presented frequently at days 4–16 of illness with presumed some involvement of coagulation system. Hence, admission coagulation study and TEG could also have depicted coagulation system involvement. This is a single-center observational study with a small sample size. All patients were on prophylactic anticoagulation when the blood sample was collected for study. Thus, exact coagulation derangement by COVID-19 disease could not be assessed. All patients received an almost homogeneous prophylactic dose of heparin or enoxaparin. Furthermore, no data were collected regarding thromboembolic or hemorrhagic events. The interleukin-6 level is an important inflammatory marker in the body and is found to be elevated in COVID-19. However, it was not done in the present study.

CONCLUSION

Elevated plasma D-dimer level was observed in nonsurvivors COVID-19 ICU patients. A large portion of COVID-19 ICU patients had a normal TEG profile. Conventional coagulation parameters and TEG profile were statistically similar between survivor and nonsurvivor COVID-19 ICU patients.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.