Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Nov 1.
Published in final edited form as: J Trauma Acute Care Surg. 2022 May 21;93(5):604–612. doi: 10.1097/TA.0000000000003655

Post-Injury Platelet Aggregation and Venous Thromboembolism

Zachary A Matthay 1, Zane J Hellmann 2, Brenda Nunez-Garcia 1, Alexander T Fields 1, Joseph Cuschieri 1, Matthew D Neal 3, Jeffrey S Berger 4, Elliot Luttrell-Williams 4, M Margaret Knudson 1, Mitchell J Cohen 5, Rachael A Callcut 6, Lucy Z Kornblith 1
PMCID: PMC9585095  NIHMSID: NIHMS1798361  PMID: 35444156

Abstract

Introduction:

Posttraumatic venous thromboembolism (VTE) remains prevalent in severely injured patients despite chemoprophylaxis. Importantly, although platelets are central to thrombosis, they are not routinely targeted in prevention of posttraumatic VTE. Further, platelets from injured patients show ex-vivo evidence of increased activation yet impaired aggregation, consistent with functional exhaustion. However, the relationship of this platelet functional phenotype with development of posttraumatic VTE is unknown. We hypothesized that following injury impaired ex-vivo platelet aggregation (PA) is associated with the development of posttraumatic VTE.

Methods:

We performed a secondary analysis of 133 severely injured patients from a prospective observational study investigating coagulation and inflammation (2011-2019). PA in response to stimulation with adenosine diphosphate (ADP), collagen, and thrombin was measured at presentation (pre-resuscitation) and 24h (post-resuscitation). Viscoelastic clot strength and lysis were measured in parallel by thromboelastography. Multivariable regression examined relationships between PA at presentation, 24h, and the delta between presentation and 24h with development of VTE.

Results:

The 133 patients were severely injured (median injury severity score 25) and 14% developed VTE (all >48 hours after admission). At presentation, platelet count and PA were not significantly different between those with and without incident VTE. However, at 24h, those who subsequently developed VTE had significantly lower platelet counts (126 x109/L vs 164 x109/L, p=0.01), and lower PA in response to ADP (p<0.05), collagen (p<0.05), and thrombin (p=0.06). Importantly, the magnitude of decrease in PA (Delta) from presentation to 24h was independently associated with development of VTE (adjusted odds ratios per 10 aggregation unit decrease: Delta-ADP 1.31, p=0.03; Delta-collagen 1.36, p=0.01, Delta-thrombin 1.41, p<0.01).

Conclusion:

Severely injured patients with decreasing ex-vivo measures of platelet aggregation despite resuscitation have an increased risk of developing VTE. This may have implications for predicting development of VTE and for studying platelet targeted chemoprophylaxis regimens.

Level of evidence:

level III, prognostic

Keywords: Platelets, platelet aggregation, platelet function tests, blood coagulation disorders, trauma, venous thromboembolism, deep vein thrombosis, pulmonary embolism

Background:

Posttraumatic venous thromboembolism (VTE) remains an important cause of morbidity and mortality despite use of chemoprophylaxis (1, 2). Deep vein thrombosis (DVT) can be detected by screening ultrasound in over a quarter of critically injured patients, and an estimated 12% of trauma deaths are attributable to fatal pulmonary embolism (PE) (2, 3). Clinical risk factors for the development of posttraumatic VTE have been well described (1, 3, 4), but our understanding of the specific changes in coagulation biology associated with posttraumatic VTE is incomplete, and as such routine standard-of-care in posttraumatic VTE prevention remains primarily limited to heparin-based chemoprophylaxis.

Severe tissue injury and hemorrhagic shock are known drivers of trauma-induced coagulopathy (TIC), encompassing a range of coagulation derangements in the platelet, endothelial, and fibrinolytic systems associated with both early hypocoagulability and subsequent hypercoagulability, thrombo-inflammatory complications, and mortality (5-9). Ex-vivo evidence of increased platelet activation yet impaired platelet aggregation is a principal finding in TIC, independent of platelet count (10-12). This is commonly identified after injury, but to what degree this reflects pathologically impaired in-vivo platelet function remains unknown (5, 9-12). It is hypothesized that this seemingly discordant biology of increased platelet activation, but impaired aggregation may represent a functionally exhausted platelet phenotype, in which the circulating platelets sampled were activated at sites of tissue injury, have degranulated, and are therefore unable to respond further to stimulation in ex-vivo assays (13, 14). Ex-vivo impairments in platelet aggregation are known to be independently associated with microvascular thrombo-inflammatory complications and mortality after severe injury (10, 11). However, whether these impairments in platelet aggregation are also associated with macrovascular thrombotic complications, including the development of posttraumatic VTE is not well understood (15). Therefore, investigating how impaired ex-vivo platelet aggregation is associated with the development of posttraumatic VTE is of particular interest given the central role platelets play in thrombosis, and the lack of routine platelet targeted chemoprophylaxis regimens (16, 17).

In this study, we examined the relationship of ex-vivo platelet aggregation immediately after injury on presentation (pre-resuscitation), at 24 hours after injury (post-resuscitation), and the change in platelet aggregation from presentation to 24 hours after injury with subsequent development of posttraumatic VTE in a cohort of severely injured trauma patients admitted to the intensive care unit (ICU). We hypothesized that following injury, impaired ex-vivo platelet aggregation (PA) is associated with the development of posttraumatic VTE.

Methods:

Patient Enrollment

Whole blood was collected in sodium citrate (3.2%) from 1,024 injured patients on arrival to the emergency department as part of a longitudinal prospective observational study examining coagulation and inflammation after injury at a Level 1 Trauma Center (2010-2019) (11, 13, 18-20). Patients who required the highest-level trauma activation were included, unless in-custody, pregnant, pediatric, suffering from >20% body surface area burns, transferred from another facility, did not require ICU level care, or were atraumatic. For the purposes of this post-hoc analysis, we further limited our study cohort to the subset of 133 patients who were admitted to the ICU ≥24 hours, had platelet aggregometry performed on presentation and at 24 hours from presentation (to be able to examine platelet aggregation both pre- and post-resuscitation), and who were admitted to the hospital for ≥3 days (to be able to examine those at risk of development of VTE) (Supplemental Figure 1). Patients were additionally excluded if they were confirmed to be taking antiplatelet or anticoagulant agents prior to admission (Supplemental Figure 1). Demographics, injury and physiologic data, and clinical outcomes were prospectively collected in parallel with laboratory assays. Consent was obtained under an initial waiver of consent with subsequent informed consent from all patients, under approval by the Committee on Human Research (IRB #19-28933). This study conforms with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines and a complete checklist has been uploaded as Supplemental Digital Content (Supplemental Table 1).

Identification and Definition of Venous Thromboembolism

Venous thromboembolism (VTE) was defined as development of radiologically confirmed pulmonary embolism (PE) or deep vein thrombosis (DVT, upper or lower extremity) diagnosed after admission. No screening protocols were used and all VTEs were considered symptomatic in this study. Chemoprophylaxis and treatment was determined by the treating clinician. We excluded isolated subsegmental pulmonary clots identified on presentation imaging from our diagnosis of pulmonary embolism, as these have been shown to have different pathophysiology and clinical implications as compared to the development of posttraumatic VTEs after presentation (21, 22).

Platelet aggregometry

Platelet aggregation (PA) was measured ex-vivo by multiple electrode platelet aggregometry (Multiplate®, Verum Diagnostica GmbH; Munich, Germany) immediately after sample collection as described previously (11, 23). Briefly, whole blood was diluted with 3mM CaCl2 in warmed normal saline, incubated at 37°C and stirred for 3 minutes in a Multiplate® test cell containing two copper electrodes. Platelet aggregation was stimulated by the surface receptor stimulants adenosine diphosphate (ADP; final concentration 6.5μM; via P2 receptors), thrombin receptor activating peptide-6 (thrombin; final concentration 32μM; via PAR receptors), and collagen (final concentration 3.2μg/mL; via GpIa/IIa and GpVI receptors). Platelet aggregation was quantified in aggregation units (AUC) by measuring the average of the change in electrical impedance at the two electrodes in each test cell over a 6-minute interval. The manufacturer provided reference ranges for the 6-minute aggregation in response to platelet stimulation by each stimulant are: ADP (36-101U), thrombin (75-137U), and collagen (24-79U). Platelet aggregation was measured on presentation and at 24 hours for all patients included in this study. The delta in platelet aggregation from presentation to 24 hours was calculated by subtracting the platelet aggregation in response to each stimulant at 24 hours from the platelet aggregation response to each stimulant at presentation.

Viscoelastic Assays

Viscoelastic testing was performed using the TEG-5000 Thrombelastography Hemostasis Analyzer according to manufacturer instructions (24) given the central role of platelet function in clot formation and breakdown. Briefly, on presentation (pre-resuscitation) and at 24 hours (post-resuscitation), citrated kaolin TEG (CK-TEG) was run in parallel with platelet aggregometry described above using a parallel sample of citrated whole blood. CK-TEG values generated include activated clotting time (ACT, min), angle (rate of clot strength increase, degrees), maximum clot strength achieved (MA, millimeters) and percent clot lysis 30 and 60 minutes after reaching MA (LY30 and LY60, %).

Statistical Analysis

Study patient characteristics, laboratory data, and outcomes are presented as mean (± standard deviation), median (interquartile range[IQR]), or percentage; univariate comparisons were made using Student’s t test for normally distributed data, Wilcoxon rank sum or Kruskal-Wallis testing for nonparametric data, and Fisher’s exact test for proportions. Based on previously published data of platelet aggregation in trauma patients (11, 23) (mean of 50 and SD ±20 aggregation units [AUs] in response to ADP), our study had 80% power to detect an effect size of 15 AUs between those with and without VTE with alpha set at 0.05. Bivariate and multivariable logistic regression tested the association of platelet aggregation and viscoelastic measures and clinical risk factors for the development of VTE. The multivariable logistic regression models were built testing risk factors for VTE and known potential confounders in the relationship of platelet aggregation with development of VTE (including platelet count, chemoprophylaxis timing, transfusions, length of stay, shock, and injury severity). The Akaike Information Criterion was used to further assess and optimize model fit on sensitivity analyses to examine the relative importance of the covariates included (25). The odds ratios reflect the estimated effect for each 10 unit change in platelet aggregation units (AUC) to allow for clinically meaningful interpretation of the results, given the known median and interquartile ranges in this population and in prior studies of trauma patients (13, 19, 23). An [alpha] < 0.05 was considered significant. Analyses were carried out in Stata version 15.1 (StataCorp, College Station, TX).

Results

The patients were young with a median age of 38, primarily male (88%), severely injured with an injury severity score of 25 (IQR 14-33), and an in-hospital mortality rate of 12% (Table 1). The majority (69%) were bluntly injured, and 49% suffered a traumatic brain injury (Table 1). Transfusion of blood products was performed in 65% of patients in the initial 24 hours, and 23% received a platelet transfusion (Table 1). Out of the 133 patients, VTE was detected in 18 patients including 8 with PE, 9 with DVT, and one with both (Table 1 and Supplemental Table 2). The median time to development of VTE was 6 days with a range of 2-29 days- 33% occurred in the first 2-5 days, 23% within 5-10 days, and 44% in 10-30 days (Supplemental Figure 2).

Table 1.

Baseline Characteristics of Patients with and without Development of Venous Thromboembolism

No VTE (n=115) Yes VTE (n=18) Total (n=133) Number with
missing data		 P-value
Age (years) 37 (26-52) 44 (29-53) 38 (26-52) 0 0.41
Sex 0 0.27
 Female 12.2% 22.2% 13.5%
 Male 88% 78% 87%
BMI (kg/m2) 26.8 (23.0-30.2) 27.6 (24.6-31.4) 26.8 (23.3-30.6) 3 0.33
Mechanism 0 0.27
 Penetrating 33% 17% 31%
 Blunt 67% 83% 69%
Injury Severity Score 22 (14-30) 29 (27-34) 25 (14-33) 0 <0.01
AIS Head 3 (0-4) 3 (2-5) 3 (0-4) 2 0.30
GCS 13 (7-15) 14 (9-15) 13 (7-15) 1 0.61
Heart Rate 96 (24) 100 (35) 96 (26) 1 0.57
Systolic Blood Pressure (mmhg) 132 (32) 128 (39) 131 (33) 1 0.70
Shock Index 0.9 (0.3) 0.9 (0.4) 0.9 (0.3) 1 0.48
pH 7.3 (7.3-7.4) 7.3 (7.2-7.3) 7.3 (7.3-7.4) 23 0.08
Base Excess (meq/ml) −2.5 (−7.0-0.2) −5.7 (−7.6--2.8) −2.6 (−7.4--0.1) 22 0.11
INR 1.1 (1.0-1.2) 1.1 (1.1-1.2) 1.1 (1.0-1.2) 5 0.48
pTT (seconds) 28 (25-30) 30 (26-34) 28 (25-31) 5 0.07
Platelet count (presentation), x109/L 263 (209-322) 260 (207-299) 263 (209-310) 0 0.59
Platelet count (24 hours), x109/L 164 (127-201) 126 (106-153) 157 (124-193) 0 0.01
Proportion with initiation of chemoprophylaxis 0
 24 hours 15% 0% 13% 0.08
 48 hours 39% 17% 36% 0.07
 72 hours 64% 39% 61% 0.04
 96 hours 78% 44% 73% <0.01
 120 hours 84% 78% 83% 0.51
Length of stay (to 28 days) 16 (9-24) 28 (21-28) 17 (10-25) 0 <0.01
ICU length of stay (to 28 days) 5 (3-11) 14 (7-23) 7 (3-14) 0 <0.01
Ventilator-free days (to 28 days) 25 (20-27) 18 (5-24) 25 (15-27) 6 <0.01
Multiple Rib Fractures (3+) 21% 33% 22% 4 0.24
Hemo- or pneumothorax 35% 50% 37% 4 0.29
Pelvic fracture 17% 28% 19% 4 0.33
Multiple long bone fractures (2+) 12% 39% 16% 4 <0.01
Spinal Cord Injury 15% 17% 16% 4 1.00
Traumatic Brain Injury 48% 56% 49% 0 0.62
Transfused any product (24h) 62% 83% 65% 0 0.11
pRBCs transfused (24h) 2.0 (0.0-4.0) 3.0 (2.0-7.0) 2.0 (0.0-5.0) 0 0.04
Transfused platelets (24h) 22% 29% 23% 0 0.54
Platelet units transfused (24h) 0.0 (0.0-0.0) 0.0 (0.0-1.0) 0.0 (0.0-0.0) 0 0.48
Mortality at discharge 10% 22% 12% 0 0.23
Open in a new tab

VTE- venous thromboembolism; AIS- abbreviated injury score; GCS- Glasgow coma scale; HR- heart rate; SBP- systolic blood pressure; shock index=HR/SBP; INR- international normalized ratio; pTT- partial thromboplastin time; chemoprophylaxis defined as initiation of low-molecular weight or unfractionated heparin; pRBCs=packed red blood cells; ICU-intensive care unit

Patients who developed VTE had significantly higher injury severity scores (29 [14-30] vs 22 [27-34]) longer hospital (28 [21-28] vs 16 days [9-24]) and ICU lengths of stay (14 [7-23] vs 5 [3-11] days), and fewer ventilator-free days (18 [5-24] vs 25 [20-27] days, Table 1, all p<0.05). Presentation platelet counts were not different between those with and without development of VTE (260x109/L [209 x109/L - 322 x109/L] vs 263x109/L [207 x109/L - 299 x109/L), but at 24 hours, those who subsequently developed VTE had significantly lower platelet counts (126 x109/L [106 x109/L - 153 x109/L] vs 164 x109/L [127 x109/L – 201 x109/L], p=0.01, Table 1). A significantly lower proportion of patients who subsequently developed VTE had initiation of VTE chemoprophylaxis by 72 and 96 hours (p<0.05, Table 1). When examining injury mechanisms, there were non-significant trends for increased rates of multiple rib fractures, and hemo- or pneumothorax among those who developed VTE, and a significantly higher proportion of patients with multiple long bone fractures (p<0.01, Table 1).

There were no significant differences in platelet aggregation parameters on presentation between those who subsequently developed VTE and those who did not (Table 2). However, at 24 hours, mean platelet aggregation responses to ADP and collagen were significantly lower in those who developed VTE (19 [12-25] vs 28 [20-40], and 17 [15-25] vs 32 [22-48] aggregation units [AU], both p<0.05), with a similar but non-significant trend for platelet aggregation in response to thrombin (47 [37-75] vs 66 [49-81] AU, p=0.06, Table 2). Given the known wide variability in platelet responsiveness to ex-vivo stimulation across individuals (26), we also examined the change in platelet aggregation for each individual from presentation to 24 hours. We found that the magnitude of the change (delta) in platelet aggregation in response to each of the surface stimulants (ADP, collagen, and thrombin) was significantly larger (more negative) in those who developed VTE (all p<0.05, Table 2). We additionally examined viscoelastic measures of clot formation and breakdown, which demonstrated that those who developed VTE had significantly lower percent clot lysis at 24 hours (p<0.05, Table 2).

Table 2.

Bivariate Comparisons of Platelet Aggregation and Viscoelastic Measures of Patients with and without Development of Venous Thromboembolism

No VTE (n=115) Yes VTE (n=18) Total (n=133) P-value
ADP AUC, presentation 55 (37-70) 53 (47-72) 54 (38-71) 0.31
Collagen AUC, presentation 48 (33-63) 50 (38-68) 49 (34-63) 0.43
Thrombin AUC, presentation 97 (81-115) 87 (80-144) 96 (80-117) 0.70
ADP AUC, 24h 28 (20-40) 19 (12-25) 26 (19-39) <0.01
Collagen AUC, 24h 32 (22-48) 17 (15-25) 29 (18-46) <0.01
Thrombin AUC, 24h 66 (49-81) 46 (37-75) 66 (46-80) 0.06
Delta ADP AUC, presentation to 24h 19 (7-36) 29 (17-58) 21 (9-40) 0.03
Delta Collagen AUC, presentation to 24h 13 (−1-25) 30 (14-53) 16 (0-26) <0.01
Delta Thrombin AUC, presentation to 24h 31 (11-49) 43 (34-90) 34 (15-51) <0.01
Clot strength (CK-MA), presentation 65.2 (61.8-68.4) 64.1 (60.4-67.9) 65.2 (61.8-68.3) 0.52
Clot lysis (CK-LY30,%), presentation 0.5 (0.1-2.0) 0.2 (0.0-1.6) 0.5 (0.0-1.9) 0.31
Clot strength (CK-MA), 24h 64.5 (61.2-67.1) 65.2 (61.4-68.4) 64.5 (61.2-67.2) 0.78
Clot lysis (CK-LY30,%), 24h 0.9 (0.3-1.8) 0.2 (0.0-1.0) 0.8 (0.2-1.8) 0.02
Open in a new tab

ADP AUC- adenosine diphosphate stimulated platelet aggregation, Collagen-collagen stimulated platelet aggregation, Thrombin AUC- thrombin stimulated platelet aggregation. Delta- Change platelet aggregation from presentation to 24 h.

We then performed multivariable logistic regression to further characterize the association of platelet aggregation with the development of VTE, controlling for key potential confounders. On presentation, there was a weak association of increasing platelet aggregation with increased odds of development of VTE, reaching statistical significance for platelet aggregation in response to thrombin (p=0.05), and similar but non-significant trends for platelet aggregation in response to ADP and collagen (Figure 1 and Supplemental Table 3). There were also no significant associations of either viscoelastic clot strength or clot lysis on presentation with subsequent development of VTE (Figure 2 and Supplemental Table 3).

Figure 1.

Figure 1.

Open in a new tab

Coefficient Plot of Association of Platelet Aggregation on Presentation, at 24 hours, and Change from 0 to 24 hours with Development of Posttraumatic Venous Thromboembolism (VTE). X axis is the odds ratio per 10 aggregation units, adjusted for length of stay, initiation of VTE chemoprophylaxis within 72 hours, platelet count, injury severity score, shock index, and whether patient was transfused in initial 24 hours.

Figure 2.

Figure 2.

Open in a new tab

Coefficient Plot of Association of Viscoelastic Clot Strength and Lysis at presentation (0 hours) and at 24 hours with Development of Posttraumatic VTE. X axis is the odds ratio per 10 aggregation units, adjusted for length of stay, initiation of chemoprophylaxis with in 72 hours, injury severity score, shock index, and whether patients were transfused in initial 24 hours.

At 24 hours however, there were some, although not all, significant associations of platelet aggregation and clot lysis with development of VTE. While there was a significant trend for the association of impaired ex-vivo platelet aggregation in response to collagen with development of VTE, there was no significant association with platelet aggregation in response to ADP or thrombin (Figure 1 and Table 3). Importantly, the relationship of development of VTE with platelet aggregation was most pronounced when examining the change (delta) in platelet aggregation from presentation to 24 hours. The larger the magnitude of decrease in platelet aggregation in response to ADP, collagen, and thrombin stimulation, the higher the odds of development of VTE were (OR 1.3-1.4 per 10u decrease in AUC, all p<0.05, Figure 1 and Table 3). Further, impaired clot lysis at 24 hours was also an important risk factor for the development of VTE with an adjusted odds ratio of 2.5 for every 1% decrease in clot lysis (p<0.05, Figure 2 and Table 3).

Table 3.

Multivariable Logistic Regression Results for Association of Platelet Aggregation and TEG values at 24h with Development of VTE

ADP Stimulated PA, 24hr Odds Ratio CI-Low CI-High P-Value
 ADP PA, (per 10 AUC units) 0.86 0.55 1.34 0.51
 Length of Stay (days) 1.09 1.02 1.17 0.01
 Initiation of chemoprophylaxis within 72 hours 0.27 0.08 0.86 0.03
 Platelet count (24 hours) 0.99 0.98 1.01 0.35
 Injury Severity Score 1.02 0.97 1.07 0.43
 Shock Index 1.46 0.27 7.96 0.66
 Transfused in initial 24 hours 1.21 0.23 6.44 0.82
 Area under ROC curve=0.84
Collagen Stimulated PA, 24h Odds Ratio CI-Low CI-High P-Value
 Collagen PA, (per 10 AUC units) 0.63 0.40 1.00 0.05
 Length of Stay (days) 1.10 1.02 1.17 0.01
 Initiation of chemoprophylaxis within 72 hours 0.40 0.12 1.30 0.13
 Platelet count (24 hours) 1.00 0.98 1.01 0.86
 Injury Severity Score 1.02 0.97 1.08 0.40
 Shock Index 0.55 0.09 3.51 0.52
 Transfused in initial 24 hours 1.55 0.26 9.37 0.63
 Area under ROC curve=0.84
Thrombin stimulated PA, 24 hours Odds Ratio CI-Low CI-High P-Value
 Thrombin PA, (per 10 AUC units) 0.91 0.68 1.23 0.55
 Length of Stay (days) 1.11 1.03 1.19 0.01
 Initiation of chemoprophylaxis within 72 hours 0.25 0.08 0.84 0.02
 Platelet count (24 hours) 0.99 0.98 1.01 0.50
 Injury Severity Score 1.03 0.98 1.09 0.27
 Shock Index 1.59 0.29 8.75 0.59
 Transfused in initial 24 hours 0.99 0.18 5.43 0.99
 Area under ROC curve=0.84
Delta-ADP presentation to 24 hours Odds Ratio CI-Low CI-High P-Value
 Delta ADP, (per 10u) 1.31 1.03 1.65 0.03
 Length of Stay (days) 1.10 1.02 1.18 0.01
 Initiation of chemoprophylaxis within 72 hours 0.45 0.13 1.54 0.20
 Platelet count (24 hours) 0.99 0.97 1.00 0.15
 Injury Severity Score 1.04 0.98 1.09 0.19
 Shock Index 0.80 0.11 5.88 0.83
 Transfused in initial 24 hours 0.89 0.16 5.10 0.90
 Area under ROC curve=0.86
Delta-Collagen presentation to 24 hours Odds Ratio CI-Low CI-High P-Value
 Delta Collagen, (per 10u) 1.36 1.07 1.73 0.01
 Length of Stay (days) 1.10 1.02 1.18 0.01
 Initiation of chemoprophylaxis within 72 hours 0.35 0.10 1.20 0.10
 Platelet count (24 hours) 0.99 0.98 1.01 0.46
 Injury Severity Score 1.04 0.98 1.10 0.21
 Shock Index 0.82 0.10 6.50 0.85
 Transfused in initial 24 hours 1.16 0.20 6.72 0.87
 Area under ROC curve=0.85
Delta-Thrombin, presentation to 24 hours Odds Ratio CI-Low CI-High P-Value
 Delta- Thrombin, (per 10u) 1.41 1.11 1.78 <0.01
 Length of Stay (days) 1.13 1.05 1.22 <0.01
 Initiation of chemoprophylaxis within 72 hours 0.18 0.04 0.72 0.02
 Platelet count (24 hours) 0.99 0.98 1.01 0.23
 Injury Severity Score 1.04 0.97 1.11 0.23
 Shock Index 1.21 0.18 8.11 0.84
 Transfused in initial 24 hours 0.55 0.08 3.67 0.53
 Area under ROC curve=0.88
Clot Strength at 24 hours Odds Ratio CI-Low CI-High P-Value
 Clot strength, (mm) 1.03 0.91 1.18 0.62
 Length of Stay (days) 1.17 1.05 1.30 <0.01
 Initiation of chemoprophylaxis within 72 hours 0.31 0.09 1.07 0.06
 Injury Severity Score 1.03 0.97 1.09 0.31
 Shock Index 1.40 0.24 8.25 0.71
 Transfused in initial 24 hours 2.97 0.49 17.88 0.23
 Area under ROC curve=0.84
Clot Lysis at 24 hours Odds Ratio CI-Low CI-High P-Value
 CK-LY30, (%) 2.49 1.14 5.49 0.02
 Length of Stay (days) 1.20 1.07 1.34 <0.01
 Initiation of chemoprophylaxis within 72 hours 0.33 0.09 1.21 0.09
 Injury Severity Score 1.03 0.97 1.09 0.33
 Shock Index 0.92 0.13 6.64 0.93
 Transfused in initial 24 hours 3.00 0.46 19.75 0.25
 Area under ROC curve=0.88
Open in a new tab

PA-Platelet aggregation in aggregation units (AUC), ADP- adenosine diphosphate, delta-change in platelet aggregation from 0 to 24 hours, CK-LY30-percent clot lysis at 30 minutes, shock index-heart rate/systolic blood pressure on arrival

Discussion

In this study, we examined the associations between post-injury ex-vivo platelet aggregation and risk of development of posttraumatic VTE in a cohort of severely injured trauma patients admitted to the intensive care unit. Our examination identified that the magnitude of decrease in ex-vivo platelet aggregation in the initial 24 hours of hospitalization for severely injured patients is a significant risk factor for development of VTE, independent of platelet count and known VTE clinical risk factors. This finding is of importance, as it may have implications for future studies examining targeted approaches to identifying patients at higher risk for development of VTE who may benefit from earlier initiation of chemoprophylaxis, and/or of the addition of platelet targeted chemoprophylaxis to current heparin-based regimens. Further, the use of the change (delta) in platelet aggregation may reduce challenges of the known wide variation in responses to ex-vivo assays of platelet aggregation across individuals and assist with a more personalized approach to assessing platelet function.

Evidence for the role of platelets in the pathobiology of VTE has been explored across disciplines (27-32), though fewer studies have examined this specifically in trauma patients (15, 33). Increased understanding of both the hemostatic and immune properties of platelets suggests that across diseases, damage associated molecular patterns triggers platelet-monocyte and platelet-neutrophil interactions leading to pro-inflammatory and hypercoagulable states, promoting endothelial injury and thrombotic complications (28, 31, 34, 35). These platelet-immune cell interactions cause release of cytokines, platelet microparticles, as well as increased tissue factor and thrombin production (31, 33). Procoagulant platelet microparticle release may be an important mechanism by which activated platelets after injury mediate the risk of hypercoagulable complications based on human and murine studies (33). Platelets also have important effects on the downregulation of fibrinolysis via release of plasminogen activator inhibitor-1 and alpha antiplasmin-2, thereby regulating clot breakdown (36, 37).

However, the changes in platelet hemostatic functions in the setting of TIC are heterogenous and dynamic with evidence that tissue injury may precipitate platelet activation and hypercoagulability, while hypoperfusion may trigger release of soluble mediators that may inhibit platelets or cause cytoskeletal and structural defects (23, 38-42). Understanding the true in-vivo biology of platelets after injury is further complicated by limitations of ex-vivo assays of platelet function and divergent findings of increased platelet activation yet impaired aggregation in ex-vivo assays from trauma patients (10, 13, 43). Important to our findings, impaired ex-vivo platelet aggregation in the setting of injury may not necessarily equate to a deficit of platelet hemostatic function in-vivo- but rather a functionally exhausted phenotype in which the sampled platelets have already been activated and degranulated at local sites of injury in-vivo, and therefore have reduced capacity to further aggregate when stimulated ex-vivo. We hypothesize from the results of our study that the decrease in ex-vivo platelet aggregation from presentation to 24 hours reflects a functionally exhausted platelet phenotype that is part of a hypercoagulable state in-vivo contributing to subsequent risk of development of VTE. Patients with the largest decrease in ex-vivo platelet aggregation following resuscitation may be those who experienced robust platelet activation and associated degranulation, leading to a population of circulating yet functionally exhausted platelets. This hypothesis is supported by prior studies of ex-vivo platelet function in trauma patients (10, 13), including one prior study based on an analysis of the Pragmatic, Randomized, Optimal Platelet and Plasma Ratios (PROPPR) study database which identified similar post-resuscitation deficits in ex-vivo platelet aggregation in trauma patients who developed VTE compared to those who did not (15). However, further studies are needed to test this hypothesis, given the lack of direct data examining platelet surface receptor markers of activation or platelet microparticle plasma profiles in this cohort.

Importantly, we also found a significant independent association between decreased viscoelastic clot lysis at 24 hours and increased risk of development of VTE, but not for overall clot strength nor for clot lysis on presentation. Under our hypothesis that patients with severe injury experience robust platelet activation, degranulation, and downstream functional exhaustion, it is plausible that platelet degranulation releasing plasminogen activator inhibitor-1 and alpha antiplasmin-2 participates in the downregulation of fibrinolysis. Several studies have detected a range of aberrations in fibrinolysis pathways after major trauma, with important associations with mortality for both patient with hyperfibrinolysis as well as those with impaired fibrinolysis or fibrinolytic shutdown (44, 45). To our knowledge, few studies have examined the relationship of impaired fibrinolysis with posttraumatic VTE development, though the Consortium of Leaders in the study of Traumatic Thromboembolism (CLOTT) Study investigators have emerging data examining impaired fibrinolysis as a significant risk factor for the development of posttraumatic VTE (21). In a study by Moore and colleagues, no association was identified between impaired fibrinolysis on presentation and development of posttraumatic VTE development, though data at 24 hours were not examined (46). Overall, because platelets are essential regulators of fibrinolysis, further examinations of the role platelets play in impaired fibrinolysis and associated worse outcomes after injury including the development of VTE is warranted (46, 47).

Limitations and future directions

This analysis has important limitations to consider, including its retrospective nature and limited sample size. The detection and diagnosis of VTE in this study was not strictly protocolized, but rather determined by clinical judgement. Therefore, our VTE rate of 14% is likely an underestimate, given prior studies using screening imaging have shown VTE in over a quarter of critically injured patients (2). The patients included in this study were composed from a cohort of severely injured patients admitted to the ICU for at least 24 hours and to the hospital for at least 3 days, and therefore the results may not be generalizable to a broader population of trauma patients, including those who expire early and those have relatively minor injuries and shorter length of stay. However, we chose to focus on this subgroup given they are at highest risk for the development of posttraumatic VTE. Although few of the patients diagnosed with PE in this study were known to have concurrent DVT, routine screening LE duplex was not performed in patients who had a diagnosis of PE. Recent data does show that some pulmonary clots in trauma patients may represent in situ thromboses, rather than embolic events, and that these have different pathophysiology than PE (21). To account for this, subsegmental pulmonary clots present on admission imaging were not included in our definition of VTE. To control for potential confounders in our multivariable regression models, we adjusted for multiple factors that influence platelet aggregation such as platelet count, injury severity, and shock that may also be associated with the development of VTE. Despite this, it is possible there is unmeasured confounding in our analysis. Furthermore, while we controlled for time to initiation of chemoprophylaxis in our models, we were not able to account for missed doses or variations in dosing and frequency. Finally, the associations of platelet aggregation with development of VTE in our study appear uniform across the surface receptor pathways tested, supporting a global or upstream mechanism mediating the relationship of post-injury platelet biology with the development of posttraumatic VTE. We hypothesize this association is due to an exhausted platelet phenotype, in which impaired ex-vivo aggregation is reflective of excessive in-vivo platelet activation and aggregation triggered by prothrombotic mediators released in the setting of extensive tissue injury (13, 38). In this setting, platelets that have been excessively activated and stimulated to degranulate in vivo may no longer capable of normal aggregation ex-vivo (14). However, this hypothesis of functional exhaustion requires further investigation including analysis of surface receptor markers of platelet activation and of platelet degranulation to provide more supporting evidence. Additionally, further study characterizing the downstream effects of altered platelet function on fibrinolysis are also warranted as this may be an important pathway through which platelets promote hypercoagulability and increased risk of posttraumatic VTE, given platelet activation can include the degranulation of platelet derived plasminogen activator inhibitor-1 and alpha antiplasmin-2.

In conclusion, this study demonstrated strong independent associations between impairments in platelet aggregation and clot lysis at 24 hours after presentation with an increased risk of development of posttraumatic VTE. We hypothesize that it is possible this reflects a functionally exhausted platelet phenotype, and believe that future studies should examine biologic markers of platelet activation and degranulation in parallel with ex-vivo aggregation assays to further characterize the platelet biology associated with the development of posttraumatic VTE. This is an important line of investigation, as these data could be used to further test whether a targeted approach to adding platelet-based chemoprophylaxis to standard-of-care heparin-based chemoprophylaxis is safe and beneficial in patients at high risk of development of posttraumatic VTE.

Supplementary Material

Supplemental Data File (.doc, .tif, pdf, etc.)

Supplemental Figure 1. Flowchart specifying inclusion and exclusion criteria. ICU-intensive care unit. LOS-length of stay.

Supplemental Figure 2. Distribution of the timing of venous thromboembolic events (VTE) in days from hospital admission.

Media Summary:

Altered platelet aggregation is common following injury, but the relationship with venous thromboembolism is unknown. We identify that the decrease in platelet aggregation from presentation to 24 hours post-resuscitation is independently associated with development of posttraumatic venous thromboembolism.

Acknowledgements:

The authors would like to thank Sue Peterson, Lilian Li, John Park, and Erin Ross for their assistance with the study. We would also like to thank our patients and their loved ones for their generous support of our research.

Disclosure information:

Dr. L Kornblith is supported by NIH 1K23GM130892-01

Footnotes

Disclosure outside the scope of this work: none

Conflicts of interest: The authors report no relevant conflicts of interest

Meeting presentation: This study was presented at the 52nd Annual Meeting of the Western Trauma Association held February 23rd to February 28thth, 2022 in Big Sky, Montana.

References

  • 1.Rogers FB. Venous thromboembolism in trauma patients: a review. Surgery. 2001;130(1):1–12. [DOI] [PubMed] [Google Scholar]
  • 2.Yumoto T, Naito H, Yamakawa Y, Iida A, Tsukahara K, Nakao A. Venous thromboembolism in major trauma patients: a single-center retrospective cohort study of the epidemiology and utility of D-dimer for screening. Acute Med Surg. 2017;4(4):394–400. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Ho KM, Burrell M, Rao S, Baker R. Incidence and risk factors for fatal pulmonary embolism after major trauma: a nested cohort study. Br J Anaesth. 2010;105(5):596–602. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Knudson MM, Ikossi DG, Khaw L, Morabito D, Speetzen LS. Thromboembolism after trauma: an analysis of 1602 episodes from the American College of Surgeons National Trauma Data Bank. Ann Surg. 2004;240(3):490–6; discussion 6-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Moore EE, Moore HB, Kornblith LZ, Neal MD, Hoffman M, Mutch NJ, et al. Trauma-induced coagulopathy. Nat Rev Dis Primers. 2021;7(1):30. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Kornblith LZ, Moore HB, Cohen MJ. Trauma-induced coagulopathy: The past, present, and future. J Thromb Haemost. 2019;17(6):852–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Johansson PI, Stensballe J, Rasmussen LS, Ostrowski SR. A high admission syndecan-1 level, a marker of endothelial glycocalyx degradation, is associated with inflammation, protein C depletion, fibrinolysis, and increased mortality in trauma patients. Ann Surg. 2011;254(2):194–200. [DOI] [PubMed] [Google Scholar]
  • 8.Johansson PI, Stensballe J, Rasmussen LS, Ostrowski SR. High circulating adrenaline levels at admission predict increased mortality after trauma. J Trauma Acute Care Surg. 2012;72(2):428–36. [DOI] [PubMed] [Google Scholar]
  • 9.Chang R, Cardenas JC, Wade CE, Holcomb JB. Advances in the understanding of trauma-induced coagulopathy. Blood. 2016;128(8):1043–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Jacoby RC, Owings JT, Holmes J, Battistella FD, Gosselin RC, Paglieroni TG. Platelet activation and function after trauma. J Trauma Acute Care Surg. 2001;51(4):639–47. [DOI] [PubMed] [Google Scholar]
  • 11.Kutcher ME, Redick BJ, McCreery RC, Crane IM, Greenberg MD, Cachola LM, et al. Characterization of platelet dysfunction after trauma. J Trauma Acute Care Surg. 2012;73(1):13. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Solomon C, Traintinger S, Ziegler B, Hanke A, Rahe-Meyer N, Voelckel W, et al. Platelet function following trauma. Thromb Hemost. 2011;106(08):322–30. [DOI] [PubMed] [Google Scholar]
  • 13.Starr NE, Matthay ZA, Fields AT, Nunez-Garcia B, Callcut RA, Cohen MJ, et al. Identification of Injury and Shock Driven Effects on Ex-Vivo Platelet Aggregometry: A Cautionary Tale of Phenotyping. J Trauma Acute Care Surg. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Pareti FI, Capitanio A, Mannucci L, Ponticelli C, Mannucci PM. Acquired dysfunction due to the circulation of "exhausted" platelets. Am J Med. 1980;69(2):235–40. [DOI] [PubMed] [Google Scholar]
  • 15.McCully BH, Connelly CR, Fair KA, Holcomb JB, Fox EE, Wade CE, et al. Onset of Coagulation Function Recovery Is Delayed in Severely Injured Trauma Patients with Venous Thromboembolism. J Am Coll Surg. 2017;225(1):42–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Hoffman M, Monroe DM 3rd. A cell-based model of hemostasis. Thromb Haemost. 2001;85(6):958–65. [PubMed] [Google Scholar]
  • 17.Ley EJ, Brown CVR, Moore EE, Sava JA, Peck K, Ciesla DJ, et al. Updated guidelines to reduce venous thromboembolism in trauma patients: A Western Trauma Association critical decisions algorithm. J Trauma Acute Care Surg. 2020;89(5):971–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Kornblith LZ, Kutcher ME, Redick BJ, Calfee CS, Vilardi RF, Cohen MJ. Fibrinogen and platelet contributions to clot formation: implications for trauma resuscitation and thromboprophylaxis. J Trauma Acute Care Surg. 2014;76(2):255–6; discussion 62-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Matthay ZA, Fields AT, Nunez-Garcia B, Patel M, Cohen MJ, Callcut RA, et al. Dynamic Effects of Calcium on In Vivo and Ex Vivo Platelet Behavior After Trauma. J Trauma Acute Care Surg. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Kornblith LZ, Robles AJ, Conroy AS, Hendrickson CM, Calfee CS, Fields AT, et al. Perhaps it's not the platelet: Ristocetin uncovers the potential role of von Willebrand factor in impaired platelet aggregation following traumatic brain injury. J Trauma Acute Care Surg. 2018;85(5):873–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Knudson MM, Moore EE, Kornblith LZ, Shui AM, Brakenridge S, Bruns BR, et al. Challenging Traditional Paradigms in Posttraumatic Pulmonary Thromboembolism. JAMA Surg. 2021:e216356. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Kearon C, Akl EA, Ornelas J, Blaivas A, Jimenez D, Bounameaux H, et al. Antithrombotic Therapy for VTE Disease: CHEST Guideline and Expert Panel Report. Chest. 2016;149(2):315–52. [DOI] [PubMed] [Google Scholar]
  • 23.Kornblith LZ, Decker A, Conroy AS, Hendrickson CM, Fields AT, Robles AJ, et al. It's About Time: Transfusion Effects on Post-Injury Platelet Aggregation Over Time. J Trauma Acute Care Surg. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Haemonetics. TEG 5000 User Manual. P/N 06-510-US, Manual Revision: AC: Haemonetics Corportation, Haemoscope Division; 2010. [Google Scholar]
  • 25.McElreath R AIC provides a surprisingly simple estimate of the average out-of-sample deviance. Boca Raton, FL: CRC Press; 2016. [Google Scholar]
  • 26.Seyfert UT, Haubelt H, Vogt A, Hellstern P. Variables influencing Multiplate(TM) whole blood impedance platelet aggregometry and turbidimetric platelet aggregation in healthy individuals. Platelets. 2007;18(3):199–206. [DOI] [PubMed] [Google Scholar]
  • 27.Falanga A, Schieppati F, Russo L. Pathophysiology 1. Mechanisms of Thrombosis in Cancer Patients. Cancer Treat Res. 2019;179:11–36. [DOI] [PubMed] [Google Scholar]
  • 28.Shih L, Kaplan D, Kraiss LW, Casper TC, Pendleton RC, Peters CL, et al. Platelet-Monocyte Aggregates and C-Reactive Protein are Associated with VTE in Older Surgical Patients. Sci Rep. 2016;6:27478. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Mitrugno A, Tassi Yunga S, Sylman JL, Zilberman-Rudenko J, Shirai T, Hebert JF, et al. The role of coagulation and platelets in colon cancer-associated thrombosis. Am J Physiol Cell Physiol. 2019;316(2):C264–C73. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Skille H, Paulsen B, Hveem K, Gabrielsen ME, Brumpton B, Hindberg K, et al. Genetic variation of platelet glycoprotein VI and the risk of venous thromboembolism. Haematologica. 2020;105(7):e358–e60. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.von Bruhl ML, Stark K, Steinhart A, Chandraratne S, Konrad I, Lorenz M, Khandoga A, et al. Monocytes, neutrophils, and platelets cooperate to initiate and propagate venous thrombosis in mice in vivo. J Exp Med. 2012;209(4):819–35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Vogel S, Bodenstein R, Chen Q, Feil S, Feil R, Rheinlaender J, et al. Platelet-derived HMGB1 is a critical mediator of thrombosis. J Clin Invest. 2015;125(12):4638–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Dyer MR, Alexander W, Hassoune A, Chen Q, Alvikas J, Liu Y, et al. Platelet-derived extracellular vesicles released after trauma promote hemostasis and contribute to DVT in mice. Thromb Haemost. 2019. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Schrottmaier WC, Kral JB, Badrnya S, Assinger A. Aspirin and P2Y12 Inhibitors in platelet-mediated activation of neutrophils and monocytes. Thromb Haemost. 2015;114(3):478–89. [DOI] [PubMed] [Google Scholar]
  • 35.Barrett TJ, Cornwell M, Myndzar K, Rolling CC, Xia Y, Drenkova K, et al. Platelets amplify endotheliopathy in COVID-19. Sci Adv. 2021;7(37):eabh2434. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 36.Brogren H, Karlsson L, Andersson M, Wang L, Erlinge D, Jern S. Platelets synthesize large amounts of active plasminogen activator inhibitor 1. Blood. 2004;104(13):3943–8. [DOI] [PubMed] [Google Scholar]
  • 37.Plow EF, Collen D. The presence and release of alpha 2-antiplasmin from human platelets. Blood. 1981;58(6):1069–74. [PubMed] [Google Scholar]
  • 38.Fields AT, Matthay ZA, Nunez-Garcia B, Matthay EC, Bainton RJ, Callcut RA, et al. Good Platelets Gone Bad: The Effects of Trauma Patient Plasma on Healthy Platelet Aggregation. Shock. 2020. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Mitchell TA, Herzig MC, Fedyk CG, Salhanick MA, Henderson AT, Parida BK, et al. Traumatic Hemothorax Blood Contains Elevated Levels of Microparticles that are Prothrombotic but Inhibit Platelet Aggregation. Shock. 2017;47(6):680–7. [DOI] [PubMed] [Google Scholar]
  • 40.Verni CC, Davila A Jr., Balian S, Sims CA, Diamond SL. Platelet dysfunction during trauma involves diverse signaling pathways and an inhibitory activity in patient-derived plasma. J Trauma Acute Care Surg. 2019;86(2):250–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Vulliamy P, Gillespie S, Armstrong PC, Allan HE, Warner TD, Brohi K. Histone H4 induces platelet ballooning and microparticle release during trauma hemorrhage. Proc Natl Acad Sci U S A. 2019;116(35):17444–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.St John AE, Newton JC, Martin EJ, Mohammed BM, Contaifer D Jr., Saunders JL, et al. Platelets retain inducible alpha granule secretion by P-selectin expression but exhibit mechanical dysfunction during trauma-induced coagulopathy. Thromb Haemost. 2019;17(5):771–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 43.Sillesen M, Johansson PI, Rasmussen LS, Jin G, Jepsen CH, Imam AM, et al. Platelet activation and dysfunction in a large-animal model of traumatic brain injury and hemorrhage. J Trauma Acute Care Surg. 2013;74(5):1252–9. [DOI] [PubMed] [Google Scholar]
  • 44.Moore HB, Moore EE, Gonzalez E, Chapman MP, Chin TL, Silliman CC, et al. Hyperfibrinolysis, physiologic fibrinolysis, and fibrinolysis shutdown: the spectrum of postinjury fibrinolysis and relevance to antifibrinolytic therapy. J Trauma Acute Care Surg. 2014;77(6):811–7; discussion 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 45.Moore HB, Moore EE, Gonzalez E, Hansen KC, Dzieciatkowska M, Chapman MP, et al. Hemolysis exacerbates hyperfibrinolysis, whereas platelolysis shuts down fibrinolysis: evolving concepts of the spectrum of fibrinolysis in response to severe injury. Shock. 2015;43(1):39–46. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 46.Nelson JT, Coleman JR, Carmichael H, Mauffrey C, Vintimilla DR, Samuels JM, et al. High Rate of Fibrinolytic Shutdown and Venous Thromboembolism in Patients With Severe Pelvic Fracture. J Surg Res. 2019;246:182–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 47.Moore HB, Moore EE, Huebner BR, Dzieciatkowska M, Stettler GR, Nunns GR, et al. Fibrinolysis shutdown is associated with a fivefold increase in mortality in trauma patients lacking hypersensitivity to tissue plasminogen activator. J Trauma Acute Care Surg. 2017;83(6):1014–22. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Data File (.doc, .tif, pdf, etc.)

Supplemental Figure 1. Flowchart specifying inclusion and exclusion criteria. ICU-intensive care unit. LOS-length of stay.

Supplemental Figure 2. Distribution of the timing of venous thromboembolic events (VTE) in days from hospital admission.

NIHMS1798361-supplement-Supplemental_Data_File___doc___tif__pdf__etc__.docx (1.4MB, docx)

RESOURCES