ΔR indicates time to initial fibrin formation in minutes.
Connelly CR, Van PY, Hart KD, Louis SG, Fair KA, Erickson AS, Rick EA, Simeon EC, Bulger EM, Arbabi S, Holcomb JB, Moore LJ, Schreiber MA. Thrombelastography-Based Dosing of Enoxaparin for Thromboprophylaxis in Trauma and Surgical PatientsA Randomized Clinical Trial. JAMA Surg. 2016;151(10):e162069. doi:10.1001/jamasurg.2016.2069
Copyright 2016 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.
Is thrombelastogram-adjusted enoxaparin better than standard-dose enoxaparin in the prevention of venous thromboembolism?
In this randomized clinical trial that included 185 trauma and surgical patients, patients receiving thrombelastogram-adjusted doses received a higher median enoxaparin dose than control group patients. Venous thromboembolism rates were similar.
Venous thromboembolism incidence was low and was similar between the groups receiving thrombelastogram-adjusted and standard enoxaparin dosing in this study population.
Prophylactic enoxaparin is used to prevent venous thromboembolism (VTE) in surgical and trauma patients. However, VTE remains an important source of morbidity and mortality, potentially exacerbated by antithrombin III or anti–Factor Xa deficiencies and missed enoxaparin doses. Recent data suggest that a difference in reaction time (time to initial fibrin formation) greater than 1 minute between heparinase and standard thrombelastogram (TEG) is associated with a decreased risk of VTE.
To evaluate the effectiveness of TEG-adjusted prophylactic enoxaparin dosing among trauma and surgical patients.
Design, Setting, and Participants
This randomized clinical trial, conducted from October 2012 to May 2015, compared standard dosing (30 mg twice daily) with TEG-adjusted enoxaparin dosing (35 mg twice daily) for 185 surgical and trauma patients screened for VTE at 3 level I trauma centers in the United States.
Main Outcomes and Measures
The incidence of VTE, bleeding complications, anti–Factor Xa deficiency, and antithrombin III deficiency.
Of the 185 trial participants, 89 were randomized to the control group (median age, 44.0 years; 55.1% male) and 96 to the intervention group (median age, 48.5 years; 74.0% male). Patients in the intervention group received a higher median enoxaparin dose than control patients (35 mg vs 30 mg twice daily; P < .001). Anti–Factor Xa levels in intervention patients were not higher than levels in control patients until day 6 (0.4 U/mL vs 0.21 U/mL; P < .001). Only 22 patients (11.9%) achieved a difference in reaction time greater than 1 minute, which was similar between the control and intervention groups (10.4% vs 13.5%; P = .68). The time to enoxaparin initiation was similar between the control and intervention groups (median [range] days, 1.0 [0.0-2.0] vs 1.0 [1.0-2.0]; P = .39), and the number of patients who missed at least 1 dose was also similar (43 [48.3%] vs 54 [56.3%]; P = .30). Rates of VTE (6 [6.7%] vs 6 [6.3%]; P > .99) were similar, but the difference in bleeding complications (5 [5.6%] vs 13 [13.5%]; P = .08) was not statistically significant. Antithrombin III and anti–Factor Xa deficiencies and hypercoagulable TEG parameters, including elevated coagulation index (>3), maximum amplitude (>74 mm), and G value (>12.4 dynes/cm2), were prevalent in both groups. Identified risk factors for VTE included older age (61.0 years vs 46.0 years; P = .04), higher body mass index (calculated as weight in kilograms divided by height in meters squared; 30.6 vs 27.1; P = .03), increased Acute Physiology and Chronic Health Evaluation II score (8.5 vs 7.0; P = .03), and increased percentage of missed doses per patient (14.8% vs 2.5%; P = .05).
Conclusions and Relevance
The incidence of VTE was low and similar between groups; however, few patients achieved a difference in reaction time greater than 1 minute. Antithrombin III deficiencies and hypercoagulable TEG parameters were prevalent among patients with VTE. Low VTE incidence may be due to an early time to enoxaparin initiation and an overall healthier and less severely injured study population than previously reported.
clinicaltrials.gov Identifier: NCT00990236.
Venous thromboembolism (VTE) is a major source of potentially preventable morbidity and mortality in critically ill trauma and surgical patients. Deep venous thrombosis (DVT) is estimated to occur in up to 60% of severely injured trauma patients and 13% to 31% of patients in intensive care who do not receive appropriate thromboprophylaxis.1,2 Pulmonary embolism (PE) occurs in approximately 2% to 22% of trauma patients and is a major cause of preventable death.3- 5 Costs associated with VTE are high.6 The significance of VTE has led to strong recommendations from the Eastern Association for the Surgery of Trauma and the American College of Chest Physicians for VTE thromboprophylaxis in surgical and trauma patients.7,8
Low-molecular-weight heparin (LMWH) is considered the standard of care for VTE prevention. It is estimated to prevent a significant number of VTE events in trauma patients and has been shown to reduce the risk of symptomatic VTE by 80% among patients who undergo abdominal or pelvic operations.9,10 Enoxaparin sodium is a well-studied and effective form of LMWH used throughout the United States.11 Despite widespread LMWH use, VTE rates remain high. The incidence of DVT among trauma patients receiving appropriate thromboprophylaxis is greater than 10% in multiple studies,12,13 and PE is the third most common cause of death in trauma patients who survive the first 24 hours.14 Pulmonary embolism is also the third most common postoperative medical complication in surgical patients in the United States.2 These observations indicate that the current approach to thromboprophylaxis is inadequate.
Strategies to monitor enoxaparin thromboprophylaxis have been studied. Low anti–Factor Xa (anti-Xa) 12-hour levels have been demonstrated in 50% of surgical patients in intensive care and are associated with a higher DVT rate.15 Despite this finding, fixed LMWH doses are generally used, and anti-Xa–guided thromboprophylaxis is not widely accepted because of considerable disagreement in the literature concerning its usefulness and its nonstandardized reference range.7,16- 23 Previously, we quantified the effects of enoxaparin by comparing the change in reaction time (time to initial fibrin formation in minutes; ∆R) between standard and heparinase thrombelastograms (TEG) and found that a ∆R greater than 1 minute was associated with decreased DVT rates.24 We also demonstrated an association between missed enoxaparin doses and higher DVT rates as well as a high incidence of anti-thrombin III (AT-III) deficiency in trauma and surgical patients.12,13 Together, these data suggest that subprophylactic enoxaparin levels, either through inadequate or missed doses, may contribute to persistent VTE incidence.
We hypothesized that a TEG-guided enoxaparin dosing strategy to increase ∆R to greater than 1 minute would result in lower VTE incidence compared with standard dosing in trauma and surgical patients. We previously performed a similar single-center study, which demonstrated a similar VTE rate between standard and TEG-guided dosing.13 However, that study was underpowered to show a significant difference, so the current study was designed as a multicenter trial.
We performed a prospective randomized clinical trial at the Oregon Health and Science University, the University of Texas Medical School at Houston, and the University of Washington from October 2012 through May 2015. Institutional review board approval was obtained at each site. Written consent was obtained from all patients or authorized representatives for research prior to enrollment. The trial protocol can be found in the Supplement.
Patients were included in this study if they were admitted to the trauma service; received a general, orthopedic, or urologic surgical procedure; were 15 years or older; had initiated standard enoxaparin thromboprophylaxis dosing (either 30 mg twice daily or 40 mg once daily); received between 3 and 5 doses; and had an expected inpatient stay of 3 or more days. Exclusion criteria included therapeutic or nonstandard enoxaparin dosing, other anticoagulation medications, the presence of intracranial hemorrhage or brain injury, and (in women) pregnancy or lactation (Figure 1). Participants were blinded to enoxaparin dose, which was based on prerandomization of each enrollment number.
Prospectively collected data included patient and injury characteristics (ie, age, sex, length of stay, body mass index [calculated as weight in kilograms divided by height in meters squared; BMI], reason for admission [trauma vs surgical procedure], Injury Severity Score, Acute Physiology and Chronic Health Evaluation score, Glasgow Coma Scale score, smoking status, and existence of comorbidities), enoxaparin dose and administration, and transfusions and procedures performed. Standard citrated and heparinase TEGs were performed using the TEG 5000 Thrombelastograph Hemostasis Analyzer System (Haemonetics Corporation). All standard TEG values were recorded, including coagulation index (CI), a calculated global index of coagulation. We calculated ∆R as the difference in time in minutes between a standard TEG and heparinase TEG for each patient sample. We calculated the TEG G value using the following equation: G = (5000 × maximum amplitude [MA])/(100 – MA).25,26 Peak anti-Xa and AT-III levels were measured 4 to 6 hours after enoxaparin administration and analyzed with the STA Compact Hemostasis System (Diagnostica Stago). Anti-thrombin III deficiency was defined as AT-III activity less than 80% (2 SDs below the mean), and anti-Xa deficiency (peak) was defined as a measurement of 0.2 IU/mL or lower.13,21- 23 Standard laboratory data, including complete blood cell counts, basic metabolic panels, arterial blood gases, and coagulation studies, were also collected.
Patients randomized to the control group received a standard enoxaparin dose (30 mg twice daily). To measure the steady-state enoxaparin effect, blood was drawn 4 to 6 hours after the third consecutive enoxaparin dose. Thrombelastograms were performed for 3 consecutive days and then twice weekly until discharge. Patients randomized to the dose-adjusted group began receiving 30 mg twice daily of enoxaparin for 3 consecutive doses, after which the dose was adjusted based on the ∆R value. If ∆R was less than 1 minute, then the dose was increased by 10 mg; if ∆R was more than 2 minutes, it was decreased by 10 mg; and if ∆R was between 1 and 2 minutes, it was unchanged. The maximum dose permitted was a therapeutic dose (1 mg/kg) and the minimum was a standard dose (30 mg twice daily). If a dose adjustment was made, 3 consecutive doses were given to again reach steady state. Additional dose adjustments could then be made. If ∆R was between 1 and 2 minutes for 3 consecutive days, then monitoring was decreased to twice weekly. If a dose was held or missed in either group, the monitoring protocol was restarted after the third consecutive dose (Figure 2).
Primary end points were VTE and bleeding complication incidence. Venous thromboembolism was defined as DVT and/or PE. Deep venous thromboses were not characterized as symptomatic or asymptomatic. To measure VTE incidence, the study protocol dictated that all patients were to receive standard, screening, bilateral, whole-leg venous duplex ultrasonography prior to discharge. In addition to the required discharge duplex ultrasonography, each institution had variable existing VTE screening protocols. At the Oregon Health and Science University, all trauma patients received weekly screenings, while surgical patients received a screening only for DVT symptoms. At the University of Washington and the University of Texas Medical School at Houston, all patients received a screening only if symptomatic. At all institutions, thoracic computed tomography was used to evaluate patients for suspected PE. Bleeding complications were defined as a bleeding event associated with need for intervention, consistent with International Society for Hemostasis and Thrombosis guidelines.27 All bleeding events were reviewed by physician study monitors.
During study development, a power analysis was performed to calculate goal enrollment. Using prior data, a 20% overall VTE rate among patients receiving enoxaparin was assumed, with a 10.9% rate for patients with ∆R greater than 1 minute and 23.0% for those with ∆R less than 1 minute.24 Three hundred patients were needed to demonstrate a significant difference in VTE rate, with β = 0.80 and α = .05. Therefore, goal study enrollment was 320 participants, to allow for patient dropout.
Data were analyzed with R version 3.1.3 (R Foundation). Comparisons were made between the control and intervention groups and between patients with and without VTE. t, Wilcoxon rank sum, Pearson χ2, and Fischer exact tests were used as appropriate. Significance was defined as P < .05.
In total, 18 612 patients were screened, and 185 were randomized (89 patients in the control group and 96 in the dose-adjusted group). All patients were analyzed, with no patients lost to follow-up (Figure 1). Baseline characteristics are shown in Table 1; characteristics were similar, although there were more males in the dose-adjusted group. The enoxaparin treatment duration was similar between the control and intervention groups, and the time to enoxaparin administration was short in both groups. Patients in the intervention group received significantly higher average prescribed and actually administered doses. A total of 97 patients (52.4%) missed at least 1 dose, but the median percentage of doses missed per patient was low (Table 1).
Thrombelastogram results and coagulation characteristics are shown in Table 2. There were no significant differences in standard TEG values (R, K time [minutes to 20 mm clot strength], α angle [rate of clot strengthening], MA, or degree of clot lysis at 30 minutes) on average or over time. Only 22 patients (11.9%) achieved ∆R greater than 1 minute, with similar incidence between groups. No significant difference was found in median ∆R overall or over time. There was no statistically significant difference between the control and intervention groups by the sixth ∆R measurement (0.08 vs 0.29; P = .09). Hypercoagulable TEG parameters at any point during hospital admission (ie, CI >3; MA >74 mm; and G value >12.4 dynes/cm2)28,29 were compared. Control patients were more often hypercoagulable using CI and MA measurements, but the incidence of hypercoagulable G values was similar. Antithrombin III deficiency was similar between the control and intervention groups (18.0% vs 19.8%; P = .85) and less prevalent than previously reported (Table 2).13 A similar percentage of patients were anti-Xa deficient, and anti-Xa levels for patients with adjusted dosing did not exceed levels in control patients until the sixth measurement (0.4 U/mL vs 0.21 U/mL; P < .001).
Venous duplex ultrasonography completion was similar between the control and intervention groups (Table 2). Overall VTE incidence was low, occurring in only 12 patients (6.5%). The median time to diagnosis was similar. The incidence of DVT, PE, VTE, or death was similar. Compared with control patients, the rate of bleeding complications in intervention patients was not statistically significant (5.6% vs 13.5%; P = .08) (Table 2). Patients with bleeding complications had a higher maximum enoxaparin dose.
Characteristics of patients with VTE are shown in Table 3. Patients with VTE were older; had higher BMI and Acute Physiology and Chronic Health Evaluation II scores; were nontrauma surgical patients; had longer enoxaparin treatment duration and lengths of stay; and had higher average prescribed and administered enoxaparin doses. While the incidence of at least 1 missed dose was not significantly higher in patients with VTE, these patients missed a higher percentage of prescribed doses than patients without VTE (median [range], 14.8% [3.8%-19.5%] vs 2.5% [0%-8.5%]; P = .05). Incidence of ∆R greater than 1 minute and all other coagulation characteristics (CI, MA, G value, and anti-Xa and AT-III deficiencies) were not significantly different between patients with and without VTE (Table 3).
An interim analysis was performed after the enrollment of 185 patients. The incidence of VTE was similar (6.7% vs 6.3%; P > .99). Assuming equivalent effect sizes for future enrollments, 3212 total patients would be needed to demonstrate statistical significance. Therefore, the study was discontinued for futility. Furthermore, if the study had reached goal enrollment (320 patients) and a similar difference in bleeding events had occurred, the rate of bleeding events would have been significantly higher in the intervention group.
This multicenter, prospective, randomized clinical trial compared the effectiveness of TEG-adjusted vs standard enoxaparin dosing to prevent VTE among trauma and surgical patients. Patients in the intervention group received significantly higher average prescribed and administered enoxaparin doses than control patients, but there was no significant difference in VTE incidence. To our knowledge, 2 other published studies13,30 have pursued TEG-adjusted strategies that increase LMWH dosing for a ∆R less than 1 minute. In our initial single-center trial,13 87 patients were randomized to TEG-adjusted vs standard enoxaparin dosing. Rates of VTE were similar between the control and intervention groups (16.2% vs 13.6%; P = .73). However, a high prevalence of AT-III deficiency (60% of patients) was observed.13 In addition, Harr et al30 recently demonstrated similar VTE rates in patients randomized to receive standard (5000 IU daily) and TEG-adjusted dalteparin dosing but suggested a role for antiplatelet therapy for VTE prophylaxis following trauma. To our knowledge, our study is the third to demonstrate that a TEG-adjusted strategy using ∆R does not improve VTE rates.
There are multiple explanations for these findings. A simple hypothesis is that missing 1 or more doses of enoxaparin, which occurred in a high percentage of patients, may counter any possible positive effects of dose adjustment. In a prior prospective observational study of trauma and general surgery patients,12 58.9% of patients missed at least 1 enoxaparin dose, and DVTs occurred in nearly a quarter of those patients. On regression analysis, a significant association was found between a missed enoxaparin dose and DVT.12 We demonstrated that 75.0% of patients with VTE missed at least 1 dose and had a higher percentage of doses missed per patient compared with non-VTE patients. These results suggest that improved compliance to minimize missed doses with any thromboprophylaxis protocol may be a very important step needed to lower VTE incidence.
As originally described, the association between ∆R less than 1 minute and VTE was observed in patients who were old (mean [SD] age, 54 [3.8] years), severely injured (mean [SD] injury severity score, 44.6 [3.2]), critically ill (mean [SD] Acute Physiology and Chronic Health Evaluation II score, 18.1 [1.5]) and obese (mean [SD] BMI, 33.8 [2.4]).24 While the association between ∆R and VTE was strong in that population, its importance may be diminished in the less critically ill trauma and surgical patients represented in our current study. To our knowledge, to date, 3 prospective randomized trials13,24 have failed to demonstrate a change in VTE incidence despite LMWH dose adjustments for ∆R less than 1 minute. In both this study and our prior trial,13 the median ∆R was less than 1 minute and was equivalent for patients with and without VTE. Also, VTE rates in this study were lower than previously described (28%).24 These observations do not support TEG-guided thromboprophylaxis among patients similar to our study population. However, it is very possible that a prospective study limited to only those patients with severe injuries and a prolonged expected length of stay with longer term follow-up could yield a different result.
Furthermore, it may be difficult to achieve a ∆R greater than 1 minute in the typical patient, given the practical limitations of the protocol and treatment duration. Prior to dose adjustment, 3 doses are required to achieve steady state after any missed or adjusted dose. In our study, the average treatment duration was only 5 to 6 days, which may not have allowed enough time for adjustments to occur to increase ∆R to greater than 1 minute. Only 22 patients (11.9%) in the current study achieved a ∆R greater than 1 minute, and the average ∆R never met or exceeded 1 minute in any study. More aggressive adjustment strategies or higher initial doses could increase the achievement of ∆R greater than 1 minute, although this may increase the risk of bleeding complications, and 30 mg twice daily is the current standard of care.
The difference in bleeding complications approached significance in this study. Sixteen of 18 bleeding complications were intraoperative or postoperative bleeding events associated with hypotension or anemia that required blood transfusion. Significant gastrointestinal bleeding from a gastric ulcer and major bleeding after inadvertent patient removal of a central venous catheter accounted for the other 2 complications. No bleeding complication resulted in death. Of interest, the median maximum dose of enoxaparin was significantly higher among patients with a bleeding complication.
We demonstrated that trauma and surgical patients had significant hypercoagulable risk factors. The CI and MA were in the hypercoagulable range for 82 (44.3%) and 99 (53.5%) patients, respectively, and were significantly higher in the control group. The G value was elevated in 123 patients (66.5%) and in 9 (75.0%) with a VTE. Previous authors31 found that increased MA is associated with PE. Others have argued that trauma patients develop a hypercoagulable state despite adequate heparin-based therapy, likely due to increased platelet activation.30 These data reinforce the importance of platelet contribution to VTE and the potential role for platelets as a target in future VTE prevention strategies.
Antithrombin III deficiency was prevalent in 35 of 185 patients (18.9%). Heparin-based anticoagulation relies on AT-III and has little pharmacologic activity with AT-III deficiency. Although our prevalence was lower than previously demonstrated, AT-III deficiency may contribute to persistent VTE incidence despite adequate heparin-based thromboprophylaxis.13 In this study, 33.3% of patients with VTE were AT-III deficient. To further investigate the effects of AT-III deficiency, particularly in those patients receiving LMWH prophylaxis, we are conducting a prospective cohort trial powered to find differences in AT-III levels between patients with and without VTE.
The overall VTE rate in this trial was lower than we demonstrated previously (6.5% vs 15.0%).13 One important difference is the time to enoxaparin administration. In this study, the median time to initiation was 1 day for both groups, whereas the mean time was 2 to 3 days in our previous study.13 This observation suggests that an earlier time to enoxaparin administration may help lower VTE incidence. Another important observation is that median BMI was significantly higher in patients with VTE (30.6 vs 27.1; P = .03). In our prior trial,13 with higher observed VTE rates, BMI ranged from 30.6 to 32.8. In a rat obesity model, McCully et al32 demonstrated that obese rats do not develop the acute coagulopathy of trauma after hemorrhagic shock and are hypercoagulable at baseline. Other authors21 suggest that obesity decreases the effectiveness of enoxaparin prophylaxis. These data suggest that the effects of obesity must be further evaluated and may contribute toward VTE formation.
This study has some limitations. First, only 185 patients were enrolled because of difficult enrollment logistics. Second, many patients in both study groups missed at least 1 enoxaparin dose. Because a missed dose is a demonstrated risk factor for DVT formation,12 any effects of dose adjustment may have been absorbed by the high rate of missed doses. Next, the physiologic effects of the dose adjustment (ie, anti-Xa and ∆R) were not observed until the sixth measurement. It is possible that the time to achieve this difference was too late because VTE events had already occurred. Furthermore, only a marginal dosing difference and no significant difference in ∆R was achieved between groups. Also, few patients in the dose-adjusted group achieved ∆R greater than 1 minute because of the constraints of the dose-adjustment schedule. Therefore, conclusions about the role of ∆R remain hypothetical. Also, variable screening protocols at participating institutions and an 80% duplex completion rate may have contributed to the lower-than-expected VTE rate because some events may have been missed. Finally, overall VTE incidence was lower than anticipated in the power analysis. Therefore, this study was underpowered to detect differences in VTE rates between study groups or coagulation differences between patients with and without VTE.
This multicenter, prospective randomized clinical trial compared TEG-adjusted vs standard prophylactic enoxaparin administration and VTE incidence in trauma and surgical patients. Few patients achieved a ∆R greater than 1 minute, and VTE rates were similar between groups. In addition, the difference in bleeding complications was not statistically significant. We conclude that TEG-adjusted enoxaparin administration based on ∆R is not supported by our current data in this study population. However, these data demonstrate that many trauma and surgical patients have hypercoagulation at baseline and that further investigations into the effects of platelet activation and obesity, time to enoxaparin administration, and AT-III deficiency are needed.
Corresponding Author: Christopher R. Connelly, MD, Division of Trauma, Critical Care, and Acute Care Surgery, Department of Surgery, Oregon Health and Science University, 3181 SW Sam Jackson Park Rd, L611, Portland, OR 97239 (email@example.com).
Accepted for Publication: April 25, 2016.
Published Online: August 3, 2016. doi:10.1001/jamasurg.2016.2069
Author Contributions: Drs Connelly and Schreiber had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Study concept and design: Connelly, Van, Louis, Fair, Holcomb, Schreiber.
Acquisition, analysis, or interpretation of data: Connelly, Hart, Louis, Fair, Erickson, Rick, Simeon, Bulger, Arabi, Moore, Schreiber.
Drafting of the manuscript: Connelly, Simeon.
Critical revision of the manuscript for important intellectual content: Connelly, Van, Hart, Louis, Fair, Erickson, Rick, Bulger, Arabi, Holcomb, Moore, Schreiber.
Statistical analysis: Connelly, Hart, Fair, Holcomb.
Obtained funding: Schreiber.
Administrative, technical, or material support: Connelly, Van, Louis, Fair, Erickson, Rick, Simeon, Arabi, Schreiber.
Study supervision: Fair, Bulger, Holcomb, Moore, Schreiber.
Conflict of Interest Disclosures: None reported.
Funding/Support: This work was funded by subaward W81XWH-11-1-0841 from the National Trauma Institute and sponsored by the US Department of the Army, Prime award W81XWH-11-1-0841. The US Army Medical Research Acquisition Activity (820 Chandler St, Fort Detrick, MD 21702-5014) was the awarding and administering acquisition office.
Role of the Funder/Sponsor: The funder had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.
Disclaimer: The opinions and assertions contained herein are the private views of the authors and are not to be construed as official or as reflecting the views of the US Department of the Army or the US Department of Defense.
Previous Presentation: This work was presented at the 87th Annual Pacific Coast Surgical Association Meeting; February 16, 2016; Kohala Coast, Hawaii.
Additional Contributions: We thank Samantha Underwood, MS (Division of Trauma, Critical Care, and Acute Care Surgery, Department of Surgery, Oregon Health and Science University, Portland), and all other research coordinators at the Oregon Health and Science University who contributed to the completion of this study. None of the contributors were compensated for their contributions.