Secondary Logo

Share this article on:

Time course and outcomes associated with transient versus persistent fibrinolytic phenotypes after injury

A nested, prospective, multicenter cohort study

Roberts, Derek J., MD, PhD; Kalkwarf, Kyle J., MD; Moore, Hunter B., MD; Cohen, Mitchell J., MD; Fox, Erin E., PhD; Wade, Charles E., PhD; Cotton, Bryan A., MD

Journal of Trauma and Acute Care Surgery: February 2019 - Volume 86 - Issue 2 - p 206–213
doi: 10.1097/TA.0000000000002099
AAST 2017 PODIUM PAPER
Editor's Choice

BACKGROUND Temporal changes in fibrinolytic activity after injury and their impact on outcomes remain poorly defined. We conducted a prospective, multicenter cohort study to determine the incidence of fibrinolytic phenotypes after injury and the trajectories and associated outcomes of these phenotypes over time.

METHODS We included adults that arrived within 6 hours of injury to three American Level I trauma centers. Clot lysis at 30 minutes (LY-30) was measured at presentation and at 3 hours, 6 hours, 12 hours, 24 hours, 48 hours, 72 hours, 96 hours, and 120 hours. LY-30 was used to categorize patients into the following fibrinolytic phenotypes: fibrinolysis shutdown (SD, LY-30 ≤0.8%), physiologic fibrinolysis (PHYS, LY-30 >0.8% to <3%), or hyperfibrinolysis (HF, LY-30 ≥3%). We used multivariable logistic regression to estimate adjusted odds ratios for mortality.

RESULTS We included 795 adults (median age, 38 years; median Injury Severity Scale score, 21). In total, 44% presented with SD, 36% with PHYS, and 21% with HF. Mortality was highest among those who presented with HF (20%) followed by SD (10%) and PHYS (7%) (p = 0.001). While mortality within the first 24 hours was highest with admission HF (14% vs. 5% SD vs. 4% PHYS; p = 0.001), both admission HF (7%) and SD (6%) had higher mortality after 24 hours compared with PHYS (3%) (p = 0.04). All patients who presented with HF switched into another phenotype or died within 24 hours. The majority of patients that presented in SD remained in that phenotype, including 71% at 24 hours and 72% at 120 hours. Persistent SD at 24 hours was independently associated with increased mortality after 24 hours (odds ratio, 3.20; 95% confidence interval, 1.51–6.67).

CONCLUSION Approximately 70% of major trauma patients who present with SD remain in this phenotype up to 120 hours postinjury. In contrast, patients presenting with HF transition into another phenotype or die within 24 hours. While early mortality is highest with the HF phenotype, persistent SD at 24 hours is associated with elevated late mortality.

LEVEL OF EVIDENCE Prognostic and epidemiological study, level II.

From the Division of Vascular and Endovascular Surgery (D.J.R.), Department of Surgery, University of Ottawa, Ottawa, Ontario, Canada; Department of Surgery (K.J.K., E.E.F., C.E.W., B.A.C.), Center for Translational Injury Research (D.J.R., K.J.K., E.E.F., C.E.W., B.A.C.), The University of Texas Health Science Center, Houston, Texas; Department of Surgery (H.B.M., M.J.C.), University of Colorado Denver, Aurora; and Denver Health Medical Center (M.J.C.), Denver, Colorado.

Submitted: September 1, 2017, Revised: October 4, 2018, Accepted: October 10, 2018, Published online: October 29, 2018.

Address for reprints: Derek J. Roberts, MD, PhD, Division of Vascular and Endovascular Surgery, Department of Surgery, University of Ottawa, The Ottawa Hospital, Civic Campus, Rm A280, 1053 Carling Ave, Ottawa, Ontario, Canada K1Y 4E9; email: Derek.Roberts01@gmail.com.

This work was completed at the Center for Translational Injury Research, The University of Texas Health Science Center, Houston, Texas.

This work was presented at the 76th Annual Meeting of The American Association for the Surgery of Trauma and the Clinical Congress of Acute Care Surgery in Baltimore, Maryland, on September 15, 2017 (oral podium presentation).

Supplemental digital content is available for this article. Direct URL citations appear in the printed text, and links to the digital files are provided in the HTML text of this article on the journal’s website (www.jtrauma.com).

Fibrinolysis is increasingly recognized as an integral component of the coagulation cascade regulating the balance between hemorrhage and thrombosis by accelerating or inhibiting clot degradation.1 The degree of fibrinolysis may be measured using point-of-care viscoelastic assays, including rapid thromboelastography (rTEG) and rotational thromboelastometry.2 Using these techniques, one of three different fibrinolytic phenotypes may be demonstrated after injury, including hyperfibrinolysis (HF), physiologic fibrinolysis (PHYS), or fibrinolysis shutdown (SD).1,3,4

Excessive clot degradation, or HF, has been reported to occur relatively infrequently shortly after severe injury, with a reported incidence of approximately 18%.3,4 This fibrinolytic phenotype may occur in combination with thrombin generation inhibition, a finding previously described as acute traumatic coagulopathy, which itself increases risk of death from hemorrhage.5–7 HF can also occur in the absence of an elevated international normalized ratio (INR), where it is associated with an increased risk of early postinjury mortality, again most often from exsanguination.3,8–12

Most patients, however, have been observed to present with the opposite fibrinolytic extreme (impaired fibrinolysis or SD).1,3,4 The reported incidence of this more common fibrinolytic phenotype ranges from 46% to 64% among severe trauma patients.1,3,4 Although its exact etiology remains unknown, it appears to occur preferentially in subjects with substantial soft tissue and/or musculoskeletal injuries.13 In contrast to HF, the development of SD after injury has been associated with a higher risk of late mortality, often resulting from traumatic brain injury (TBI) or multisystem organ failure.1,3,4 In patients with severe extremity trauma, it has also been associated with increased venous thromboembolic (VTE) complications.14

To date, temporal changes in fibrinolytic activity after injury in adults, and the associated impact of these changes on mortality, remain poorly defined. One recent single-center cohort study reported that up to 44% of injured adults with SD upon admission to the intensive care unit (ICU) remain in this phenotype up to 1 week post-ICU admission.1 This persistent SD cohort had an associated fourfold increase in mortality compared with patients that did not recover from SD.1 However, as TEGs were not performed serially to assess for SD after injury, and persistent SD was defined as ongoing SD at 1 week,1 knowledge gaps remain about the time course and outcomes associated with fibrinolytic phenotypes after injury.

In this study, we sought to determine the incidence and associated characteristics of fibrinolytic phenotypes after injury and the trajectories and associated outcomes of these phenotypes over time. We hypothesized that HF would either resolve quickly or be rapidly lethal in most patients while SD would frequently persist and be associated with increased odds of late mortality.

Back to Top | Article Outline

METHODS

Design and Setting

This 14-month, prospective, multicenter cohort study was set across three American College of Surgeons-verified Level I trauma centers located in the United States: The Red Duke Trauma Institute at Memorial Hermann Hospital (UTHealth, Houston, TX), San Francisco General Hospital (UCSF, San Francisco, CA), and Oregon Health & Sciences University (OHSU, Portland, OR). The institutional review boards of these hospitals and their affiliated medical schools provided ethical approval for the study, which was nested in a larger cohort study of the overall clinical utility of serial rTEG in injured adults.15,16 The study was funded, in part, by Haemonetics Corporation (Braintree, MA) and is registered online at ClinicalTrials.gov (NCT01228058). It is reported according to the Strengthening of Reporting of Observational Studies in Epidemiology statement.17

Back to Top | Article Outline

Participants

Between December 1, 2010, and January 31, 2012, we assessed all injured adults (those 18 years or older) who met criteria for the above institution's highest level of trauma team activation. We included adults that arrived within 6 hours of injury and excluded those who were younger than 18 years, pregnant, or incarcerated before admission; received cardiopulmonary resuscitation before hospital arrival; suffered burns covering greater than 20% of their total body surface area; died within 30 minutes of hospital arrival; or arrived longer than 6 hours after injury. Delayed informed consent was obtained from all patients, or from their legally authorized representative, for inclusion of their data in the study.

Back to Top | Article Outline

VTE Prophylaxis and Antifibrinolytic Administration

During the study period, all of the three study trauma centers initiated VTE prophylaxis upon admission for all patients except those with American Association for the Surgery of Trauma grade III or higher solid organ injuries (which had VTE prophylaxis held for 48 hours) and TBI (which had VTE prophylaxis held for 48–72 hours). We did not screen for VTE complications (deep vein thrombosis [DVT] and/or pulmonary embolism [PE]) and only ordered VTE diagnostic studies if these diagnoses were clinically suspected. As study patients were enrolled during a period where none of the above trauma centers were using antifibrinolytics, none of the injured patients in this study received tranexamic acid or other antifibrinolytics.

Back to Top | Article Outline

Data Sources

Immediately after arrival, dedicated research personnel collected prehospital data. These personnel also collected data on patient vital signs and resuscitation fluids provided in the emergency department, operating room, interventional radiology suite, and ICU in “real time” through direct observation. Data after 24 hours were collected on a daily basis by these personnel by reviewing patient medical records. We collected data on Abbreviated Injury Scale scores and Injury Severity Scale (ISS) score from institutional trauma registries. Complication and outcome data were obtained prospectively from hospital medical records.

Back to Top | Article Outline

Collection, Processing, and Analysis of Blood and rTEG Specimens

rTEG generates several measures that provide a detailed description of the clotting cascade and may be displayed on a bedside computer monitor while the test is being run in the laboratory.2,18 Blood and rTEG samples were collected immediately after presentation and at 3 hours, 6 hours, 12 hours, 24 hours, 48 hours, 72 hours, 96 hours, and 120 hours in 2.7-mL vacutainer (Becton-Dickinson, Franklin Lanes, NJ) tubes containing sodium citrate. rTEG samples were subsequently processed as previously described.19 Briefly, blood samples for rTEG were immediately transported to a TEG 5000 Thrombelastograph Hemostasis Analyzer System (Haemonetics Corporation) where anticoagulation was reversed by adding calcium chloride to the citrated whole blood contained in the rTEG cup. Samples were subsequently activated with tissue factor and kaolin and then assayed as per manufacturer's recommendations. Quality controls were performed on rTEG analyzers every 8 hours as suggested by Haemonetics Corporation. To avoid potential for observer bias, clinicians were blinded to all rTEG tracings and measures produced during the study.

Back to Top | Article Outline

Exposure and Outcome Variables

We used rTEG LY-30 measures and previously published and validated definitions to classify patients into one of three fibrinolysis phenotypes at each rTEG collection time: SD (LY-30 ≤0.8%), PHYS (LY-30 0.9% to 2.9%), or HF (LY-30 ≥3%).1,3,4,20 LY-30 is an rTEG measure that reflects the degree of fibrinolysis at 30 minutes after clot has achieved maximum amplitude (MA, the point at which clot strength reaches its maximal measure in millimeters on the rTEG tracing).2

The co-primary outcomes were early (≤24 hours) and late (>24 hours) mortality. Mortality was stratified in this fashion as studies have suggested that many patients with HF die early while those with SD experience a more delayed mortality.3,4 Secondary outcomes included 30-day mortality, development of coagulopathy (defined using conventional coagulation tests as an INR >1.321 or rTEG as an activated clotting time ≥128 seconds, K time ≥2.5 minutes, α-angle ≤56°, MA ≤55 mm, or LY-30 ≥3%18,22), requirement for red blood cell (RBC) transfusion, total volume of RBCs administered, administration of ≥10 U of RBC in the first 24 hours, VTE complications (DVT and/or PE), and ICU- and hospital-free days out of 30 days.

Back to Top | Article Outline

Statistical Analyses

We summarized dichotomous and continuous data using counts (percentages) and medians (with interquartile ranges [IQR]). These statistics were compared using Fisher's exact and Wilcoxon rank sum or Kruskal-Wallis tests, respectively.

We calculated areas under the receiving operation characteristic (ROC) curves to determine which time cutoff for defining persistent SD had the highest accuracy for predicting mortality after 24 hours. Multivariable logistic regression was used to estimate adjusted odds ratios (ORs) (with 95% confidence intervals [CIs]) comparing late mortality between patients who did versus did not remain in SD at this time cutoff. This model used data from the subset of patients who presented with SD and included variables for age, sex, mechanism of injury (blunt or penetrating), and study site. We assessed model goodness-of-fit using a Hosmer-Lemeshow test with the generally recommended 10 groups.23

Using data from the subset of patients who presented with SD, we then used a mixed-effects model with a study site-specific random intercept to identify potential independent predictors of persistent SD at the same time cutoff used in the above logistic regression model.24 As insufficient literature existed to prespecify a comprehensive list of “well-motivated” potential predictors of persistent SD, we included all unique variables with less than 10% missingness (to maintain power) that were associated with persistent SD (p < 0.2) at the above time cutoff as fixed effects.23 We also included age and male sex as fixed effects in this analysis to build a model comparable with that created in a previous study of persistent SD after major injury.1 We assessed prediction accuracy of the mixed-effects model by calculating the area under the ROC curve after estimation.

We considered two-sided p values less than 0.05 significant. Stata MP version 13.1 (Stata Corp., LP, College Station, TX) was used for statistical analyses.

Back to Top | Article Outline

RESULTS

Participants

In total, 795 injured adults were enrolled in the study and had serial rTEGs drawn after injury. The median age of the patients was 38 (IQR, 26–53), 76% were males, and 73% were injured by blunt mechanisms, resulting in a median ISS score of 21 (IQR, 10–29). Sixty-seven percent were severely injured (ISS, ≥15), 12% had a systolic blood pressure (BP) less than 90 mm Hg at presentation, and 53% received an RBC transfusion within the first 24 hours.

Back to Top | Article Outline

Comparison of Patients With SD, PHYS, and HF at Admission

A total of 694 (87%) patients had an rTEG performed after arrival that provided LY-30 data to determine presenting fibrinolytic phenotype. Of these patients, 143 (21%) presented with HF, 247 (36%) with PHYS, and 304 (44%) with SD.

Characteristics of the patients stratified by presenting fibrinolytic phenotype are shown in Table 1. Patients who presented with SD were more likely to be injured by a blunt mechanism than those who presented with HF and PHYS (p = 0.007). These patients also had higher ISS (p < 0.001) and lower Glasgow Coma Scale (GCS) (p = 0.008) scores. In contrast, adults who presented with HF were more often hypotensive compared with those with admission PHYS and SD (p = 0.02). Finally, patients who presented with SD, and especially HF, were more likely than those who presented with PHYS to have an associated coagulopathy (defined using an INR >1.3 [p = 0.01] and rTEG [p < 0.001]).

TABLE 1

TABLE 1

Patients who presented with HF and SD were more likely to receive an RBC (p = 0.003), plasma (p = 0.03), and platelet transfusion (p < 0.001) than those who presented with PHYS (Table 2). However, patients who presented with HF received the most RBCs within the first 24 hours. They were also more likely to receive 10 or more units of RBCs than those with admission PHYS and SD during this period (p = 0.001).

TABLE 2

TABLE 2

Forty-eight (6%) patients developed VTE complications during the study period, resulting in 34 (4%) DVTs and 17 (2%) PEs. The risk of VTE was similar between patients who presented with HF (6%), PHYS (4%), and SD (7%) (p = 0.19). Median ICU-free days were less in patients who presented with SD (24; IQR, 15–28) than HF (27; IQR, 15–29) or PHYS (28; IQR, 21–29) (p < 0.001). In addition, median hospital-free days were less in those who presented with SD (17; IQR, 6–24) than HF (20; IQR, 5–26) or PHYS (21; IQR, 12–26) (p < 0.001).

Overall, 30-day mortality was highest among those who presented with HF (20%) followed by SD (10%) and PHYS (7%) (p = 0.001). Mortality within the first 24 hours was also highest with admission HF (14% vs. 4% with PHYS vs. 5% with SD, p = 0.001). Both admission HF (7%) and SD (6%) had higher mortality after 24 hours when compared with those with admission PHYS (3%) (p = 0.04).

Back to Top | Article Outline

Time Course of HF and SD and Defining Persistent SD

Figure 1 and Supplemental Digital Content 1 (see Figure, Supplemental Digital Content 1, http://links.lww.com/TA/B224) display temporal plots of LY30 and all other rTEG measures after injury by presenting fibrinolytic phenotype, respectively. Most patients who presented with SD remained in this phenotype, including 80% at 3 hours, 79% at 6 hours, 77% at 12 hours, 71% at 24 hours, 69% at 48 hours, 67% at 72 hours, and 72% at both 96 hours and 120 hours. Using only the data on patients who presented in SD and had complete serial rTEG data at all of the nine blood collection time points (n = 65), a similar 77% remained in shutdown at 24 hours, 74% at 48 hours, 71% at 72 hours, 76% at 96 hours, and 68% at 120 hours. Defining persistent SD as SD at 24 hours had the highest accuracy for predicting late mortality (after 24 hours) after injury, with an area under the ROC curve of 0.62 (95% CI, 0.55–0.69). In contrast to those who presented with SD, all of the patients who presented with HF switched into another fibrinolysis phenotype or died within the first 24 hours of admission.

Figure 1

Figure 1

Back to Top | Article Outline

Predictors of Persistent SD

Characteristics of the patients that presented in SD and remained in SD at 24 hours (persistent SD) versus those with SD that resolved by 24 hours (transient SD) are presented in Table 3. Patients with persistent SD had higher ISS scores (p = 0.01), were more hypocoagulable by admission INR (p = 0.004) and PTT (p = 0.001), and had lower LY-30 values at admission and at 24 hours (p < 0.001 for both) than those with transient SD. Patients with persistent SD also had a higher incidence of platelet transfusion in the first 24 hours (p = 0.001) and received more RBCs (p = 0.02), plasma (p = 0.01), and platelets (p = 0.002) during this period than their transient SD counterparts (Table 4).

TABLE 3

TABLE 3

TABLE 4

TABLE 4

Using mixed-effects models adjusting for study site, independent predictors of persistent SD at 24 hours included admission systolic BP (adjusted OR per each 1 mm Hg increase in systolic BP, 0.99; 95% CI, 0.97–0.99; p = 0.04) and LY-30 percentage (adjusted OR per each 1% increase in LY-30, 0.075; 95% CI, 0.018–0.31; p < 0.001) and transfusion of platelets within the first 24 hours of admission (adjusted OR, 7.92; 95% CI, 1.51–41.53; p = 0.01) (Table 5). The area under the ROC curve for this prediction model was 0.80 (95% CI, 0.73–0.87).

TABLE 5

TABLE 5

Back to Top | Article Outline

Persistent SD and Outcomes

The risk of VTE was similar between patients with persistent (9%) versus transient (7%) SD (p = 0.48). However, both median ICU-free days (23; IQR, 13–27 vs. 27; IQR, 19–29) and median hospital-free days (13; IQR, 8–23 vs. 20; IQR, 11–26) were lower in those with persistent versus transient SD groups, respectively (both p < 0.001).

Patients with persistent SD at 24 hours were more likely to die than those with transient SD (8% vs. 3%, p = 0.048). When compared with patients with transient SD, those with persistent SD at 24 hours had an associated increase in late mortality (>24 hours) after adjusting for age, sex, mechanism of injury, and study site (adjusted OR = 3.20; 95% CI = 1.51–6.67; p = 0.002) (see Table, Supplemental Digital Content 2 for the full logistic regression model, http://links.lww.com/TA/B225). The Hosmer-Lemeshow test for this regression model ruled out a gross lack of fit (p = 0.44). All deaths in those with persistent SD were from TBI, while all deaths in the transient SD group were from cardiac causes.

Back to Top | Article Outline

DISCUSSION

This study is the first to determine both the incidence of different fibrinolytic phenotypes after injury in adults and the associated trajectory and outcomes of these phenotypes over time in patients. Importantly, our findings are based on a sample of injured adults that were not treated with tranexamic acid or other antifibrinolytics. Similar to other studies, SD was the most common fibrinolytic phenotype in adults presenting after major injury.1,3,4 Overall mortality was also again observed to follow a U-shaped distribution, being highest among those who presented with HF followed by SD and then PHYS.1,3,4 Novel findings of this study suggest that the majority of adults who present with SD after major injury remain in this phenotype at 24 hours and up to 120 hours postinjury. Patients presenting with HF, on the other hand, quickly transition into an SD or PHYS phenotype or die within 24 hours of presentation. Finally, while early mortality again appears to be greatest among trauma patients with the HF phenotype, in this study those who remained in SD at 24 hours after injury had an increased late mortality beyond 24 hours.

Similar to other studies (which reported an HF incidence of approximately 18% in injured patients admitted to the emergency department), we observed that HF was an infrequent yet lethal fibrinolytic phenotype after injury.3,4 In this study, the incidence of HF was 20%, and these patients had an associated mortality of 20%. Preclinical and clinical studies suggest that HF likely results from an overwhelming release of tissue plasminogen activator (without a concomitant rise in circulating levels of its cognate inhibitor, plasminogen activator inhibitor-1) in patients with hemorrhagic shock.9,12,25–27 In support of the above, we observed that more patients with HF than other fibrinolytic phenotypes presented with lower BPs and were hypotensive. Moreover, nearly one half required an RBC transfusion and approximately one third required >10 U of RBCs in the first 24 hours. These observations confirm those reported by Moore et al.3 in 2014.

As in two previous studies of admission fibrinolytic phenotypes,3,4 SD was the most commonly observed phenotype after injury. Although the exact etiology of SD remains unknown, studies conducted in nonhuman primates suggest that substantial soft tissue and/or musculoskeletal injury promotes transition into an SD phenotype.13 In this study, more patients with SD presented after blunt trauma, which resulted in higher ISS scores in this group than in those with HF. These patients may have had a greater degree of soft tissue and/or musculoskeletal injury as blunt trauma is frequently associated with the dispersion of a large amount of energy to a greater body surface area than stab or gunshot wounds.28 These patients also had a relatively high incidence of coagulopathy, a finding previously reported in other studies.4 They were also more likely to receive an RBC, plasma, and platelet transfusion in the first 24 hours than those presenting with PHYS.

The current study demonstrated that approximately 70% of major trauma patients who present with SD remain in this phenotype at 24 hours and up to 120 hours postinjury. Adults presenting with HF, on the other hand, quickly died or transitioned into a PHYS or SD phenotype. In a prospective cohort study published in 2017, Leeper et al.20 similarly observed that while most children with SD upon admission to the ICU at the Children's Hospital of Pittsburgh remained in this phenotype during the 7 day study period, those admitted with HF transitioned into another phenotype in the first several days or died. Further, when a phenotypic change did occur in these patients, it was most common during the first 24 to 48 hours after injury.20 In the only other study to date that assessed whether fibrinolytic shutdown was persistent or transient, Meizoso et al.1 reported that up to 44% of injured adults who presented with SD upon admission to the ICU remain in this phenotype at 1 week post-ICU admission. Collectively, the above findings suggest that SD is a relatively persistent fibrinolytic phenotype after injury while HF often occurs transiently and is rapidly lethal unless corrected. Importantly, this finding may not be dependent on delivery of antifibrinolytics. In support of this, a recent cohort study reported that HF may be rapidly reversed in hypotensive trauma patients within 2 hours if successfully resuscitated, which appears related to release of large amounts of plasminogen activator inhibitor-1 during that time frame.29

The current multicenter study observed that a higher admission systolic BP, lower admission LY-30 percentage, and the transfusion of platelets in the first 24 hours were independently associated with persistent SD at 24 hours. These findings mirror closely those of Meizoso et al.,1 who reported that both LY-30 at admission and the transfusion of platelets during the first week of admission to the ICU were both associated with persistent SD 1 week after ICU admission. As patients presenting with lower LY-30 values at admission likely have more profound degrees of SD early after injury, we hypothesize that these patients may be more prone to the persistence of SD over time. If this is true, lower LY-30 values at admission may provide a method for identifying those at higher risk for persistent SD after injury. Further, as outlined by Meizoso et al.,1 the finding of an association between platelet transfusion and SD may be related to the prothrombotic effects of platelets.1

Defining persistent SD as SD at 24 hours was an accurate predictor of mortality beyond 24 hours. Moreover, patients who remained in SD at 24 hours in this study had an increased late mortality. These findings persisted after adjusting for potential confounding factors and are similar to those reported by both Leeper et al.20 and Meizoso et al.1 In the study by Leeper et al.,20 children who arrived in SD and remained in this phenotype in the days after ICU admission, as well as those who developed SD later during their ICU admission, had a higher risk of death or disability. Meizoso et al.1 also reported that persistent SD at 1 week after ICU admission was independently associated with increased odds of mortality. In this study and others, causes of death in the persistent SD group most often included TBI (and/or multisystem organ failure).1,3,4

This study has potential limitations. First, rTEGs were only obtained from patients who met the highest level of trauma activation during the study. Our findings may therefore not be generalizable to less severely injured patients. Second, although not all patients had rTEG measurements performed at all nine rTEG measurement time points, our findings of the temporal course of SD after injury were nearly identical between patients with complete versus incomplete rTEG data. Third, as we conducted a number of statistical tests, some of our findings may represent type I error. However, as our findings are similar to those reported by previous investigations, we believe them to be worth reporting. Fourth, although we found an association between persistent SD and TBI, we did not collect data on the type of TBI in this study, and differences may exist between TBI subtypes and development of SD or other fibrinolytic phenotypes. Finally, both HF and SD, while strongly associated with negative outcomes in trauma patients, may not actually be causative but instead represent surrogates of severe injury and disturbed physiology. Randomized controlled trials using presentation phenotypes to guide therapy are therefore needed before any definitive recommendations can be afforded.

Back to Top | Article Outline

CONCLUSION

Our findings suggest that almost 70% of major trauma patients who present with SD remain in this phenotype at 24 hours and up to 120 hours postinjury. Patients presenting with HF, on the other hand, quickly transition into an SD or PHYS phenotype or die within 24 hours of presentation. While early mortality is highest with the HF phenotype, persistent SD at 24 hours is associated with an elevated late mortality. These findings may have implications for improving injured patient prognostication, the design of future studies, and potentially the treatment of future trauma patients.

Back to Top | Article Outline

AUTHORSHIP

D.J.R. participated in the literature search. All authors participated in the study concept and design. B.A.C. participated in the acquisition of study funding. All authors participated in the data collection. D.J.R. and B.A.C. participated in the analysis of data. All authors participated in the interpretation of data analyses. D.J.R. participated in the drafting of the article. All authors participated in the critical revision of the article. B.A.C. participated in the study supervision.

Back to Top | Article Outline

ACKNOWLEDGMENTS

We would like to acknowledge personnel critical to the accomplishment of the current multicenter study for their assistance with this project: University of California, San Francisco group (Dr. Matthew E. Kutcher, Dr. Brittney J. Redick, Dr. Lucy Z. Kornblith, and Ms. Mary F. Nelson); Oregon Health Science University (Dr. Martin A. Schreiber and Ms. Samantha Underwood); and The University of Texas Health Science Center at Houston group (Dr. John B. Holcomb, Ms. Jeanette Podbielski BSN, and Mr. Timothy Welch). We also thank Dr. Craig N. Jenne for his assistance with creation of Figure 1 and the Figure in Supplementary Digital Content 1.

Back to Top | Article Outline

DISCLOSURE

This study was supported, in part, by Haemonetics Corporation (Braintree, Massachusetts). These funders had no role in the design or conduct of the study; collection, management, analysis, or interpretation of the data; or preparation, review, or approval of the resultant article. Dr. Cotton has served as a consultant for Haemonetics Corporation since 2015. The other authors have no conflicts of interest to declare.

Back to Top | Article Outline

REFERENCES

1. Meizoso JP, Karcutskie CA, Ray JJ, Namias N, Schulman CI, Proctor KG. Persistent fibrinolysis shutdown is associated with increased mortality in severely injured trauma patients. J Am Coll Surg. 2017;224(4):575–582.
2. Gonzalez E, Moore EE, Moore HB. Management of trauma-induced coagulopathy with thrombelastography. Crit Care Clin. 2017;33(1):119–134.
3. Moore HB, Moore EE, Gonzalez E, Chapman MP, Chin TL, Silliman CC, Banerjee A, Sauaia A. Hyperfibrinolysis, physiologic fibrinolysis, and fibrinolysis shutdown: the spectrum of postinjury fibrinolysis and relevance to antifibrinolytic therapy. J Trauma Acute Care Surg. 2014;77(6):811–817; discussion 817.
4. Moore HB, Moore EE, Liras IN, Gonzalez E, Harvin JA, Holcomb JB, Sauaia A, Cotton BA. Acute fibrinolysis shutdown after injury occurs frequently and increases mortality: a multicenter evaluation of 2,540 severely injured patients. J Am Coll Surg. 2016;222(4):347–355.
5. Davenport RA, Guerreiro M, Frith D, Rourke C, Platton S, Cohen M, Pearse R, Thiemermann C, Brohi K. Activated protein C drives the hyperfibrinolysis of acute traumatic coagulopathy. Anesthesiology. 2017;126(1):115–127.
6. Brohi K, Singh J, Heron M, Coats T. Acute traumatic coagulopathy. J Trauma. 2003;54(6):1127–1130.
7. MacLeod JB, Lynn M, McKenney MG, Cohn SM, Murtha M. Early coagulopathy predicts mortality in trauma. J Trauma. 2003;55(1):39–44.
8. Kutcher ME, Cripps MW, McCreery RC, Crane IM, Greenberg MD, Cachola LM, Redick BJ, Nelson MF, Cohen MJ. Criteria for empiric treatment of hyperfibrinolysis after trauma. J Trauma Acute Care Surg. 2012;73(1):87–93.
9. Schochl H, Cadamuro J, Seidl S, Franz A, Solomon C, Schlimp CJ, Ziegler B. Hyperfibrinolysis is common in out-of-hospital cardiac arrest: results from a prospective observational thromboelastometry study. Resuscitation. 2013;84(4):454–459.
10. Moore HB, Moore EE, Morton AP, Gonzalez E, Fragoso M, Chapman MP, Dzieciatkowska M, Hansen KC, Banerjee A, Sauaia A, et al. Shock-induced systemic hyperfibrinolysis is attenuated by plasma-first resuscitation. J Trauma Acute Care Surg. 2015;79(6):897–903; discussion 903–4.
11. Cotton BA, Harvin JA, Kostousouv V, Minei KM, Radwan ZA, Schochl H, Wade CE, Holcomb JB, Matijevic H. Hyperfibrinolysis at admission is an uncommon but highly lethal event associated with shock and prehospital fluid administration. J Trauma Acute Care Surg. 2012;73(2):365–370; discussion 370.
12. Chapman MP, Moore EE, Moore HB, Gonzalez E, Gamboni F, Chandler JG, Mitra S, Ghasabyan A, Chin TL, Sauaia A, et al. Overwhelming tPA release, not PAI-1 degradation, is responsible for hyperfibrinolysis in severely injured trauma patients. J Trauma Acute Care Surg. 2016;80(1):16–23; discussion 23–5.
13. Macko AR, Moore HB, Cap AP, Meledeo MA, Moore EE, Sheppard FR. Tissue injury suppresses fibrinolysis after hemorrhagic shock in nonhuman primates (rhesus macaque). J Trauma Acute Care Surg. 2017;82(4):750–757.
14. Gary JL, Schneider PS, Galpin M, Radwan Z, Munz JW, Achor TS, Prasarn ML, Cotton BA. Can thrombelastography predict venous thromboembolic events in patients with severe extremity trauma? J Orthop Trauma. 2016;30(6):294–298.
15. Pommerening MJ, Schwartz DA, Cohen MJ, Schreiber MA, del Junco DJ, Camp EA, Wade CE, Holcomb JB, Cotton BA. Hypercoagulability after injury in premenopausal females: a prospective, multicenter study. Surgery. 2014;156(2):439–447.
16. Pommerening MJ, Rahbar E, Minei K, Holcomb JB, Wade CE, Schreiber MA, Cohen MJ, Underwood SJ, Nelson M, Cotton BA. Splenectomy is associated with hypercoagulable thrombelastography values and increased risk of thromboembolism. Surgery. 2015;158(3):618–626.
17. von Elm E, Altman DG, Egger M, Pocock SJ, Gøtzsche PC, Vandenbroucke JP; STROBE Initiative. The strengthening the reporting of observational studies in epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61(4):344–349.
18. Holcomb JB, Minei KM, Scerbo ML, Radwan ZA, Wade CE, Kozar RA, Gill BS, Albarado R, McNutt MK, Khan S, et al. Admission rapid thrombelastography can replace conventional coagulation tests in the emergency department: experience with 1974 consecutive trauma patients. Ann Surg. 2012;256(3):476–486.
19. Cotton BA, Faz G, Hatch QM, Radwan ZA, Podbielski J, Wade C, Kozar RA, Holcomb JB. Rapid thrombelastography delivers real-time results that predict transfusion within 1 hour of admission. J Trauma. 2011;71(2):407–414; discussion 414–7.
20. Leeper CM, Neal MD, McKenna CJ, Gaines BA. Trending fibrinolytic dysregulation: fibrinolysis shutdown in the days after injury is associated with poor outcome in severely injured children. Ann Surg. 2017;266(3):508–515.
21. Leeper CM, Kutcher M, Nasr I, McKenna C, Billiar T, Neal M, Sperry J, Gaines BA. Acute traumatic coagulopathy in a critically injured pediatric population: definition, trend over time, and outcomes. J Trauma Acute Care Surg. 2016;81(1):34–41.
22. Folkerson LE, Sloan D, Davis E, Kitagawa RS, Cotton BA, Holcomb JB, Tomasek JS, Wade CE. Coagulopathy as a predictor of mortality after penetrating traumatic brain injury. Am J Emerg Med. 2018;36(1):38–42.
23. Vittinghoff E, Glidden DV, Shiboski SC, McCulloch CE. Chapter 10. Predictor Selection. In: Vittinghoff E, Glidden DV, Shiboski SC, McCulloch CE, eds. Regression Methods in Biostatistics: Linear, Logistic, Survival, and Repeated Measures Models. 2nd ed. Springer Science+Business Media: New York, NY; 2012:395–429.
24. Ware JH. Linear models for the analysis of longitudinal studies. Am Stat. 1985;39:95–101.
25. Cardenas JC, Matijevic N, Baer LA, Holcomb JB, Cotton BA, Wade CE. Elevated tissue plasminogen activator and reduced plasminogen activator inhibitor promote hyperfibrinolysis in trauma patients. Shock. 2014;41(6):514–521.
26. Brohi K, Cohen MJ, Ganter MT, Schultz MJ, Levi M, Mackersie RC, Pittet JF. Acute coagulopathy of trauma: hypoperfusion induces systemic anticoagulation and hyperfibrinolysis. J Trauma. 2008;64(5):1211–1217; discussion 1217.
27. Duvekot A, Viersen VA, Dekker SE, Geeraedts LM Jr., Schwarte LA, Spoelstra-Man AM, van de Ven PM, van den Brom CE, de Waard MC, Loer SA, et al. Low cerebral oxygenation levels during resuscitation in out-of-hospital cardiac arrest are associated with hyperfibrinolysis. Anesthesiology. 2015;123(4):820–829.
28. Mattox KL, Moore EE, Feliciano DV, eds. Trauma. China: McGraw Hill Companies Inc.; 2013:1224.
29. Moore HBME, Gonzalez E, Huebner BJ, Sheppard F, Banerjee A, Sauaia A, Silliman CC. Reperfusion shutdown: delayed onset of fibrinolysis resistance after resuscitation from hemorrhagic shock is associated with increased circulating levels of plasminogen activator inhibitor-1 and postinjury complications. Blood. 2016;128:206.
Keywords:

Fibrinolysis; fibrinolysis shutdown; hyperfibrinolysis; thromboelastography; wounds and injuries

Supplemental Digital Content

Back to Top | Article Outline
© 2019 Lippincott Williams & Wilkins, Inc.