Trauma remains the leading cause of death in both military and civilian populations younger than 45 years, with hemorrhagic shock accounting for half of these deaths (1–4). Many of these deaths are deemed potentially preventable with appropriate resuscitation and care. This requires not only an early recognition of hemorrhagic shock but also an effective and quick intervention to reverse the damages resulting from shock. Recent clinical studies, such as the Prospective Observational Multicenter Major Trauma Transfusion (PROMMTT) study, have shown that most deaths occur very early in severely injured trauma patients; 88% of deaths within the first hour were due to hemorrhage (5). Furthermore, among all hemorrhagic deaths observed in the PROMMTT study, 81% occurred within the first 6 h of admission (5). It is well known that death after hemorrhagic shock is strongly associated with hypoperfusion, tissue ischemia, cellular dysfunction, and microvascular collapse (6). Most efforts to reduce hemorrhagic deaths have focused on improvements to hemorrhage control devices and procedures; resuscitation methods, such as more efficient volume replacement; increasing blood oxygen to ischemic tissues; reduction of humoral and cellular inflammatory mediators; and correction of coagulopathy. Nonetheless, there remains no criterion standard for guiding resuscitation early after injury (6, 7). While blood lactate and base deficit values are generally reliable and have been used more frequently to guide resuscitation, they do not fully capture the status of a dynamic microcirculation (8–11). Thus, there is a need for expanding our knowledge of the microcirculatory status in severely injured patients, particularly those with hemorrhagic shock.
Hemorrhagic shock is defined as cellular dysfunction due to lack of perfusion in the microcirculation (12). Therefore, to correct hemorrhagic shock, one must restore both the macrocirculation and microcirculation. The restoration of the macrocirculation refers to the replacement of blood volume lost as well as restoring homeostatic vitals including blood pressure, heart rate, and cardiac output. Maintenance of arterial pressure and cardiac output by infusing sufficient volume of resuscitative fluids is vitally important for restoring tissue perfusion, but it does not always ensure adequate microcirculatory flow or tissue oxygenation. In addition, to replenishing blood volume and tissue oxygenation, restoration of the endothelial glycocalyx layer is also necessary for improved outcomes. Kozar et al. (13) demonstrated partial restoration of the glycocalyx layer with plasma resuscitation in a rodent model. Furthermore, Kozar et al. (13), Pati et al. (14), and Peng et al. (15) recently demonstrated an association between hyperpermeability, inflammation, and the loss of syndecan 1 from the endothelial glycocalyx layer. Although the exact role of the endothelial glycocalyx in hemorrhagic shock and resuscitation remains unknown, it is believed that maintaining the glycocalyx layer is necessary for improved vascular and inflammatory responses. Thus, it is critical to not only monitor the systemic responses to resuscitation, but also consider the numerous downstream effects on fluid balance at the microcirculation (i.e., Starling forces and permeability), coagulopathy, endothelial integrity, and inflammation, all which are extremely important in the management of hemorrhagic shock and resuscitation. The Starling forces refer to hydrostatic pressure, and colloid osmotic pressures (COPs), also known as oncotic pressure. The Starling forces along with cellular and vascular permeability are the driving forces regulating fluid balance at the microcirculation.
We performed a prospective observational study on trauma patients to evaluate the changes in the hemodynamics, plasma COP, serum protein, and syndecan 1 levels, as a measure of endothelial dysfunction, as well as measures of coagulopathy on patient outcomes. Given the effect of resuscitation on fluid balance and the endothelial glycocalyx layer, we hypothesized that changes in protein flux and oncotic pressure gradients could serve as an early predictor of poor patient outcomes and severity of injury and shock. We hypothesized that low plasma oncotic pressures would correlate with increased damage of the glycocalyx layer, resulting in increased transfusion requirements and poorer outcomes.
MATERIALS AND METHODS
This prospective observational study was conducted under an approved institutional review board (Universal Study, HSC-GEN-12-0059), which included all adult trauma patients (≥16 years) at the highest level of activation at Memorial Hermann Hospital Texas Medical Center (MHH-TMC). This was an opportunistic prospective study that ran parallel to three other clinical studies at MHH-TMC; patients admitted to the other studies were not included in this analysis. Pregnant women and prisoners were excluded from the study. Institutional review board approval was obtained for delayed consent, given the nature of trauma patients; therefore, consent was obtained from the patient or their legally authorized representative within 48 to 72 h of admission, or as soon as possible. If patients were discharged within 24 h of admission, a waiver of consent was obtained. Blood was collected from each patient in citrated tubes on admission. Samples from patients who refused consent were destroyed. Blood was also collected from 10 healthy volunteers to serve as a control group.
Osmolality, oncotic pressure, and serum protein
Blood samples were centrifuged, and plasma was stored for future use at −80°C. Osmolality and COP (i.e., oncotic pressure) were measured from fresh plasma at room temperature using the Wescor Vapro Osmometer (model 5520; Logan, Utah) and Wescor Colloid Osmometer (model 4420), respectively. Plasma COP (i.e., plasma oncotic pressure) was measured using the minimal volume method (∼125–150 µL of plasma), dictated in the Wescor manual. The membrane used in the colloid osmometer was impermeable to proteins exceeding 30,000 molecular weight. Both oncotic and osmolality measurements were run in duplicate. Serum protein was measured from fresh plasma at room temperature using a clinical refractometer (RHC-200/ATC; C&A Scientific, Manassas, Va), and these values were compared with the hospital laboratory values for total protein to ensure accuracy.
Thrombelastography (TEG, model TEG5000; Haemonetics, Braintree, Mass) and platelet function via impedance aggregometry (Multiplate; DiaPharma, West Chester, Ohio) were performed on all blood samples to characterize clot strength and platelet function. Standard tissue factor and kaolin activated rapid TEG was performed according to manufacturer’s instructions, as previously described (16–19). The TEG tracings are characterized by the following parameters: activated clotting time (ACT), split point (SP), r value, k time, α angle, maximum amplitude (MA), G value, and LY30. The r value (i.e., reaction time) or ACT is defined as the time from the start of the test and fibrin formation, which is representative of clotting factors (reference range, 0–118 s). The k time (i.e., coagulation time) starts shortly after the r value is calculated until the clot reaches a predetermined strength (usually 20 mm) and is increased with hypofibrinogenemia or platelet deficiency (reference range, 1–2 min). The SP is the short time between the end of r time and start of k time. The α angle is the slope of the tracing that represents the rate of clot formation and decreases with hypofibrinogenemia or platelet deficiency (reference range, 66–82 degrees). The MA is the greatest amplitude of the tracing and represents both fibrinogen and platelet contributions to clot strength (reference range, 54–72 mm). The G value (reference range, 5.3 × 103 to 12 × 103 dyn/cm2) is a measure of absolute clot strength, incorporating both enzymatic and platelet contributions, and is usually decreased in hypocoagulable states. Finally, LY30 is the percent amplitude reduction at 30 min after MA measurement and represents fibrinolysis (reference range, 0.0%–7.5%). Platelet function was assessed by Multiplate in response to the following five agonists: collagen, adenosine diphosphate, arachidonic acid, thrombin receptor-activating peptide, and ristocetin.
Commercial enzyme-linked immunosorbent assays for soluble syndecan 1 were performed as a measure of endothelial dysfunction (catalog ab46506; Abcam, Cambridge, England). Frozen citrated plasma samples were used for all enzyme-linked immunosorbent assays. All samples were run in duplicate. Samples that were over the detection range of the assay were diluted and rerun.
Clinical laboratory results from the time of admission were collected into a database for the consented patients. This included complete blood count test results, pH, base excess, admission vital signs, prehospital fluids, 24-h blood transfusions, 24-h crystalloid infusions, complications, Injury Severity Score (ISS), and patient outcomes. Patient demographics such as age, sex, race, type, and mechanism of injury were documented.
Our primary interest was to determine if admission plasma COP would be predictive of transfusion volume requirements. Therefore, the primary outcome was 24-h total blood units transfused. We also were interested in patient outcomes, such as complications, in-hospital mortality, and length of hospital stay. These were evaluated as part of a secondary analysis. Statistical significance was set at the 5% level. One-way analysis of variance and Kruskal-Wallis rank-sum tests were used for univariate analyses for subgroup comparisons. A purposeful regression approach was taken to adjust for necessary confounders (20). This included conducting pairwise correlations between all collected variables, and those that were both clinically and statistically significant (P < 0.05) were included in the regression model. Furthermore, checks for multicolinearity and interactions were made and included in the regression model. A negative binomial regression was used to assess the effect of plasma oncotic pressure on the number of units of blood products administered within 24 h, while adjusting for potential confounders and interactions.
A total of 556 patients at the highest activation were admitted to MHH-TMC between April and August 2012; 493 were 16 years or older; 107 patients consented to this prospective study, but only 104 had plasma oncotic pressure measurements (three were missing because of limited plasma volumes and thus excluded from the study). The overall patient demographics were consistent with highest-level-activation patients at most level I trauma centers (76% male, median age of 44 years, median ISS of 14, and an overall in-hospital mortality rate of 18%; Table 1).
Average serum protein levels were significantly lower in trauma patients than in healthy volunteers (control group), 6.1 ± 0.8 g/dL in trauma patients compared with 6.6 ± 0.4 g/dL (P < 0.05). Normal serum protein is generally considered between 6 and 8.5 g/dL. Therefore, 43% of our admitted trauma patients had low protein levels upon admission. Furthermore, average plasma COP was 17.7 ± 2.6 mmHg in trauma patients compared with 20.7 ± 2.1 mmHg in control subjects. We subgrouped patients into two categories based on their plasma oncotic pressures: low COP of 16.5 mmHg or less (n = 34) and normal COP of greater than 16.5 mmHg (n = 70). This cutoff for low COP was determined by the 2-SD mark from our healthy volunteers. It should be noted that there is no established universal cutoff for low plasma COP, but normal adults range between 20 and 26 mmHg (21). Using this cutoff, we observed that 32% of our cohort had reduced plasma oncotic pressures. Distributions of serum protein and plasma oncotic pressures from our cohort are illustrated in Figure 1. In addition, we observed a strong association between serum protein values and plasma oncotic pressure (R 2 = 0.7), where plasma COP equals approximately three times the serum protein value (Fig. 2).
Patient characteristics of each subgroup are listed in Table 1 in comparison to the overall population. Of note, patients with low plasma oncotic pressure were significantly more injured, demonstrating higher ISS despite having similar vital signs and weighted revised trauma scores (w-RTS). They were also more acidotic (greater base deficit and lower pH) and had significantly lower red blood cell (RBC) counts, lower hemoglobin, lower hematocrit, and fewer platelets compared with the normal COP group (Tables 1 and 2). Patients with low plasma oncotic pressure were significantly more likely to receive blood products, with a median of 4 units of total blood products within the first 24 h of admission compared with no blood products (0 units) in the normal COP group (Table 3). Total blood product was defined as the sum of RBC units, plasma units, and platelet units transfused. There were no significant differences in mortality across groups (Table 1). The causes of death were mainly head trauma/injuries (21%), hemorrhagic shock (16%), and multiorgan failure (11%). There were only two patients (<2%) in our cohort who suffered from acute respiratory distress syndrome (ARDS).
Syndecan 1 levels were significantly elevated in patients with low plasma oncotic pressure, 184.04 ± 383.89 ng/mL, compared to 52.41 ± 56.24 ng/mL in patients with normal COP (P = 0.03). We observed a significant pairwise correlation (inversely related) between plasma oncotic pressure and syndecan 1 (correlation coefficient, −0.3615; P = 0.0002), suggesting an exponential increase in syndecan 1 shedding with decreasing plasma oncotic pressure.
There were no significant differences in platelet function based on the Multiplate assays between both groups (data not shown). Rapid TEG values only revealed a significant increase in SP time and LY30 in the patients from the low COP group compared with the normal COP group (Table 4). This suggests that patients with low plasma oncotic pressure (i.e., hypoproteinemia) are more likely to be hyperfibrinolytic and may demonstrate higher levels of lysis.
Multivariable negative binomial regression revealed a significant effect of low plasma COP (incidence risk ratio of 4.73 with 95% confidence interval of 1.07–20.92) on total blood transfusion volumes within 24 h of admission, while adjusting for potential confounders (Table 5). This means that for patients with low plasma COP (≤16.5 mmHg) there is increased likelihood to receive blood product transfusions, including RBC and plasma products. It is important to note that we observed significant pairwise correlations between base excess and plasma COP values (r = 0.36, P = 0.0023), plasma COP and ISS values (r = −0.39, P = 0.0002), plasma COP and prehospital crystalloid (r = −0.48, P = 0.0274), and plasma COP and prehospital blood (r = −0.26, P = 0.0211). Despite the fact that plasma COP, base excess, and ISS are independent factors, all three of these factors tended to have strong associations between them. Therefore, we adjusted for these factors and their interactions with each other in the aforementioned multivariable negative binomial regression.
Fluid resuscitation after massive hemorrhage may result in extensive hemodilution and altered coagulopathy, which are multifactorial in nature. While resuscitation efforts have improved tremendously over the past decade, it is frequently difficult to decide when to start and stop resuscitation. Furthermore, there remains no universal standard or optimized resuscitation protocol or ideal resuscitative fluid; the answers to questions circling what resuscitative fluid to give and how much to give remain elusive. We believe the rudimentary understanding of the multifactorial nature of hemorrhagic shock and resuscitation is responsible for the disparities in resuscitation methods. In this article, we attempted to evaluate basic Starling forces of transcapillary fluid flux along with changes in coagulopathy and glycocalyx shedding in relation to transfusion requirements and patient outcomes in hopes to better understand the mechanisms of hemorrhagic shock and resuscitation.
Starling’s principle of fluid exchange states that transendothelial filtration is driven by capillary pressure (P c) and interstitial protein osmotic pressure (πi), while an opposing absorptive force is exerted by plasma protein oncotic pressure (πp) and interstitial pressure (Pi) (22, 23). Therefore, as a trauma patient bleeds and is then transfused with resuscitative fluids, one alters the balance of hydrostatic and oncotic pressures at the microcirculatory level. When infusing large volumes of crystalloids (e.g., normal saline or lactated Ringer’s solution) that have no oncotic potential, one observes a greater fluid imbalance often resulting in greater hemodilution and edema, whereas resuscitation with plasma, which has some inherent oncotic potential and other factors, has been shown to mitigate the development of pulmonary edema (21). These results are similar to the animal models resuscitated with hyperviscous and hyperoncotic solutions (24–27). Our cohort further supports this, where patients at our center in hemorrhagic shock are resuscitated using a 1:1:1 ratio of RBCs, plasma, and platelets with minimal crystalloids, and we observed only two cases of ARDS (28).
We observed a significant association between plasma COP and clinical outcomes in our cohort. Patients with low plasma COP at admission were significantly more injured and required longer hospital stays than patients with normal COP, despite having no significant differences in admission vital signs and w-RTSs. This was demonstrated by their significantly increased ISS, greater base deficits, and need of blood transfusions. These results support the notion of resuscitating the microcirculation and having a microcirculatory marker for hemorrhagic shock, because systemic vital signs were not different between groups. Similar to other reports, we observed significant associations between base excess, pH, and reduced plasma COP to poor patient outcomes, regardless of blood pressure or heart rate levels (8, 9, 11, 29). Furthermore, low plasma oncotic pressure strongly correlated with reduced serum protein levels (i.e., hypoproteinemia), exhibiting a 3:1 ratio where plasma COP is approximately three times serum protein. This convenient ratio could potentially be used to guide resuscitation in a manner such that emergency medical staff could quickly determine serum protein values via a clinical refractometer and multiply by 3 to determine an estimate for plasma COP, where a COP of 16.5 mmHg or less could potentially trigger the start of resuscitation, possibly in the prehospital setting.
In terms of coagulopathy, we did not observe any differences in clot strength, as indicated by the MA value from TEG or platelet function measured by impedance aggregometry. However, TEG showed significant differences in SP and LY30 (i.e., % lysis at 30 min) in patients with low COP compared with the normal COP group. This finding is somewhat intuitive because lysis involves the breakdown of proteins, and therefore hyperfibrinolytic patients would most likely demonstrate lower serum protein levels and consequently lower COP. Furthermore, we did not detect any significant differences in platelet function (via impedance aggregometry) across the groups, despite significantly lower platelet counts in patients with low plasma COP. This may be due to the fact that median platelet count was 200,000 cells/µL in our low plasma COP group, which is still within reference ranges; significant impairment in platelet function is generally observed when platelet counts drop below 150,000 cells/ µL (30). These data suggest that although trauma-induced coagulopathy is an important contributor to the progression of shock, plasma COP and base deficit are stronger early indicators of injury severity or transfusion requirements in this cohort.
Furthermore, as we suspected, decreasing plasma oncotic pressures strongly correlated with an exponential increase in shed syndecan 1. This supports our hypothesis that more of the glycocalyx layer is shed with increasing injury severity, in agreement with Johansson et al. (31). We were unable to verify if shed syndecan 1 levels decreased over time with fresh frozen plasma resuscitation because blood samples were collected only at the time of admission. Future work in this area could provide vital information regarding glycocalyx restoration during resuscitation in trauma patients. Other glycosaminoglycan proteins that could have been shed include heparan sulfate, chondroitin sulfate, and hyaluronic acid, which could provide additional information regarding the inflammatory and coagulation cascades after trauma; however, this was beyond the scope of this study. Nonetheless, we were able to demonstrate significant associations between plasma COP, syndecan 1 shedding, and clinical outcomes. We believe that this is the first step in fully characterizing the interplay between the hemodynamics, coagulopathy, and endothelial dysfunction as well as clinical outcomes during hemorrhagic shock and resuscitation.
One of the main goals of this study was to highlight the role of Starling forces and the microcirculation in the progression of hemorrhagic shock in severely injured trauma patients. A similar study was done by Lucas et al. (32, 33) showing reduced plasma oncotic pressure in injured patients, but it was limited by its focus on the fluid exchange only; no measurements of endothelial function or glycocalyx thickness were performed. Furthermore, although we did not measure interstitial fluid space volumes, our reduction in plasma proteins and COP were consistent with the findings of Lucas et al. (32–34), and we agree that hemorrhagic shock leads to a cellular insult, which consequently affects the intravascular and extravascular volumes and resuscitation needs. It is known that the microcirculation provides the greatest surface area between the endothelium, platelets, and other key signaling proteins involved in vascular permeability, coagulopathy, and inflammatory cascades after trauma (35, 36). Therefore, it is truly necessary to understand all three main responses and their interactions with one another during injury and resuscitation. While plasma COP was the only Starling force measured in this study, significant association with injury severity and transfusion requirements is still demonstrated. Hypoproteinemia in trauma patients was a significant indicator of injury severity and strongly correlated with blood transfusion volumes. These results closely parallel clinical studies that have shown hypoproteinemia to be the strongest indicator of pulmonary edema, acute lung injury, or ARDS in patients admitted to the ICU (37, 38). We were unable to demonstrate an association between hypoproteinemia, plasma COP, and ARDS because of the fact that only two patients developed ARDS in our cohort. We speculate that the low incidence of ARDS can be attributed to the relatively low volume of crystalloid resuscitation at our center (28). Future work can be done to measure the other Starling forces, such as capillary hydrostatic pressure, to better illustrate the imbalance in Starling forces during shock and resuscitation. It would also be of great benefit to perform a longitudinal study where we monitor patients over time during their hospital stay to optimize resuscitation strategies.
A limitation of this study was the inclusion of transfer patients and patients who received resuscitation prehospital. A total of 30 transfer patients were included in this study; nearly one-fourth (n = 8) of the patients from the low COP group received prehospital blood products, including RBCs and plasma, but it is interesting to note that their prehospital RBC or plasma transfusions did not restore their plasma COP, as they were still low upon emergency department admission. This may require us to reevaluate and consider using hyperoncotic and hypertonic resuscitative fluids for early resuscitation to correct the reduced COP in the early/acute setting. Even when these transfer patients were excluded, the trends between low plasma COP and injury severity and transfusion requirements remained statistically significant. Although we admit that measurement of plasma COP is not standard-of-care practice, its measurement in addition to other standard hospital laboratory values, such as base deficit, could potentially aid in guiding resuscitation in severely injured patients.
New approaches on the treatment of hemorrhagic shock with resuscitation must go beyond volume replacement, normalization of macrohemodynamic parameters, vital signs, and systemic oxygen transport. Resuscitation must mitigate the “lethal triad” of acidosis, hypothermia, and coagulopathy (39). In fact, this article suggests a possibly new triad of balancing the hemodynamics, inflammation, and coagulopathy at the microcirculatory level for improved outcomes after hemorrhagic shock. Although the concept of restoring the microcirculation via resuscitation is not new, using biomarkers that reflect the injury and repair of the endothelium or microvascular cues as predictors for guiding and monitoring resuscitation is fairly fresh. This article warrants further investigation of the Starling forces in the progression of hemorrhagic shock and resuscitation, where one could potentially use hypoproteinemia and/or reduced plasma COP (≤16.5 mmHg) in conjunction with base deficit for the initial indication of hemorrhagic shock and transfusion/resuscitation needs.
In this prospective study, we have shown that plasma COP and serum protein levels are significantly reduced in trauma patients upon admission. Reduced plasma COP (≤16.5 mmHg) was shown to be a strong indicator of injury severity and transfusion volume requirements, as well as endothelial dysfunction via increased syndecan 1 shedding, regardless of platelet function and altered coagulopathy. The data from this study shed light on the importance of the dynamics between Starling forces, vascular permeability via shedding of the endothelial glycocalyx layer, and resuscitation requirements.
1. Campbell BT, Saleheen H, Borrup K, McQuay J, Luk S, Hiscoe J, Lapidus G: Epidemiology of trauma at a level 1 trauma center. Conn Med 73 (7): 389–394, 2009.
2. Breeze J, Bryant D: Current concepts in the epidemiology and management of battlefield head, face and neck trauma. J R Army Med Corps 155 (4): 274–278, 2009.
3. Kauvar DS, Lefering R, Wade CE: Impact of hemorrhage on trauma outcome: an overview of epidemiology, clinical presentations, and therapeutic considerations. J Trauma 60 (Suppl 6): S3–S11, 2006.
4. Acosta JA, Yang JC, Winchell RJ, Simons RK, Fortlage DA, Hollingsworth-Fridlund P, Hoyt DB: Lethal injuries and time to death in a level I trauma center. J Am Coll Surg 186 (5): 528–533, 1998.
5. Holcomb JB, del Junco DJ, Fox EE, Wade CE, Cohen MJ, Schreiber MA, Alarcon LH, Bai Y, Brasel KJ, Bulger EM, et al.: The Prospective, Observational, Multicenter, Major Trauma Transfusion (PROMMTT) study: comparative effectiveness of a time-varying treatment with competing risks. Arch Surg 1–10, 2012.
6. Paes-da-Silva F, Gonzalez AP, Tibirica E: Effects of fluid resuscitation on mesenteric microvascular blood flow and lymphatic activity after severe hemorrhagic shock in rats. Shock 19 (1): 55–60, 2003.
7. Jarrar D, Chaudry IH, Wang P: Organ dysfunction following hemorrhage and sepsis: mechanisms and therapeutic approaches [review]. Int J Mol Med 4 (6): 575–583, 1999.
8. Ducrocq N, Kimmoun A, Levy B: LACTATE or ScvO2 as an endpoint in resuscitation of shock states? Minerva Anestesiol 79 (9): 1049–1058, 2013.
9. Vandromme MJ, Griffin RL, Weinberg JA, Rue LW 3rd, Kerby JD: Lactate is a better predictor than systolic blood pressure for determining blood requirement and mortality: could prehospital measures improve trauma triage? J Am Coll Surg 210 (5): 861–867–869, 867–869, 2010.
10. Pearse RM: Extending the role of lactate measurement into the prehospital environment. Crit Care 13 (1): 115, 2009.
11. Vincent JL: End-points of resuscitation: arterial blood pressure, oxygen delivery, blood lactate, or…? Intensive Care Med 22 (1): 3–5, 1996.
12. Mattox KL, Moore EE, Feliciano DV: Trauma. McGraw-Hill Professional, New York, 2012.
13. Kozar RA, Peng Z, Zhang R, Holcomb JB, Pati S, Park P, Ko TC, Paredes A: Plasma restoration of endothelial glycocalyx in a rodent model of hemorrhagic shock. Anesth Analg 112 (6): 1289–1295, 2011.
14. Pati S, Matijevic N, Doursout MF, Ko T, Cao Y, Deng X, Kozar RA, Hartwell E, Conyers J, Holcomb JB: Protective effects of fresh frozen plasma on vascular endothelial permeability, coagulation, and resuscitation after hemorrhagic shock are time dependent and diminish between days 0 and 5 after thaw. J Trauma 69 (Suppl 1): S55–S63, 2010.
15. Peng Z, Pati S, Potter D, Brown R, Holcomb JB, Grill R, Wataha K, Park PW, Xue H, Kozar RA: Fresh frozen plasma lessens pulmonary endothelial inflammation and hyperpermeability after hemorrhagic shock and is associated with loss of syndecan 1. Shock 40 (3): 195–202, 2013.
16. Vogel AM, Radwan ZA, Cox CS Jr, Cotton BA: Admission rapid thrombelastography delivers real-time “actionable” data in pediatric trauma. J Pediatr Surg 48 (6): 1371–1376, 2013.
17. Lee TH, McCully BH, Underwood SJ, Cotton BA, Cohen MJ, Schreiber MA: Correlation of conventional thrombelastography and rapid thrombelastography in trauma. Am J Surg 205 (5): 521–527; discussion 527, 2013.
18. Holcomb JB, Minei KM, Scerbo ML, Radwan ZA, Wade CE, Kozar RA, Gill BS, Albarado R, McNutt M, Khan S, et al.: Admission rapid thrombelastography can replace conventional coagulation tests in the emergency department: experience with 1974 consecutive trauma patients. Ann Surg 256 (3): 476–486, 2012.
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 71(2): 407–414; discussion 414–417, 2011.
20. Bursac Z, Gauss CH, Williams DK, Hosmer DW: Purposeful selection of variables in logistic regression. Source Code Biol Med 3: 17, 2008.
21. Morissette MP: Colloid osmotic pressure: its measurement and clinical value. Can Med Assoc J 116 (8): 897–900, 1977.
22. Levick JR, Michel CC: Microvascular fluid exchange and the revised Starling principle. Cardiovasc Res 87 (2): 198–210, 2010.
23. Starling EH: On the absorption of fluids from the connective tissue spaces. J Physiol 19: 312–326, 1896.
24. Cabrales P, Intaglietta M, Tsai AG: Increase plasma viscosity sustains microcirculation after resuscitation from hemorrhagic shock and continuous bleeding. Shock 23 (6): 549–555, 2005.
25. Cabrales P, Tsai AG, Intaglietta M: Hyperosmotic-hyperoncotic versus hyperosmotic-hyperviscous: small volume resuscitation in hemorrhagic shock. Shock 22 (5): 431–437, 2004.
26. Mazzoni MC, Borgstrom P, Arfors KE, Intaglietta M: Dynamic fluid redistribution in hyperosmotic resuscitation of hypovolemic hemorrhage. Am J Physiol 255 (3 Pt 2): H629–H637, 1988.
27. Wettstein R, Erni D, Intaglietta M, Tsai AG: Rapid restoration of microcirculatory blood flow with hyperviscous and hyperoncotic solutions lowers the transfusion trigger in resuscitation from hemorrhagic shock. Shock 25 (6): 641–646, 2006.
28. Cotton BA, Reddy N, Hatch QM, LeFebvre E, Wade CE, Kozar RA, Gill BS, Albarado R, McNutt MK, Holcomb JB: Damage control resuscitation is associated with a reduction in resuscitation volumes and improvement in survival in 390 damage control laparotomy patients. Ann Surg 254 (4): 598–605, 2011.
29. Permpikul C, Noppakaorattanamanee K, Tongyoo S, Ratanarat R, Poompichet A: Dynamics of central venous oxygen saturation and serum lactate during septic shock resuscitation. J Med Assoc Thai 96 (Suppl 2): S232–S237, 2013.
30. Stissing T, Dridi NP, Ostrowski SR, Bochsen L, Johansson PI: The influence of low platelet count on whole blood aggregometry assessed by Multiplate. Clin Appl Thromb Hemost 17 (6): E211–E217, 2011.
31. 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 254 (2): 194–200, 2011.
32. Lucas CE, Benishek DJ, Ledgerwood AM: Reduced oncotic pressure after shock: a proposed mechanism. Arch Surg 117 (5): 675–679, 1982.
33. Lucas CE, Ledgerwood AM, Rachwal WJ, Grabow D, Saxe JM: Colloid oncotic pressure and body water dynamics in septic and injured patients. J Trauma 31 (7): 927–931; discussion 931–933, 1991.
34. Lucas CE, Ledgerwood AM: Physiology of colloid-supplemented resuscitation from shock. J Trauma 54 (Suppl 5): S75–S81, 2003.
35. Ward KR: The microcirculation: linking trauma and coagulopathy. Transfusion 53 (Suppl 1): 38S–47S, 2013.
36. Granger DN, Kubes P: The microcirculation and inflammation: modulation of leukocyte-endothelial cell adhesion. J Leukoc Biol 55 (5): 662–675, 1994.
37. Arif SK, Verheij J, Groeneveld AB, Raijmakers PG: Hypoproteinemia as a marker of acute respiratory distress syndrome in critically ill patients with pulmonary edema. Intensive Care Med 28 (3): 310–317, 2002.
38. Mangialardi RJ, Martin GS, Bernard GR, Wheeler AP, Christman BW, Dupont WD, Higgins SB, Swindell BB: Hypoproteinemia predicts acute respiratory distress syndrome development, weight gain, and death in patients with sepsis. Ibuprofen in Sepsis Study Group. Crit Care Med 28 (9): 3137–3145, 2000.
39. Shaz BH, Dente CJ, Harris RS, MacLeod JB, Hillyer CD: Transfusion management of trauma patients. Anesth Analg 108 (6): 1760–1768, 2009.
Keywords:© 2014 by the Shock Society
Trauma; resuscitation; serum protein; Starling forces; syndecan 1