WTA 2014 PLENARY PAPER
Differences between blunt and penetrating trauma after resuscitation with hydroxyethyl starch Casey J. Allen, MD, Evan J. Valle, MD, Jassin M. Jouria, MD, Carl I. Schulman, MD, PhD, Nicholas Namias, MD, Alan S. Livingstone, MD, and Kenneth G. Proctor, PhD, Miami, Florida
The purpose of this study was to test the hypothesis that a single bolus of 6% hydroxyethyl starch (HES 450/0.7 in lactated electrolyte injection) during initial resuscitation has a differential effect in blunt and penetrating trauma patients. METHODS: Consecutive admissions to the trauma service were reviewed. Patients who died within 24 hours were excluded. Multivariate analysis defined individual predictors for the primary outcomes, acute kidney injury (AKI) and mortality within 90 days. Data were expressed as mean T SD, and significance was assessed at p G 0.05. RESULTS: There were 1,410 patients (76% male; mean T SD, age 43 T 18 years; 68% blunt trauma; mean T SD Injury Severity Score [ISS] 14 T 11; AKI, 4.4%; and mortality, 3.4%). HES (0.5Y1.5 L) was administered to 216 patients (15.3%). After multiple logistic regression, HES remained a significant independent predictor of AKI after blunt trauma (odds ratio [OR], 2.54; 95% confidence interval [CI], 1.24Y5.19; area under the receiver operating characteristic curve [AUROC], 0.809) but not penetrating trauma (OR, 0.90; 95% CI, 0.23Y3.60; AUROC, 0.849). In separate logistic regression models, HES was a significant predictor of mortality after blunt trauma (OR, 3.77; 95% CI, 0.91Y0.97; AUROC, 0.921) but not penetrating trauma (OR, 0.72; 95% CI, 0.13Y3.94; AUROC, 0.904). CONCLUSION: HES is an independent risk factor for AKI and death after blunt, but not penetrating, trauma, which underscores a fundamental difference between these two injury types. (J Trauma Acute Care Surg. 2014;77: 859Y864. Copyright * 2014 by Lippincott Williams & Wilkins) LEVEL OF EVIDENCE: Epidemiologic study, level III. KEY WORDS: Hextend; acute kidney injury; mechanism of injury. BACKGROUND:
T
rauma is broadly classified by the mechanism of injury (MOI). Depending on the site and amount of energy transfer, blunt and penetrating forces produce different characteristic injury patterns and often require different therapeutic approaches. Nevertheless, Advance Trauma Life Support (ATLS) guidelines currently recommend crystalloids as the first-line therapy for hypovolemic shock, regardless of MOI.1,2 Hydroxyethyl starches (HES) are occasionally used for volume expansion in civilian practice, but virtually every randomized controlled trial (RCT) has failed to find a survival benefit of any colloid(includingvariousHESformulations)versuscrystalloids.3,4 For battlefield fluid resuscitation, the Department of Defense Committee on Tactical Combat Casualty Care recommends a half-liter bolus of HES5 in hypotensive or obtunded patients if blood products are not immediately available.6Y8 We confirmed short-term salutary effects when small-volume HES was used in a civilian Level 1 center, particularly after penetrating trauma.9,10 Submitted: January 15, 2014, Revised: April 3, 2014, Accepted: June 6, 2014, Published online: September 22, 2014. From the Divisions of Trauma, Surgical Critical Care, Dewitt-Daughtry Department of Surgery, University of Miami Miller School of Medicine, Miami, Florida. Portions of these data were presented at the 2013 Florida Committee on Trauma and at the 44th Annual Meeting of the Western Trauma Association, March 2Y7, 2014, in Steamboat Springs, Colorado. Address for reprints: Kenneth G. Proctor, PhD, Divisions of Trauma and Surgical Critical Care, Daughtry Family Department of Surgery, University of Miami School of Medicine, Ryder Trauma Center, 1800 NW 10th Ave, Miami, FL 33136; email: [email protected]. DOI: 10.1097/TA.0000000000000422
In June 2013, the US Food and Drug Administration (FDA) issued a black box warning on the use of HES based on reports of increased rates of mortality and acute kidney injury (AKI) after large-volume resuscitation in septic, critically ill medical patients.11Y13 For this reason, because of the inherent differences between blunt and penetrating injuries and because of the differences in volume and time administered between trauma and medical patients, we tested the hypothesis that a one-time, low-volume HES bolus during initial resuscitation has a differential effect in blunt and penetrating trauma patients.
PATIENTS AND METHODS The University of Miami and Jackson Memorial Hospital Institutional Review Board approved this retrospective study with waiver of informed consent. The data registry of the Ryder Trauma Center, an academic tertiary care regional referral center, was queried for all patients admitted to the trauma surgery service between October 1, 2011, and December 31, 2012. Patients who were admitted to the floor, intensive care unit (ICU), or operating room were included. Patients admitted to orthopedics or neurosurgery, younger than 18 years, with known kidney disease, pregnant, incarcerated, and/or died within 24 hours of arrival were excluded. Demographics, Injury Severity Score (ISS), admission vital signs and laboratory values, MOI, initial diagnostic imaging, and outcome variables including survival were reviewed. Patients that received 6% HES 450/0.7 in a lactated electrolyte injection (Hextend, http://www.hospira.com/) and/or blood
J Trauma Acute Care Surg Volume 77, Number 6
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
859
J Trauma Acute Care Surg Volume 77, Number 6
Allen et al.
RESULTS
TABLE 1. Study Population Age, y Sex Male Female Race Latin White Black MOI Blunt Penetrating ISS TBI Disposition Floor ICU Operating room Transfusion HES AKI Mortality
43 T 18 76% 24% 45% 23% 32% 68% 32% 14 T 11 14.0% 43% 25% 32% 12.4% 15.3% 4.4% 3.4%
products within 24 hours after arrival to the hospital were identified. The variables included patient age, initial systolic blood pressure (SBP), heart rate (HR), hematocrit (Hct), ISS, presence of traumatic brain injury (TBI), Glasgow Coma Scale (GCS) score, requirement of blood transfusion during initial resuscitation, need for surgery within 24 hours of arrival or direct ICU admission, and the use of HES. TBI was defined by standard DRG International Classification of DiseasesV9th Rev. coding, documented mental status, and/or evidence of intracranial hemorrhage on computed tomography (CT). AKI was defined as a rise in creatinine (CR) more than 1.5 times above the baseline level for at least 48 consecutive hours, according to RIFLE [Risk, Injury, Failure, Loss, and End stage] and AKIN [Acute Kidney Injury Network] criteria.14,15 The practice among our group of experienced trauma surgeons is relatively consistent, but no fixed resuscitation protocols were in place during the study period. The general treatment strategy is to resuscitate according to ATLS guidelines at physician’s discretion. The use of HES as a one-time intravenous (IV) bolus during initial trauma resuscitation, although not part of standard ATLS guidelines, was adopted at our institution because of the evidence of benefit.9,10 Populations were divided into blunt and penetrating injury MOIs. Univariate analysis was performed to determine differences in demographics, injury severity, physiologic variables, laboratory values, and the categorical outcomes of AKI and mortality. Fisher’s exact tests were used for categorical variables, and Student’s t tests were used for continuous variables. To control for the confounding variables, multiple logistic regression models were used to identify the significant predictors for death or AKI within 90 days of admission. Variables were added to the multiple logistic regression if p G 0.2. For all other comparisons, p G 0.05 was considered statistically significant. 860
Table 1 describes a study population of 1,410 consecutive trauma admissions characterized as 76% male, mean T SD age of 43 T 18 years, 68% blunt force injuries, ISS of 14 T 11, with a 4.4% rate of AKI and 3.4% rate of mortality (excluding deaths G 24 hours). The causes of death (n = 48) include pulmonary embolus (Days 25Y50) in 4 patients, multiorgan failure (Days 5Y80) in 29 patients, and TBI (Days 2Y7) in 15 patients. HES was administered to 216 patients (15.3%). The median dose of HES within the first 24 hours was 500 mL (range, 250Y1,500 mL). Although a sevenfold dose range exists, the vast majority of patients received 500 mL. Of the 216 patients that received HES, only 3 patients (1%) received G 500 mL, 40 patients (19%) received 9 500 mL, and 173 patients (80%) received 500 mL. For this reason, a dose response could not be adequately assessed. Table 2 compares those who received HES with those who did not. The rate of AKI was four times higher (12.1% vs. 3.0%, p G 0.01) and the mortality rate was six times higher (11.5% vs. 1.9%, p G 0.01) in those that received HES. Those who received HES were older, with lower initial SBP, higher initial HR, lower Hct, lower GCS score, and received more transfusions (all p G 0.01). These factors (alone or in combination) may have accounted for increased rates of AKI and mortality, so that multivariate analysis was used to determine the independent predictors. For blunt MOI, AKI and mortality rates were 2.6% and 1.9% in those that did not receive HES and 14.4% (p G 0.01) and 12.9% (p G 0.01) in those that received HES, respectively. For penetrating MOI, AKI and mortality rates were 3.8% and 1.9% without HES and 6.6% and 8.1% with HES, respectively (p = nonsignificant [NS] and p G 0.01). There was no significant difference between blunt and penetrating MOIs in the proportion of patients that received
TABLE 2. Initial Physiologic and Laboratory Values of HES and no-HES Groups mean T SD Age, y SBP, mm Hg HR, beats/min Hct, % GCS score ISS Blunt injury Initial Cr, mg/dL Disposition Floor ICU Operating room AKI Day of AKI Transfusion TBI Hemodialysis Mortality
No HES, n = 1,194
HES, n = 216
P
42 T 18 138 T 30 93 T 20 40 T 6 14 T 3 13 T 10 68% 1.01 T 0.6
46 T 19 112 T 34 105 T 24 37 T 6 12 T 4 21 T 12 70% 1.08 T 0.38
0.012 G0.01 G0.01 G0.01 G0.01 G0.01 NS NS
48.1% 22.6% 29.4% 3.0% 9 T 11 6.9% 12.0% 1.2% 1.9%
16.7% 39.7% 43.5% 12.1% 9 T 10 42.6% 26.3% 2.9% 11.5%
G0.01
G0.01 NS G0.01 G0.01 NS G0.01
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 77, Number 6
Allen et al.
HES (16% vs. 14%, p = NS). However, blunt trauma patients were more than twice as likely to receive either a CT with IV contrast (87% vs. 41%, p G 0.001) or both HES and a CT (14% vs. 6%, p G 0.001). Table 3 shows that regardless of MOI, AKI was associated with Hct, GCS score, ISS, and transfusion (all p G 0.05). Hypovolemic shock (as reflected by lower SBP in penetrating trauma [p = 0.017] and higher HR in blunt trauma [p G 0.01]) was also associated with AKI. Furthermore, more than 90% of the cases developed in patients transferred to the ICU or operating room after resuscitation (both p G 0.02). It is noteworthy that HES was significantly associated with AKI after blunt trauma (p G 0.01), but not after penetrating trauma. Table 3 highlights the factors associated with mortality after blunt and penetrating trauma. Regardless of MOI, mortality was associated with SBP, HR, Hct, GCS score, ISS, HES, and transfusion (all p G 0.05). To control for the variables significantly associated with mortality or AKI, multiple logistic regression models were used to identify the significant independent predictors. Figure 1 shows
Figure 1. Independent predictors of AKI, results of multiple logistic regression; conventional indicators of injury severity. Blunt and penetrating populations. Variables run in each regression model include age (y), SBP (mm Hg), HR (beats/min), Hct (%), GCS, ISS, disposition, HES use, and transfusion use.
the odds ratio (OR) for various predictors of AKI. Figure 2 shows the corresponding data for various predictors of mortality.
TABLE 3. Factors Associated With AKI and Mortality, Univariate Analysis, Blunt and Penetrating Populations AKI Blunt mean T SD Age, SD, y SBP, SD, mm Hg HR, beats/min Hct, % GCS score, ISS Initial Cr, mg/dL Disposition Floor ICU Operating room HES Transfusion
Penetrating
No AKI
AKI
P
No AKI
AKI
p
45 T 18 138 T 30 94 T 21 40 T 6 14 T 3 16 T 11 1.0 T 0.6
49 T 17 128 T 43 108 T 22 37 T 8 10 T 5 29 T 15 1 T 0.2
NS NS G0.01 0.011 G0.01 G0.01 NS
36 T 14 129 T 31 96 T 20 39 T 6 14 T 2 10 T 9 1.1 T 0.5
39 T 17 109 T 44 99 T 25 34 T 8 11 T 5 22 T 10 1.1 T 0.4
NS 0.017 NS G0.01 G0.01 G0.01 NS
48% 34% 18% 14% 10%
7% 44% 49% 51% 27%
G0.01
37% 8% 55% 14% 16%
6% 6% 89% 22% 39%
G0.01 G0.01
0.016
NS 0.011
Mortality Blunt Age, y SBP, mm Hg HR, beats/min Hct, % GCS score, SD ISS Initial Cr, mg/dL Disposition Floor ICU Operating room HES Transfusion
Penetrating
Survival 45 T 18 138 T 31 94 T 21 40 T 6 14 T 3 15 T 10 1.0 T 0.6
Death 47 T 23 117 T 43 110 T 28 37 T 7 7T5 34 T 15 1.1 T 0.6
P NS G0.01 G0.01 G0.01 G0.01 G0.01 NS
Survival 36 T 14 129 T 31 96 T 20 39 T 6 14 T 2 10 T 9 1.1 T 0.5
Death 49 T 17 94 T 38 101 T 28 35 T 6 8T5 23 T 11 1.4 T 0.8
P 0.02 G0.01 G0.01 0.05 G0.01 G0.01 0.04
48% 32% 20% 14% 9%
3% 68% 29% 56% 44%
G0.01
37% 8% 55% 13% 16%
8% 8% 83% 42% 50%
G0.01
G0.01 G0.01
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
G0.01 G0.01
861
J Trauma Acute Care Surg Volume 77, Number 6
Allen et al.
These conventional indicators of injury severity (SBP, Hct, ISS, and GCS score) have consistent values for predicting the respective outcome, regardless of MOI. In contrast, the effect of HES on AKI or mortality depends on the MOI. Within the same set of logistic regression models, HES was a significant independent predictor of mortality in the blunt population (OR, 3.77; 95% CI, 1.63Y8.69; area under the receiver operating characteristic curve [AUROC], 0.921) but not the penetrating population (OR, 0.72; 95% CI, 0.13Y3.94; AUROC, 0.904), and HES remained a significant predictor of AKI in the blunt population (OR, 2.54; 95% CI, 1.24Y5.19; AUROC, 0.809) but not the penetrating population (OR, 0.90; 95% CI, 0.23Y3.60; AUROC, 0.849). Figure 3 underscores the point that HES has no significant risk on either mortality or AKI in penetrating trauma patients but significantly increases the risk of both AKI and mortality threefold to fivefold in blunt patients (both p G 0.05). Figure 3 also illustrates how, after multiple logistic regression to control for injury severity and initial hemodynamic status, blood transfusion does not have significant risk, rather trends toward a benefit.
DISCUSSION The new findings from this study are that HES use during initial trauma resuscitation increases the risk of AKI and death after blunt injury, but no harmful effects were observed in the penetrating population. HES is used at the same time following injury, in the same quantity, and for the same purpose, yet there is clear evidence of harm within only the blunt population. This underscores a fundamental difference between blunt and penetrating injury. We previously reviewed 1,714 consecutive trauma admissions and observed that overall mortality (early + late deaths) in patients who received an HES bolus was significantly lower compared with those who received only crystalloids.9 A major criticism of that study was that only the univariate, and not the
Figure 2. Independent predictors of mortality, results of multiple logistic regression; conventional indicators of injury severity. Blunt and penetrating populations. Variables run in each regression model include age (y), SBP (mm Hg), HR (beats/min), Hct (%), GCS, ISS, disposition, HES use, and transfusion use. 862
Figure 3. Independent predictive value of HES use on AKI and mortality, results of multiple logistic regression. Blunt and penetrating populations. Variables run in each regression model include age (y), SBP (mm Hg), HR (beats/min), Hct (%), GCS, ISS, disposition, HES use, and transfusion use.
multivariate, analysis was significant. Moreover, the inclusion of early deaths may have caused a selection bias. In a follow-up study, we attempted to reconcile these issues. The data showed that in those with penetrating injury, despite having a higher HR, worse base deficit, and lower Hct, HES significantly reduced transfusion requirements and improved 24-hour urine output and fluid balance when compared with the standard of care.10 In contrast, HES had no detectable effect after blunt trauma.10 We are not the first to demonstrate a differential effect of HES in blunt versus penetrating trauma. James et al.3 performed a prospective RCT comparing crystalloid versus HES on renal function. They found that in those who sustained penetrating trauma, AKI occurred significantly less frequently and lactate clearance was significantly better with HES versus the standard of care. No benefits were seen in the blunt trauma patients.3 Lissauer et al.16 retrospectively evaluated the long-term effects of HES on renal function and mortality in trauma patients. They found a significant association of AKI and mortality in those who received HES. However, their population overwhelmingly consisted of blunt injury (85%), and no differences between blunt and penetrating trauma population were addressed. Regardless, the Pharmacovigilance Risk Assessment Committee in the European Union concluded that there is a clear indication of harm when HES is used for fluid resuscitation with no evidence of a greater benefit when compared with the standard of care in a wide array of patient groups.17 Because the risks HES poses to patients were considered to outweigh the benefits in all clinical settings, the use of HES was suspended in the United Kingdom in 2013.17 Recently, the US FDA convened a public workshop in collaboration with the National Heart, Lung, and Blood Institute at the National Institutes of Health, the US Army Medical Materiel Command, Department of Defense, and the Office of the Assistant Secretary of Health to discuss the risks and benefits of HES solutions. Panelists presented data from RCTs, metaanalyses, and observational studies that showed increased mortality and/or renal injury requiring renal replacement therapy * 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 77, Number 6
when HES was used in certain critically ill adult patients.11Y13 The FDA concluded that HES should not be used in some patient populations. In June 2013, a black box warning was issued to include the risk of mortality and severe renal injury.18 When considered in context, these present results act as a bridge between the studies showing long-term risks4,11Y13,16 and the few showing potential benefit in particular surgical subpopulations.3,9,10 The increased risk of AKI and death in the blunt, but not the penetrating, population is consistent with both schools of thought regarding HES. Although there is a global risk of using HES, numerous studies directly addressing the trauma population show short-term benefit within the penetrating population alone. This present study is the first to evaluate the difference of long-term effects of HES use related to MOI. There is always a balance of risk and benefit with any medication or treatment intervention; the risk-to-benefit ratio of using HES seems to shift with respect to the blunt and penetrating trauma populations. Hopefully, this basic observation will not be considered as simply one more data point comparing colloid with crystalloid. At least 32 RCTs have compared albumin with crystalloid solutions in patients with hypovolemia or burns, with no evidence that albumin reduces mortality when compared with cheaper alternatives.19 There have been at least 55 RCTs comparing colloids (including 16 trials with various forms of HES) with crystalloids and no evidence that resuscitation with colloids reduces the risk of death compared with crystalloids.4 No single-center retrospective review can properly address specific drug safety or efficacy issues that were already highlighted in multicenter RCTs. Rather, data from this present study underscore the fundamental differences between blunt and penetrating injury. We have no unequivocal explanation for the difference between blunt and penetrating injury or for how these results reveal that difference, but we can propose a plausible mechanism based on available evidence. First, there is no question that HES is nephrotoxic in some conditions.11,13,16 Second, blunt force injury can cause a significant release of nephrotoxic metabolites, such as myoglobin and creatine kinase,20 and can evoke proportional increases in serum interleukin 6 and interleukin 8, which are significantly associated with ICU stay, hospital stay, infection rate, systemic inflammatory response syndrome, multiple-organ failure score, and mortality.21,22 Third, blunt trauma patients are much more likely to obtain a CTwith nephrotoxic IV contrast upon initial workup when compared with that of penetrating patients. Therefore, it is logical to postulate that the blunt population is inherently more at risk to potential nephrotoxicities, with the exposure to HES as possibly the factor that ultimately induces clinically significant AKI. Even after adjusting for baseline severity of illness, case mix, race, sex, and age, AKI is independently associated with increased mortality.23,24 Although we will not postulate why other studies have shown a benefit of HES use in only the penetrating population, it is plausible that this early benefit may prevent some of the long-term risks association. So, in effect, there is a risk-tobenefit ratio with the use of any intervention, and we propose this balance may shift with the use of HES in the blunt and penetrating populations. There are several limitations to this study. This is a single-institution retrospective review, and therefore, baseline
Allen et al.
differences between groups could not be directly controlled. Since those who received HES tended to be more critically injured, there was an association of harm within the treatment group. However, these associations are also seen within those that received a transfusion, and no risk was seen within this group. Moreover, there remains a clear discrepancy of harm between the blunt and penetrating populations. Given the adverse effect profile of HES, it seems plausible that these differences are related to MOI. Moreover, resuscitative strategies were based on physician discretion instead of a standard resuscitation protocol. Given the sample size and that many physicians likely follow similar ATLS guidelines and resuscitate to similar hemodynamic parameters, these differences are likely minimal. Long term follow-up is another potential weakness. We have only in-hospital outcomes, so we would not have records to reflect those who might have developed organ dysfunction or died outside the hospital. Still, patients discharged are hemodynamically stable and, thus, unlikely to experience organ dysfunction or death. Moreover, TBI was defined by DRG International Classification of DiseasesV9th Rev. coding, but despite this, a diagnostic discrepancy between the subpopulations is unlikely to exist. Finally, dividing the population into injury groups may weaken the study strength by creating smaller samples sizes for statistical analysis. Regardless, the regression models were always strongly predictive, and the risk factors remained internally consistent with the exception to HES use between the blunt and the penetrating populations. In summary, within the context of some limitations, the data suggest that after blunt trauma, HES independently increases risks of AKI and death, whereas after penetrating trauma, there was no evidence of risk. This finding reveals a fundamental difference between blunt and penetrating injuries, but further study is necessary to understand how the MOI interacts with the biologic action of HES. ACKNOWLEDGMENT We thank the nursing and administrative staff at Ryder Trauma Center for their cooperation and assistance with the patients and their families. AUTHORSHIP C.J.A. is directly responsible for all aspects of this study. He participated in the collection, analysis, and interpretation of data as well as drafting and revision of the manuscript, figures, and tables. E.J.V. and J.M.J. participated in the experimental design, collection of data, as well as revision of the manuscript, figures, and tables. C.I.S., N.N., and A.S.L. were medically responsible for the patients and participated in the review and revision of the manuscript, figures, and tables. K.G.P. had overall responsibility for the study, including conception and experimental design; analysis and interpretation of data; drafting and revision of the manuscript, figures, and tables; statistical expertise and evaluation; obtaining funding for this project; and supervision. DISCLOSURE This study was supported by Grants #N140610670 from the Office of Naval Research and #09078015 from US Army Medical Research and Materiel Command.
REFERENCES 1. American College of Surgeons. Advanced Trauma Life Support for Doctors ATLS: Manuals for Coordinators and Faculty. 8th ed. Chicago, IL: American College of Surgeons; 2008.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
863
J Trauma Acute Care Surg Volume 77, Number 6
Allen et al.
2. Krug EG, Sharma GK, Lozano R. The global burden of injuries. Am J Public Health. 2000;90(4):523Y526. 3. James MF, Michell WL, Joubert IA, Nicol AJ, Navsaria PH, Gillespie RS. Resuscitation with hydroxyethyl starch improves renal function and lactate clearance in penetrating trauma in a randomized controlled study: the FIRST trial (Fluids in Resuscitation of Severe Trauma). Br J Anaesth. 2011;107(5):693Y702. 4. Perel P, Roberts I, Ker K. Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Database Syst Rev. 2013;2:CD000567. 5. Tactical Combat Casualty Care Guidelines. September 17, 2012. Available at: http://www.health.mil/Libraries/120917_TCCC_Course_Materials/ TCCC-Guidelines-120917.pdf. Accessed January 10, 2014. 6. Butler F. Fluid resuscitation in tactical combat casualty care: brief history and current status. J Trauma. 2011;70(Suppl 5):S11YS12. 7. Holcomb JB. Fluid resuscitation in modern combat casualty care: lessons learned from Somalia. J Trauma. 2003;54(Suppl 5):S46YS51. 8. Butler FKJr., Holcomb JB, Giebner SD, McSwain NE, Bagian J. Tactical combat casualty care 2007: evolving concepts and battlefield experience. Mil Med. 2007;172(Suppl 11):1Y19. 9. Ogilvie MP, Pereira BM, McKenney MG, McMahon PJ, Manning RJ, Namias N, Livingstone AS, Schulman CI, Proctor KG. First report on safety and efficacy of hetastarch solution for initial fluid resuscitation at a level 1 trauma center. J Am Coll Surg. 2010;210(5):870Y880, discussion 880Y882. 10. Ryan ML, Ogilvie MP, Pereira BM, Gomez-Rodriguez JC, Livingstone AS, Proctor KG. Effect of hetastarch bolus in trauma patients requiring emergency surgery. J Spec Oper Med. 2012;12(3):57Y67. ˚ neman 11. Perner A, Haase N, Guttormsen AB, Tenhunen J, Klemenzson G, A A, Madsen KR, Møller MH, Elkj&r JM, Poulsen LM, et al.; 6S Trial Group; Scandinavian Critical Care Trials Group. Hydroxyethyl starch 130/ 0.42 versus Ringer’s acetate in severe sepsis. N Engl J Med. 2012;367(2): 124Y134. Epub 2012 Jun 27. Erratum in: N Engl J Med. 2012 Aug 2; 367(5):481. 12. Guidet B, Martinet O, Boulain T, Philippart F, Poussel JF, Maizel J, Forceville X, Feissel M, Hasselmann M, Heininger A, et al. Assessment of hemodynamic efficacy and safety of 6% hydroxyethylstarch 130/0.4 vs. 0.9% NaCl fluid replacement in patients with severe sepsis: the CRYSTMAS study. Crit Care. 2012;16(3):R94. 13. Myburgh JA, Finfer S, Bellomo R, Billot L, Cass A, Gattas D, Glass P, Lipman J, Liu B, McArthur C, McGuinness S, Rajbhandari D, Taylor CB, Webb SA; CHEST Investigators; Australian and New Zealand Intensive Care Society Clinical Trials Group. Hydroxyethyl starch or saline for fluid resuscitation in intensive care. N Engl J Med. 2012;367(20):1901Y1911. 14. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P; Acute Dialysis Quality Initiative workgroup. Acute renal failure - definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care. 2004;8(4):R204YR212. 15. Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, Levin A; Acute Kidney Injury Network. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care. 2007;11(2):R31. 16. Lissauer ME, Chi A, Kramer ME, Scalea TM, Johnson SB. Association of 6% hetastarch resuscitation with adverse outcomes in critically ill trauma patients. Am J Surg. 2011;202(1):53Y58. 17. PRAC confirms that hydroxyethyl-starch solutions (HES) should no longer be used in patients with sepsis or burn injuries or in critically ill patients. European Medicines Agency. Press release October 11, 2013. Available at: http://www.ema.europa.eu/docs/en_GB/document_library/Press_release/ 2013/10/WC500151964.pdf. Accessed January 10, 2014. 18. FDA Safety Communication: Boxed Warning on increased mortality and severe renal injury, and additional warning on risk of bleeding, for use of hydroxyethyl starch solutions in some settings. Press release November 25, 2013. Available at: http://www.fda.gov/biologicsbloodvaccines/safetyavailability/ ucm358271.htm. Accessed January 10, 2014. 19. Liberati A, Moja L, Moschetti I, Gensini GF, Gusinu R. Human albumin solution for resuscitation and volume expansion in critically ill patients. Intern Emerg Med. 2006;1(3):243Y245.
864
20. Subramanian A, Sukheeja D, Trikha V, Pandey A, Albert V, Pandey R. Evaluation of serum creatine kinase and urinary myoglobin as markers in detecting development of acute renal failure in severely injured trauma patients. ISRN Emerg Med. 2013. Article ID 241036, 8 pages. Available at http://dx.doi.org/10.1155/2013/241036. 21. Strecker W, Gebhard F, Perl M, Rager J, Buttenscho¨n K, Kinzl L, Beck A. Biochemical characterization of individual injury pattern and injury severity. Injury. 2003;34(12):879Y887. 22. Strecker W, Gebhard F, Rager J, Bru¨ckner UB, Steinbach G, Kinzl L. Early biochemical characterization of soft-tissue trauma and fracture trauma. J Trauma. 1999;47(2):358Y364. 23. Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, Kellum JA. RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care. 2006;10(3):R73. 24. Hoste EA, Kellum JA. RIFLE criteria provide robust assessment of kidney dysfunction and correlate with hospital mortality. Crit Care Med. 2006; 34(7):2016Y2017.
EDITORIAL CRITIQUE The discussion about the use of colloids for the care of the critically ill or injured dates back decades. While there are a large number of manuscripts on the subject, the issue still remains unresolved. Surgeons vociferously defend one or the other as superior. Those espousing colloids as better often prefer the synthetic colloids as they are less expensive and perhaps safer. In this article, the group from Miami advocate for the use of hydroxyethyl starch (HES) in the resuscitation of patients with penetrating injury. There are many limitations to this study, virtually all of which are noted in the manuscript. The authors explanation for their results seems plausible. All shock is not the same. The tissue crush injury, particularly muscle crush injury that accompanies severe blunt trauma, may put patients at risk for acute renal failure, differently than they are with pure hemorrhage. The nephrotoxicity of HES may then be amplified in such patients. Our understanding of resuscitation continues to become more refined. While some have used colloids for initial trauma resuscitation, crystalloids use is much more common. Those of us who have been around a while remember the days of the use of high volume crystalloid resuscitation based on belief that only a relatively small portion remained intravascular. We then migrated to a more restricted crystalloid restriction strategy using permissive hypotension. Patients with significant hemorrhage are now given blood and plasma early, though the optimal ratios remain poorly defined. Throughout this period of time, HES has gotten a black eye, as many reports have linked its use with renal failure and death. This report from the University of Miami should make us step back and re-evaluate. I, for one, am not willing to embrace the widespread use of HES, even for penetrating trauma, based on this single study. This data is controversial. It will raise questions. The debate about colloids versus crystalloids will again be re-fueled. This would be a perfect topic for a large prospective trial. Thomas M. Scalea, MD R Adams Cowley Shock Trauma Center University of Maryland School of Medicine Baltimore, Maryland
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
Differences between blunt and penetrating trauma after resuscitation with hydroxyethyl starch Casey J. Allen, MD, Evan J. Valle, MD, Jassin M. Jouria, MD, Carl I. Schulman, MD, PhD, Nicholas Namias, MD, Alan S. Livingstone, MD, and Kenneth G. Proctor, PhD, Miami, Florida
The purpose of this study was to test the hypothesis that a single bolus of 6% hydroxyethyl starch (HES 450/0.7 in lactated electrolyte injection) during initial resuscitation has a differential effect in blunt and penetrating trauma patients. METHODS: Consecutive admissions to the trauma service were reviewed. Patients who died within 24 hours were excluded. Multivariate analysis defined individual predictors for the primary outcomes, acute kidney injury (AKI) and mortality within 90 days. Data were expressed as mean T SD, and significance was assessed at p G 0.05. RESULTS: There were 1,410 patients (76% male; mean T SD, age 43 T 18 years; 68% blunt trauma; mean T SD Injury Severity Score [ISS] 14 T 11; AKI, 4.4%; and mortality, 3.4%). HES (0.5Y1.5 L) was administered to 216 patients (15.3%). After multiple logistic regression, HES remained a significant independent predictor of AKI after blunt trauma (odds ratio [OR], 2.54; 95% confidence interval [CI], 1.24Y5.19; area under the receiver operating characteristic curve [AUROC], 0.809) but not penetrating trauma (OR, 0.90; 95% CI, 0.23Y3.60; AUROC, 0.849). In separate logistic regression models, HES was a significant predictor of mortality after blunt trauma (OR, 3.77; 95% CI, 0.91Y0.97; AUROC, 0.921) but not penetrating trauma (OR, 0.72; 95% CI, 0.13Y3.94; AUROC, 0.904). CONCLUSION: HES is an independent risk factor for AKI and death after blunt, but not penetrating, trauma, which underscores a fundamental difference between these two injury types. (J Trauma Acute Care Surg. 2014;77: 859Y864. Copyright * 2014 by Lippincott Williams & Wilkins) LEVEL OF EVIDENCE: Epidemiologic study, level III. KEY WORDS: Hextend; acute kidney injury; mechanism of injury. BACKGROUND:
T
rauma is broadly classified by the mechanism of injury (MOI). Depending on the site and amount of energy transfer, blunt and penetrating forces produce different characteristic injury patterns and often require different therapeutic approaches. Nevertheless, Advance Trauma Life Support (ATLS) guidelines currently recommend crystalloids as the first-line therapy for hypovolemic shock, regardless of MOI.1,2 Hydroxyethyl starches (HES) are occasionally used for volume expansion in civilian practice, but virtually every randomized controlled trial (RCT) has failed to find a survival benefit of any colloid(includingvariousHESformulations)versuscrystalloids.3,4 For battlefield fluid resuscitation, the Department of Defense Committee on Tactical Combat Casualty Care recommends a half-liter bolus of HES5 in hypotensive or obtunded patients if blood products are not immediately available.6Y8 We confirmed short-term salutary effects when small-volume HES was used in a civilian Level 1 center, particularly after penetrating trauma.9,10 Submitted: January 15, 2014, Revised: April 3, 2014, Accepted: June 6, 2014, Published online: September 22, 2014. From the Divisions of Trauma, Surgical Critical Care, Dewitt-Daughtry Department of Surgery, University of Miami Miller School of Medicine, Miami, Florida. Portions of these data were presented at the 2013 Florida Committee on Trauma and at the 44th Annual Meeting of the Western Trauma Association, March 2Y7, 2014, in Steamboat Springs, Colorado. Address for reprints: Kenneth G. Proctor, PhD, Divisions of Trauma and Surgical Critical Care, Daughtry Family Department of Surgery, University of Miami School of Medicine, Ryder Trauma Center, 1800 NW 10th Ave, Miami, FL 33136; email: [email protected]. DOI: 10.1097/TA.0000000000000422
In June 2013, the US Food and Drug Administration (FDA) issued a black box warning on the use of HES based on reports of increased rates of mortality and acute kidney injury (AKI) after large-volume resuscitation in septic, critically ill medical patients.11Y13 For this reason, because of the inherent differences between blunt and penetrating injuries and because of the differences in volume and time administered between trauma and medical patients, we tested the hypothesis that a one-time, low-volume HES bolus during initial resuscitation has a differential effect in blunt and penetrating trauma patients.
PATIENTS AND METHODS The University of Miami and Jackson Memorial Hospital Institutional Review Board approved this retrospective study with waiver of informed consent. The data registry of the Ryder Trauma Center, an academic tertiary care regional referral center, was queried for all patients admitted to the trauma surgery service between October 1, 2011, and December 31, 2012. Patients who were admitted to the floor, intensive care unit (ICU), or operating room were included. Patients admitted to orthopedics or neurosurgery, younger than 18 years, with known kidney disease, pregnant, incarcerated, and/or died within 24 hours of arrival were excluded. Demographics, Injury Severity Score (ISS), admission vital signs and laboratory values, MOI, initial diagnostic imaging, and outcome variables including survival were reviewed. Patients that received 6% HES 450/0.7 in a lactated electrolyte injection (Hextend, http://www.hospira.com/) and/or blood
J Trauma Acute Care Surg Volume 77, Number 6
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
859
J Trauma Acute Care Surg Volume 77, Number 6
Allen et al.
RESULTS
TABLE 1. Study Population Age, y Sex Male Female Race Latin White Black MOI Blunt Penetrating ISS TBI Disposition Floor ICU Operating room Transfusion HES AKI Mortality
43 T 18 76% 24% 45% 23% 32% 68% 32% 14 T 11 14.0% 43% 25% 32% 12.4% 15.3% 4.4% 3.4%
products within 24 hours after arrival to the hospital were identified. The variables included patient age, initial systolic blood pressure (SBP), heart rate (HR), hematocrit (Hct), ISS, presence of traumatic brain injury (TBI), Glasgow Coma Scale (GCS) score, requirement of blood transfusion during initial resuscitation, need for surgery within 24 hours of arrival or direct ICU admission, and the use of HES. TBI was defined by standard DRG International Classification of DiseasesV9th Rev. coding, documented mental status, and/or evidence of intracranial hemorrhage on computed tomography (CT). AKI was defined as a rise in creatinine (CR) more than 1.5 times above the baseline level for at least 48 consecutive hours, according to RIFLE [Risk, Injury, Failure, Loss, and End stage] and AKIN [Acute Kidney Injury Network] criteria.14,15 The practice among our group of experienced trauma surgeons is relatively consistent, but no fixed resuscitation protocols were in place during the study period. The general treatment strategy is to resuscitate according to ATLS guidelines at physician’s discretion. The use of HES as a one-time intravenous (IV) bolus during initial trauma resuscitation, although not part of standard ATLS guidelines, was adopted at our institution because of the evidence of benefit.9,10 Populations were divided into blunt and penetrating injury MOIs. Univariate analysis was performed to determine differences in demographics, injury severity, physiologic variables, laboratory values, and the categorical outcomes of AKI and mortality. Fisher’s exact tests were used for categorical variables, and Student’s t tests were used for continuous variables. To control for the confounding variables, multiple logistic regression models were used to identify the significant predictors for death or AKI within 90 days of admission. Variables were added to the multiple logistic regression if p G 0.2. For all other comparisons, p G 0.05 was considered statistically significant. 860
Table 1 describes a study population of 1,410 consecutive trauma admissions characterized as 76% male, mean T SD age of 43 T 18 years, 68% blunt force injuries, ISS of 14 T 11, with a 4.4% rate of AKI and 3.4% rate of mortality (excluding deaths G 24 hours). The causes of death (n = 48) include pulmonary embolus (Days 25Y50) in 4 patients, multiorgan failure (Days 5Y80) in 29 patients, and TBI (Days 2Y7) in 15 patients. HES was administered to 216 patients (15.3%). The median dose of HES within the first 24 hours was 500 mL (range, 250Y1,500 mL). Although a sevenfold dose range exists, the vast majority of patients received 500 mL. Of the 216 patients that received HES, only 3 patients (1%) received G 500 mL, 40 patients (19%) received 9 500 mL, and 173 patients (80%) received 500 mL. For this reason, a dose response could not be adequately assessed. Table 2 compares those who received HES with those who did not. The rate of AKI was four times higher (12.1% vs. 3.0%, p G 0.01) and the mortality rate was six times higher (11.5% vs. 1.9%, p G 0.01) in those that received HES. Those who received HES were older, with lower initial SBP, higher initial HR, lower Hct, lower GCS score, and received more transfusions (all p G 0.01). These factors (alone or in combination) may have accounted for increased rates of AKI and mortality, so that multivariate analysis was used to determine the independent predictors. For blunt MOI, AKI and mortality rates were 2.6% and 1.9% in those that did not receive HES and 14.4% (p G 0.01) and 12.9% (p G 0.01) in those that received HES, respectively. For penetrating MOI, AKI and mortality rates were 3.8% and 1.9% without HES and 6.6% and 8.1% with HES, respectively (p = nonsignificant [NS] and p G 0.01). There was no significant difference between blunt and penetrating MOIs in the proportion of patients that received
TABLE 2. Initial Physiologic and Laboratory Values of HES and no-HES Groups mean T SD Age, y SBP, mm Hg HR, beats/min Hct, % GCS score ISS Blunt injury Initial Cr, mg/dL Disposition Floor ICU Operating room AKI Day of AKI Transfusion TBI Hemodialysis Mortality
No HES, n = 1,194
HES, n = 216
P
42 T 18 138 T 30 93 T 20 40 T 6 14 T 3 13 T 10 68% 1.01 T 0.6
46 T 19 112 T 34 105 T 24 37 T 6 12 T 4 21 T 12 70% 1.08 T 0.38
0.012 G0.01 G0.01 G0.01 G0.01 G0.01 NS NS
48.1% 22.6% 29.4% 3.0% 9 T 11 6.9% 12.0% 1.2% 1.9%
16.7% 39.7% 43.5% 12.1% 9 T 10 42.6% 26.3% 2.9% 11.5%
G0.01
G0.01 NS G0.01 G0.01 NS G0.01
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 77, Number 6
Allen et al.
HES (16% vs. 14%, p = NS). However, blunt trauma patients were more than twice as likely to receive either a CT with IV contrast (87% vs. 41%, p G 0.001) or both HES and a CT (14% vs. 6%, p G 0.001). Table 3 shows that regardless of MOI, AKI was associated with Hct, GCS score, ISS, and transfusion (all p G 0.05). Hypovolemic shock (as reflected by lower SBP in penetrating trauma [p = 0.017] and higher HR in blunt trauma [p G 0.01]) was also associated with AKI. Furthermore, more than 90% of the cases developed in patients transferred to the ICU or operating room after resuscitation (both p G 0.02). It is noteworthy that HES was significantly associated with AKI after blunt trauma (p G 0.01), but not after penetrating trauma. Table 3 highlights the factors associated with mortality after blunt and penetrating trauma. Regardless of MOI, mortality was associated with SBP, HR, Hct, GCS score, ISS, HES, and transfusion (all p G 0.05). To control for the variables significantly associated with mortality or AKI, multiple logistic regression models were used to identify the significant independent predictors. Figure 1 shows
Figure 1. Independent predictors of AKI, results of multiple logistic regression; conventional indicators of injury severity. Blunt and penetrating populations. Variables run in each regression model include age (y), SBP (mm Hg), HR (beats/min), Hct (%), GCS, ISS, disposition, HES use, and transfusion use.
the odds ratio (OR) for various predictors of AKI. Figure 2 shows the corresponding data for various predictors of mortality.
TABLE 3. Factors Associated With AKI and Mortality, Univariate Analysis, Blunt and Penetrating Populations AKI Blunt mean T SD Age, SD, y SBP, SD, mm Hg HR, beats/min Hct, % GCS score, ISS Initial Cr, mg/dL Disposition Floor ICU Operating room HES Transfusion
Penetrating
No AKI
AKI
P
No AKI
AKI
p
45 T 18 138 T 30 94 T 21 40 T 6 14 T 3 16 T 11 1.0 T 0.6
49 T 17 128 T 43 108 T 22 37 T 8 10 T 5 29 T 15 1 T 0.2
NS NS G0.01 0.011 G0.01 G0.01 NS
36 T 14 129 T 31 96 T 20 39 T 6 14 T 2 10 T 9 1.1 T 0.5
39 T 17 109 T 44 99 T 25 34 T 8 11 T 5 22 T 10 1.1 T 0.4
NS 0.017 NS G0.01 G0.01 G0.01 NS
48% 34% 18% 14% 10%
7% 44% 49% 51% 27%
G0.01
37% 8% 55% 14% 16%
6% 6% 89% 22% 39%
G0.01 G0.01
0.016
NS 0.011
Mortality Blunt Age, y SBP, mm Hg HR, beats/min Hct, % GCS score, SD ISS Initial Cr, mg/dL Disposition Floor ICU Operating room HES Transfusion
Penetrating
Survival 45 T 18 138 T 31 94 T 21 40 T 6 14 T 3 15 T 10 1.0 T 0.6
Death 47 T 23 117 T 43 110 T 28 37 T 7 7T5 34 T 15 1.1 T 0.6
P NS G0.01 G0.01 G0.01 G0.01 G0.01 NS
Survival 36 T 14 129 T 31 96 T 20 39 T 6 14 T 2 10 T 9 1.1 T 0.5
Death 49 T 17 94 T 38 101 T 28 35 T 6 8T5 23 T 11 1.4 T 0.8
P 0.02 G0.01 G0.01 0.05 G0.01 G0.01 0.04
48% 32% 20% 14% 9%
3% 68% 29% 56% 44%
G0.01
37% 8% 55% 13% 16%
8% 8% 83% 42% 50%
G0.01
G0.01 G0.01
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
G0.01 G0.01
861
J Trauma Acute Care Surg Volume 77, Number 6
Allen et al.
These conventional indicators of injury severity (SBP, Hct, ISS, and GCS score) have consistent values for predicting the respective outcome, regardless of MOI. In contrast, the effect of HES on AKI or mortality depends on the MOI. Within the same set of logistic regression models, HES was a significant independent predictor of mortality in the blunt population (OR, 3.77; 95% CI, 1.63Y8.69; area under the receiver operating characteristic curve [AUROC], 0.921) but not the penetrating population (OR, 0.72; 95% CI, 0.13Y3.94; AUROC, 0.904), and HES remained a significant predictor of AKI in the blunt population (OR, 2.54; 95% CI, 1.24Y5.19; AUROC, 0.809) but not the penetrating population (OR, 0.90; 95% CI, 0.23Y3.60; AUROC, 0.849). Figure 3 underscores the point that HES has no significant risk on either mortality or AKI in penetrating trauma patients but significantly increases the risk of both AKI and mortality threefold to fivefold in blunt patients (both p G 0.05). Figure 3 also illustrates how, after multiple logistic regression to control for injury severity and initial hemodynamic status, blood transfusion does not have significant risk, rather trends toward a benefit.
DISCUSSION The new findings from this study are that HES use during initial trauma resuscitation increases the risk of AKI and death after blunt injury, but no harmful effects were observed in the penetrating population. HES is used at the same time following injury, in the same quantity, and for the same purpose, yet there is clear evidence of harm within only the blunt population. This underscores a fundamental difference between blunt and penetrating injury. We previously reviewed 1,714 consecutive trauma admissions and observed that overall mortality (early + late deaths) in patients who received an HES bolus was significantly lower compared with those who received only crystalloids.9 A major criticism of that study was that only the univariate, and not the
Figure 2. Independent predictors of mortality, results of multiple logistic regression; conventional indicators of injury severity. Blunt and penetrating populations. Variables run in each regression model include age (y), SBP (mm Hg), HR (beats/min), Hct (%), GCS, ISS, disposition, HES use, and transfusion use. 862
Figure 3. Independent predictive value of HES use on AKI and mortality, results of multiple logistic regression. Blunt and penetrating populations. Variables run in each regression model include age (y), SBP (mm Hg), HR (beats/min), Hct (%), GCS, ISS, disposition, HES use, and transfusion use.
multivariate, analysis was significant. Moreover, the inclusion of early deaths may have caused a selection bias. In a follow-up study, we attempted to reconcile these issues. The data showed that in those with penetrating injury, despite having a higher HR, worse base deficit, and lower Hct, HES significantly reduced transfusion requirements and improved 24-hour urine output and fluid balance when compared with the standard of care.10 In contrast, HES had no detectable effect after blunt trauma.10 We are not the first to demonstrate a differential effect of HES in blunt versus penetrating trauma. James et al.3 performed a prospective RCT comparing crystalloid versus HES on renal function. They found that in those who sustained penetrating trauma, AKI occurred significantly less frequently and lactate clearance was significantly better with HES versus the standard of care. No benefits were seen in the blunt trauma patients.3 Lissauer et al.16 retrospectively evaluated the long-term effects of HES on renal function and mortality in trauma patients. They found a significant association of AKI and mortality in those who received HES. However, their population overwhelmingly consisted of blunt injury (85%), and no differences between blunt and penetrating trauma population were addressed. Regardless, the Pharmacovigilance Risk Assessment Committee in the European Union concluded that there is a clear indication of harm when HES is used for fluid resuscitation with no evidence of a greater benefit when compared with the standard of care in a wide array of patient groups.17 Because the risks HES poses to patients were considered to outweigh the benefits in all clinical settings, the use of HES was suspended in the United Kingdom in 2013.17 Recently, the US FDA convened a public workshop in collaboration with the National Heart, Lung, and Blood Institute at the National Institutes of Health, the US Army Medical Materiel Command, Department of Defense, and the Office of the Assistant Secretary of Health to discuss the risks and benefits of HES solutions. Panelists presented data from RCTs, metaanalyses, and observational studies that showed increased mortality and/or renal injury requiring renal replacement therapy * 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
J Trauma Acute Care Surg Volume 77, Number 6
when HES was used in certain critically ill adult patients.11Y13 The FDA concluded that HES should not be used in some patient populations. In June 2013, a black box warning was issued to include the risk of mortality and severe renal injury.18 When considered in context, these present results act as a bridge between the studies showing long-term risks4,11Y13,16 and the few showing potential benefit in particular surgical subpopulations.3,9,10 The increased risk of AKI and death in the blunt, but not the penetrating, population is consistent with both schools of thought regarding HES. Although there is a global risk of using HES, numerous studies directly addressing the trauma population show short-term benefit within the penetrating population alone. This present study is the first to evaluate the difference of long-term effects of HES use related to MOI. There is always a balance of risk and benefit with any medication or treatment intervention; the risk-to-benefit ratio of using HES seems to shift with respect to the blunt and penetrating trauma populations. Hopefully, this basic observation will not be considered as simply one more data point comparing colloid with crystalloid. At least 32 RCTs have compared albumin with crystalloid solutions in patients with hypovolemia or burns, with no evidence that albumin reduces mortality when compared with cheaper alternatives.19 There have been at least 55 RCTs comparing colloids (including 16 trials with various forms of HES) with crystalloids and no evidence that resuscitation with colloids reduces the risk of death compared with crystalloids.4 No single-center retrospective review can properly address specific drug safety or efficacy issues that were already highlighted in multicenter RCTs. Rather, data from this present study underscore the fundamental differences between blunt and penetrating injury. We have no unequivocal explanation for the difference between blunt and penetrating injury or for how these results reveal that difference, but we can propose a plausible mechanism based on available evidence. First, there is no question that HES is nephrotoxic in some conditions.11,13,16 Second, blunt force injury can cause a significant release of nephrotoxic metabolites, such as myoglobin and creatine kinase,20 and can evoke proportional increases in serum interleukin 6 and interleukin 8, which are significantly associated with ICU stay, hospital stay, infection rate, systemic inflammatory response syndrome, multiple-organ failure score, and mortality.21,22 Third, blunt trauma patients are much more likely to obtain a CTwith nephrotoxic IV contrast upon initial workup when compared with that of penetrating patients. Therefore, it is logical to postulate that the blunt population is inherently more at risk to potential nephrotoxicities, with the exposure to HES as possibly the factor that ultimately induces clinically significant AKI. Even after adjusting for baseline severity of illness, case mix, race, sex, and age, AKI is independently associated with increased mortality.23,24 Although we will not postulate why other studies have shown a benefit of HES use in only the penetrating population, it is plausible that this early benefit may prevent some of the long-term risks association. So, in effect, there is a risk-tobenefit ratio with the use of any intervention, and we propose this balance may shift with the use of HES in the blunt and penetrating populations. There are several limitations to this study. This is a single-institution retrospective review, and therefore, baseline
Allen et al.
differences between groups could not be directly controlled. Since those who received HES tended to be more critically injured, there was an association of harm within the treatment group. However, these associations are also seen within those that received a transfusion, and no risk was seen within this group. Moreover, there remains a clear discrepancy of harm between the blunt and penetrating populations. Given the adverse effect profile of HES, it seems plausible that these differences are related to MOI. Moreover, resuscitative strategies were based on physician discretion instead of a standard resuscitation protocol. Given the sample size and that many physicians likely follow similar ATLS guidelines and resuscitate to similar hemodynamic parameters, these differences are likely minimal. Long term follow-up is another potential weakness. We have only in-hospital outcomes, so we would not have records to reflect those who might have developed organ dysfunction or died outside the hospital. Still, patients discharged are hemodynamically stable and, thus, unlikely to experience organ dysfunction or death. Moreover, TBI was defined by DRG International Classification of DiseasesV9th Rev. coding, but despite this, a diagnostic discrepancy between the subpopulations is unlikely to exist. Finally, dividing the population into injury groups may weaken the study strength by creating smaller samples sizes for statistical analysis. Regardless, the regression models were always strongly predictive, and the risk factors remained internally consistent with the exception to HES use between the blunt and the penetrating populations. In summary, within the context of some limitations, the data suggest that after blunt trauma, HES independently increases risks of AKI and death, whereas after penetrating trauma, there was no evidence of risk. This finding reveals a fundamental difference between blunt and penetrating injuries, but further study is necessary to understand how the MOI interacts with the biologic action of HES. ACKNOWLEDGMENT We thank the nursing and administrative staff at Ryder Trauma Center for their cooperation and assistance with the patients and their families. AUTHORSHIP C.J.A. is directly responsible for all aspects of this study. He participated in the collection, analysis, and interpretation of data as well as drafting and revision of the manuscript, figures, and tables. E.J.V. and J.M.J. participated in the experimental design, collection of data, as well as revision of the manuscript, figures, and tables. C.I.S., N.N., and A.S.L. were medically responsible for the patients and participated in the review and revision of the manuscript, figures, and tables. K.G.P. had overall responsibility for the study, including conception and experimental design; analysis and interpretation of data; drafting and revision of the manuscript, figures, and tables; statistical expertise and evaluation; obtaining funding for this project; and supervision. DISCLOSURE This study was supported by Grants #N140610670 from the Office of Naval Research and #09078015 from US Army Medical Research and Materiel Command.
REFERENCES 1. American College of Surgeons. Advanced Trauma Life Support for Doctors ATLS: Manuals for Coordinators and Faculty. 8th ed. Chicago, IL: American College of Surgeons; 2008.
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
863
J Trauma Acute Care Surg Volume 77, Number 6
Allen et al.
2. Krug EG, Sharma GK, Lozano R. The global burden of injuries. Am J Public Health. 2000;90(4):523Y526. 3. James MF, Michell WL, Joubert IA, Nicol AJ, Navsaria PH, Gillespie RS. Resuscitation with hydroxyethyl starch improves renal function and lactate clearance in penetrating trauma in a randomized controlled study: the FIRST trial (Fluids in Resuscitation of Severe Trauma). Br J Anaesth. 2011;107(5):693Y702. 4. Perel P, Roberts I, Ker K. Colloids versus crystalloids for fluid resuscitation in critically ill patients. Cochrane Database Syst Rev. 2013;2:CD000567. 5. Tactical Combat Casualty Care Guidelines. September 17, 2012. Available at: http://www.health.mil/Libraries/120917_TCCC_Course_Materials/ TCCC-Guidelines-120917.pdf. Accessed January 10, 2014. 6. Butler F. Fluid resuscitation in tactical combat casualty care: brief history and current status. J Trauma. 2011;70(Suppl 5):S11YS12. 7. Holcomb JB. Fluid resuscitation in modern combat casualty care: lessons learned from Somalia. J Trauma. 2003;54(Suppl 5):S46YS51. 8. Butler FKJr., Holcomb JB, Giebner SD, McSwain NE, Bagian J. Tactical combat casualty care 2007: evolving concepts and battlefield experience. Mil Med. 2007;172(Suppl 11):1Y19. 9. Ogilvie MP, Pereira BM, McKenney MG, McMahon PJ, Manning RJ, Namias N, Livingstone AS, Schulman CI, Proctor KG. First report on safety and efficacy of hetastarch solution for initial fluid resuscitation at a level 1 trauma center. J Am Coll Surg. 2010;210(5):870Y880, discussion 880Y882. 10. Ryan ML, Ogilvie MP, Pereira BM, Gomez-Rodriguez JC, Livingstone AS, Proctor KG. Effect of hetastarch bolus in trauma patients requiring emergency surgery. J Spec Oper Med. 2012;12(3):57Y67. ˚ neman 11. Perner A, Haase N, Guttormsen AB, Tenhunen J, Klemenzson G, A A, Madsen KR, Møller MH, Elkj&r JM, Poulsen LM, et al.; 6S Trial Group; Scandinavian Critical Care Trials Group. Hydroxyethyl starch 130/ 0.42 versus Ringer’s acetate in severe sepsis. N Engl J Med. 2012;367(2): 124Y134. Epub 2012 Jun 27. Erratum in: N Engl J Med. 2012 Aug 2; 367(5):481. 12. Guidet B, Martinet O, Boulain T, Philippart F, Poussel JF, Maizel J, Forceville X, Feissel M, Hasselmann M, Heininger A, et al. Assessment of hemodynamic efficacy and safety of 6% hydroxyethylstarch 130/0.4 vs. 0.9% NaCl fluid replacement in patients with severe sepsis: the CRYSTMAS study. Crit Care. 2012;16(3):R94. 13. Myburgh JA, Finfer S, Bellomo R, Billot L, Cass A, Gattas D, Glass P, Lipman J, Liu B, McArthur C, McGuinness S, Rajbhandari D, Taylor CB, Webb SA; CHEST Investigators; Australian and New Zealand Intensive Care Society Clinical Trials Group. Hydroxyethyl starch or saline for fluid resuscitation in intensive care. N Engl J Med. 2012;367(20):1901Y1911. 14. Bellomo R, Ronco C, Kellum JA, Mehta RL, Palevsky P; Acute Dialysis Quality Initiative workgroup. Acute renal failure - definition, outcome measures, animal models, fluid therapy and information technology needs: the Second International Consensus Conference of the Acute Dialysis Quality Initiative (ADQI) Group. Crit Care. 2004;8(4):R204YR212. 15. Mehta RL, Kellum JA, Shah SV, Molitoris BA, Ronco C, Warnock DG, Levin A; Acute Kidney Injury Network. Acute Kidney Injury Network: report of an initiative to improve outcomes in acute kidney injury. Crit Care. 2007;11(2):R31. 16. Lissauer ME, Chi A, Kramer ME, Scalea TM, Johnson SB. Association of 6% hetastarch resuscitation with adverse outcomes in critically ill trauma patients. Am J Surg. 2011;202(1):53Y58. 17. PRAC confirms that hydroxyethyl-starch solutions (HES) should no longer be used in patients with sepsis or burn injuries or in critically ill patients. European Medicines Agency. Press release October 11, 2013. Available at: http://www.ema.europa.eu/docs/en_GB/document_library/Press_release/ 2013/10/WC500151964.pdf. Accessed January 10, 2014. 18. FDA Safety Communication: Boxed Warning on increased mortality and severe renal injury, and additional warning on risk of bleeding, for use of hydroxyethyl starch solutions in some settings. Press release November 25, 2013. Available at: http://www.fda.gov/biologicsbloodvaccines/safetyavailability/ ucm358271.htm. Accessed January 10, 2014. 19. Liberati A, Moja L, Moschetti I, Gensini GF, Gusinu R. Human albumin solution for resuscitation and volume expansion in critically ill patients. Intern Emerg Med. 2006;1(3):243Y245.
864
20. Subramanian A, Sukheeja D, Trikha V, Pandey A, Albert V, Pandey R. Evaluation of serum creatine kinase and urinary myoglobin as markers in detecting development of acute renal failure in severely injured trauma patients. ISRN Emerg Med. 2013. Article ID 241036, 8 pages. Available at http://dx.doi.org/10.1155/2013/241036. 21. Strecker W, Gebhard F, Perl M, Rager J, Buttenscho¨n K, Kinzl L, Beck A. Biochemical characterization of individual injury pattern and injury severity. Injury. 2003;34(12):879Y887. 22. Strecker W, Gebhard F, Rager J, Bru¨ckner UB, Steinbach G, Kinzl L. Early biochemical characterization of soft-tissue trauma and fracture trauma. J Trauma. 1999;47(2):358Y364. 23. Hoste EA, Clermont G, Kersten A, Venkataraman R, Angus DC, De Bacquer D, Kellum JA. RIFLE criteria for acute kidney injury are associated with hospital mortality in critically ill patients: a cohort analysis. Crit Care. 2006;10(3):R73. 24. Hoste EA, Kellum JA. RIFLE criteria provide robust assessment of kidney dysfunction and correlate with hospital mortality. Crit Care Med. 2006; 34(7):2016Y2017.
EDITORIAL CRITIQUE The discussion about the use of colloids for the care of the critically ill or injured dates back decades. While there are a large number of manuscripts on the subject, the issue still remains unresolved. Surgeons vociferously defend one or the other as superior. Those espousing colloids as better often prefer the synthetic colloids as they are less expensive and perhaps safer. In this article, the group from Miami advocate for the use of hydroxyethyl starch (HES) in the resuscitation of patients with penetrating injury. There are many limitations to this study, virtually all of which are noted in the manuscript. The authors explanation for their results seems plausible. All shock is not the same. The tissue crush injury, particularly muscle crush injury that accompanies severe blunt trauma, may put patients at risk for acute renal failure, differently than they are with pure hemorrhage. The nephrotoxicity of HES may then be amplified in such patients. Our understanding of resuscitation continues to become more refined. While some have used colloids for initial trauma resuscitation, crystalloids use is much more common. Those of us who have been around a while remember the days of the use of high volume crystalloid resuscitation based on belief that only a relatively small portion remained intravascular. We then migrated to a more restricted crystalloid restriction strategy using permissive hypotension. Patients with significant hemorrhage are now given blood and plasma early, though the optimal ratios remain poorly defined. Throughout this period of time, HES has gotten a black eye, as many reports have linked its use with renal failure and death. This report from the University of Miami should make us step back and re-evaluate. I, for one, am not willing to embrace the widespread use of HES, even for penetrating trauma, based on this single study. This data is controversial. It will raise questions. The debate about colloids versus crystalloids will again be re-fueled. This would be a perfect topic for a large prospective trial. Thomas M. Scalea, MD R Adams Cowley Shock Trauma Center University of Maryland School of Medicine Baltimore, Maryland
* 2014 Lippincott Williams & Wilkins
Copyright © 2014 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited.
