Performance Improvement Program Review of Institutional Massive Transfusion Protocol Adherence: An Opportunity for Improvement

Abstract

Background

Given the acuity of patients who receive MTPs and the resources they require, MTPs are a compelling target for performance improvement. This study evaluated adherence with our MTP’s plasma:red blood cell ratio (FFPR) of 1:2 and platelet:red blood cell ratio (PLTR) of 1:12, to test the hypothesis that ratio adherence is associated with lower inpatient mortality.

Materials and Methods

The registry of an urban level I trauma center was queried for adult patients who received at least 6 units of packed red blood cells within 4 hours of presentation. Patients were excluded for interfacility transfer, cardiac arrest during the prehospital phase or within one hour of arrival, or for head AIS ≥5. Univariate analysis and multiple logistic regressions were performed to identify variables associated with early transfusion protocol noncompliance and the effect on inpatient mortality.

Results

Three hundred and eighty-three patients were included, with mean ISS of 25.9 ± 13.3 and inpatient mortality of 28.5%. Increasing age, ISS, INR, and total units of blood product transfused were associated with increased odds of mortality, while an increase in revised trauma score was associated with a decreased odds ratio of mortality. Achieving our goal ratios were protective against mortality, with OR of .451 (P = .013) and .402 (P=.003), respectively.

Discussion

Large proportions of critically injured patients were transfused fewer units of plasma and platelets than our MTP dictated; failure to achieve intended ratios at 4 hours was strongly associated with inpatient mortality. MTP processes and outcomes should be critically assessed on a regular basis as part of a mature performance improvement program to ensure protocol adherence and optimal patient outcome.

Keywords

transfusion ratios massive transfusion protocol protocol adherence performance improvement damage control resuscitation

Background

Hemorrhage is the most common cause of potentially preventable mortality following injury.^1-5 Damage control resuscitation (DCR) has become the dominant paradigm in the initial management of patients with exsanguinating hemorrhage.⁶ While DCR attempts to correct the components of the “lethal triad” of acidosis, coagulopathy, and hypothermia, its key component is using whole blood or balanced transfusion to avoid dilution of procoagulant factors and replenish them as they are consumed. To accomplish DCR, trauma centers adopted massive transfusion protocols (MTPs) to facilitate balanced transfusions; trauma centers must now have MTPs in order to maintain American College of Surgeons verification.⁷

Hospitals expend considerable cost and resources using MTPs, as well as for other aspects of care for patients with exsanguinating trauma. Patients requiring MTP activation are among the most critically injured; their resource-intensive care justifies much of the work of establishing, verifying, and maintaining trauma centers.

Twenty years ago, the Institute of Medicine published Crossing the Quality Chasm, which analyzed the quality of health care delivery in the United States and called for significant improvements in health care.⁸ Mortality has typically been used as a quality indicator for programs but as mortality following trauma has decreased significantly it is a less useful quality indicator. Gruen and colleagues reviewed quality indicators for trauma programs and suggested the development of quality indicators beyond mortality to determine the effectiveness of a program.⁹ The American College of Surgeons’ Trauma Quality Improvement Project enumerates a comprehensive quality improvement program to compare mortality and many processes of care, including MTP.¹⁰ Prior studies have shown higher mortality when predefined ratios are not achieved.^11-13

This study was undertaken to evaluate institutional adherence with our protocol’s intended plasma to red blood cell ratio (FFPR) and platelet to red blood cell ratio (PLTR), and to test the hypothesis that ratio adherence is associated with lower inpatient mortality. Our research plan tested this hypothesis using retrospective data collection taking advantage of data collected by the institutional trauma registry as this will allow re-evaluation as part of an iterative PI process.

Materials and Methods

Setting, Patients, and Data Sources

This study was approved by the Institutional Review Board of the University of Tennessee Health Science Center (#19-07085-XP), and a waiver of informed consent was obtained according to federal guidelines.

This study was conducted at the Elvis Presley Trauma Center, a state-designated level 1 adult trauma center in Memphis, Tennessee. PRBC and FFP units are stored in the trauma bay for immediate use. The blood fridge contains 10 O+ PRBC, 4 O- PRBC, and 3 thawed FFP units. Since 2016, our institution’s MTP has consisted of coolers containing 6 units of PRBC and 3 units of FFP, with a unit (“sixpack”) of platelets in alternating coolers. Other hemostatic products agents (eg, additional FFP or platelets and cryoprecipitate) are given at the discretion of the trauma team and are not part of a resuscitation protocol. Baseline labs, including arterial blood gas, lactate, hematocrit, international normalized ratio (INR), and thromboelastogram (TEG) are drawn at presentation for all patients presenting as highest-level trauma activations.

The institutional trauma registry was queried for all patients aged 15-89 presenting between January 1, 2017 and July 31, 2020 who received at least 6 units of PRBC within 4 hours of presentation. Patients were excluded for interfacility transfer, if they experienced a cardiac arrest during the prehospital phase or within one hour of arrival, or for head AIS ≥5.

Data was abstracted from the trauma center registry for demographic information, admission physiology markers, injury severity, coagulopathy, early blood product transfusion, operations, and outcomes. Inpatient complications, abstracted from the trauma center registry, were recorded in accordance with the National Trauma Data Standard. Missing data was obtained from the Electronic Medical Record (EMR) when available. Demographic information included age, sex, and race. Admission vital signs, Glasgow Coma Scale (GCS), base excess, lactic acid, and international normalized ratio (INR) were abstracted from the registry. Anatomic injury severity was measured by injury severity score (ISS) and physiologic severity by the revised trauma score (RTS).¹⁴ TEG parameters were obtained from the EMR. LY30 was not routinely obtained on the admission TEG. While TEG is checked for all trauma activations, it was rarely used to guide resuscitation during the study period given the lack of software to provide real-time TEG data. Blood component transfusion totals in the first four hours after trauma center presentation were abstracted from the trauma registry.

Figure 1.

Flow diagram of study patients. FFP indicates fresh frozen plasma; PRBC, packed red blood cell.

Statistical Analysis

Our registry captured transfusion data in the form of units (“six packs” for platelets) through 2019, and milliliters beginning January 1, 2020. For the purposes of analysis, PRBC, FFP, and platelet transfusions were normalized to units, assuming that each unit’s volume was approximately 250-300 mL for PRBC and FFP and 150-250 mL for platelets. Transfusion ratios were analyzed both as continuous variables and as binary variables, that is, whether or not patients received at least the MTP protocol ratios, 1:2 (.5) for FFP:PRBC (FFPR) and 1:12 (.0833) for platelet:PRBC (PLTR) in the first four hours following trauma center presentation.

All statistical analyses were performed using STATA/BE 17.0 (StataCorp LLC, College Station, TX). Univariate analyses were performed using two-tailed t-tests for normally distributed continuous data, Mann-Whitney U tests for non-normally distributed data, and χ² tests for categorical data. Multiple logistic regression was performed for inpatient mortality. Variables with a P value <.1 were used to construct models, which were then refined to remove irrelevant variables and improve goodness of fit. Sensitivity analyses were performed to model the potential effect of missing-not at random data; since this did not lead to any significant changes in effect size within the regression model, mean imputation was used for missing data. The final regression models were tested for multicollinearity and specification error, the Hosmer-Lemeshow test was used to evaluate goodness of fit, and Pearson residuals and Pregibon leverage statistics were investigated to make sure the regression model was not inappropriately affected by outliers.

Results

Five hundred sixty-two patients met inclusion criteria; 179 were excluded, yielding a study population of 383 patients (Figure 1). The study population had a mean age of 38.1 years; 80.9% of participants were male, and 71.0% were African American. Six patients (1.6%) were taking anticoagulant or antiplatelet medications (not including aspirin) prior to injury. These patients were included in the analysis as there were no qualitative differences in the outcomes when they were removed. A majority (56.1%) presented following a penetrating mechanism. The mean ISS was 25 (17, 34), with a lactate of 5.53 (3.47, 8.7). Inpatient mortality was 28.5%.

Patients received a median of 10 (7, 16) units of PRBC, 4 (3, 8) units of FFP, and 1 (0, 2) units of platelets. The median FFPR was .42 (.31, .53), and the median PLTR was .11 (0, .17). In terms of compliance with goal transfusion ratios, 148 patients (38.6%) had an FFPR of at least 1:2, and 240 patients (62.7%) had a PLTR of at least 1:12. See Table 1 for overall patient population characteristics.

Table 1.

Demographics of the Study Cohort.

	N	%	Median (IQR)	Mean ± SD
Age	383		38.1
Male sex	310	80.9
African American race	272	71.0
AC/AP prior to admission	6	1.6
Penetrating mechanism	215	56.1
ISS			25 (17, 34)
SBP (mmHg)			111 (88, 140)
HR (/min)				110.1 ± 25.4
GCS score			15 (8, 15)
Lactate (mmol/L)			5.53 (3.47, 8.7)
INR			1.27 (1.15, 1.49)
PRBC units (first 4 hours)			10 (7, 16)
FFP units (first 4 hours)			4 (3, 8)
Platelet units (first 4 hours)			1 (0, 2)
FFP:PRBC ratio			.42 (.31, .53)
FFP:PRBC ratio ≥.5 (%)	148	38.6
Platelet:PRBC ratio			.11 (0, .17)
Platelet:PRBC ratio ≥.833 (%)	240	62.7
ICU LOS (survivors) (d)			8 (4, 19)
Ventilator (survivors) (d)			4 (2, 12)
Hospital LOS (survivors) (d)			19 (10, 32)
Inpatient mortality (%)	109	28.5

AC indicates anticoagulation; AP, antiplatelet therapy; ISS, injury severity score; SBP, systolic blood pressure; HR, heart rate; GCS, Glasgow Coma Scale; INR, international normalized ratio; PRBC, packed red blood cell; FFP, fresh frozen plasma; LOS, length of stay.

All initial demographic, physiologic, and coagulation parameters were similar between patients above and below the goal FFPR (Table 2). Mortality was lower among patients meeting goal FFPR (20.3% versus 33.6%, P = .005).

Table 2.

Demographics, Admission Physiologic Parameters, and Outcomes of Patients Who Received an FFP:PRBC Ratio <1:2 Versus ≥1:2 in the First 4 Hours After Presentation. Median (IQR), Mean ± SD, or % (N).

	FFP:PRBC <1:2	FFP:PRBC ≥1:2	p
Age	35 (25, 50)	35 (26, 50)	.789
Male sex	80.5% (N = 190)	81.1% (N = 120)	.955
African American race	68.9% (N = 162)	74.3% (N = 110)	.509
Penetrating mechanism	54.9% (N = 129)	58.1% (N = 86)	.537
ISS	25 (16, 34)	25 (17, 34)	.527
AIS head >1	17.0% (N = 40)	24.3% (N = 36)	.081
SBP (mmHg)	112 (88, 140)	111 (88, 135)	.638
HR (/min)	109.4 ± 25.4	111.2 ± 25.5	.524
GCS score	15 (8, 15)	15 (8, 15)	.408
RTS	7.55 (5.68, 7.84)	7.11 (5.66, 7.84)	.903
Lactate (mmol/L)	5.8 (3.5, 9.2)	5.1 (3.4, 7.9)	.300
Base excess	−9 (−15 to −5)	−8 (−11, −5)	.068
INR	1.3 (1.2, 1.5)	1.2 (1.1, 1.5)	.159
PRBC units in first 4 h	10 (7, 16)	9 (6, 14.5)	.106
FFP units in first 4 h	3 (2, 6)	6 (4, 10)	<.001*
Platelet units in first 4 h	1 (0, 2)	1 (1, 2)	<.001*
ICU LOS (survivors) (d)	6 (4, 20)	8 (5, 18)	.291
Hospital LOS (survivors) (d)	18 (8, 30)	19 (12, 36)	.239
Inpatient mortality	33.6% (N = 79)	20.3% (N = 30)	.005*

*denotes statistical significance.

FFP indicates fresh frozen plasma; PRBC, packed red blood cell; ISS, injury severity score; AIS, abbreviated injury score; SBP, systolic blood pressure; HR, heart rate; GCS, Glasgow coma scale; RTS, revised trauma score; INR, international normalized ratio; LOS, length of stay.

Likewise, patients who achieved goal PLTR were similar to those who did not achieve goal ratio (Table 3). However, while the VTE rate was not different (5.8% versus 2.8%, P = .174), the composite complication rate was significantly higher in the group achieving goal PLTR, 22.9% versus 12.6% (P = .013) (supplemental table).

Table 3.

Demographics, Admission Physiologic Parameters, and Outcomes of Patients Who Received a platelet:PRBC Ratio <1:12 Versus ≥1:12 in the First 4 Hours After Presentation. Median (IQR), Mean ± SD, or % (N).

	Platelet:PRBC <1:12	Platelet:PRBC ≥1:12	p
Age	36 (26, 50)	34 (25, 50)	.740
Male sex	76.2% (N = 109)	83.8% (N = 201)	.070
African American race	68.5% (N = 98)	72.5% (N = 174)	.408
Penetrating mechanism	53.2% (N = 76)	57.9% (N = 139)	.363
ISS	22 (14, 34)	26 (17, 34)	.114
AIS head >1	23.1% (N = 33)	17.9% (N = 43)	.221
SBP (mmHg)	118 (90, 140)	109 (86, 137)	.214
HR (/min)	112.5 ± 25.6	108.6 ± 25.2	.143
GCS score	15 (8, 15)	15 (8, 15)	.708
RTS	7.55 (5.44, 7.84)	7.11 (5.68, 7.84)	.733
Lactate (mmol/L)	5.5 (3.4, 8.2)	5.7 (3.5, 8.8)	.649
Base excess	−7 (−14, −4)	−9 (−13, −5)	.330
INR	1.3 (1.2, 1.5)	1.3 (1.1, 1.5)	.377
PRBC units in first 4 h	8 (6, 15)	10 (8, 17)	.004*
FFP units in first 4 h	3 (1, 6)	5 (3, 8)	<.001*
Platelet units in first 4 h	0 (0, 0)	2 (1, 3)	<.001*
ICU LOS (survivors) (d)	6 (3, 17)	8 (4, 20)	.072
Hospital LOS (survivors) (d)	17 (7, 31)	19 (12, 33)	.069
Inpatient mortality	36.4% (N = 52)	23.8% (N = 57)	.008*

*denotes statistical significance.

PRBC, packed red blood cell; ISS, injury severity score; AIS, abbreviated injury score; SBP, systolic blood pressure; HR, heart rate; GCS, Glasgow coma scale; RTS, revised trauma score; INR, international normalized ratio; FFP, fresh frozen plasma; LOS, length of stay.

The multiple logistic regression model for inpatient mortality is presented in Table 4. Increasing age, ISS, INR, and total units of blood product transfused were associated with increased odds of mortality, while an increase in RTS was associated with a decreased odds ratio of mortality. Eleven patients were missing data for INR. Since these values may have been missing-not at random, sensitivity analysis was performed by assuming the missing INR values were all equal to .8 (below the 1^st percentile) or to 3.0 (above the 99th percentile). In neither case did odds ratios become nonsignificant. In the model where missing INR values were assumed to be 3.0, the odds ratio for INR decreased by 59.7%, but all other variables under both assumptions changed by .0% to 5.9%. Since the model was not substantively changed under either assumption for missing INR, mean imputation was used and subjects with missing INR data were assigned the mean INR value of 1.41. The Hosmer-Lemeshow test for the final regression model yielded a P value of .220, consistent with appropriate fit. In the final model, achieving our goal FFPR and PLTR was associated with improved mortality, with odds ratios of .470 (P = .0136) and .402 (P = .003), respectively.

Table 4.

Multiple Logistic Regression for Mortality. N = 383.

	OR	95% CI	p
FFP:PRBC ratio ≥1:2	.470	.254, .869	.016
Platelet:PRBC ratio ≥1:12	.402	.221, .731	.003
Age	1.043	1.025, 1.061	<.001
ISS	1.024	1.003, 1.046	.025
RTS	.722	.612, .851	<.001
INR	5.180	2.123, 12.637	<.001
PRBC + FFP + platelet units in first 4 hours	1.048	1.030, 1.067	<.001

FFP indicates fresh frozen plasma; PRBC, packed red blood cell; ISS, injury severity score; RTS, revised trauma score; INR, international normalized ratio.

Discussion

We evaluated compliance with our institutional MTP component ratios during the first four hours after presentation to assess the effect of adherence on inpatient mortality. Only 40.8% of patients achieved the goal FFPR of ≥.5, and only 63.4% of patients achieved the goal PLTR of ≥.0833. Failure to achieve our target ratios resulted in a statistically significant mortality increase. Using multiple logistic regression to control for known confounding variables, achieving goal FFPR and PLTR were associated with odds ratios for mortality of .470 and .402, suggesting that early transfusion ratios are an important process measure.

Donabedian has described three distinct aspects of quality care: structure, process, and outcome.¹⁵ In his model, which has been validated for trauma care,¹⁶ improvements in structure lead to improvement in processes which, in turn, improve outcomes. Compliance with MTP transfusion ratios can be studied to determine if it leads to improvement in patient outcomes.

Only two publications have investigated the effects of compliance with hospital MTPs as part of a performance improvement process. Cotton et al investigated compliance with seven process elements of their institutional trauma exsanguination protocol.¹⁷ As in our study, they found no statistical correlation between patient factors and protocol compliance. Additionally, protocol compliance was associated with a significant decrease in mortality at both 24 hours and 30 days. In another single-institution study, Bawazeer et al assessed compliance with thirteen MTP process measures.¹⁸ As in our study, they found that the prescribed transfusion ratios were only attained in 53% of MTP activations. They stratified patients according to the degree of compliance with process measures: <60%, 60-80%, or >80% compliance. Compared to the middle group, the group of patients with >80% compliance had a significant reduction in mortality by logistic regression. Like our study, both studies were retrospective, making it possible that the effect of protocol compliance on mortality was due to unmeasured confounding by patient acuity, since process compliance may be reduced for the sickest patients at highest risk for mortality.

This study was a pragmatic retrospective investigation of our MTP process. Despite lacking the granularity and precision of prospectively collected data, this methodology will facilitate repeat assessments during a PI process. A key strength of the methodology was the acuity and case mix of the patient population. With a median ISS of 25 and a 56.1% penetrating mechanism, these patients were at substantial risk for inpatient mortality. However, the high acuity and penetrating index in our single-institution study may limit applicability to other settings, as well as to populations that were excluded from our study.

This study is subject to several limitations. Although all measured covariates were equivalent between groups, they may still have been subject to unmeasured confounding. The most important determinant of mortality from exsanguination may be the rate of hemorrhage; this eludes precise quantification in the clinical setting and is impossible to estimate from a retrospective review. Patients may have been transfused with a preponderance of PRBC before MTP coolers are delivered; this could introduce survivorship bias as patients who die rapidly may not have received as much hemostatic blood product. Since this study did not have prospective data collection, it lacks granularity as to the timing of each transfusion, making survivorship bias difficult to detect. The critical administration threshold of at least 3 units of PRBC transfused within 60 minutes would have been a more robust massive transfusion definition for this study,¹⁹ but without prospective data collection this was difficult to determine, limiting the repeatability of our study. Additionally, differences in initial resuscitation practices between team leaders prior to MTP initiation likely impacted the transfusion ratios. Finally, our data only captured transfusions occurring during the first four hours. This early resuscitation period is clinically significant, but this four-hour limit may miss relevant data for patients in whom hemostasis takes longer.

This study highlights the potential for applying the principles of Donabedian’s model to the PI process in our mature trauma program institutional MTP. MTPs have multiple stakeholders, including the trauma team, blood bank, anesthesiology, nursing, and non-trauma services. While our performance improvement intervention will require buy-in from multiple hospital stakeholders, we plan to incorporate goal-directed hemostatic resuscitation into our MTP and engage in staff education throughout the trauma center.

Several improvements in our institutional process have been implemented following the results of this study. Based on our institutional protocol during the study period, a single admission TEG was drawn for all trauma activations. Given the lack of appropriate TEG software at the time of this study, real-time data was unavailable. Thus, the timing for TEG results did not make TEG guided resuscitation possible. We now have TEG software and are utilizing TEG for guided resuscitation in both the operating room as well as the ICU for these critically ill patients. During the study period, there were inadequacies in the MTP records from our institutional blood bank which made it difficult to discern why our goal ratios were not met. Additionally, prehospital blood administration was likely in a 1:1 ratio. As well, blood administration in our resuscitation bay was at the discretion of the attending surgeon and may not have represented a 2:1 or even a 1:1 ratio. Based on the analysis of this data, our MTP was reviewed and we have corrected the inadequacies in our blood bank record collection and are moving toward implementing a ratio of 1:1:1.

To be effective, a PI program should continually examine the structure of the program and the processes of care to determine the effect on patient outcomes. Since MTP adherence appears to be associated with improved inpatient mortality, this process fits nicely into Donabedian’s model. Compliance with MTP processes and outcomes should be regularly assessed to optimize patient outcomes.

Supplemental Material

Supplemental Material - Performance Improvement Program Review of Institutional Massive Transfusion Protocol Adherence: An Opportunity for Improvement

Supplemental Material for Performance Improvement Program Review of Institutional Massive Transfusion Protocol Adherence: An Opportunity for Improvement by Thomas Easterday, Saskya Byerly, Louis Magnotti, Peter Fischer, Kinjal Shah, Martin Croce, Andrew Kerwin, and Isaac Howley in The American Surgeon

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Supplemental Material

Supplemental material for this article is available online.

References

Spinella

. Zero preventable deaths after traumatic injury: An achievable goal. J Trauma Acute Care Surg. 2017;82(6S suppl 1):S2-S8.

Evans

van Wessem

McDougall

Lee

Lyons

Balogh

. Epidemiology of traumatic deaths: comprehensive population-based assessment. World J Surg. 2010;34(1):158-163.

Kashuk

Moore

Johnson

, et al.

Postinjury life threatening coagulopathy: is 1:1 fresh frozen plasma:packed red blood cells the answer?

J Trauma. 2008;65(2):261-270.

Kauvar

Lefering

Wade

. Impact of hemorrhage on trauma outcome: an overview of epidemiology, clinical presentations, and therapeutic considerations. J Trauma. 2006;60(6 suppl l):S3-S11.

Teixeira

Inaba

Hadjizacharia

, et al. Preventable or potentially preventable mortality at a mature trauma center. J Trauma. 2007;63(6):1338-1346.

Holcomb

Jenkins

Rhee

, et al. Damage control resuscitation: directly addressing the early coagulopathy of trauma. J Trauma. 2007;62(2):307-310.

American College of Surgeons Committee on Trauma . Resources for the Optimal Care of the Injured Patient. 7th ed. Chicago: American College of Surgeons; 2022.

Institute of Medicine (US) . Committee on Quality of Health Care in America. Crossing the Quality Chasm: A New Health System for the 21st Century. Washington, DC: National Academies Press (US); 2001.

Gruen

Gabbe

Stelfox

Cameron

. Indicators of the quality of trauma care and the performance of trauma systems. Br J Surg. 2012;99(suppl 1):97-104.

10.

American College of Surgeons Committee on Trauma . ACS TQIP Massive Tranfusion in Trauma Guidelines. Chicago: American College of Surgeons; 2014.

11.

Borgman

Spinella

Perkins

, et al. The ratio of blood products transfused affects mortality in patients receiving massive transfusions at a combat support hospital. J Trauma. 2007;63(4):805-813.

12.

Cotton

Gunter

Isbell

, et al. Damage control hematology: the impact of a trauma exsanguination protocol on survival and blood product utilization. J Trauma. 2008;64(5):1177-1182.

13.

Gunter

Jr Au

Isbell

Mowery

Young

Cotton

. Optimizing outcomes in damage control resuscitation: identifying blood product ratios associated with improved survival. J Trauma. 2008;65(3):527-534.

14.

Champion

Sacco

Copes

Gann

Gennarelli

Flanagan

. A revision of the Trauma Score. J Trauma. 1989;29(5):623-629.

15.

Donabedian

. An Introduction to Quality Assurance in Health Care. New York, NY: Oxford University Press; 2003.

16.

Moore

Lavoie

Bourgeois

Lapointe

. Donabedian's structure-process-outcome quality of care model: Validation in an integrated trauma system. J Trauma Acute Care Surg. 2015;78(6):1168-1175.

17.

Cotton

Dossett

Nunez

Robertson

Young

. Room for (performance) improvement: provider-related factors associated with poor outcomes in massive transfusion. J Trauma. 2009;67(5):1004-1012.

18.

Bawazeer

Ahmed

Izadi

McFarlan

Nathens

Pavenski

. Compliance with a massive transfusion protocol (MTP) impacts patient outcome. Injury. 2015;46(1):21-28.

19.

Savage

Sumislawski

Zarzaur

Dutton

Croce

Fabian

. The new metric to define large-volume hemorrhage: results of a prospective study of the critical administration threshold. J Trauma Acute Care Surg. 2015;78(2):224-229.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.20 MB