Abstract
Introduction
Crises like the COVID-19 pandemic create blood product shortages. Patients requiring transfusions are placed at risk and institutions may need to judiciously administer blood during massive blood transfusions protocols (MTP). The purpose of this study is to provide data-driven guidance for the modification of MTP when the blood supply is severely limited.
Methods
This is a retrospective cohort study of 47 Level I and II trauma centers (TC) within a single healthcare system whose patients received MTP from 2017 to 2019. All TC used a unifying MTP protocol for balanced blood product transfusions. The primary outcome was mortality as a function of volume of blood transfused and age. Hemoglobin thresholds and measures of futility were also estimated. Risk-adjusted analyses were performed using multivariable and hierarchical regression to account for confounders and hospital variation.
Results
Proposed MTP maximum volume thresholds for three age groupings are as follows: 60 units for ages 16-30 y, 48 units for ages 31-55 y, and 24 units for >55 y. The range of mortality under the transfusion threshold was 30%-36% but doubled to 67-77% when the threshold was exceeded. Hemoglobin concentration differences relative to survival were clinically nonsignificant. Prehospital measures of futility were prehospital cardiac arrest and nonreactive pupils. In hospital risk factors of futility were mid-line shift on brain CT and cardiopulmonary arrest.
Conclusions
Establishing MTP threshold practices under blood shortage conditions, such as the COVID pandemic, could sustain blood availability by following relative thresholds for MTP use according to age groups and key risk factors.
Keywords: Blood transfusion, Massive transfusion protocol, Mortality, Patient outcomes, Traumatic injury
Background
Crises, such as the COVID-19 pandemic, weather events, natural disasters, terrorist acts, and war, can create both short-term and long-term shortages in blood supply.1 The COVID-19 pandemic, for instance, led to a worldwide shortage of blood products, likely resulting from social distancing, policies implemented to protect donors from COVID-19 infections, and individual misunderstanding of donation risk.2 Early in the pandemic, the American Red Cross reported there were 2700 canceled blood drives and 86,000 fewer donations.3 These cancelations were alarming considering that roughly 80% of the blood collected by the Red Cross comes from these blood drives. In times of blood scarcity, regardless of the cause, patients in need of transfusions may be placed at risk, and institutions may face challenges in maintaining standard blood utilization practices.
Most institutions have a structured massive transfusion protocol (MTP) that guides large volume, ratio-specific blood product transfusion for patients in hemorrhagic shock–a practice that has been shown to improve outcomes.4 , 5 Though MTP is used in many conditions such as major surgery, peripartum hemorrhage, and vascular emergencies, injured patients are by far the largest consumers of blood products. According to the American College of Surgeons (ACS), trauma patients consume 70% of all blood transfused at a trauma center.6 Though MTPs guide when and how to administer blood products based on physiologic response and hemoglobin (Hb) levels, few take into account patient-specific characteristics, like age. Even fewer protocols have concrete recommendations on futility thresholds—when blood transfusion should be terminated as it is unlikely to provide benefit.
Under conditions of limited blood supply, clinicians may benefit from structured, data-driven guidance for discontinuation of the MTP to conserve the limited blood supply to ultimately benefit the maximum number of patients. This study aims to identify contributing factors for mortality among patients with massive transfusion, suggest potential “futility thresholds” for blood transfusion according to different age groups, and provide guidance on Hb target levels in times of limited blood product availability.
Methods
The study sample consisted of inpatients aged 16 y and older who arrived as a full trauma activation and received the MTP at one of 47 Level I or II trauma centers within a large US health care system between the years of 2017-2019. The hospitals were experiencing blood product shortages for reasons unrelated to the COVID-19 pandemic. Patients who arrived dead (no discernible blood pressure, pulseless, apneic) were excluded. Data from the National Trauma Data Standard (NTDS) were submitted by each trauma center to a centralized, system-wide registry. Registry data were joined with transfusion and laboratory data from the hospital system's clinical data warehouse and deidentified to create the study dataset. Death was determined using hospital discharge disposition status. A formal institutional review board (IRB) exemption was obtained for this analysis and the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines were followed to report this research (eTable 1; Supplemental Digital Content; STROBE Statement).7
Most trauma centers indicated administration of the MTP as a hospital related event in the trauma registry. For those that administered MTPs without recording MTP as an event, MTP was defined as receiving >5 units of blood product within the first 4 h of admission or ≥10 units of blood products within the first 24 h of admission. This process of screening for MTP in the large clinical data warehouse was validated by independent verification against the number of MTP events in the sample of trauma centers who recorded MTP as a hospital event.
All trauma centers followed the health care system's existing unified MTP guideline to direct their massive transfusion practices, which is based on a 1:1:1 transfusion ratio of packed red blood cells (pRBC), fresh frozen plasma (FFP), and apheresis platelets. No whole blood data were collected. Transfusion is typically initiated with 4-6 units pRBC, 4-6 units FFP, and one pack of apheresis platelets, which are then alternated in rounds of 4-6 units pRBC and 4-6 units FFP.
The primary outcome of this study was mortality, as a function of the amount of blood product transfused, stratified by age. Therefore, for the analysis, patients were grouped by a cross-tabulation of age and transfusion amount. In clinical practice, a “round” of MTP is generally equivalent to 12 units of component products (pRBC, FFP, and apheresis platelets). For this study, the quantity of blood component transfused was therefore stratified by multiples of 12 units. The final strata were set at <24 units (<2 MTP rounds), 25-36 units (3 rounds), 37-48 units (4 rounds), 49-60 units (5 rounds), 61-72 units (6 rounds), 73-84 units (7 rounds), and >84 units (8 or more rounds). As age alone can be an independent predictor of mortality, age groups were initially defined as 16-30, 31-55, 56-64, 65-74, 75-84, and 85-100 y per standard hospital network practice. After analysis, these groupings were further collapsed to three comparable groupings of 16-30, 31-55, and ≥56 y due to a clear linear relationship between mortality and age, with similar mortality results noted within each age stratum. The newly defined transfusion strata were then filtered by age groups to create a cross tabulation of amount of transfusion by age group, with the raw mortality calculated for each age and transfusion combination. The sample size required for 80% power with a 95% confidence interval to determine threshold differences in 16-30 y old group was 86 patients when comparing a prethreshold mortality of 55% and a post-threshold mortality rate of 75%. This group had 593 patients. For the 31-55 y old group the sample size needed was also 86 when comparing 52.4% versus 72.7% mortality rates. This cohort had 575 patients. For the ≥56 y old group when comparing 71.4 versus 100% mortality rates the sample sized needed was 20. This cohort had 437 patients.
Logistic regression was performed to evaluate how predefined transfusion thresholds affect inpatient mortality for each age group. The univariate and adjusted multivariable models were used to calculate the odds ratio (OR) of mortality when comparing patients above the predefined transfusion threshold to those below the threshold. Confounders were considered for the multivariable risk adjusted analysis if it was reasonable to assume that these variables had an independent effect on mortality in trauma patients. The final risk adjusted multivariable regression model for trauma patients who received MTP included gender, race, injury mechanism (i.e., blunt or penetrating), comorbidities (via the Charlson comorbidity index), the presence or absence of a code event, and International Classification of Diseases, 10th Revision (ICD-10) derived injury severity score (ICISS). The regression model also underwent a reliability adjustment to account for variations in hospital practices among the different hospitals through hierarchical linear regression.8 , 9 Missing data were imputed using Rubin's rules for multiple imputation.10 The variable with the most missing data was Hb, with 17% data missing.
A separate analysis was performed to determine if there was a group of prehospital, hospital, and comorbidity variables that could refine mortality predictions and serve as markers of futility in initiating or continuing MTP. Model variables included age, prearrival cardiac arrest, in-hospital cardiac arrest, initial ED pulse rate >120 bpm or <60 bpm, traumatic brain injury (TBI) with midline shift, fixed and dilated pupils, transfusion amount, initial hospital GCS score, and injury type (blunt or penetrating).
Large differences in Hb concentration noted between MTP survivors and nonsurvivors may suggest differences in resuscitation or major bleeding complications. To determine if these differences existed, we compared the overall average Hb levels for the hospital length of stay, Hb concentrations on the day of admission, and Hb concentrations on the last day of hospitalization. To determine a minimum Hb threshold that maintains blood conservation in a scenario where the blood supply is limited, as was the state of the hospital system during the study period, the average lowest Hb concentrations were compared between survivors and nonsurvivors.
All data were analyzed using SAS version 9.4 (SAS Institute Inc., Cary, NC, USA). To describe the study sample, parametric data expressed as proportions were evaluated by Chi Square tests and the appropriate t-test or ANOVA for continuous data. Nonparametric data were evaluated by Fisher's exact test for proportions and the Wilcoxon Rank-Sum Test for continuous data.
Results
From 2017 to 2019, a total of 1605 trauma patients required MTP. Table 1 shows the socio-demographic differences between inpatient transfusion groups stratified by their characteristics and risk factors. There were no significant differences between transfusion threshold groups except for variables associated with mortality and injury severity. The pRBC: FFP: Platelets transfusion ratios were 1.1:0.9:1 in the 16-30 age group, 1:0.7:1 in the 31-55 age group, and 0.6:0.4:1 in the 56-100 age group.
Table 1.
Socio-demographics and covariates between MTP cohorts from 2016 to 2019 among 47 ACS verified level I and II trauma centers.
| Socio-demographics and covariates | Age 16-30 (n = 593) | Age 31-55 (n = 575) | Age 56-100 (n = 437) | P Value |
|---|---|---|---|---|
| Gender | ||||
| Male | 81.6% | 77.4% | 67.7% | <0.0001 |
| Female | 18.4% | 22.6% | 32.3% | <0.0001 |
| Race | ||||
| White | 50.8% | 60.9% | 75.3% | <0.0001 |
| Black | 29.7% | 22.8% | 11.0% | <0.0001 |
| Asian | 1.0% | 1.9% | 1.8% | 0.40 |
| Other | 18.6% | 14.4% | 11.9% | 0.01 |
| Insurance status | ||||
| Medicaid | 14.7% | 14.1% | 6.0% | <0.0001 |
| Medicare | 1.4% | 5.4% | 32.0% | <0.0001 |
| Other | 5.1% | 4.0% | 1.6% | 0.01 |
| Government | 1.9% | 3.3% | 1.6% | 0.13 |
| Private or Commercial | 34.1% | 35.8% | 38.7% | 0.31 |
| Self-Pay | 43.0% | 37.4% | 20.1% | <0.0001 |
| Disposition | ||||
| Home or Self Care or Home with Services | 30.9% | 26.6% | 9.8% | <0.0001 |
| Expired | 32.2% | 36.4% | 45.5% | <0.0001 |
| Continuing care (SNF, long term acute care, in-patient rehab) | 19.4% | 20.0% | 25.6% | 0.03 |
| Transfer | 5.4% | 3.3% | 2.1% | 0.02 |
| Other | 12.1% | 13.7% | 16.9% | 0.09 |
| Injury mechanism | ||||
| Blunt | 96.1% | 96.7% | 99.1% | 0.01 |
| Penetrating | 3.9% | 3.3% | 0.9% | 0.01 |
| ISS | ||||
| <9 | 1.9% | 2.8% | 3.4% | 0.28 |
| 9-15 | 9.6% | 11.3% | 11.0% | 0.61 |
| 16-24 | 21.6% | 23.3% | 22.9% | 0.77 |
| ≥25 | 67.0% | 62.6% | 62.7% | 0.22 |
| CCI | ||||
| 0 | 96.8% | 95.7% | 92.0% | 0.002 |
| 1 | 2.2% | 2.3% | 4.8% | 0.02 |
| 2 | 0.3% | 1.0% | 1.1% | 0.27 |
| ≥3 | 0.7% | 1.0% | 2.1% | 0.12 |
| ICISS | ||||
| <0.5 | 56.7% | 54.3% | 56.8% | 0.64 |
| 0.5-0.6 | 13.7% | 14.1% | 13.5% | 0.96 |
| 0.6-0.7 | 12.8% | 13.9% | 11.2% | 0.44 |
| 0.7-0.8 | 8.3% | 8.4% | 9.8% | 0.62 |
| 0.8-0.9 | 5.7% | 5.7% | 6.6% | 0.80 |
| >0.9 | 2.9% | 3.7% | 2.1% | 0.33 |
| Head AIS score, mean (SD) | 4.0 (1.5) | 3.7 (1.6) | 3.4 (1.5) | 0.0001 |
| Head AIS score, median | 4 | 3 | 3 | 0.0002 |
MTP = massive blood transfusion protocols; ACS = The American College of Surgeons; SNF = skilled nursing facility; ISS = injury severity score; CCI = Charlson comorbidity index; ICISS = international classification of disease injury severity score; AIS = abbreviated injury scale; SD = standard deviation.
As expected, a higher number of units of blood products transfused were correlated with increased mortality and injury severity. There was an inverse relationship between age and transfusion thresholds and a positive relationship between age and raw mortality; younger patients had lower mortality at higher transfusion values compared to older patients. In comparing the raw mortality rates for the combination of age and transfusion groupings, a transfusion cutoff (or threshold) was defined by the largest percent difference in mortality from one threshold to the next highest threshold. The transfusion volume threshold for each age grouping was above 60 units for those aged 16-30, above 48 units for the 31-55 age group, and above 24 units for patients 56-100 y (Table 2 ).
Table 2.
Raw mortality percentage stratified by age and MTP transfusion cohort.
| Age | ≤24 units | 25-36 units | 37-48 units | 49-60 units | 61-72 units | 73-84 units | >84 units |
|---|---|---|---|---|---|---|---|
| 16-30 | n = 418 | n = 86 | n = 38 | n = 20 | n = 12 | n = 9 | n = 10 |
| 25.1% | 40.7% | 50.0% | 55.0% | 75.0% | 44.4% | 80.0% | |
| 31-55 | n = 404 | n = 78 | n = 42 | n = 22 | n = 6 | n = 11 | n = 12 |
| 27.0% | 52.6% | 52.4% | 72.7% | 83.3% | 54.6% | 83.3% | |
| 56-100 | n = 337 | n = 43 | n = 21 | n = 11 | n = 10 | n = 6 | n = 9 |
| 36.2% | 69.8% | 71.4% | 100% | 100% | 50.0% | 88.9% |
MTP = massive blood transfusion protocols.
Mortality was assessed after risk adjustment for gender, comorbidities, ICISS, cardiopulmonary resuscitation (CPR) event, injury mechanism, and reliability adjustment. The risk adjusted mortality rate was compared for those receiving less than a specific threshold amount of blood to those receiving more than the threshold amount. For all age groups, the range of the mortality rate for transfusion volumes under the transfusion threshold was 30.3%-36.2%. The mortality rate more than doubled (67.7%-77%) once the transfusion threshold was exceeded. The overthreshold values showed significantly higher adjusted OR for mortality, with the highest being the 55-100 age group (adjusted odds ratio [aOR]: 8.8, 95% CI: 4.49-17.24) (Table 3 ).
Table 3.
Risk adjusted mortality OR by transfusion cutoff (before and after threshold) by age group∗.
| Age | Transfusion threshold | Mortality below threshold | Mortality above threshold | Odds ratio | Adjusted odds Ratio† | Adjusted odds Ratio‡ |
|---|---|---|---|---|---|---|
| 16-30 | 60 units | 30.3% | 67.7% | 4.82 (2.23, 10.50) | 4.31 (1.61, 11.56) | 4.92 (1.92, 12.65) |
| 31-55 | 48 units | 32.8% | 72.6% | 5.41 (2.85, 10.27) | 4.69 (2.10, 10.47) | 5.54 (2.58, 11.93) |
| 56-100 | 24 units | 36.2% | 77.0% | 5.90 (3.52, 9.88) | 8.80 (4.49, 17.24) | 8.39 (4.37, 16.13) |
OR = odds ratio; ICISS = international classification of disease injury severity; ScoreCPR = cardiopulmonary resuscitation.
Transfusion defined as total unit volume from an approximate 1:1:1 ratio of PRBC:FFP:PLTS.
Risk adjusted by gender, comorbidities, ICISS, CPR event, injury mechanism, reliability adjustment.
Risk adjusted by gender, comorbidities, ISS, CPR event, injury mechanism, reliability adjustment.
Hb concentrations in g/dL were compared between survivors and nonsurvivors. There were slight differences in the average Hb levels between the groups, with 9.92 (±1.58) among survivors compared to 9.72 (±1.91) among nonsurvivors (P = 0.027). Hb concentrations were higher among survivors compared to nonsurvivors for the first hospital day (11.34, standard deviation [SD]: 1.94 versus 10.30, SD: 2.09; P < 0.001), and only slightly higher for the last hospital day (9.8, SD: 1.73 versus 9.24, SD: 2.17; P < 0.001). To explore possible lower Hb thresholds, we examined the lowest Hb concentrations among those who lived and those who died. For those patients, there were no significant differences between those who lived and those who died (7.20, SD: 1.97 versus 7.40, SD: 2.36; P = 0.943).
Figure 1 shows the strength of association between mortality and prehospital, hospital, and comorbidity variables based on findings from our sample. Only brain midline shift and fixed and dilated pupils had higher ORs compared to blood transfusion. Prearrival cardiac arrest, in-hospital cardiac arrest, and pupillary response were the highest predictors for mortality, even when blood transfusion was included in the adjustment.
Fig. 1.
Prehospital, hospital and comorbidity variables for mortality among those who received MTP. MTP = massive blood transfusion protocols; GCS = glasgow coma scale; CPR = cardiopulmonary resuscitation.
Combining the results of the transfusion thresholds stratified by age group and risk factors generated a tiered algorithm for MTP under austere conditions (Fig. 2 ).
Fig. 2.
Recommended maximum mass transfusion protocol volume thresholds under austere conditions.
Discussion
This study represents a large multicenter experience with current massive transfusion practices in routine health care delivery, in which resources are plentiful and readily available as conditions require. This work provides a window into defining the point at which blood product transfusion volumes do not significantly improve survival under normal conditions, and thus can inform a more disciplined approach to the MTP when blood is scarce. A data-driven guideline for MTP use in the resource-constrained environment anticipates the possibility of this context and is particularly credible when based on analysis of a large dataset.
The results from this study suggest that in conditions of inadequate blood supply it is possible to limit the amount of blood products during MTP use and have anticipated outcomes that achieve the “greatest good for the greatest number” when stratified by age groups. We recommend a tiered response tailored to the degree of regional blood product shortage where patients receiving MTP would not be transfused above the critical threshold specific to their age group. On average, the raw mortality rate below the threshold is approximately 30% for each age cohort and doubles to over 60% once transfused over the volume threshold (Table 3).
Restrictive transfusion has been studied in multiple types of clinical scenarios. However, in most studies, the transfusion goal is to meet a physiologic laboratory benchmark such as an Hb level. Transfusion thresholds, both restrictive and liberal, have been discussed in different disease processes including critically ill adults,11 cardiac surgery patients,12 and pediatric patients.13 Most trials demonstrated no significant differences in outcome when a restrictive approach is taken to transfusion.14 Trauma patients can likely be managed in a similar manner once source control of the bleeding is achieved and the patient is not severely coagulopathic. In this study, there were only slight differences in the average Hb concentration between those who survived and those who died (9.92, SD: 1.58 versus 9.72, SD: 1.91; P = 0.027). Our results also suggest that a lower Hb threshold may be tolerated, as there were no significant differences in the minimum Hb level between those who survived and those who died (7.20, SD: 1.97 versus 7.40, SD: 2.36; P = 0.943). Though not specifically supported by our data, we speculate that a lower threshold, especially for younger, healthier patients, could be 5.5 to 7.0 g/dL. For the protocol (Fig. 2), we proposed 7 g/dL as the acceptable Hb threshold in moderately limited conditions. For more austere situations (major and critical), the acceptable Hb transfusion threshold was 6 g/dL, based on the range of the SD.
Studies of MTP have evaluated volume of pRBC transfusion as it relates to mortality. Early studies did not find mortality differences between large volume blood resuscitation, such as pRBC transfusion of 20 units to 68 units, or even between 50 units and >75 units.15 , 16 Instead, clinical and treatment variables, such as the use of inotropes, low systolic pressures, and a large base deficit, were more predictive of death. Similarly, our study found that some patient-level clinical and treatment variables were significantly associated with mortality among patients who received MTP. Some of these variables (e.g., cardiac arrest after arrival to the hospital with ongoing CPR and traumatic brain injury with midline shift on brain CT scan) had a stronger association with mortality than blood transfusion volumes and should be considered in decisions for discontinuation of MTP under limited resource conditions (Fig. 1). Prearrival variables that may help in the decision not to initiate MTP under limited resource conditions include prearrival cardiac arrest, a GCS motor score of 1, and advanced age.
An important difference between earlier MTP studies and this study is that current transfusion practices in trauma follow a balanced 1:1:1 transfusion ratio. This balanced blood product transfusion ratio may improve survival.4 A prospective, randomized trial showed that a balanced transfusion ratio led to both a decrease in exsanguination and better hemostasis.5 In our study, adherence to a balanced ratio was most apparent for patients in the young, that is 16-30, and middle age, that is 31-55 y old, age groups (1.1:0.9:1 and 1:0.7:1, respectively). For the older age group, that is, 56-100 y, a higher proportion of platelets was transfused. Given the common use of antiplatelet agents in the geriatric population,17 the observed increase in platelet transfusion may reflect an attempt to correct perceived platelet dysfunction.
An important strength of this study is that it is one of the largest studies to examine modern day massive transfusion practices from ACS verified trauma centers distributed across the country. The differences in characteristics of the patients in each transfusion cohort were not significant and allow for generalizability of these results to most trauma patients (Table 1). We were able to estimate an optimal volume of balanced blood transfusion unique to each age grouping, which provides maximal survival benefit without excessive blood transfusion. In addition, significant risk factors associated with higher mortality were reported, which should help bedside clinicians in making decisions to not initiate the MTP or to terminate an ongoing MTP in the setting of blood shortages.
Limitations of this study include its retrospective design and very high proportion of blunt trauma patients. No definitive interpretation of volume of transfusion as an intervention for improved mortality can be made. We are reporting the strength of the association of specific blood transfusion levels with mortality and the observed differences in raw and adjusted mortality at specific threshold levels to help guide clinical decision making, particularly with a limited blood supply. In addition, without randomization, there is the potential for significant differential bias between groups, which is unaccounted for in our multivariable regression models. For example, other predictors of mortality such as hypothermia, coagulopathy, acidosis, and a measured shock index were not readily available in our dataset. While patients who require MTP will likely have abnormal values consistent with shock, the degree in which these abnormal values may influence mortality may not be interpreted from our study. Another limitation is that the definition for massive transfusion is different from prior studies. We report a combination of blood products, not just pRBC transfusion. Thus, the transfusion volume reported may reflect fewer pRBC units compared to previous studies. However, since transfusion practices have changed within the last several years, a potential strength of this work is that it reflects current practices in MTP. Most patients received balanced blood product transfusion, particularly in the young, that is 16-30 y old, and middle aged, that is 31-55 y old age groups; this observation offers further validation for our definition of MTP.
An additional limitation is the choice of mortality outcome. A consensus statement for hemorrhage-related trials recommends the use of 3-h, 6-h, or 24-h all-cause mortality time points.18 However, obtaining mortality data at these specific time points was unobtainable due to the retrospective nature of our study. It should also be noted that the recommendations made in this work apply to MTP practices when blood products are both restricted and limited at multiple levels. This study also may lack external validity to other trauma populations. Considering variable results in health centers, we encourage other health systems to define their own futility thresholds during austere times.
Lastly, we acknowledge the limitations related to the somewhat arbitrary cutoff of a 1/3 chance of death threshold as a futility threshold. The applicability of this threshold may vary depending on patient factors, such as age and life expectancy. These thresholds should be viewed as a general guideline rather than a strict rule, as surgical judgment remains crucial in decision-making. The complexity of individual cases, including differences in injury severity and patient comorbidities, should be considered when assessing futility and determining appropriate transfusion thresholds.
Conclusion
This study identified age specific volume-to-mortality associations for balanced contemporary MTP practices. Using these data-driven guidelines, MTP practices under blood shortage conditions could follow thresholds for limiting transfusion volumes according to age group. Our findings support previous work, including that the decision to start the MTP should take into consideration whether or not the patient had a prearrival cardiac arrest or presented with nonreactive pupils related to severe TBI. Discontinuation of the MTP should be considered if the patient has a severe TBI with midline shift on brain CT scan or has sustained a cardio-pulmonary arrest (“code”) with CPR ongoing on arrival to the trauma bay.
Level of Evidence
Level II (therapeutic/care management).
Author Contributions
All authors certified they made contributions in design, data acquisition, data analysis and interpretation, and drafting and revising the manuscript. Authors have given approval of the version to be published.
Disclosure
The authors have no conflicts of interest to declare. All coauthors have seen and agreed with the contents of the manuscript and there is no financial interest to report. This research was supported (in whole or in part) by HCA Healthcare and/or an HCA Healthcare affiliated entity. The views expressed in this publication represent those of the author(s) and do not necessarily represent the official views of HCA Healthcare or any of its affiliated entities.
Funding
There was no funding provided to perform or complete this research.
Meeting Presentation
This work was presented as a poster presentation at the 80th Annual Meeting of AAST and Clinical Congress of Acute Care Surgery on September 29-October 2, 2021.
Footnotes
Supplementary data related to this article can be found at https://doi.org/10.1016/j.jss.2023.05.028.
Supplementary Materials
References
- 1.World Health Organization . World Health Organization; Geneva: 2021. Maintaining a Safe and Adequate Blood Supply and Collecting Convalescent Plasma in the Context of the COVID-19 Pandemic: Interim Guidance, 17 February 2021. [Google Scholar]
- 2.Miskeen E., Yahia A.I.O., Eljack T.B., Karar H.K. The impact of COVID-19 pandemic on blood transfusion services: a Perspective from health Professionals and donors. J Multidiscip Healthc. 2021;14:3063–3071. doi: 10.2147/JMDH.S337039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.American Red Cross American Red Cross Faces Severe Blood Shortage as Coronavirus Outbreak Threatens Availability of Nation’s Supply. 2020. https://www.redcross.org/about-us/news-and-events/press-release/2020/american-red-cross-faces-severe-blood-shortage-as-coronavirus-outbreak-threatens-availability-of-nations-supply.html Available at:
- 4.Holcomb J.B., del Junco D.J., Fox E.E., et al. The prospective, observational, multicenter, major trauma transfusion (PROMMTT) study: comparative effectiveness of a time-varying treatment with competing risks. JAMA Surg. 2013;148:127–136. doi: 10.1001/2013.jamasurg.387. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Holcomb J.B., Tilley B.C., Baraniuk S., et al. Transfusion of plasma, platelets, and red blood cells in a 1:1:1 vs a 1:1:2 ratio and mortality in patients with severe trauma: the PROPPR randomized clinical trial. JAMA. 2015;313:471–482. doi: 10.1001/jama.2015.12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.TQIP A. ACS; Chicago: 2013. Massive transfusion in trauma guidelines. [Google Scholar]
- 7.von Elm E., Altman D.G., Egger M., Pocock S.J., Gøtzsche P.C., Vandenbroucke J.P. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. J Clin Epidemiol. 2008;61:344–349. doi: 10.1016/j.jclinepi.2007.11.008. [DOI] [PubMed] [Google Scholar]
- 8.Dimick J.B., Ghaferi A.A., Osborne N.H., Ko C.Y., Hall B.L. Reliability adjustment for reporting hospital outcomes with surgery. Ann Surg. 2012;255:703–707. doi: 10.1097/SLA.0b013e31824b46ff. [DOI] [PubMed] [Google Scholar]
- 9.Dimick J.B., Staiger D.O., Birkmeyer J.D. Ranking hospitals on surgical mortality: the importance of reliability adjustment. Health Serv Res. 2010;45(6 Pt 1):1614–1629. doi: 10.1111/j.1475-6773.2010.01158.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Marshall A., Altman D.G., Holder R.L., Royston P. Combining estimates of interest in prognostic modelling studies after multiple imputation: current practice and guidelines. BMC Med Res Methodol. 2009;9:57. doi: 10.1186/1471-2288-9-57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Hébert P.C., Wells G., Blajchman M.A., et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. transfusion requirements in critical care investigators, Canadian critical care trials group. N Engl J Med. 1999;340:409–417. doi: 10.1056/NEJM199902113400601. [DOI] [PubMed] [Google Scholar]
- 12.Murphy G.J., Pike K., Rogers C.A., et al. Liberal or restrictive transfusion after cardiac surgery. N Engl J Med. 2015;372:997–1008. doi: 10.1056/NEJMoa1403612. [DOI] [PubMed] [Google Scholar]
- 13.Lacroix J., Hébert P.C., Hutchison J.S., et al. Transfusion strategies for patients in pediatric intensive care units. N Engl J Med. 2007;356:1609–1619. doi: 10.1056/NEJMoa066240. [DOI] [PubMed] [Google Scholar]
- 14.Carson J.L., Guyatt G., Heddle N.M., et al. Clinical practice guidelines from the AABB: red blood cell transfusion thresholds and storage. JAMA. 2016;316:2025–2035. doi: 10.1001/jama.2016.9185. [DOI] [PubMed] [Google Scholar]
- 15.Velmahos G.C., Chan L., Chan M., et al. Is there a limit to massive blood transfusion after severe trauma? Arch Surg. 1998;133:947–952. doi: 10.1001/archsurg.133.9.947. [DOI] [PubMed] [Google Scholar]
- 16.Vaslef S.N., Knudsen N.W., Neligan P.J., Sebastian M.W. Massive transfusion exceeding 50 units of blood products in trauma patients. J Trauma. 2002;53:291–295. doi: 10.1097/00005373-200208000-00017. discussion 295-6. [DOI] [PubMed] [Google Scholar]
- 17.Ang D., Kurek S., McKenney M., et al. Outcomes of geriatric trauma patients on preinjury anticoagulation: a multicenter study. Am Surg. 2017;83:527–535. [PubMed] [Google Scholar]
- 18.Holcomb J.B., Moore E.E., Sperry J.L., et al. Evidence-based and clinically relevant outcomes for hemorrhage control trauma trials. Ann Surg. 2021;273:395–401. doi: 10.1097/SLA.0000000000004563. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.