Age-Related Mortality in Trauma Patients Requiring Massive Transfusion

Abstract

Background

Massive transfusion (MT) protocols improve survival in trauma patients. Elderly trauma patients requiring MT represent a high-risk population, yet outcome data remain limited. Understanding age-related differences is critical to guide resuscitation and resource allocation.

Objective

To characterize the association between MT and mortality in elderly vs younger trauma patients.

Participants

Trauma patients at a Level I trauma center (1/2013-09/2024) who required MT (≥10 units of whole blood (WB) and/or packed red blood cells (pRBC) within 24 h) were included. Patients ≥65 (elderly) were compared to <65 (non-elderly). The primary outcome was 30-day mortality. Secondary outcomes included ICU and hospital length of stay (LOS), and ventilator days.

Results

Of 368 patients meeting inclusion criteria, 30 (8%) were elderly. Elderly patients were equally likely to be male (70% vs 83%, P = 0.06), but were significantly less likely to present with GCS ≤8 (20% vs 45%, P < 0.01) despite a significantly higher incidence of severe head trauma (AIS head >3 [53% vs 32%, P = 0.02]). There was no difference in median blood products transfused within the first 24 hours (23 vs 22 units, P = 0.95). Overall mortality was 51%, higher in elderly patients (73% vs 49%, P < 0.01), with shorter time to death (median 14 vs 34 days, P < 0.01). Adjusted Cox regression confirmed significantly higher adjusted mortality in the elderly (HR 1.25, P = 0.04).

Conclusion

Elderly trauma patients requiring MT experience earlier and significantly higher mortality than younger patients, highlighting the need for improved risk stratification and tailored resuscitation strategies.

Keywords

trauma trauma acute care

Key Takeaways

• Elderly trauma patients (≥65 years) requiring massive transfusion experience earlier and significantly higher mortality than younger patients.

• Early identification of high-risk elderly patients and incorporation of structured goals-of-care discussions and multidisciplinary decision-making may therefore be particularly important in this population.

Introduction

Hemorrhage remains one of the leading causes of preventable death following traumatic injury, and massive transfusion (MT) protocols have become a cornerstone of modern trauma resuscitation.¹ Early, balanced transfusion strategies prioritizing plasma, platelets, and packed red blood cells (pRBC) have been shown to improve hemostasis and survival in severely injured patients.^1-3

While these advances have improved outcomes in the general trauma population, elderly trauma patients remain particularly vulnerable. Older adults experience higher mortality after injury compared with younger patients, even after adjustment for injury severity, and are less tolerant of hemorrhagic shock and physiologic stress.^4,5 Age-related declines in physiologic reserve, along with a higher prevalence of comorbidities and frailty, may limit the effectiveness of standard resuscitation strategies in this population.^6,7 Traditional trauma scoring systems may inadequately account for these factors, potentially underestimating mortality risk in elderly patients.^5-7

The growing adoption of whole blood-based resuscitation further complicates interpretation of transfusion practices and outcomes. While whole blood provides a balanced hemostatic product, variability in its use and in the transition to component therapy may result in transfusion patterns that differ from those evaluated in major clinical trials.^3,8 The PROPPR (Pragmatic, Randomized Optimal Platelet and Plasma Ratios) trial demonstrated improved hemostasis with early balanced component therapy using a 1:1:1 ratio.² Additionally, practice patterns surrounding massive transfusion have evolved over time, with variability in institutional adoption potentially influencing observed outcomes.^2,3

Despite recognition that elderly patients experience worse outcomes after trauma, the interaction between massive transfusion practices and mortality in this population remains incompletely defined. In particular, transfusion characteristics, timing of death, and independent predictors of mortality among elderly patients receiving massive transfusion have not been well characterized. A better understanding of these factors is essential to inform resuscitation strategies, guide prognostication, and optimize resource utilization in this growing patient population.

The objective of this study was to characterize the association between massive transfusion and mortality in elderly trauma patients compared with younger patients at a Level I trauma center. We hypothesized that elderly patients requiring MT would experience significantly higher adjusted 30-day mortality compared to their younger counterparts.

Patients and Methods

This retrospective observational study was approved by the institutional review board. The trauma registry of our Level 1 trauma center was queried for consecutive patients who received MT from January 1, 2013 through September 30, 2024 and who met the following criteria:

(1) Age 18 years or older

(2) Received MT, defined as cumulative total of 10 units of WB and pRBC within the first 24 h from admission.

The definition of massive transfusion used in this study incorporated both whole blood and pRBC to reflect contemporary resuscitation practices. As whole blood provides a balanced hemostatic product, inclusion of both components allows for more accurate characterization of transfusion burden across evolving clinical protocols. Patients defined as receiving MT, and analyzed at two time points, 4 h and 24 h. Completion of massive transfusion within 4 h was defined as receipt of ≥10 units of whole blood and/or pRBC within the first 4 h of admission.

Tracked variables included demographics, mechanism of injury, admission vital signs, initial Emergency Department Glasgow Coma Scale scores (GCS) Injury Severity Score (ISS), Abbreviated Injury Scale (AIS) score for all body regions, intensive care unit (ICU) admission, hospital length of stay (LOS), ICU LOS, ventilator days, transfusion data at 4 h and 24 h, and mortality. Variables were selected a priori based on clinical relevance and prior literature. Patients were then categorized into two cohorts by age, with patients ≥65 defined as elderly.

Statistical analysis was performed using IBM SPSS for Windows, version 29.0.2.0 (IBM Corp., Armonk, NY). Continuous variables are presented as medians with interquartile ranges (IQRs), and categorical variables as counts with percentages. Elderly (≥65 years) and non-elderly (<65 years) patients were compared using the χ² test or Fisher’s exact test for categorical variables, while means were compared using t-test or Mann-Whitney U test, as appropriate. The primary outcome was 30-day mortality, while secondary outcomes included hospital and ICU LOS and ventilation days. Time to death was evaluated using Kaplan-Meier methods, and group differences were assessed using the log-rank test.

Multivariable Cox proportional hazards regression was used to estimate adjusted associations with time to death, with covariates selected based on univariable associations (P < 0.2). Collinearity between covariates was assessed using correlation matrices. Highly collinear variables (eg, GCS and head AIS, ISS and regional AIS scores) were not included in the same model; selection was guided by clinical relevance and prior literature. Results were reported as hazard ratios (HRs) with 95% confidence intervals (CIs). A binary logistic regression was performed to identify independent predictors of 30-day mortality in the subgroup of patients completing MT within 24 h and in those completing it within 4 h. Results were reported as odds ratios (ORs) with corresponding 95% CIs. Statistical significance was defined as P < 0.05.

Results

Over the approximately 10-year study period, a total of 368 patients met inclusion criteria. Of these, 30 (8%) were elderly (age ≥65 years) (Figure 1). The median age of the elderly group was 73 (IQR 69-81 years), and non-elderly was 33 (IQR 25-46). Sex distribution was similar, with the majority being male in both groups (70% vs 83%, P = 0.06). There was a statistically significant difference in the race of the groups, with elderly patients significantly more likely to be white (Table 1). Blunt injury was the primary mechanism of injury for elderly patients at 93% vs 58% for non-elderly (P < 0.01). Within the blunt injury category, elderly patients were significantly more likely to require MTP in auto vs pedestrian accidents (82.1% vs 35%, P < 0.01). The median ISS was comparable for both groups (elderly: 34 [IQR 30-42] vs non-elderly: 29 [IQR 29-41], P = 0.07). Elderly patients were significantly more likely to sustain severe head trauma, defined as AIS head ≥3 (53% vs 32%, P = 0.02). Table 1 shows further demographic and injury characteristic data. Vital signs on admission demonstrate that a majority of patients in both groups were presenting with hemodynamic instability, with median shock indices of 1.6 for both groups (Table 2). Initial GCS scores taken and elderly patients were less likely to have a GCS ≤8 (20% vs 48%, P = 0.01). There was no difference in median number of pRBC and WB transfused within the first 4 h, though there was a significant difference between median number of units of plasma, with elderly receiving 3.9 (IQR 1.6-6.1), and non-elderly 6.3 (IQR 2-13.4) units. However, there was no difference in the median number of pRBC, WB, and plasma transfused within the first 24 h (Table 3).

Figure 1.

Study flow chart of included patients

Table 1.

Patient Demographics and Injury Characteristics

	≥65	<65	P value
Age	73 (69-81)	33 (25-46)	<0.01
Sex
Male	70%	83%	0.06
Race
White	73%	31%	<0.01
Black/African American	3%	25%
Hispanic/Latino	3%	8%
Asian	3%	7%
Other	17%	29%
Mechanism of injury
Blunt	93%	58%	<0.01
Type of blunt injury
AvP	82%	35%	<0.01
MVC	11%	34%
ISS	34 (30-42)	29 (29-41)	0.07
AIS head ≥3	53%	32%	0.02
AIS chest ≥3	73%	55%	0.05
AIS A/P ≥3	63%	54%	0.30

ISS, injury severity score; AIS abbreviated injury score; A/P, abdomen/pelvis; AvP, auto vs pedestrian; MVC, motor vehicle collision.

Table 2.

Initial Emergency Department Vitals

	≥65	<65	P value
SBP <90 mmHg	83%	74%	0.25
Heart rate bpm	100 (81-122)	115 (87-130)	0.67
Lowest SBP mmHg	66 (54-75)	69 (57-86)	0.12
Shock index	1.6 (1.1-1.9)	1.6 (1.1-2.0)	0.66
GCS ≤8	20%	48%	0.01

SBP, systolic blood pressure; GCS, Glasgow Coma Scale.

Table 3.

Patient Massive Transfusion Data

	≥65	<65	P value
4 h
#Units of PRBC + WB	12.3 (9.3-15.2)	13.7 (9.3-20.8)	0.30
#Units of FFP	3.9 (1.6-6.1)	6.3 (2-13.4)	<0.01
24 h
#Units of PRBC + WB	14.0 (11.7-17.7)	16.7 (12-23)	0.95
#Units of FFP	6.7 (4.1-9.9)	9.1 (5.4-16.2)	0.40

PRBC, packed red blood cells; WB, whole blood; FFP, fresh frozen plasma.

Primary Outcomes

The overall 30-day mortality was 51% (n = 189). Kaplan-Meier survival analysis demonstrated a clear separation between the two age groups (Figure 2). Elderly patients exhibited significantly higher mortality compared with non-elderly patients (73% [n = 22] vs 49% [n = 167], respectively, P=<0.01). In addition, median time to death was markedly shorter in the elderly cohort (3 days) compared with the non-elderly cohort (34 days). A multivariable Cox proportional hazards model adjusting for differences between the two cohorts (Chest AIS ≥3, Head AIS ≥3, ISS, ED GCS ≤8, age ≥65 years, and lowest SBP), which demonstrated that elderly patients had significantly higher mortality with an adjusted hazard ratio of 1.25, P < 0.04.

Figure 2.

Survival curve for patients receiving massive transfusion by age group

Secondary Outcomes

The median hospital LOS was significantly shorter for the elderly group (3.5 days [IQR, 1-14.8]) compared to the non-elderly cohort (11 days [IQR, 1-30], (P < 0.01). The median ICU LOS was similar for both cohorts (5.5 days [IQR 3-14.8] vs 6 days [IQR, 2-14], P = 0.8). The median ventilation days for both cohorts were the same at 3 days (IQR, 1-11) (Table 4). When excluding patients that died during admission, the median hospital LOS for the groups was similar (elderly 22.5 days [IQR 17-28.8] vs non-elderly 28 days [IQR 14.5-47], P = 0.48).

Table 4.

Secondary Outcomes

	≥65	<65	P value
Hospital LOS days	3.5 (1-14.8)	11 (1-30)	<0.01
Hospital LOS excluding deaths	22.5 (17-28.8)	28 (14.5-47)	0.48
ICU LOS days	5.5 (3-14.8)	6 (2-14)	0.83
Ventilator days	3 (1-11)	3 (1-11)	0.55

LOS, Length of Stay; ICU, intensive care unit.

Independent Predictors of Mortality for Patients Completing MT Within 24 hours

A binary logistic regression was performed to identify independent predictors of mortality for patients who completed MT by 24 h. The model included Chest AIS ≥3, Head AIS ≥3, Injury Severity Score (ISS), admission Glasgow Coma Scale (GCS) ≤8, age ≥65 years, and lowest systolic blood pressure (SBP) as predictors. Age ≥65 years and admission GCS ≤8 were independently associated with mortality; however, age ≥65 years was the strongest predictor, with 5.5-fold higher odds of mortality (95% CI [2.2-13.8], P < 0.01), while GCS ≤8 demonstrated less pronounced association (Table 5). In contrast, Head AIS ≥3 (P = 0.2), Chest AIS ≥3 (P = 0.7), and lowest SBP (P = 0.8) did not predict mortality.

Table 5.

Independent Predictors of Mortality for Patients Completing MT Within 24 h

Predictor	P value	OR (95% CI)
Age ≥65 years	<0.01	5.5 (2.2-13.8)
GCS ≤8	<0.01	6.7 (3.9-11.4)

GCS, Glasgow Coma Scale.

Independent Predictors of Mortality for Patients Completing MT Within 4 hours

A binary logistic regression was performed to identify independent predictors of mortality for patients who completed MT by 4 hours, which was not significantly associated with mortality (OR = 1.37, 95% CI [0.56-3.35], P = 0.5). Among patients completing massive transfusion within 4 hours, age ≥65 years was the strongest predictor of mortality (OR = 4.8, 95% CI [1.6-13.8], P < 0.01). Admission GCS ≤8 demonstrated more modest association and should be interpreted with caution (Table 6). Systolic blood pressure <90 mmHg, head AIS ≥3, and chest AIS ≥3 were not statistically significant predictors (all P > 0.05).

Table 6.

Independent Predictors of Mortality for Patients Completing MT Within 4 h

Predictor	P value	OR (95% CI)
Age ≥65 years	<0.01	4.8 (1.6-13.8)
GCS ≤8	<0.01	6.6 (3.1-14.1)

GCS, Glasgow Coma Scale.

Discussion

In this retrospective analysis, elderly patients experienced significantly higher and early mortality compared with younger patients, despite receiving similar volumes of blood products. Advanced age remained an independent predictor of mortality after adjustment for injury severity, physiologic derangement, and anatomic injury patterns, suggesting that the survival benefits observed with massive transfusion in the general population may be attenuated in elderly patients.

Notably, elderly patients demonstrated worse survival despite presenting with less severe neurologic compromise, as evidenced by a lower proportion of patients with admission GCS ≤8, and similar overall injury severity compared with younger patients. Although elderly patients were more likely to sustain severe head injury, head AIS ≥3 was not independently associated with mortality after adjustment, suggesting that chronological age and global physiologic vulnerability may outweigh isolated anatomic injury in determining outcomes. Additionally, comorbidities, medication use (eg, sedatives or anticoagulants), and limitations of GCS as a neurologic assessment tool in older adults may contribute to this discrepancy. These findings are consistent with prior literature demonstrating that traditional injury scoring systems incompletely capture physiologic reserve, frailty, and comorbidity burden in older adults.^4-6 Chronologic age may therefore serve as a surrogate for diminished tolerance to hemorrhagic shock, coagulopathy, and the systemic inflammatory response to injury.⁷

Despite comparable transfusion requirements at both early (4-h) and late (24-h) time points, elderly patients experienced markedly earlier mortality, with a median time to death of only 3 days. This finding suggests that mortality in elderly patients requiring MT is driven primarily by early physiologic failure rather than late complications such as infection or prolonged critical illness. Similar patterns of early mortality in geriatric trauma populations have been reported previously.^8,9 The shorter hospital and ICU lengths of stay observed among elderly patients in this study likely reflect early mortality rather than improved recovery and should therefore be interpreted with caution.¹⁰

Importantly, neither early completion of MT nor transfusion volume was independently associated with mortality in multivariable analyses, whereas age, injury severity, and admission neurologic status remained dominant predictors. Interpretation of analyses based on completion of massive transfusion within 4 h should be made with caution. Although a majority of patients in both elderly and younger cohorts completed massive transfusion within this timeframe (63% and 71%, respectively), this approach may still introduce survivorship bias, as patients who die early may not survive long enough to meet transfusion thresholds. These findings suggest that while balanced resuscitation strategies such as those validated in the PROPPR trial are necessary to address hemorrhagic shock, they may be insufficient to overcome the reduced physiologic resilience associated with advanced age.^2,3,8 Differences in early transfusion practices may also contribute to observed outcomes. Early balanced resuscitation with plasma, platelets, and red blood cells has been associated with improved hemostasis; however, variability in initial transfusion ratios and timing of plasma administration may influence outcomes. Early identification of high-risk elderly patients and incorporation of structured goals-of care discussions and multidisciplinary decision-making may therefore be particularly important in this population.¹¹

This study has several limitations. Its retrospective design limits causal inference and introduces potential selection bias, and the relatively small elderly cohort may reduce statistical power. Important factors such as frailty, comorbidity burden, anticoagulant use, and goals-of-care decisions were not captured and may confound outcomes; in particular, withdrawal of care among elderly patients may influence observed mortality. Additionally, data on the source of hemorrhage and cause of death were not consistently available. Although multivariable modeling was performed, residual confounding, including potential collinearity among injury severity measures, remains possible. Finally, transfusion practices evolved over the study period, including increased adoption of whole blood and variability in implementation of massive transfusion protocols. These temporal changes may influence both transfusion patterns and outcomes and should be considered when interpreting the results.

Future research should focus on identifying modifiable contributors to excess mortality among elderly trauma patients requiring massive transfusion. Prospective studies incorporating frailty indices, pre-injury functional status, comorbidity burden, and anticoagulant or antiplatelet use may improve risk stratification beyond chronological age.^5,6,12 Large, multicenter studies with adequate representation of elderly patients will be essential to validate these findings and guide optimization of resuscitation strategies and resource utilization in this growing and vulnerable trauma population.^4,8

In conclusion, elderly trauma patients requiring massive transfusion experience significantly higher and earlier mortality than younger patients, independent of transfusion volume and traditional injury severity metrics. This higher early mortality has important implications for prognostication and goals-of-care discussions. Incorporating age-informed risk stratification into early trauma decision-making may support timely, patient-centered conversations about expected clinical trajectories and alignment of treatment intensity with patient values, while continuing to prioritize aggressive hemorrhage control when appropriate.¹³

Footnotes

ORCID iDs

Sophie Gonzalez

Andrew Maneval

Tejal Brahmbhatt

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

Declaration of Conflicting Interests

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

References

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.

Holcomb

Tilley

Baraniuk

, et al. Transfusion of plasma, platelets, and red blood cells in a 1:1:1 vs 1:1:2 ratio and mortality in patients with severe trauma: the PROPPR randomized clinical trial. JAMA. 2015;313(5):471-482. doi:10.1001/jama.2015.12.

Holcomb

del Junco

Fox

, et al. The prospective, observational PROMMTT study: comparative effectiveness of a time-varying treatment with competing risks. JAMA Surg. 2013;148(2):127-136.

Callcut

Kornblith

Conroy

, et al. The why & how our trauma patients die: a prospective multicenter Western Trauma association study. J Trauma Acute Care Surg. 2019;86(5):864-870.

Hashmi

Kaji

Nathens

. Practical guide to geriatric trauma: a review of the literature. J Trauma Acute Care Surg. 2013;75(6 suppl 3):S56-S63.

Joseph

Pandit

Zangbar

, et al. Validating trauma-specific frailty index for geriatric trauma patients: a prospective analysis. J Am Coll Surg. 2014;219(1):10-17.

Cook

Joseph

Inaba

, et al. Multicenter external validation of the geriatric trauma outcome score: a prognostic tool for elderly trauma patients. J Trauma Acute Care Surg. 2016;80(2):204-209.

Lord

Midwinter

Chen

, et al. The systemic immune response to trauma: an overview of pathophysiology and treatment. Lancet. 2014;384(9952):1455-1465.

Plummer

Gordon

Gillman

, et al. Outcomes of massive transfusion in geriatric trauma patients compared with younger cohorts. Injury. 2014;45(9):1296-1300.

10.

Evans

Cook

Christy

, et al. Geriatric trauma: predictors of mortality and patterns of care. J Trauma. 2010;68(2):345-350.

11.

Taylor

Tracy

Meyer

Pasquale

Napolitano

. Trauma in the elderly: intensive care unit resource use and outcome. J Trauma. 2002;53(3):407-414.

12.

Mosenthal

Murphy

Barker

Lavery

Retano

Livingston

. Changing the culture around end-of-life care in the trauma intensive care unit. J Trauma. 2008;64(6):1587-1593.

13.

Maxwell

Mion

Mukherjee

, et al. Preinjury physical frailty and trauma outcomes in older adults. J Trauma Acute Care Surg. 2019;86(6):1039-1045.