Abstract
Purpose
Although balanced transfusion protocols (platelet concentrate (PC): red blood cell (RBC) = 1:1) are standard for severe trauma, the impact of a higher PC: RBC ratio (> 1:1) on survival remains unclear. This study aimed to investigate the effectiveness of a high PC to RBC among severely injured patients requiring massive transfusion.
Methods
We performed a retrospective cohort study using the Japan Trauma Data Bank (2019–2023). Adult patients receiving massive transfusion (> = 10 units of RBC) within the first 24 h of injury were included. Patients were classified into high PC (> 1) and low PC groups (< = 1). The primary outcome was 24-hour mortality, analyzed using inverse probability of treatment weighting and modified Poisson regression. Additionally, we performed spline curve analysis to investigate the nonlinear relationship between the PC: RBC ratio and outcome.
Results
Among 3,067 patients (mean age 57.1 years, 67.2% male), 934 were in the high PC group and 2,133 in the low PC group. The high PC group had lower 24-hour mortality (8.9% vs. 15.7%). Adjusted analysis showed a significantly lower risk of mortality in the high PC group (risk ratio 0.54; 95% CI 0.43–0.69). A nonlinear analysis suggested that increasing PC: RBC ratio was associated with lower mortality.
Conclusion
A high PC: RBC ratio was associated with lower 24-hour mortality in severely injured patients requiring massive transfusion, suggesting that a higher PC strategy may improve outcomes.
Supplementary Information
The online version contains supplementary material available at 10.1007/s00068-026-03171-y.
Keywords: Transfusion, Hemorrhagic shock, Trauma, Platelet, Ratio
Introduction
Traumatic hemorrhagic shock is one of the major causes of mortality in severely injured patients [1–4]. Trauma induced coagulopathy (TIC) develops in 25% of these severely injured patients [5] and is the one component of lethal triad of trauma death, and its incidence is directly associated with injury severity and mortality [6]. The damage control resuscitation (DCR) has been introduced to mitigate the effect of the lethal triad including TIC through hemorrhage control, minimize crystalloid and a balanced transfusion strategy, and contributed to the survival of hemorrhagic shock patients at risk of mortality [1, 7]. Balanced transfusion is a key component of DCR, and providing blood components in a fixed ratio of fresh frozen plasma (FFP): platelet concentrate (PC): red blood cell (RBC) = 1:1:1 has been reported to reduce mortality in patients with traumatic hemorrhagic shock [8, 9].
Regarding balanced transfusion, platelet has been particularly important in hemostatic role. Pivotal randomized controlled PROPPR study investigated the effect of a high ratio of FFP, PC, and RBC in a 1:1:1 configuration compared to a 1:1:2 ratio, and concluded that more patients in the 1:1:1 group achieved hemostasis and that there was fewer mortality due to exsanguination within 24 h [10]. Therefore, the Eastern Association for the Surgery of Trauma guideline recommended high ratios of PC to RBC, as close as possible 1:1 [11]. In addition, the European guideline on the management of major bleeding and coagulopathy recommended a ratio of PC to RBC, of at least 1:2, as needed [12].
However, the optimal ratio of PC to RBC was not fully understood. In particular, evidence is lacking on whether a much higher ratio of PC to RBC (above 1:1) could improve survival in severely injured patients. Therefore, we aimed to investigate the effectiveness of a high PC to RBC ratio among severely injured patients who required massive transfusion. Our hypothesis is that a higher ratio of PC to RBC is associated with lower mortality. Based on this hypothesis, we explored the optimal ratio of PC to RBC to reduce patient mortality.
Methods
This retrospective observational study used data from the Japan Trauma Data Bank (JTDB). This study conforms with the Strengthening the Reporting of Observational Studies in Epidemiology guidelines, and a complete checklist has been uploaded as Supplementary Table 1. We conducted the study based on an approved by institutional review board of National Defense Medical College Hospital (Number 4971), which waived the requirement for informed consent because of the retrospective nature of the study using anonymized data.
Data source
The JTDB is a natonwide trauma registry established in 2003 by the Japanese Association for the Surgery of Trauma and the Japanese Association for Acute Medicine to ensure and improve the quality of trauma care in Japan. The data set currently encompasses information from 302 acute care hospitals in Japan [13]. Participating hospitals are required to register all patients admitted for trauma with an Injury Severirty Score (ISS) of ≧ 9.
Study population
We used the JTDB between January 1, 2019 and December 31, 2023, we idenfied patients 16 years or older who were admitted to hospital for trauma and received massive transfusion within the first 24 h of injury. We defined massive transfusion as receiving at least 10 units of red blood cell transfusion. In Japan, each unit of RBC and FFP is derived from 200 ml of whole blood, containing approximately 140 ml and 120 ml, respectively. A single unit of PC contains approximately 0.2 × 1011 platelets with a volume of approximately 20 ml [14]. Therefore, one unit of each blood component product in Japan corresponds to approximately half of the dose typically used in Europe and the United States. We defined the following exclusion criteria: (1) Indirectly transferred from the scene; (2) Trauma mechanism except for blunt or penetrating; (3) Cardiopulmonary arrest at hospital arrival; and (4) Abbreviated injury scale (AIS) = 6 in any body region.
Study variables
We collected the following variables: demographics; charlson comorbidity index (CCI) [15]; blunt or penetrating trauma, mechanism of injury, characteristics at arrival including systolic blood pressure (SBP), heart rate (HR), respiratory rate (RR) amd Glasgow coma scale (GCS); lactate value; diagnostic procedure including whole body computed tography (CT) scan and angiography for chest, abdomen or pelvis; trauma scores such as AIS and injury severity score (ISS); treatment within 24 h including tranexamic acid use and blood transfusion; hemostatic procedure for thorax and abdomen, damage control surgery, mortalities including 24-hour mortality, in-hospital mortality and emergency department (ED) mortality, and complications including pulmonary embolism, acute respiratory distress syndrome (ARDS), acute kidney injury (AKI), and pulmonary edema.
Outcome
We defined the primary outcome as 24-hour mortality to focus on hemostasis. Secondary outcomes included in-hospital mortality, ED mortality, transfusion volume within 24 h, hemostatic procedures and complications.
Data management
We identified the missingness for variables and imputed the missing variables by the random forest method (Supplementary Table 2) [16]. First, we conducted a descriptive survey of the characteristics among studied patients. Continuous variables are presented as mean and standard deviation (SD), and categorical variables are presented as numbers and percentages.
Investigation of the effect of a high platelet to red blood cell ratio on outcomes
To investigate the association of a high PC to RBC ratio on outcomes, we defined the exposure group as patients who received a high PC ratio (PC to RBC ratio > 1) and the comparison as those receiving a low ratio (PC to RBC ratio = < 1). After a descriptive analysis of patient characteristics between the high and low PC groups, we applied inverse probability of treatment weighting (IPTW) with stabilized weights to account for baseline differences and to create a pseudo-population in which transfusion ratio was independent of the measured covariates [17].
We calculated the propensity score for receiving a high PC transfusion using a multivariable logistic regression model based on the variables including age, sex, CCI, trauma mechanism, mechanism of injury, vital signs on hospital arrival, lactate level, use of whole body CT, angiography for the chest, abdomen or pelvis, AIS for the head/neck, chest, abdomen, and pelvis/lower extremity, ISS and administration of tranexamic acid referencing previous literatures [18, 19].
If properly constructed, IPTW can control for confounding variables measured across study groups. We confirmed adequate overlap in propensity scores between the high and low PC groups. Covariates balance in the weighted cohort was assessed using P-value. Finally, we constructed an IPTW modified Poisson regression model to estimate the risk ratio (RR) and 95% confidence intervals (CIs) for each outcome, using the low PC group as the reference [20].
Subgroup analysis
Additionally, we defined six subgroups to explore the heterogeneity of trauma-induced coagulopathy within the study population and to identify populations that may benefit from a high PC strategy in terms of mortality reduction. The subgroups included patients who: (1) presented with shock (systolic blood pressure < 90mmHg at arrival), (2) underwent hemorrhage control procedures, (3) received more than 20 units of RBC within 24 h of arrival, (4) had a thorax AIS > = 3, (5) had an abdomen AIS > = 3, (6) had a lower extremity AIS > = 3, and (7) (8) with and without TXA. We conducted IPTW analysis using the same covariates and methodology as in the primary analysis to evaluate the association between the PC to RBC ratio and mortality in each predefined subgroup.
Sensitivity analysis
PC are generally administered later than RBC and/or FFP in massively transfused patients [21], potentially introducing immortal time bias [22]. To mitigate this bias, we conducted five sensivitity analyses. Sensitivity analysis 1: excluded cases of ED mortality Sensitivity analysis 2: excluding cases without platelet transfusion. These sensitivity analysis aimed to examine the potential impact of a scenario in which patients categorized to low-ratio group died in ED before receiving sufficient amount of platelet transfusion or they had no chance to receive the platelet transfusion such as die prior to the platelet transfusion. Moreover, we performed two following sensitivity analyses to investigate the scenario in which transfusion volume was considered a major confounder, adjusted for RBC transfusion volume (Sensitivity analysis 3) and adjusted for RBC and FFP transfusion volumes (Sensitivity analysis 4). Since transfusion timing may also act as a confounder, we conducted another analysis (Sensitivity analysis 5) adjusted for the time to RBC transfusion. For all five sensitivity analyses, we applied an IPTW model consistent with the primary analysis. Next, we performed doubly robust estimation to account for possible misspecification in the IPTW model [23]. Finally, because transfusion practice could vary by institution, we constructed a multilevel model to incorporate the heterogeneity in the institutions. In this model, the variables in original analysis were treated as fixed effects, and the institution identifier was included as a random effect.
Investigation of the association of a high platelet to plasma ratio on outcomes
In addition, to explore the association between high platelet to plasma ratio on outcomes, we conducted a similar modified Poisson regression model with IPTW by defining the exposure as a high PC ratio (PC to to FFP ratio ≥ 1) and the comparison as a low ratio (PC to FFP ratio < 1).
Investigation of the optimal ratio of the platelet to red blood cell
To identify the PC to RBC ratio associated with the highest probability of survival, we constructed modified Poisson regression models using detalied categories: PC ratios as < 0.5, 0.5-1.0, 1.0-1.5, and ≥ 1.5, with the < 0.5 group used as the reference. The same covariates as in the original IPTW model were adjusted for in this models. In addition, we created the spline curve analysis to investigate the nonlinear relationship between the PC to RBC ratio and 24-hour mortality in overall and the subgroup who received surgical procedure [18].
We performed all statistical analyses using R software (version 4.2.2; R Foundation for Statistical Computing, Vienna, Austria).
Results
During the study periods, 3,894 patients received massive transfusions. After applying the predefined exclusion criteria, 3,067 patients from 188 hospitals remained for analysis. Supplementary Fig. 1 presents a flow diagram of patient selection.
Supplementary Table 3 summarizes the other baseline characteristics and outcomes of the overall study population. The mean age was 57.1 years (22.5), and 67.2% of the patients were male. Blunt trauma accounted for 92.5% of cases. The mean SBP was 100.8 (39.8) mmHg, mean HR was 105.5 (28.5) beats per minute, and the mean lactate level was 6.1 (5.3) mmol/L. Whole body CT was performed in 2,771 (90.4%) patients. The mean ISS was 29.3 (13.3). TXA was administered to 1,427 (46.5%) patients. The mean number of units transfused was 22.4 (18.9) for RBC, 26.1 (75.4) for FFP, and 16.7 (14.8) for PC. Hemostatic procedures were performed in 85.1% of patients. The primary outcome was 24-hour mortality, which occurred in 13.6% of patients (417/3,067). The secondary outcome was in-hospital mortality, which occurred in 32.3% of patients (991/3,067).The overall complication rate was 38.2% (1,171/3.067). Mean PC units of study patients was 16.7 (13.5) units and frequnecy of PC units is shown in Fig. 1 and Supplementary Fig. 2. The distribution of PC to RBC ratios across hospitals is shown in Supplementary Fig. 3. The variables with the highest rates of missing data were lactate (20.8%), and respiratory rate (12.0%), while missingness for all other variables was below 10% (Supplementary Table 2).
Fig. 1.
Histogram of PC units of overall study patients. PC, platelet concentrate
Results of investigation of the association of high-platelet ratio on outcomes
Among the study patients, 934 patients (30.5%) were classified into the high PC group and 2,133 patients (69.5%) into the low PC group. Table 1 presents the baseline characteristics of the low PC and high PC groups before and after IPTW. The distribution of propensity scores is shown in Supplementary Fig. 4. After IPTW, all SMD between the groups were below 0.1, indicating covariates were well balanced. Table 2 shows the interventions and outcomes for the low and high PC groups before IPTW. The mean transfusion volumes within the first 24 h were as follows: RBC, 23.6 (21.5) units in the low PC groups vs. 19.7 (10.7) units in the high PC group; FFP, 25.7 (84.5) units vs. 27.2 (48.7) units; and PC, 12.6 (11.1) units vs. 26.1 (13.7) units, respectively. The proportions of 24-hour mortality was 15.7% (334/2,133) in the low PC group and 8.9% (83/934) in the high PC group.
Table 1.
Baseline characteristics of studied patients between low and high ratio of platelet concentrate (PC) to red blood cell (RBC)
| Variables | Before IPTW | After IPTW | ||||
|---|---|---|---|---|---|---|
| Low PC N = 2,133 |
High PC N = 934 |
P-value | Low PC N = 3,069 |
High PC N = 3,057 |
P-value | |
| Demographics | ||||||
| Age, y, mean (SD) | 55.9 (22.3) | 59.5 (22.6) | < 0.01 | 57.0 (22.2) | 56.9 (22.8) | 0.88 |
| Sex, male, n (%) | 1,450 (68.0) | 612 (65.5) | 0.19 | 2,064 (67.2) | 2,046 (66.9) | 0.86 |
| CCI, mean (SD) | 0.4 (0.9) | 0.5 (1.1) | < 0.01 | 0.4 (1.0) | 0.4 (0.9) | 0.84 |
| Trauma mechanism | ||||||
| Blunt, n (%) | 1,966 (92.2) | 872 (93.4) | 0.26 | 2,839 (92.5) | 2,731 (92.3) | 0.84 |
| Mechanism of injury, n (%) | ||||||
| Pedestrian (TA) | 339 (15.9) | 188 (20.1) | < 0.01 | 527 (17.2) | 527 (17.3) | 0.97 |
| Bicycle (TA) | 166 (7.8) | 65 (7.0) | 0.46 | 228 (7.5) | 219 (7.2) | 0.81 |
| Motorbike (TA) | 266 (13.4) | 95 (10.2) | 0.01 | 122 (10.5) | 111 (9.5) | 0.91 |
| Car (TA) | 339 (15.9) | 146 (15.6) | 0.87 | 485 (15.8) | 479 (15.7) | 0.94 |
| Slip | 44 (2.1) | 35 (3.7) | < 0.01 | 78 (2.6) | 78 (2.6) | 0.99 |
| Fall | 142 (6.7) | 84 (9.0) | 0.02 | 228 (7.4) | 228 (7.5) | 0.96 |
| Free fall | 439 (20.6) | 199 (21.3) | 0.66 | 637 (20.8) | 631 (20.7) | 0.95 |
| Others | 378 (17.7) | 122 (13.1) | < 0.01 | 501 (16.3) | 504 (16.5) | 0.91 |
| Vital signs at arrival, mean (SD) | ||||||
| SBP, mmHg | 100.8 (39.0) | 101.0 (41.6) | 0.90 | 100.7 (39.2) | 100.3 (41.2) | 0.77 |
| HR, bpm | 106.0 (28.8) | 104.4 (28.0) | 0.16 | 105.5 (28.6) | 105.8 (27.8) | 0.74 |
| RR, bpm | 24.1 (8.7) | 25.0 (9.4) | 0.01 | 24.3 (8.8) | 24.3 (8.9) | 0.98 |
| GCS | 9.8 (4.7) | 9.8 (4.6) | 0.89 | 9.7 (4.6) | 9.7 (4.6) | 0.77 |
| Lactate value, mmol/L, mean (SD) | 6.1 (5.3) | 6.0 (5.5) | 0.76 | 6.1 (5.3) | 6.1 (5.4) | 0.77 |
| Diagnostic procedure, n (%) | ||||||
| Whole body CT | 1,917 (89.9) | 854 (91.4) | < 0.01 | 2,757 (89.8) | 2,791 (91.3) | 0.37 |
| Angiography | 787 (36.9) | 436 (46.7) | < 0.01 | 1,226 (40.0) | 1,236 (40.5) | 0.81 |
| Trauma scores, mean (SD) | ||||||
| AIS for head and neck | 1.8 (2.0) | 2.1 (2.0) | < 0.01 | 1.8 (2.0) | 1.8 (2.0) | 0.81 |
| AIS for chest | 2.2 (1.7) | 2.3 (1.7) | 0.03 | 2.2 (1.7) | 2.2 (1.7) | 0.89 |
| AIS for abdomen | 1.7 (1.6) | 1.8 (1.6) | 0.06 | 1.7 (1.6) | 1.7 (1.6) | 0.84 |
| AIS for pelvis, extremities | 2.3 (1.7) | 2.3 (1.8) | 0.88 | 2.2 (1.7) | 2.2 (1.7) | 0.97 |
| ISS | 28.6 (13.5) | 31.0 (12.7) | < 0.01 | 29.4 (13.7) | 29.6 (12.5) | 0.68 |
| TXA, n (%) | 949 (44.5) | 478 (51.2) | < 0.01 | 1,427 (46.5) | 1,414 (46.3) | 0.90 |
IPTW, inverse probability of treatment weighting; SMD, standardized mean difference; SD, standard deviation; CCI, Charlson comorbidity index; TA, traffic accident; SBP, systolic blood pressure; HR, heart rate; RR, respiratory rate; GCS, Glasgow Coma Scale; AIS, abbreviated injury scale; ISS, injury severity score; TXA, tranexamic acid
Table 2.
Interventions and outcomes between low and high ratio of platelet concentrate (PC) to red blood cell (RBC)
| Variables | Low PC N = 2,133 |
High PC N = 934 |
P-value |
|---|---|---|---|
| Transfusion within 24 h | |||
| RBC, units, mean (SD) | 23.6 (21.5) | 19.6 (10.7) | < 0.01 |
| FFP, units, mean (SD) | 25.7 (84.5) | 27.2 (48.7) | 0.60 |
| PC, units, mean (SD) | 12.6 (11.1) | 26.1 (13.7) | < 0.01 |
| Hemostatic procedure, n (%) | |||
| Overall | 1,816 (85.1) | 793 (84.9) | 0.87 |
| Thorax | 350 (16.4) | 149 (16.0) | 0.79 |
| Abdomen | 725 (34.0) | 323 (34.6) | 0.77 |
| Outcomes, n (%) | |||
| 24-hour mortality | 334 (15.7) | 83 (8.9) | < 0.01 |
| ED-mortality | 118 (5.5) | 14 (1.5) | < 0.01 |
| In-hospital mortality | 719 (33.7) | 272 (29.1) | 0.01 |
| Overall complication* | 744 (34.9) | 427 (45.7) | < 0.01 |
| PE* | 28 (1.3) | 14 (1.5) | 0.73 |
| ARDs* | 21 (1.0) | 16 (1.7) | 0.11 |
| AKI* | 54 (2.5) | 43 (4.6) | < 0.01 |
| Pulmonary edema* | 15 (0.7) | 10 (1.1) | 0.29 |
RBC, red blood cell; FFP, fresh frozen plasma; PC, platelet concentrate; SD, standard deviation; PE, pulmonary embolism; ARDS, acute respiratory distress syndrome; AKI, acute kidney injury
༊Exclusion for 24-hour mortality case
Table 3 presents the result of the IPTW analysis, demonstrating that the high PC group had a lower risk of 24-hour mortality (risk ratio [RR], 0.54; 95% confidence interval [CI], 0.43–0.69), ED-mortality (RR, 0.32; 95% CI, 0.18–0.57), and in-hospital mortality (RR, 0.78; 95% CI, 0.69–0.88) compared with the low PC group. However, the high PC group showed a higher risk of overall complications (RR, 1.15; 95% CI, 1.04–1.27). In contrast, the high PC group did not demonstrate a significantly increased risk of pulmonary embolism (RR, 1.38; 95% CI, 0.71–2.65), acute respiratory distress syndrome (ARDS) (RR, 1.43; 95% CI, 0.71–2.91), acute kidney injury (AKI) (RR, 1.43; 95% CI, 0.94–2.18), or pulmonary edema (RR, 1.28; 95% CI, 0.55–2.99).
Table 3.
Results of modified Poison regression analysis weighted by inverse probability of treatment weighting
| Outcomes | RR | 95% CI |
|---|---|---|
| Primary outcome | ||
| 24-hour mortality | 0.54 | 0.43–0.69 |
| Secondary outcomes | ||
| ED-mortality | 0.32 | 0.18–0.57 |
| In-hospital mortality | 0.78 | 0.69–0.88 |
| Overall complication* | 1.15 | 1.04–1.27 |
| PE* | 1.38 | 0.71–2.65 |
| ARDs* | 1.43 | 0.71–2.91 |
| AKI* | 1.43 | 0.94–2.18 |
| Pulmonary edema* | 1.28 | 0.55–2.99 |
RR, risk ratio; CI, confidence interval; ED, emergency department; PE, pulmonary embolism; ARDS, acute respiratory distress syndrome; AKI, acute kidney injury
Figure 2 and Supplementary Table 4 illustrate the effect of a high PC ratio on 24-hour mortality in the overall cohort and across subgroups after IPTW. All subgroups showed a potential survival benefit assoiated with a high PC ratio. Sensitivity analyses confirmed the robustness of these findings, demonstrating that a high PC ratio was consistently associated with a lower risk of 24-hours mortality (Supplementary Table 5). Additionally, a higher PC to plasma ratio was also linked to a lower risk of 24-hour mortality (RR, 0.74; 95% CI, 0.57–0.96).
Fig. 2.
Forest plot of risk ratios and 95% confidence interval of 24-hour mortality for main and subgroup analysis. Each row represents a different subgroup, with point estimates of the risk ratio (RR) and whiskers indicating the 95% confidence interval (CI). The outcome was 24-hour mortality, and RR closer to 0 indicate a lower risk of 24-hour mortality. Each subgroup is defined as follows: Subgroup 1: patients presented with shock (systolic blood pressure < 90mmHg at arrival), Subgroup 2: patients underwent hemorrhage control procedures, Subgroup 3: patients received more than 20 units of RBC within 24 h of arrival, Subgroup 4: patients had a thorax AIS > = 3, Subgroup 5: patients had an abdomen AIS > = 3, Subgroup 6: patients had a lower extremity AIS > = 3, Subgroup 7: patients with TXA, and Subgroup 8: patients without TXA. RR, risk ratio; CI, confidence interval; RBC, red blood cell; AIS, abbreviated injury scale; TXA, tranexamic acid
Result of investigation of the optimal ratio of the platelet to red blood cell
Table 4 presents the interventions and outcomes across different PC to RBC ratio groups. Adjusted outcomes for 24-hour and in-hospital mortality among these ratio groups are shown in Table 5. Compared with a reference ratio < 0.5, the 0.5-1.0, 1.0-1.5, and 1.5-2.0 ratio groups were associated with lower risk of both 24-hour and in-hospital mortality.
Table 4.
Interventions and outcomes between ratio of PC to RBC groups
| Variables | Ratio < 0.5 N = 942 |
0.5 < = Ratio < 1 N = 1,780 |
1 < = Ratio < 1.5 N = 863 |
1.5 < = Ratio N = 309 |
P-value |
|---|---|---|---|---|---|
| Transfusion within 24 h | |||||
| RBC, units, mean (SD) | 25.2 (33.3) | 22.6 (14.3) | 20.6 (11.4) | 16.7 (6.9) | < 0.01 |
| FFP, units, mean (SD) | 29.8 (166.1) | 24.3 (17.6) | 30.3 (123.4) | 28.7 (81.6) | 0.42 |
| PC, units, mean (SD) | 4.7 (7.4) | 16.5 (10.5) | 24.2 (13.2) | 31.7 (13.8) | < 0.01 |
| Hemostatic procedure, n (%) | |||||
| Overall | 792 (84.1) | 1,485 (83.4) | 734 (85.1) | 265 (85.8) | 0.61 |
| Thorax | 196 (20.8) | 291 (16.3) | 156 (18.1) | 46 (14.9) | 0.02 |
| Abdomen | 324 (34.4) | 591 (33.2) | 311 (36.0) | 95 (30.7) | 0.31 |
| Outcomes, n (%) | |||||
| 24-hour mortality | 273 (29.0) | 257 (14.4) | 84 (9.7) | 17 (5.5) | < 0.01 |
| ED-mortality | 158 (16.8) | 73 (4.1) | 14 (1.6) | 2 (0.6) | < 0.01 |
| In-hospital mortality | 410 (43.5) | 611 (34.3) | 252 (29.2) | 82 (26.5) | < 0.01 |
| Overall complication | 291 (30.9) | 616 (34.6) | 402 (46.6) | 142 (46.0) | < 0.01 |
| PE | 9 (1.0) | 20 (1.1) | 12 (1.4) | 5 (1.6) | 0.73 |
| ARDS | 6 (0.6) | 17 (1.0) | 12 (1.4) | 5 (1.6) | 0.30 |
| AKI | 22 (2.3) | 48 (2.7) | 36 (4.2) | 17 (5.5) | < 0.01 |
| Pulmonary edema | 8 (0.8) | 12 (0.7) | 7 (0.8) | 5 (1.6) | 0.41 |
FFP, fresh frozen plasma; SD, standard deviation; ED, emergency department; PE, pulmonary embolism; ARDS, acute respiratory distress syndrome; AKI, acute kidney injury
Table 5.
Adjusted outcomes of 24-hour mortality and in-hospital mortality between ratio of PC to RBC groups
| Variable | Ratio of PC to RBC group | |||
|---|---|---|---|---|
| n | Ratio 1 (< 0.5) n = 942 |
Ratio 2 (< 1) n = 1,780 |
Ratio 3 (< 1.5) n = 863 |
Ratio 4 n = 309 |
| 24-hour mortality | 273 (29.0) | 257 (14.4) | 84 (9.7) | 17 (5.5) |
| Risk ratio (95% CI) | ||||
| Crude | Reference | 0.41 (0.34–0.50) | 0.26 (0.20–0.35) | 0.14 (0.09–0.24) |
| Adjusted model | Reference | 0.48 (0.38–0.60) | 0.28 (0.21–0.37) | 0.19 (0.11–0.32) |
| In-hospital mortality | 410 (43.5) | 611 (34.3) | 252 (29.2) | 82 (26.5) |
| Risk ratio (95% CI) | ||||
| Crude | Reference | 0.68 (0.58–0.80) | 0.54 (0.44–0.65) | 0.47 (0.35–0.62) |
| Adjusted model | Reference | 0.72 (0.59–0.88) | 0.44 (0.35–0.56) | 0.50 (0.36–0.70) |
Modified Poisson regression models were adjusted for age, sex, trauma mechanism, sBP, HR, RR, GCS, lactate value, diagnostic procedure, AIS for head, chest, abdomen, and peripheral injuries, ISS and TXA
SBP, systolic blood pressure; HR, heart rate; RR, respiratory rate; GCS, Glasgow Coma Scale; AIS, abbreviated injury scale; ISS, injury severity score; TXA, tranexamic acid
Cubic spline for PC to RBC ratios
A cubic spline analysis demonstrated a monotonic. dose-dependent association between increasing PC to RBC ratios and lower 24-hour mortality in the overall cohort as well as in the subgroup of patients who underwent surgical procedures (Fig. 3).
Fig. 3.
Restricted cubic spline for platelet concentrate (PC) to red blood cell (RBC) ratio. Association between the PC to RBC ratio and 24-hour mortality in the overall cohort (Fig. 3A) and in the subgroup of the patients who received surgical procedure (Fig. 3B). The model was fitted with a cubic spline of the FFP to RBC ratio adjusted for age, sex, CCI, trauma mechanism, mechanism of injury, vital signs on hospital arrival, lactate level, use of whole body CT, angiography for the chest, abdomen or pelvis, AIS for the head/neck, chest, abdomen, and pelvis/lower extremity, ISS and administration of tranexamic acid, with a ratio of 1 as a reference. PC, platelet concentrate; RBC, red blood cell; FFP, fresh frozen plasma; CCI, charlson comorbidity index; CT, computed tomography; AIS, abbreviated injury scale; ISS, injury severity score
Discussion
In summary, this study demonstrated several important findings. First, A high ratio of PC to RBC/FFP was associated with lower mortality among massively transfused patients. Second, investigation of the optimal ratio of PC to RBC also suggested that a higher ratio of PC to RBC was associated with lower mortality. Third, a high ratio of PC to RBC was associated with increased overall frequency of complication.
Two systematic reviews of investigating the effect of a high ratio of PC to RBC reported that a high ratio of PC to RBC was associated with 24-hour mortality [24, 25]. However, most of the included studies defined a high ratio of PC to RBC as below 1.0, and currently there is a lack of investigation of the effect of higher ratios of PC to RBC, particularly those above 1.0. The Implementing Treatment Algorithms for the Correction of Trauma-Induced Coagulopathy (ITACTIC) trial, which evaluated the efficacy of a viscoelastic hemostatic assay augmented protocol, reported that the viscoelastic assay group had a PC to RBC ratio of 1.27 compared with 0.62 in conventional coagulation test group, and concluded there was no statistically significant difference in 24-hour mortality [26]. In the aforementioned PROPPR trial, the high ratio of PC to RBC group had a ratio of 1.26 and showed lower mortality due to exsanguination within 24 h [10].
Our study added the evidence of the effect of a much higher ratio of PC to RBC on decreased mortality among massively transfused patients. Furthermore, additional analysis investigating the optimal ratio of PC to RBC suggested that the higher the ratio of PC to RBC, the better the prognosis. The pivotal function of platelet is to achieve successful hemostasis and contribute to reducing early mortality due to hemorrhage. Our cohort included massively transfused patients, and approximately 85% patients required a hemostatic procedure; therefore, the effect of high PC strategy may be particularly pronounced in reduced mortality.
Within the concept of TIC, platelet dysfunction after trauma has been studied in previous literatures. A recent sub-analysis of a randomized controlled trial from Australia reported that approximately 70% of severely injured patients had platelet hypofunction as measured by whole blood aggregometry [27]. While it remains unclear whether this impaired platelet function can be addressed with PC transfusion, balanced PC administration has been reported to have the potential to mitigate coagulopathy. Previous literature reported that patients who received balanced FFP and PC transfusions had a reduced risk of mortality compared with patients who received balanced FFP but unbalanced PC transfusions. This was explained by the fact that PC acts as a facilitator of coagulation cascade propagation and may help potentiate the hemostatic benefits of coagulation proteins provided with FFP [9]. Our study aligns with this previous study in showing that a high ratio of PC to FFP was also associated with reduced 24-hour mortality.
However, a high PC-to-RBC ratio may also have potential drawbacks in massively transfused patients. In this study, a higher PC to RBC ratio was associated with an increased risk of overall complications. The apparent paradox of lower mortality but higher complication rates warrants careful interpretation. One possible explanation is survivor bias, as patients who survived the initial resuscitation phase were more likely to live long enough to develop and be diagnosed with complications. To mitigate survivor bias, we performed a sensitivity analysis excluding patients who died in the emergency department. In addition, the findings may reflect a temporal trade-off between early survival benefit and later morbidity. Early and aggressive platelet administration may enhance hemostasis and reduce hemorrhage-related mortality during the acute phase of trauma. However, patients who survive this critical period may remain at risk for subsequent complications. Platelet transfusion has been associated with pro-inflammatory and pro-thrombotic effects, which could contribute to organ dysfunction, including AKI and ARDS.
Consistent with this hypothesis, previous literatures reported the increased risk of complications after PC administration. A secondary analysis of the prospective observational Massive Transfusion in Children (MATIC) study reported that every 10% increase in the PC to RBC was associated with an increased risk of AKI [28]. Similarly, the Glue Grant database study found that platelet transfusion was associated with an increased risk of ARDS, and transfusion-related acute lung injury (TRALI) has also recognized as a potential pulmonary complication of platelet administration [29]. Taken together, this study suggests that a higher ratio of PC to RBC could be beneficial for massively transfused patients requiring hemostatic procedure. However, clinicians should refrain from administering additional PC transfusion after hemostasis, given the potential risk of complications.
Limitations
This study has several limitations. First, we retrospectively defined massive transfusion within 24 h of injury; therefore, the study population may differ from other definitions used to evaluate high PC strategies.
Second, selection bias between the high- and low-PC groups is possible, as patients who survived initial resuscitation may have received higher PC-to-RBC ratios. Although time-dependent analysis would better address this issue, data on transfusion timing were unavailable. We performed sensitivity analyses excluding emergency department deaths and patients without PC transfusion to assess robustness; however, residual bias cannot be excluded.
Third, as a retrospective observational study, unmeasured confounding is inevitable. In particular, TXA administration was not fully protocolized during the study period and depended on physician discretion, which may have influenced outcomes.
Fourth, external validity may be limited. Platelet units in Japan generally contain fewer platelets than standard apheresis units used in the United States and Europe (approximately 3–4 × 10¹¹ platelets per dose) [30], and thus equivalent PC-to-RBC ratios may not represent comparable absolute platelet doses.
Fifth, cause-specific mortality and detailed neurological data were unavailable. Mortality in massively transfused patients may also be influenced by intracranial hemorrhage and neurological deterioration. Although improved in-hospital survival was also observed in the high PC group, the impact of transfusion strategy on neurological outcomes remains uncertain.
Despite these limitations, our findings suggest a potential strategy to reduce mortality in severely injured patients and may inform future randomized controlled trials.
Conclusions
This nationwide observational study suggested that a PC to RBC greater than 1 was associated with lower mortality. In addition, an increasing PC to RBC ratio was associated with improved survival outcomes. Further RCTs focusing on the optimal transfusion strategy for PC to RBC are warranted.
Supplementary Information
Below is the link to the electronic supplementary material.
Acknowledgements
We thank all hospitals which participated JTDB and registered the patient’s data in JTDB.
Author contributions
Study concept, design: Aoki, Okada, Fujiwara, Katsura, Matsushima. Data collection and analysis: Aoki, Okada, Fujiwara, Katsura, Matsumoto, Matsushima Writing: Aoki, Okada, Fujiwara, Katsura, Matsumoto, Kiyozumi, Tomura, Matsushima. Critical revision: Aoki, Okada, Fujiwara, Katsura, Matsumoto, Kiyozumi, Tomura, Matsushima.
Funding
Open Access funding provided by National Defense Medical College
Data availability
The data that support the findings of this study are available from the corresponding author upon reasonable request.
Declarations
Competing interests
The authors declare no competing interests.
Footnotes
Publisher’s note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
References
- 1.Cole E, Weaver A, Gall L, West A, Nevin D, Tallach R, et al. A Decade of Damage Control Resuscitation: New Transfusion Practice, New Survivors, New Directions. Ann Surg. 2021;273(6):1215–20. [DOI] [PubMed] [Google Scholar]
- 2.Davis JS, Satahoo SS, Butler FK, Dermer H, Naranjo D, Julien K, et al. An analysis of prehospital deaths: Who can we save? J Trauma Acute Care Surg. 2014;77(2):213–8. [DOI] [PubMed] [Google Scholar]
- 3.Eastridge BJ, Holcomb JB, Shackelford S. Outcomes of traumatic hemorrhagic shock and the epidemiology of preventable death from injury. Transfusion. 2019;59(S2):1423–8. [DOI] [PubMed] [Google Scholar]
- 4.Tisherman SA, Schmicker RH, Brasel KJ, Bulger EM, Kerby JD, Minei JP, et al. Detailed description of all deaths in both the shock and traumatic brain injury hypertonic saline trials of the Resuscitation Outcomes Consortium. Ann Surg. 2015;261(3):586–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Moore EE, Moore HB, Kornblith LZ, Neal MD, Hoffman M, Mutch NJ, et al. Trauma-induced coagulopathy. Nat Rev Dis Primers. 2021;7(1):30. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Buzzard L, Schreiber M. Trauma-induced coagulopathy: What you need to know. J Trauma Acute Care Surg. 2024;96(2):179–85. [DOI] [PubMed] [Google Scholar]
- 7.Lammers DT, Holcomb JB. Damage control resuscitation in adult trauma patients: What you need to know. J Trauma Acute Care Surg. 2023;95(4):464–71. [DOI] [PubMed] [Google Scholar]
- 8.Douglas M, Obaid O, Castanon L, Reina R, Ditillo M, Nelson A, et al. After 9,000 laparotomies for blunt trauma, resuscitation is becoming more balanced and time to intervention shorter: Evidence in action. J Trauma Acute Care Surg. 2022;93(3):307–15. [DOI] [PubMed] [Google Scholar]
- 9.Dorken Gallastegi A, Naar L, Gaitanidis A, Gebran A, Nederpelt CJ, Parks JJ, et al. Do not forget the platelets: The independent impact of red blood cell to platelet ratio on mortality in massively transfused trauma patients. J Trauma Acute Care Surg. 2022;93(1):21–9. [DOI] [PubMed] [Google Scholar]
- 10.Holcomb JB, Tilley BC, Baraniuk S, Fox EE, Wade CE, Podbielski JM, 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(5):471–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Cannon JW, Khan MA, Raja AS, Cohen MJ, Como JJ, Cotton BA, et al. Damage control resuscitation in patients with severe traumatic hemorrhage: A practice management guideline from the Eastern Association for the Surgery of Trauma. J Trauma Acute Care Surg. 2017;82(3):605–17. [DOI] [PubMed] [Google Scholar]
- 12.Rossaint R, Afshari A, Bouillon B, Cerny V, Cimpoesu D, Curry N, et al. The European guideline on management of major bleeding and coagulopathy following trauma: sixth edition. Crit Care. 2023;27(1):80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Committee TJAftSoTTR. Japan Trauma Data Bank Report 2023. 2024. Available from: https://www.jtcr-jatec.org/traumabank/dataroom/data/JTDB2023.pdf.
- 14.Shibahashi K, Aoki M, Hikone M, Sugiyama K. Association between transfusion volume and survival outcome following trauma: Insight into the limit of transfusion from an analysis of nationwide trauma registry in Japan. J Trauma Acute Care Surg. 2024;96(5):742–8. [DOI] [PubMed] [Google Scholar]
- 15.Charlson ME, Pompei P, Ales KL, MacKenzie CR. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40(5):373–83. [DOI] [PubMed] [Google Scholar]
- 16.Stekhoven DJ, Bühlmann P. MissForest–non-parametric missing value imputation for mixed-type data. Bioinformatics. 2012;28(1):112–8. [DOI] [PubMed] [Google Scholar]
- 17.Austin PC. The performance of different propensity-score methods for estimating differences in proportions (risk differences or absolute risk reductions) in observational studies. Stat Med. 2010;29(20):2137–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Fujiwara G, Okada Y, Ishii W, Echigo T, Shiomi N, Ohtsuru S. High Fresh Frozen Plasma to Red Blood Cell Ratio and Survival Outcomes in Blunt Trauma. JAMA Surg. 2024;159(11):1272–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Aoki M, Abe T, Komori A, Katsura M, Matsushima K. Association between whole blood ratio and risk of mortality in massively transfused trauma patients: retrospective cohort study. Crit Care. 2024;28(1):253. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Zou G. A modified poisson regression approach to prospective studies with binary data. Am J Epidemiol. 2004;159(7):702–6. [DOI] [PubMed] [Google Scholar]
- 21.Aoki M, Fukushima K, Hayakawa M, Kudo D, Tagami T, Oshima K, et al. Adherence to balanced transfusion among severely injured patients: A post hoc analysis of the RESTRIC trial. Acute Med Surg. 2024;11(1):e70016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Yadav K, Lewis RJ. Immortal Time Bias in Observational Studies. JAMA. 2021;325(7):686–7. [DOI] [PubMed] [Google Scholar]
- 23.Tao Y, Fu H. Doubly robust estimation of the weighted average treatment effect for a target population. Stat Med. 2019;38(3):315–25. [DOI] [PubMed] [Google Scholar]
- 24.Kleinveld DJB, van Amstel RBE, Wirtz MR, Geeraedts LMG, Goslings JC, Hollmann MW, et al. Platelet-to-red blood cell ratio and mortality in bleeding trauma patients: A systematic review and meta-analysis. Transfusion. 2021;61(Suppl 1Suppl 1):S243–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Rijnhout TWH, Duijst J, Noorman F, Zoodsma M, van Waes OJF, Verhofstad MHJ, et al. Platelet to erythrocyte transfusion ratio and mortality in massively transfused trauma patients. A systematic review and meta-analysis. J Trauma Acute Care Surg. 2021;91(4):759–71. [DOI] [PubMed] [Google Scholar]
- 26.Baksaas-Aasen K, Gall LS, Stensballe J, Juffermans NP, Curry N, Maegele M, et al. Viscoelastic haemostatic assay augmented protocols for major trauma haemorrhage (ITACTIC): a randomized, controlled trial. Intensive Care Med. 2021;47(1):49–59. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Durkin ZJ, Whebell SF, Wullschleger M, McCullough J, Czuchwicki S, Wake E, et al. Platelet dysfunction in trauma: a sub study of the FEISTY pilot trial analysing whole blood aggregometry data. Injury. 2025;56(8):112468. [DOI] [PubMed] [Google Scholar]
- 28.Vaizer R, Leeper CM, Lu L, Josephson CD, Leonard JC, Brown JB, et al. Increased platelet to red blood cell transfusion ratio associated with acute kidney injury in children with life-threatening bleeding. Transfusion. 2024;64(Suppl 2):S62–71. [DOI] [PubMed] [Google Scholar]
- 29.Kasotakis G, Starr N, Nelson E, Sarkar B, Burke PA, Remick DG, et al. Platelet transfusion increases risk for acute respiratory distress syndrome in non-massively transfused blunt trauma patients. Eur J Trauma Emerg Surg. 2019;45(4):671–9. [DOI] [PubMed] [Google Scholar]
- 30.Friedman M, Costa V, Rafiee B, Hilbert T, Jafri M, Wu DW. Platelet transfusions: current practices and emerging alternatives in the United States. Life (Basel). 2025;15(6). [DOI] [PMC free article] [PubMed]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.