Abstract
Background:
Massive transfusion protocols (MTPs) to address hemorrhage are under-studied in children. The objective was to determine the effect of MTP implementation on outcomes of injured children.
Study Design and Methods:
This retrospective comparison of injured children before and after MTP implementation for children <18 yo who presented in 2005–2014 and received red blood cells (RBC) within 24 hours of arrival. Children were divided into groups based on pre-/post-MTP implementation and sub-grouped based on receipt of massive transfusion (≥40 ml/kg PRBC or ≥80 ml/kg total blood products at 24 hours from arrival). The primary outcome was in-hospital mortality and secondary outcomes were total blood product use, intensive care unit (ICU)/ventilator/pressor-free days, composite morbidity and Glasgow Outcome Score (GOS).
Results:
11,995 children presented for trauma care over 9 years; 235 received RBCs. 120 were in the pre-MTP group and 115 in the post-MTP, of which 26 and 17 received massive transfusion in the pre and post-MTP groups, respectively; 11 had MTP activations. Children massively transfused post-MTP received mean plasma:RBC and platelet:RBC ratios greater than 1:1 at both 6 and 24 hours with no significant difference in total admission blood product use. There was no difference in in-hospital mortality between pre- and post-MTP groups (24% vs. 19%) or massive transfusion subgroups (54% vs. 47%). There were no differences in secondary outcomes.
Conclusions:
While we were not able to show improvements in outcome, MTP implementation led to higher plasma and platelet:RBC ratios without an associated change in blood product use or composite morbidity.
Keywords: pediatric trauma, massive transfusion protocol, hemorrhage
Background
Acute hemorrhage is the most common preventable cause of death in the first few hours following injury.1–7 Up to 40% of adults and children arrive at trauma centers coagulopathic, and the coagulopathy is associated with increased mortality.8–11 The recent Pragmatic, Randomized Optimal Platelet and Plasma Ratios (PROPPR) trial showed decreased death from hemorrhage in adults who received transfusion with high plasma (FFP) and platelet to red blood cell (RBC) ratios (balanced transfusion of 1:1:1 plasma:platelets:red blood cells).1 This study emphasized the importance of giving plasma, platelets, and red blood cells in a 1:1:1 ratio early in the management of hemorrhagic shock and coagulopathy in trauma.
To expedite transfusion with plasma, platelets, and RBCs close to a 1:1:1 ratio, trauma centers have adopted massive transfusion protocols (MTPs), which specify early preparation and administration of platelets and plasma and delivery of blood products in predefined ratios until the patient is hemodynamically stable. Implementation of MTPs in non-pediatric trauma centers has been associated with improved mortality, decreased risk of multi-organ failure and sepsis, and decreased total blood use.12–15 However, there is evidence that balanced massive transfusion in adults who are at low risk of life-threatening hemorrhage is associated with increased organ failure and ventilator days.16
The utility of MTPs in the management of injured children is understudied. One prior study of MTP implementation in a pediatric setting showed no difference in mortality after instituting a MTP.17 Given the paucity of data evaluating MTPs in children, the purpose of our study was to determine the effect of MTP implementation on resuscitation and outcomes for injured children. The primary outcome examined was in-hospital mortality and the secondary outcomes were intensive care unit (ICU)/ventilator/pressor-free days and Glasgow Outcome Score (GOS). We also investigated adherence to the MTP and overall blood product utilization by evaluating administered blood product volumes and ratios.
Materials and Methods
This was a retrospective study of injured children < 18 years old who presented to an American College of Surgeons verified level-1 pediatric trauma center, from May 1, 2005 to February 28, 2014. We conducted a comparison of outcomes in transfused children before and after the MTP implementation on October 14, 2009. We included children who presented to the emergency department (ED) for trauma care and received red blood cells (RBC) within 24 hours of arrival. Eligible children were identified through the trauma registry using ICD-9 trauma E-codes and cross-referenced against the blood bank database. Exclusion criteria included: death within 1 hour of arrival, presentation 24 hours after injury, or those with injuries unrelated to blunt or penetrating trauma such as drownings, hangings, or isolated burns. The study population was divided into six groups based on admission before or after MTP implementation and whether they received massive transfusion (≥40 ml/kg RBC or ≥80 ml/kg total blood products at 24 hours from arrival): 1) massively transfused prior to MTP implementation (“pre-MTP massive”); 2) non-massively transfused prior to MTP implementation (“pre-MTP non-massive”); 3) massively transfused with MTP activation (“MTP activated massive”); 4) non-massively transfused with MTP activation (“MTP activated non-massive”); 5) massively transfused without MTP activation after MTP implementation (“MTP not activated massive”) and 6) non-massively transfused without MTP activation after MTP implementation (“MTP not activated non-massive).
The decision to activate the MTP was made by an attending level surgeon, emergency medicine physician, intensivist, or anesthesiologist. Per the protocol, indicators of massive hemorrhage included: 1) acute transfusion of 40ml/kg of blood in a child or 2) clinical concern for acute imminent blood loss. Once activated, blood was prepared in packs based on the child’s weight until the attending physician gave the order to stop the MTP (Table 1). The goal was to give all products in one pack prior to moving on to the next, but the attending could alter the amount of products given.
Table 1.
Definition for the massive transfusion protocol (MTP) defined packs delivered based on weight.
| Children < 30 kg | |||
| MTP Pack 1 | MTP Pack 2 | MTP Pack 3 | MTP Pack 4 |
| 2 units PRBCs | 2 units PRBCs | 2 units PRBCs | 2 units PRBCs |
| 2 unit FFP | 2 unit FFP | 2 unit FFP | 2 unit FFP |
| ½ SDP | ½ SDP | ½ SDP | ½ SDP |
| 2 units cryoprecipitate | 2 units cryoprecipitate | 2 units cryoprecipitate | |
| Children > 30 kg | |||
| MTP Pack 1 | MTP Pack 2 | MTP Pack 3 | MTP Pack 4 |
| 4 units PRBCs | 4 units PRBCs | 4 units PRBCs | 4 units PRBCs |
| 4 unit FFP | 4 unit FFP | 4 unit FFP | 4 unit FFP |
| 1 SDP | 1 SDP | 1 SDP | 1 SDP |
| 5 pack pre-pooled cryoprecipitate | 5 pack pre-pooled cryoprecipitate | 5 pack pre-pooled cryoprecipitate | |
PRBC = packed red blood cells
FFP = fresh frozen plasma
SDP = single donor platelets
Using a strict protocol, data was abstracted from direct review of the emergency medical services records, hospital records, trauma registry, and blood bank database. We collected data for age, gender, mechanism of injury, injury severity score (ISS), abbreviated injury scale (AIS), time to arrival in ED, vital signs on arrival (heart rate, blood pressure (BP), temperature), Glasgow Coma Scale (GCS), initial laboratory values (hemoglobin (hgb), hematocrit (hct), platelet count, prothrombin time (PT), activated partial thromboplastin time (aPTT), international normalized ratio (INR), base deficit (BD), pH, BIG score (BD+[2.5×INR]+[15-GCS]), length of hospital stay, length of pediatric intensive care unit (PICU) stay, number of days on vasopressors, number of ventilator days, need for damage control surgery, patient death, cause of death, composite morbidity (included multi-organ failure, sepsis, acute respiratory distress syndrome (ARDS), transfusion related acute lung injury (TRALI), abdominal compartment syndrome, and thrombotic complications), Glasgow Outcome Scale (GOS), intravenous fluid (IVF) use, and timing and amount of blood products given. Blood product volumes were recorded in milliliters (mL) when possible and converted to units using predefined definitions: 1 unit plasma=200mL, 1 unit RBC=300mL, and 1 unit platelets=50mL apheresis platelets.18,19 Plasma and platelet deficits were calculated assuming a balanced transfusion ratio of 1:1:1 units of plasma:platelets:RBCs. The abstractor was blinded from transfusion group assignment by first abstracting from the charts of all children who received blood products and then abstracting blood product data to identify the massively transfused and MTP activation groups.
Data was managed using REDCap (Research Electronic Data Capture), a secure, web-based application.20 Statistical analyses were completed using SAS 9.3. Continuous variables were reported as means ± standard deviation (SD) and analyzed using ANOVA, logistic regression, or the t-test. Categorical variables were reported as frequencies and percentages and analyzed using Chi-square or Fisher’s exact test. If a patient was missing a variable, they were excluded from the analysis of that specific variable. We performed a comparison of patients who died versus survived to see which variables were associated with mortality and adjusted the outcome analysis using those variables. The Washington University in St. Louis Institutional Review Board approved this study with waiver of consent.
Results
Over nine years, 11,995 children presented for trauma management, with 235 (2.0%) meeting study criteria. 120 children were in the pre-MTP group and 115 in the post-MTP group of which 26 and 17 children received massive transfusion, respectively. Of the 17 children who received massive transfusion in the post-MTP group, 11 had MTP activations (group 3). 12 children had MTP activated but never received massive transfusion per our definition (group 4).
All study groups were comparable in age, gender, time from injury to hospital arrival and whether they were transferred (Table 2). The massive transfusion groups (groups 1, 3, and 5) had significantly higher ISS, greater proportion with AIS head ≥ 3 and GCS ≤ 8, lower hemoglobin, lower platelets, higher INR, and higher base deficit (Table 2). Groups 1, 3, and 5 received more pre-arrival IVF; there were no significant differences in the amount of pre-arrival RBCs the groups received, and none of the groups received pre-arrival platelets. Only patients in groups 5 and 6 received plasma prior to arrival. Group 4, patients who had MTP activated but did not receive massive volumes of blood, had higher ISS than the groups that did not receive massive transfusion but not as high as the groups that did (Table 2).
Table 2.
Comparison of demographic and initial clinical characteristics of all cohorts (if a variable is missing, reported the n in each group for that variable)
| Pre-MTP | Post-MTP | ||||||
|---|---|---|---|---|---|---|---|
| Variable | Group 1 Massive N=26 |
Group2 Non-massive N=94 |
Group3 + MTP Massive N=11 |
Group4 + MTP Non-massive N=12 |
Group5 − MTP Massive N=6 |
Group6 − MTP Non-massive N=86 |
p value* |
| Age (years) | 9.1 ± 5.3 | 10.2 ± 6.0 | 8.5 ± 6.5 | 11.0 ± 6.4 | 4.7 ± 6.3 | 8.6 ± 6.1 | 0.164 |
| Gender (male) | 20 (76.9%) | 58 (61.7%) | 5 (45.5%) | 10 (83.3%) | 5 (83.3%) | 51 (59.3%) | 0.209 |
| Race | <0.001 | ||||||
| Caucasian | 10 (38.5%) | 53 (56.4%) | 9 (63.6%) | 4 (36.3%) | 3 (50.0%) | 63 (73.3%) | |
| African American | 11 (42.3%) | 39 (41.5%) | 4 (36.4%) | 8 (66.7%) | 2 (33.3%) | 20 (23.3%) | |
| Hispanic/Latino | 1 (3.8%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | 0 (0%) | |
| Other/Unknown | 4 (15.4%) | 2 (2.1%) | 0 (0%) | 0 (0%) | 1 (16.7%) | 3 (3.5%) | |
| Hours from injury until arrival to ED | 1.29 ± 1.10 | 2.78 ± 3.81 | 1.52 ± 0.83 | 1.20 ± 1.22 | 4.18 ± 3.25 | 3.14 ± 3.83 | 0.067 |
| Transfer (yes) † | 9 (34.6%) | 34 (36.2%) | 3 (27.3%) | 4 (33.3%) | 4 (66.7%) | 46 (53.5%) | 0.098 |
| Mechanism of injury | 0.042 | ||||||
| Fall | 1 (3.8%) | 11 (11.7%) | 0 (0%) | 2 (16.7%) | 0 (0%) | 11 (12.8%) | |
| MVC‡ | 5 (19.2%) | 28 (29.8%) | 5 (45.5%) | 2 (16.7%) | 1 (16.7%) | 34 (39.5%) | |
| MVC non-vehicle‡ | 8 (30.8%) | 24 (25.5%) | 2 (18.2%) | 2 (16.7%) | 1 (16.7%) | 9 (10.5%) | |
| Other | 3 (11.5%) | 10 (10.6%) | 2 (18.2%) | 0 (0%) | 3 (50.0%) | 17 (19.8%) | |
| Penetrating | 9 (34.6%) | 21 (22.3%) | 2 (18.2%) | 6 (50.0%) | 1 (16.7%) | 15 (17.4%) | |
| Suspected non-accidental trauma | 3 (11.5%) | 11 (11.7%) | 2 (18.2%) | 1 (8.3%) | 3 (50.0%) | 14 (16.3%) | 0.225 |
| ISS | 32.8 ± 16.9 (n=26) |
22.7 ± 13.7 (n=93) |
35.8 ± 16.4 (n=11) |
29.0 ± 20.5 (n=12) |
32.3 ± 10.0 (n=6) |
22.1 ± 13.5 (n=86) |
<0.001 |
| AIS head (≥3) | 16 (61.5%) (n=26) | 40 (43.0%) (n=93) | 9 (81.8%) (n=11) | 4 (33.3%) (n=12) | 6 (100%) (n=6) | 46 (53.5%) (n=86) | 0.008 |
| Pre-arrival IVF (ml/kg) | 26.0 ± 27.8 | 15.7 ± 18.0 | 30.8 ± 37.5 | 8.0 ± 7.3 | 32.0 ± 33.0 | 19.5 ± 22.0 | 0.025 |
| Pre-arrival RBC (ml/kg) | 1.8 ± 7.5 | 0.6 ± 3.0 | 1.4 ± 4.6 | 0.5 ± 1.6 | 5.2 ± 9.2 | 0.6 ± 2.1 | 0.059 |
| Pre-arrival FFP (ml/kg) | 0 | 0 | 0 | 0 | 3.0 ± 7.4 | 0.1 ± 0.5 | <0.001 |
| Pre-arrival platelets (ml/kg) | 0 | 0 | 0 | 0 | 0 | 0 | |
| Arrival heart rate | 112 ± 45 (n=26) |
119 ± 39 (n=93) |
91 ± 47 (n=11) |
111 ± 50 (n=12) |
115 ± 30 (n=6) |
121 ± 36 (n=86) |
0.258 |
| Arrival systolic BP | 91 ± 44 (n=26) |
114 ± 24 (n=93) |
112 ± 33 (n=11) |
92 ± 33 (n=12) |
111 ±23 (n=6) |
112 ± 23 (n=86) |
0.001 |
| Arrival diastolic BP | 60 ± 33 (n=26) |
67 ± 20 (n=93) |
75 ± 28 (n=11) |
58 ± 28 (n=12) |
73 ± 28 (n=6) |
65 ± 20 (n=86) |
0.330 |
| Arrival temperature | 35.5 ± 0.8 (n=18) |
36.1 ± 1.2 (n=81) |
35.8 ± 1.8 (n=9) |
35.8 ± 1.6 (n=11) |
34.7 ± 2.4 (n=6) |
36.2 ± 1.2 (n=73) |
0.023 |
| Arrival GCS (≤8) | 19 (73.1%) (n=26) | 33 (37.5%) (n=88) | 9 (81.8%) (n=11) | 6 (54.6%) (n=11) | 5 (83.3%) (n=6) | 39 (48.2%) (n=81) | 0.002 |
| Admit hemoglobin | 9.5 ± 2.1 (n=26) |
10.5 ± 2.0 (n=93) |
7.0 ± 2.8 (n=11) |
10.2 ± 3.2 (n=11) |
8.8 ± 2.5 (n=6) |
10.3 ± 2.1 (n=86) |
<0.001 |
| Admit hematocrit | 28.5 ± 6.5 (n=26) |
31.0 ± 5.8 (n=93) |
21.5 ± 8.7 (n=11) |
30.4 ± 9.5 (n=11) |
26.4 ± 6.4 (n=6) |
30.6 ± 5.9 (n=86) |
<0.001 |
| Admit platelets | 245 ± 140 (n=26) |
272 ± 96 (n=93) |
186 ± 97 (n=11) |
260 ± 146 (n=11) |
167 ± 81 (n=6) |
280 ± 106 (n=86) |
0.020 |
| Admit INR | 2.64 ± 2.28 (n=25) |
1.32 ± 0.29 (n=83) |
2.79 ± 2.32 (n=11) |
2.06 ± 1.98 (n=11) |
2.01 ± 0.87 (n=6) |
1.26 ± 0.24 (n=64) |
<0.001 |
| Admit base deficit | 10.0 ± 8.4 (n=26) |
5.8 ± 4.1 (n=81) |
10.6 ± 7.2 (n=11) |
6.9 ± 5.2 (n=10) |
11.6 ± 10.7 (n=6) |
5.3 ± 3.9 (n=64) |
<0.001 |
| Admit BIG score | 26.9 ± 12.4 (n=25) |
14.0 ± 7.8 (n=72) |
27.0 ± 13.4 (n=11) |
19.4 ± 11.2 (n=10) |
26.5 ± 13.3 (n=6) |
16.5 7.3 (n=54) |
<0.001 |
| OR by 6 hours from arrival (yes) | 19 (95.0%) (n=20) |
42 (59.2%) (n=71) |
7 (87.5%) (n=8) |
7 (77.8%) (n=9) |
1 (50.0%) (n=2) |
47 (71.2%) (n=66) |
0.018 |
| Damage control surgery (yes) | 15 (75.0%) (n=20) |
8 (11.3%) (n=71) |
5 (62.5%) (n=8) |
1 (11.1%) (n=9) |
1 (50.0%) (n=2) | 9 (13.6%) (n=66) |
<0.001 |
p-values refer to comparison of demographic and initial characteristics of all groups
Transferred from another hospital to the trauma center
MVC = motor vehicle collision; MVC non-vehicle refers to mechanisms where the victim is not enclosed in a vehicle such as motorcyles, all-terrain vehicles, bicycles, and pedestrians
We examined blood product use in both the children who received massive transfusion and all children who received transfusion without achieving massive transfusion volumes to see if MTP implementation generally changed transfusion practices. Patients received most of the plasma, platelets, and cryoprecipitate within 24 hours. Very few patients (25 of the 235) in our study population received cryoprecipitate. There were significant differences in the transfusion parameters for those children who received massive transfusion post-MTP compared to those pre-MTP implementation (Table 3): time to first plasma was shorter, the mean plasma and platelet to RBC ratios were greater than 1:1 at both 6 and 24 hours, and the plasma and platelet deficits were decreased. When comparing all transfused patients pre/post-MTP implementation, the plasma:PRBC ratios were also higher and platelet and plasma deficits were significantly lower, but the platelet:PRBC ratios at 6 and 24 hours were not significantly different (Table 3). Of note, while children who received massive transfusion post-MTP implementation received more platelets and plasma earlier in their management, they received less RBCs, so the mean amount of blood product used was not increased during admission (158.2 ± 112.7 ml/kg pre-MTP implementation and 158.6 ± 73.7 ml/kg post-MTP implementation).
Table 3.
Description of number of patients (N) who received platelets, plasma, and cryoprecipitate in each comparison group and also unadjusted and adjusted logistic regression (for BIG and ISS) comparing blood product usage pre and post-MTP implementation with all transfused children and with only the subgroups who received massive transfusion (if a variable is missing, reported the n in each group for that variable)
| Variables | All transfused | Massive transfusion subgroup | ||||||
|---|---|---|---|---|---|---|---|---|
| Pre-MTP Group 1,2 N=120 |
Post-MTP Group 3 – 6 N=115 |
Unadjusted p-value |
Adjusted p-value |
Pre-MTP Group 1 N=26 |
Post-MTP Group 3,5 N=17 |
Unadjusted p-value |
Adjusted p-value |
|
| N received plasma by 24 hr | 45 of 120 (37.5%) |
54 of 115 (47.0%) |
NA | NA | 23 of 26 (88.5%) |
16 of 17 (94.1%) |
NA | NA |
| N received plasma during admission | 49 of 120 (40.8%) |
57 of 115 (49.6%) |
NA | NA | 23 of 26 (88.5%) |
16 of 17 (94.1%) |
NA | NA |
| N received platelets by 24 hr | 22 of 120 (18.3%) |
23 of 115 (20.0%) |
NA | NA | 17 of 26 (65.4%) |
15 of 17 (88.2%) |
NA | NA |
| N received platelets during admission | 31 of 120 (25.8%) |
27 of 115 (23.5%) |
NA | NA | 18 of 26 (69/2%) |
15 of 17 (88.2%) |
NA | NA |
| N received cryoprecipitate by 24 hr | 14 of 120 (11.7%) |
10 of 115 (8.7%) |
NA | NA | 10 of 26 (38.5%) |
6 of 17 (35.3%) |
NA | NA |
| N received cryoprecipitate during admission | 14 of 120 (11.7%) |
11 of 115 (9.6%) |
NA | NA | 10 of 26 (38.5%) |
7 of 17 (41.2%) |
NA | NA |
| Total 24 hr IVF (ml/kg) | 132.3 ± 68.3 | 113.6 ± 60.3 | 0.027 | 0.024 | 192.5 ± 86.2 | 161.3 ± 102.2 | 0.288 | 0.139 |
| Plasma:PRBC 6 hr | 0.4 ± 0.9 (n=85) |
0.7 ± 0.8 (n=86) |
0.027 | 0.006 | 0.5 ± 0.4 (n=26) | 1.2 ± 0.7 (n=16) | 0.004 | 0.001 |
| Platelet:PRBC 6 hr | 0.2 ± 0.4 (n=85) |
0.3 ± 0.7 (n=86) |
0.136 | 0.554 | 0.4 ± 0.4 (n=26) | 1.2 ± 1.2 (n=16) | 0.019 | 0.006 |
| Plasma deficit 6 hr (units) | 1.5 ± 2.7 | 0.2 ± 1.7 | <0.001 | <0.001 | 3.9 ± 3.8 | −0.2 ± 3.4 | 0.001 | 0.800 |
| Platelet deficit 6 hr (units) | 1.8 ± 2.6 | 0.7 ± 2.6 | 0.001 | 0.009 | 4.1 ± 3.7 | −1.4 ± 5.8 | 0.002 | 0.001 |
| Plasma:PRBC 24 hr | 0.4 ± 0.7 | 0.6 ± 0.8 | 0.036 | 0.004 | 0.5 ± 0.3 | 1.2 ± 1.0 | 0.010 | 0.002 |
| Platelet:PRBC 24 hr from arrival | 0.2 ± 0.5 | 0.5 ± 1.6 | 0.077 | 0.065 | 0.6 ± 0.4 | 1.6 ± 1.9 | 0.037 | 0.014 |
| Plasma deficit 24 hr (units) | 1.9 ± 2.8 | 0.5 ± 1.8 | < 0.001 | <0.001 | 4.1 ± 3.8 | −0.4 ± 3.2 | <0.001 | <0.001 |
| Platelet deficit 24 hr (units) | 2.0 ± 2.6 | 0.6 ± 3.5 | <0.001 | 0.004 | 3.1 ± 4.4 | −3.3 ± 6.1 | <0.001 | 0.001 |
| Total cryoprecipitate 24 hr from arrival (ml/kg) | 0.2 ± 0.6 | 0.4 ± 1.6 | 0.137 | * | 0.6 ± 0.8 | 2.0 ± 3.4 | <0.001 | * |
| Total admission blood product used (ml/kg) | 58.6 ± 78.0 | 50.2 ± 58.6 | 0.350 | 0.622 | 158.2 ± 112.7 | 158.6 ± 73.7 | 0.990 | 0.751 |
| Hours from arrival to 1st blood product | 4.8 ± 6.2 | 3.5 ± 5.4 | 0.097 | 0.003 | 0.8 ± 1.1 | 0.9 ± 1.4 | 0.709 | 0.688 |
| Hours from arrival to 1st PRBC | 5.2 ± 6.5 | 4.1 ± 5.5 | 0.147 | 0.012 | 0.8 ± 1.1 | 1.4 ± 1.8 | 0.236 | 0.180 |
| Hours from arrival to 1st FFP† | 18.2 ± 82.8 (n=49) |
36.4 ± 245.2 (n=58) |
0.601 | 0.639 | 2.7 ± 2.2 (n=23) | 1.0 ± 0.6 (n=17) | 0.001 | 0.005 |
| Hours from arrival to 1st platelets‡ | 39.8 ± 103.8 (n=31) |
165.8 ± 575.2 (n=27) |
0.271 | 0.279 | 6.0 ± 8.1 (n=18) | 4.1 ± 5.2 (n=15) | 0.444 | 0.421 |
NA: Not applicable; statistical comparison not performed
Insufficient sample size to perform adjusted analysis
Patients who did not receive FFP were excluded from this analysis
Patients who did not receive platelets were excluded from this analysis
There were no significant differences in patient outcomes (Table 4); however, while not significant, all hemorrhagic deaths occurred pre-MTP implementation. When comparing massively transfused patients who were alive versus dead at 24 hours, there were significant differences in ISS, temperature, PT, aPTT, INR, base deficit, and BIG score. Since a number of these parameters overlap, we performed an adjusted logistic regression using ISS and the BIG score. After adjustment for ISS and BIG score, the decrease in hours from arrival to first blood product and PRBC when comparing all transfused patient pre and post-MTP became significant, but not in the patients who received massive transfusion. There remained no significant differences in patient outcomes after adjustment for ISS and BIG scores (Tables 3 and 4).
Table 4.
Unadjusted and adjusted logistic regression (for BIG and ISS) comparing outcomes pre and post-MTP implementation with all transfused children and with only the subgroups who received massive transfusion
| Variables | All transfused |
Massive transfusion subgroup |
||||||
|---|---|---|---|---|---|---|---|---|
| Pre-MTP Group 1,2 N=120 |
Post-MTP Group 3–6 N=115 |
Unadjusted p-value |
Adjusted p-value |
Pre-MTP Group 1 N=26 |
Post-MTP Group 3,5 N=17 |
Unadjusted p-value |
Adjusted p-value |
|
| 24-hour mortality | 16 (13.3%) | 12 (10.4%) | 0.493 | 0.875 | 10 (38.5%) | 6 (35.3%) | 0.834 | 0.944 |
| In-hospital mortality | 29 (24.2%) | 22 (19.1%) | 0.349 | 0.623 | 14 (53.8%) | 8 (47.1%) | 0.663 | 0.729 |
| Cause of death | 0.092 | 0.934 | 0.306 | 0.937 | ||||
| Hemorrhage | 5 (17.2%) | 0 (0%) | 3 (21.4%) | 0 (0%) | ||||
| Brain injury | 21 (72.4%) | 18 (81.8%) | 9 (64.3%) | 7 (87.5%) | ||||
| Organ failure | 1 (3.4%) | 1 (4.6%) | 1 (7.1%) | 0 (0%) | ||||
| Other | 2 (6.9%) | 1 (4.6%) | 1 (7.1%) | 0 (0%) | ||||
| Unknown | 0 (0%) | 2 (9.1%) | 0 (0%) | 1 (12.5%) | ||||
| Hospital LOS (n= survivors) |
26.5 ± 38.3 (n=91) |
36.8 ± 128.6 (n=93) |
0.464 | 0.606 | 39.0 ± 30.1 (n=12) | 45.8 ± 30.9 (n=9) | 0.619 | 0.656 |
| PICU free days | 5.6 ± 5.9 | 6.2 ± 6.7 | 0.447 | 0.180 | 4.3 ± 5.8 | 6.0 ± 7.6 | 0.420 | 0.330 |
| Ventilator-free day | 8.1 ± 7.2 | 8.2 ± 7.6 | 0.944 | 0.594 | 7.0 ± 8.1 | 8.3 ± 9.3 | 0.614 | 0.584 |
| Pressor-free days | 11.7 ± 10.2 | 12.0 ± 10.1 | 0.851 | 0.462 | 10.3 ± 12.3 | 11.8 ± 11.5 | 0.695 | 0.719 |
| Composite morbidity (yes) | 29 (24.2%) | 21 (18.3%) | 0.269 | 0.946 | 11 (42.3%) | 7 (41.2%) | 0.941 | 0.964 |
| Glasgow Outcome Score | 0.640 | 0.623 | 0.779 | 0.729 | ||||
| 1 | 29 (24.2%) | 22 (19.1%) | 14 (53.8%) | 8 (47.1%) | ||||
| 2 and 3 | 16 (13.3%) | 17 (14.8%) | 5 (19.2%) | 5 (29.4%) | ||||
| 4 and 5 | 75 (62.5%) | 76 (66.1%) | 7 (26.9%) | 4 (23.5%) | ||||
Discussion
In this study of MTP implementation, we were not able to demonstrate reduced mortality, but we did demonstrate increased plasma and platelet to RBC ratios, reduced plasma and platelet deficits assuming a balanced transfusion ratio and decreased time to first plasma administration. These are factors believed to address coagulopathy and theoretically improve the management of hemorrhagic shock. Equally important, MTP did not lead to increased blood product use. Finally, we did not see an increase in the composite morbidity rate associated with the changes in transfusion practice that occurred with MTP implementation.
Multiple adult studies on MTP in trauma have shown improved survival and decreased blood product use. Cotton et al. showed a 74% reduction in odds of mortality and a decrease in overall blood product use at 24 hours.13 In addition to improved mortality, in a second study they showed that patients who received MTP were less likely to develop septic shock, ventilator-associated pneumonia, and multi-organ failure.12 Dente et al. demonstrated that transfusion under MTP improved survival to discharge in patients who suffered blunt trauma, however patients who suffered penetrating trauma had improved 24-hour survival, but not survival to discharge.21 O’Keefe et al did not show improvement in mortality, but their study did find that the MTP group received fewer blood products, which translated to a savings of $2270 per patient.
While the adult literature is promising, literature on pediatric MTP is limited and thus far has not shown any improvement in outcome after MTP implementation. Dressler et al. and Pickett et al. described the use of MTP in a child who suffered from intraoperative bleeding during a right hepatectomy and a pediatric victim of a gunshot wound to the chest, respectively.22,23 Neither were coagulopathic upon arrival to the ICU after being in the operating room. While these case studies illustrated the prevention of coagulopathy, larger case series have not demonstrated improved survival with pediatric MTP. Two prior studies examined the use of pediatric MTP and its effect on outcome in pediatric hemorrhage. Chidester et al. prospectively examined the impact of MTP implementation at their institution and reported the outcomes on 8 surgical and 47 traumatic patients, of which 22 received blood products with MTP activation and 33 without MPT activation.24 They found no difference in mortality between the two groups, but thromboembolic complications occurred more frequently in the non-MTP group.
Hendrickson et al. reported outcomes for 102 children transfused for traumatic hemorrhage of which 43 (17 pre-MTP and 26 post-MTP) were massively transfused (received ≥ 70 ml/kg blood products), a similar sample size to ours.17 Similarly, they found no difference in mortality, but improved median time to plasma transfusion. Unlike our study, they only compared blood product ratios at 24 hours from admission. It is important to identify whether MTP can facilitate improved ratios at earlier time periods, so we also studied values at 6 hours. The Prospective, Observational, Multicenter, Major Trauma Transfusion (PROMMT) study, a prospective cohort study that examined the effect on outcome of the timing of blood transfusions during trauma resuscitation in adults found that high ratio transfusion is most helpful in the first 6 hours when exsanguination is the most common cause of death.7 While it did not improve mortality in our study, MTP implementation resulted in higher plasma and platelet to RBC ratios at 6 hours.
Despite the fact that previous adult studies indicate that higher plasma and platelet to RBC transfusion improves survival in trauma, there is also concern that transfusion of these blood products may lead to complications such as multi-organ failure, sepsis, TRALI, and ARDS, especially in patients not at risk for massive hemorrhage.16 Because the rate of each potential complication is low, we created a composite morbidity variable. In our study population, children cared for post-MTP received more plasma and platelets, yet there was no significant difference in composite morbidity.
We specifically compared all transfused children and then only the children who received massive volumes of blood product, omitting children with MTP activations who did not receive massive volumes (group 4). While group 4 was more severely injured than the other non-massive transfusion groups but did not receive massive transfusion volumes, it is unclear whether MTP was activated unnecessarily in patients in group 4 or if MTP activation enabled the achievement of hemostasis earlier and thus eliminated the need for massive volumes of blood products. Group 4 had a lower in-hospital mortality rate than the children who had MTP activations and received massive transfusion (group 3), 1 (20.0%) vs. 7 (38.9%), respectively. Given the limitations of a retrospective study, it is difficult to determine causality, whether children in group 4 survived because of the MTP activation or because they were not as severely injured and MTP did not play a role in their survival. Group 4 was not included in our analysis of outcomes for children who received massive transfusion pre and post-MTP implementation in order to prevent falsely reporting improved mortality post-MTP.
A survey of MTPs at pediatric hospitals revealed a wide variation in the use of MTPs and also the ratios of plasma and platelets:RBC, with most utilizing transfusion ratios ≥ 1:2 but not all MTPs provided products at a 1:1:1 ratio.25 Also, MTP activations were rare events in pediatric hospitals. In the surveyed hospitals, Horst et al. found that hospitals who were able to report an exact number of activations reported a median of six activations in one year, and these were not limited to activations for trauma patients, making MTP activations a rare event in the care of injured children.25 Given that MTP activations are uncommon, the optimal transfusion ratio has not yet been determined, and given the wide variation in MTP at pediatric hospitals, it is difficult at this time to conclude definitively whether MTP should be used in children.
This study had several limitations. It is a single-center study with a sample size that may have been insufficient to examine infrequent events. Retrospective chart reviews have a potential for ascertainment bias, however, we limited this bias by blinding the abstractor to patient group assignment. We also limited survivor treatment bias by excluding patients who died within 1 hour of arrival, but no patients fell under this exclusion. We also assumed an optimal ratio of 1:1:1 units of plasma:platelets:RBC to be used to address hemorrhagic shock but this has yet to be demonstrated in children. This ratio was obtained from adult studies. In addition to the optimal blood ratio, we recognize that the definition of massive transfusion, especially in children, has been debated. We used the definition that was prevalent at the time of our study and used in our institutional MTP.
MTPs are a promising method of resuscitation to prevent hemorrhagic death in trauma, a time-dependent event. This study demonstrated the ability of our specific MTP to facilitate transfusion of plasma and platelets at higher ratios early in resuscitation, which in adults has been shown to improve early mortality from trauma. However, despite the improved blood ratios, we were unable to demonstrate improved mortality with implementation of MTP. A larger multi-center study is needed to determine if MTPs translate into improved mortality in children without increasing morbidity.
Acknowledgments
Source of Funding:
This study was funded by the Training of the Pediatric Emergency Physician-Scientist Grant (NIH Grant#T32HD049338) and Washington University Institute of Clinical and Translational Sciences which is, in part, supported by the NIH/National Center for Advancing Translational Sciences (NCATS) (CTSA grant #UL1TR000448).
Footnotes
Conflicts of Interest
There were no conflicts of interest to disclose.
References
- 1.Holcomb JB, Tilley BC, 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(5):471–82. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Baker CC, Oppenheimer L, Stephens B, et al. Epidemiology of trauma deaths. Am J Surg. 1980;140:144–50. [DOI] [PubMed] [Google Scholar]
- 3.Dutton RP, Stansbury LG, Leone S, et al. Trauma mortality in mature trauma systems: Are we doing better? An analysis of trauma mortality patterns, 1997–2008. J Trauma. 2010;69(3):620–6. [DOI] [PubMed] [Google Scholar]
- 4.Kauvar DS, Lefering R, Wade CE. Impact of hemorrhage on trauma outcome: An overview of epidemiology, clinical presentations, and therapeutic considerations. J Trauma. 2006;60:S3–S11. [DOI] [PubMed] [Google Scholar]
- 5.Spinella PC, Holcomb JB. Resuscitation and transfusion principles for traumatic hemorrhagic shock. Blood Rev. 2009;23(6):231–40. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Zink KA, Sambasivan CN, Holcomb JB, et al. A high ratio of plasma and platelets to packed red blood cells in the first 6 hours of massive transfusion improves outcomes in a large multicenter study. Am J Surg. 2009;197(5):565–70. [DOI] [PubMed] [Google Scholar]
- 7.Holcomb JB, del Junco DJ, Fox EE, et al. The Prospective, Observational, Multicenter, Major Trauma Transfusion (PROMMTT) Study. JAMA Surgery. 2013;148(2):127. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hendrickson JE, Shaz BH, Pereira G, et al. Coagulopathy is prevalent and associated with adverse outcomes in transfused pediatric trauma patients. J Pediatr. 2012;160(2):204–9.e3. [DOI] [PubMed] [Google Scholar]
- 9.MacLeod JBA, Lynn M, McKenney MG, et al. Early coagulopathy predicts mortality in trauma. J Trauma. 2003;55(1):39–44. [DOI] [PubMed] [Google Scholar]
- 10.Niles SE, McLaughlin DF, Perkins JG, et al. Increased mortality associated with the early coagulopathy of trauma in combat casualties. J Trauma. 2008;64(6):1459–65. [DOI] [PubMed] [Google Scholar]
- 11.Patregnani JT, Borgman MA, Maegele M, et al. Coagulopathy and shock on admission is associated with mortality for children with traumatic injuries at combat support hospitals. Pediatr Crit Care Med. 2012;13(3):1–5. [DOI] [PubMed] [Google Scholar]
- 12.Cotton BA, Au BK, Nunez TC, et al. Predefined massive transfusion protocols are associated with a reduction in organ failure and postinjury complications. J Trauma. 2009;66(1):41–9. [DOI] [PubMed] [Google Scholar]
- 13.Cotton BA, Gunter OL, Isbell J, et al. Damage control hematology: The impact of a trauma exsanguination protocol on survival and blood product utilization. J Trauma. 2008;64(5):1177–83. [DOI] [PubMed] [Google Scholar]
- 14.Dente CJ, Shaz BH, Nicholas JM, et al. Improvements in early mortality and coagulopathy are sustained better in patients with blunt trauma after institution of a massive transfusion protocol in a civilian level I trauma center. J Trauma. 2009;66(6):1616–24. [DOI] [PubMed] [Google Scholar]
- 15.O’Keeffe T, Refaai M, Tchorz K. A massive transfusion protocol to decreased blood component use and costs. Arch Surg. 2008;143(7):686–91. [DOI] [PubMed] [Google Scholar]
- 16.Borgman MA, Spinella PC, Holcomb JB, et al. The effect of FFP:RBC ratio on morbidity and mortality in trauma patients based on transfusion prediction score. Vox Sang. 2011;101(1):44–54. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Hendrickson JE, Shaz BH, Pereira G, et al. Implementation of a pediatric trauma massive transfusion protocol: one institution’s experience. Transfusion. 2012;52(6):1228–36. [DOI] [PubMed] [Google Scholar]
- 18.Palmieri TL, Greenhalgh DG, Sen S. Prospective comparison of packed red blood cell–to–fresh frozen plasma transfusion ratio of 4:1 versus 1:1 during acute massive burn excision. J Trauma Acute Care Surg. 2013;74(1):76–83. [DOI] [PubMed] [Google Scholar]
- 19.Fuenfer MM, Creamer KM, editors. Pediatric Surgery and Medicine for Hostile Environments. Washington, DC: Walter Reed Army Medical Center, Borden Institute; 2010. [Google Scholar]
- 20.Harris P, Taylor R, Thielke R, et al. Research electronic data capture (REDCap) - A metadata-driven methodology and workflow process for providing translational research informatics. J Biomed Inform. 2009;42(2):377–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Dente CJ, Shaz BH, Nicholas JM, et al. Early predictors of massive transfusion in patients sustaining torso gunshot wounds in a civilian level I trauma center. J Trauma. 2010;68(2):298–304. [DOI] [PubMed] [Google Scholar]
- 22.Dressler AM, Finck CM, Carroll CL, et al. Use of a massive transfusion protocol with hemostatic resuscitation for severe intraoperative bleeding in a child. J Pediatr Surg. 2010;45(7):1530–3. [DOI] [PubMed] [Google Scholar]
- 23.Pickett PM, Tripi PA. Massive transfusion protocol in pediatric trauma. Int Anesthesiol Clin. 2011;49(2):62–7. [DOI] [PubMed] [Google Scholar]
- 24.Chidester SJ, Williams N, Wang W, et al. A pediatric massive transfusion protocol. J Trauma Acute Care Surg. 2012;73(5):1273–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Horst J, Leonard JC, Vogel A, et al. A survey of US and Canadian hospitals’ paediatric massive transfusion protocol policies. Transfus Med. 2016;26(1):49–56. [DOI] [PubMed] [Google Scholar]