Abstract
Objective:
The purpose of our study was to describe children with life-threatening bleeding.
Design:
We conducted a prospective observational study of children with life-threatening bleeding events.
Setting:
Twenty-four children’s hospitals in the United States, Canada and Italy participated.
Subjects:
Children ages 0–17 years who received >40 mL/kg total blood products over 6 hours or were transfused under massive transfusion protocol (MTP) were included.
Interventions:
Children were compared according bleeding etiology: trauma, operative or medical.
Measurements:
Patient characteristics, therapies administered, and clinical outcomes were analyzed.
Results:
Among 449 enrolled children, 55.0% were male and the median age was 7.3 years. Bleeding etiology was 46.1% trauma, 34.1% operative and 19.8% medical. Prior to the life-threatening bleeding event, most had age-adjusted hypotension (61.2%) and 25% were hypothermic. Children with medical bleeding had higher median Pediatric Risk of Mortality scores (18) compared to children with trauma (11) and operative bleeding (12). Median Glasgow Coma Scores were lower for children with trauma (3) compared to operative (14) or medical bleeding (10.5). Median time from bleeding onset to first transfusion was 8 minutes for red blood cells, 34 minutes for plasma and 42 minutes for platelets. Post-event acute respiratory distress syndrome (20.3%) and acute kidney injury (18.5%) were common. Twenty-eight-day mortality was 37.5% and higher among children with medical bleeding (65.2%) compared to trauma (36.1%) and operative (23.8%). There were 82 hemorrhage deaths; 65.8% occurred by 6 hours and 86.5% by 24 hours.
Conclusions:
Patient characteristics and outcomes among children with life-threatening bleeding varied by cause of bleeding. Mortality was high and death from hemorrhage in this population occurred rapidly.
Keywords: Hemorrhage, Blood Transfusion, Resuscitation, Trauma Centers, Emergency Medicine, Operating Rooms, Critical Care, Pediatrics, Child
Introduction
Over the past decade, management of life-threatening bleeding has incorporated damage control resuscitation (DCR) strategies.1 Principles of DCR include permissive hypotension; rapid surgical control; avoidance of hemodilution, acidosis, hypocalcemia, and hypothermia; and early empiric, balanced transfusion with red blood cells (RBCs), plasma, and platelets, or whole blood when available.2,3 Medical centers use massive transfusion protocols (MTPs) to standardize the resuscitation of patients with life-threatening bleeding using DCR principles.4,5
Evidence for DCR is rooted in military and adult civilian trauma populations, however through the implementation of MTPs, these strategies are being used to manage all causes of hemorrhage, including medical and surgical bleeding, and for children.4 There are no prospective pediatric multi-center data published on the demographic and clinical characteristics of children with life-threatening bleeding, which DCR therapies are most commonly used in children, or associated outcomes according to cause bleeding.
It is essential that DCR principles are studied in children.6 Children have different disease burden and pathophysiology than adults leading to different etiologies for bleeding and response to therapy. These differences may affect which resuscitative practices are efficacious and safe for children with life-threatening bleeding and warrant pediatric-specific clinical trials. The purpose of this study was to prospectively describe the patient characteristics, management, and outcomes of children with life-threatening bleeding as a requisite step to inform clinical trial design. We hypothesized that there would be differences in these data based on the etiology of bleeding which can inform future lines of inquiry.
Materials and Methods
We conducted a prospective, observational study of children presenting with life-threatening bleeding at 24 medical centers in the United States, Canada, and Italy from January 2014 through October of 2018. The month that each participating center began enrolling varied from January 2014 to May 2017.
Children ages 0–17 years were eligible for enrollment if they received >40 mL per kg of total blood products over 6 hours or if they were transfused under MTP activation. The threshold of > 40ml/kg of blood products is based on literature that it is predictive of increased risk of death in children with traumatic injury.7 Children were excluded if a MTP was activated, but no blood products were transfused. Hospital blood bank documentation was used to identify eligible subjects.
Data extracted from hospital medical records included demographics, cause of bleeding, Glasgow Coma Scale (GCS), vital signs, measures of end-organ function, fluids, blood products, hemostatic adjuncts, morbidities, and mortality. The timing of mortality was based on when the first life-threatening event occurred. For patients with multiple events only the first event was analyzed. The injury severity score (ISS) was obtained from the hospital trauma registry. The Pediatric Risk of Mortality score (PRISM III) was calculated for the day of the bleeding event. Pre-event data were collected for the 24-hour period prior to the event. Data categorized as “during” the event were collected from the initiation of the massive transfusion (MT) until the end of the MT. The initiation of the MT was defined as either the time the MTP was activated or the time of administration of the first blood product for children without MTP activation. The end of a MT was defined as the time MTP was terminated or the time that the last blood product was transfused with no other blood products being administered in the subsequent 60 minutes. Post-event data were collected at the time closest to the end of the event with a maximum of 12 hours. Patients were categorized into three groups according to the etiology of their bleeding: traumatic injury, operative, and medical. Traumatic injury was defined as bleeding due to blunt or penetrating trauma. Operative was defined as any bleeding related to an operative procedure in the procedural or immediate post-procedural period, including interventional radiology procedures. Medical bleeding was defined as being a result of the pathophysiology of medical illness (e.g. gastrointestinal bleeding).
Missing weight values were estimated based on age and sex from the Centers for Disease Control and Prevention Clinical Growth Charts in order to standardize volume of blood product administered.8 When calculating PRISM III scores, all missing values were assumed to be within the normal range according to the methods for calculating the score.9 During data quality monitoring, one site (10 patients) had missing data for many key variables and was eliminated from analysis.
This study was conducted under waiver of informed consent granted by each participating center’s institutional review board.
Statistical analyses were conducted using SAS/STAT® software, v 9.4. When comparing between the three groups, tests of association included ANOVA and chi-square unless distributional assumptions were not met in which case Kruskal-Wallis H and Fisher’s exact were employed. If an overall test was significant (P < 0.05), pairwise tests were performed applying the Bonferroni correction (P = 0.05 / 3 = 0.017). The Kaplan-Meier method was used to calculate 28-day survival.
Results
A total of 449 children with life-threatening bleeding met inclusion criteria. The median (interquartile range (IQR)) age was 7.3 years (1.7–14.7) (Table 1). For those with documented race and ethnicity, 28.7% were black and 11.0% Hispanic. There were more males (55.0%), however this related to a higher proportion of males with traumatic bleeding (62.3%). The causes of bleeding were traumatic injury in 207 children (46.1%), operative in 153 (34.1%) and medical in 89 (20.2%). Children with traumatic bleeding were older, 10.4 years, (4.7–15.4) and those with operative bleeding were younger, 2.1 years, (0.3–11.8). Twenty-nine children had more than one life-threatening bleeding event; traumatic injury (3), operative (19), and medical (7).
Table 1.
Demographic and blood type measures
| Overall (N = 449) | Trauma (N = 207) | Operative (N = 153) | Medical (N = 89) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Measure | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | P |
| Age (yr)a, b, c | 449 | 7.3[1.7–14.7] | 207 | 10.4 [4.7–15.4] | 153 | 2.1 [0.3–11.8] | 089 | 7.96 [1.2–14.7] | <.01 |
| Age < 1 yeara, b, c | 449 | 94 (20.9) | 207 | 16 (7.7) | 153 | 58 (37.9) | 089 | 20 (22.5) | <.01 |
| Sexa, b | 449 | 207 | 153 | 089 | 0.02 | ||||
| Male | 247 (55.0) | 129 (62.3) | 76 (49.7) | 42 (47.2) | |||||
| Female | 202 (45.0) | 78 (37.7) | 77 (50.3) | 47 (52.8) | |||||
| Racea, b | 380 | 173 | 131 | 076 | <.01 | ||||
| White | 242 (63.7) | 98 (56.6) | 91 (69.5) | 53 (69.7) | |||||
| Black | 109 (28.7) | 72 (41.6) | 20 (15.3) | 17 (22.4) | |||||
| Other | 29 (7.6) | 3 (1.7) | 20 (15.3) | 6 (7.9) | |||||
| Hispanic|Latino | 392 | 43 (11.0) | 181 | 16 (8.8) | 134 | 21 (15.7) | 077 | 6 (7.8) | <.10 |
| Blood group | 436 | 197 | 152 | 087 | 0.98 | ||||
| A | 157 (36.0) | 67 (34.0) | 58 (38.2) | 32 (36.8) | |||||
| B | 65 (14.9) | 30 (15.2) | 22 (14.5) | 13 (14.9) | |||||
| AB | 12 (2.8) | 5 (2.5) | 5 (3.3) | 2 (2.3) | |||||
| O | 202 (46.3) | 95 (48.2) | 67 (44.1) | 40 (46.0) | |||||
| Rhesus factor | 434 | 196 | 151 | 087 | 0.87 | ||||
| + | 359 (82.7) | 163 (83.2) | 123 (81.5) | 73 (83.9) | |||||
| − | 75 (17.3) | 33 (16.8) | 28 (18.5) | 14 (16.1) | |||||
Abbreviations M median; N number; P probability; Q quartile.
P < 0.017 trauma v operative
P < 0.017 trauma v medical
P < 0.017 operative v medical
Children with traumatic injuries had a median ISS (IQR) of 29 (20–38), indicating severe injury. Blunt trauma accounted for 66% of injuries and penetrating 34%. The surgical services for children with operative bleeding were cardiothoracic surgery (34%), neurosurgery (24%), general surgery (19%), liver transplant (8%) and other (23%). The conditions in children with medical bleeding were gastrointestinal bleeding (33%), sepsis (19%), oncologic (12%), and other (36%).
Most children (61.2%) had age-adjusted hypotension and approximately a quarter were hypothermic (35°C) prior to the bleeding event (Table 2). Children with medical bleeding had higher PRISM III scores [Median (IQR), 18 (12–24)] when compared to children with traumatic [11 (5–23)] and operative bleeding [12 (6–21)]. However, GCS was lower for children with traumatic bleeding [3 (3–13)] compared to children with operative [14 (3–15)] and medical bleeding [10.5 (3–14)].
Table 2.
Clinical measures pre life-threatening bleeding
| Overall (N = 449) | Trauma (N = 207) | Operative (N = 153) | Medical (N = 89) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Measure | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | P |
| Pediatric risk of mortality IIIb, c | 428 | 12 [6–22] | 193 | 11 [5–23] | 150 | 12 [6–21] | 85 | 18 [12–24] | <.01 |
| Lowest Glasgow Coma Scalea | 255 | 7 [3–14] | 165 | 3 [3–13] | 50 | 14.0 [3–15] | 40 | 10.5 [3–14] | <.01 |
| Fixed, dilated pupilsa, b | 362 | 64 (17.7) | 180 | 54 (30.0) | 116 | 3 (2.6) | 66 | 7 (10.6) | <.01 |
| Mechanical ventilation#b | 432 | 255 (59.0) | 196 | 106 (54.1) | 152 | 88 (57.9) | 84 | 61 (72.6) | 0.02 |
| Age-adjusted hypotension | 410 | 251 (61.2) | 180 | 107 (59.4) | 145 | 85 (58.6) | 85 | 59 (69.4) | 0.22 |
| Lowest body temperature (°C) | 339 | 36.2 [35.2–36.6] | 123 | 36.0 [35.0–36.5] | 136 | 36.2 [35.4–36.6] | 80 | 36.2 [35.6–36.7] | 0.13 |
| Highest body temperature (°C)a, b | 352 | 36.9 [36.3–37.5] | 129 | 36.5 [35.7–37.0] | 143 | 37.2 [36.8–37.7] | 80 | 37.1 [36.7–38.0] | <.01 |
| Lowest power of hydrogen (pH)a, c | 308 | 7.20 [7.10–7.30] | 102 | 7.17 [7.00–7.24] | 135 | 7.30 [7.20–7.30] | 71 | 7.20 [7.00–7.30] | <.01 |
| Base deficit|excess (mmol/L) | 272 | −.45 [−5.0–7.00] | 89 | −1.0 [−7.0–8.00] | 123 | 0.00 [−4.0–5.00]. | 60 | 1.10 [−7.0–6.50]. | 0.89 |
| Age-adjusted abnormally high serum creatinineb, c | 299 | 90 (30.1) | 110 | 30 (27.3) | 112 | 24 (21.4) | 77 | 36 (46.8) | <.01 |
| Lowest hemoglobin (g/dL)b, c | 339 | 9.5 [7.4–11.3] | 130 | 9.8 [8.3–11.8] | 130 | 9.8 [7.7–11.6] | 79 | 7.90 [6.00–9.70] | <.01 |
| Lowest white blood cell count (K/cumm)a, b | 323 | 10.4 [5.8–17.0] | 130 | 12.5 [8.2–17.7] | 117 | 8.6 [5.9–14.6] | 76 | 6.6 [2.3–19.4] | <.01 |
| Lowest platelet count (x 109/L)a, b, c | 324 | 152 [76–242] | 132 | 203 [149–282]. | 116 | 137 [78.5–230] | 76 | 49 [20–126]. | <.01 |
| Highest international normalized ratiob, c | 250 | 1.50 [1.20–2.03] | 98 | 1.50 [1.26–1.90] | 95 | 1.39 [1.15–1.70] | 57 | 1.78 [1.35–3.70] | <.01 |
| Highest prothrombin time (sec)c | 246 | 17.6 [15.0–21.9] | 100 | 17.6 [15.0–21.7] | 89 | 17.1 [14.0–20.3] | 57 | 19.8 [15.7–31.4] | 0.01 |
| Highest partial thromboplastin time (sec)b | 259 | 42.9 [31.2–91.0] | 103 | 38.0 [28.0–70.0] | 98 | 43.3 [32.9–62.0] | 58 | 61.9 [31.7–150]. | 0.02 |
| Fibrinogen (mg/dL)a, b | 131 | 158 [95.0–261] | 27 | 121 [60.0–176] | 63 | 181 [127–275]. | 41 | 178 [126–261]. | 0.01 |
Abbreviations M median; N number; P probability; Q quartile.
P < 0.017 trauma v operative
P < 0.017 trauma v medical
P < 0.017 operative v medical
For children with operative bleeding, mechanical ventilation status was assessed immediately prior to the bleeding event but did not include intubation and ventilation for the procedure.
Children with traumatic and medical bleeding had lower serum pH values [Median (IQR), 7.17 (7.00–7.24) and 7.20 (7.00–7.30) respectively], compared to children with operative bleeding [7.30 (7.20–7.30)]. More children with medical bleeding [46.8%] had elevated creatinine levels compared to traumatic [27.3%] and operative bleeding [21.4%]. Children with medical bleeding had lower hemoglobin levels [7.9 g/dL (6.0–9.7)] and higher international normalized ratios (INRs) [1.78 (1.35–3.70)] compared to those with traumatic [9.8 g/dL (8.3–11.8); 1.50 (1.26–1.90)] and operative bleeding [9.8 g/dL (7.7–11.6); 1.39 (1.15–1.70)].
Regardless of the cause of bleeding, children frequently received RBC transfusion prior to MT (41.4%) (Table 3). Those with operative bleeding received higher volumes of RBCs pre-event [35.1 mL/kg (15.3–60.0)] compared to those with traumatic [17.9 mL/kg (10.6–28.8)] or medical bleeding [18.9 mL/kg (11.9–31.9)]. Plasma, platelet and cryoprecipitate administration prior to MT was more common for children with operative (24.8%, 24.8%, 14.4% respectively) and medical bleeding (32.6%, 28.1%, 7.9%) compared to those with traumatic bleeding (6.8%, 2.9%, 1.0.9%). Most children (62.8%) received crystalloid prior to MT with a median volume administered of 25 mL/kg (12.4–41.4). Pre-event use of tranexamic acid, aminocaproic acid, and recombinant factor VIIa was uncommon.
Table 3.
Measures of blood product use and hemostatic agent administration pre life-threatening bleeding
| Overall (N = 449) | Trauma (N = 207) | Operative (N = 153) | Medical (N = 89) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Measure | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | P |
| Any red blood cells | 449 | 186 (41.4) | 207 | 86 (41.5) | 153 | 60 (39.2) | 89 | 40 (44.9) | 0.68 |
| Red blood cells (mL/kg)a, c | 186 | 20.2 [11.2–41.5] | 86 | 17.9 [10.6–28.8] | 60 | 35.1 [15.3–60.0] | 40 | 18.9 [11.9–31.9] | <.01 |
| Any plasmaa, b | 449 | 81 (18.0) | 207 | 14 (6.8) | 153 | 38 (24.8) | 89 | 29 (32.6) | <.01 |
| Plasma (mL/kg)a, b | 81 | 13.1 [6.0–29.7] | 14 | 5.2 [3.9–10.4] | 38 | 18.8 [9.4–41.0] | 29 | 18.0 [7.6–36.9] | <.01 |
| Any plateletsa, b | 449 | 69 (15.4) | 207 | 6 (2.9) | 153 | 38 (24.8) | 89 | 25 (28.1) | <.01 |
| Platelets (mL/kg)c | 69 | 12.3 [6.7–22.5] | 6 | 17.0 [6.8–22.5] | 38 | 16.8 [8.3–33.3] | 25 | 8.3 [5.4–12.3] | 0.01 |
| Any cryoprecipitatea, b | 449 | 31 (6.9) | 207 | 2 (1.0) | 153 | 22 (14.4) | 89 | 7 (7.9) | <.01 |
| Cryoprecipitate (mL/kg) | 31 | 5.17 [2.9–15.2] | 2 | 16.0 [2.9–29.1] | 22 | 5. 9 [3.3–15.2] | 7 | 4.7 [2.2–8.8] | 0.54 |
| Any blood product | 449 | 212 (47.2) | 207 | 87 (42.0) | 153 | 77 (50.3) | 89 | 48 (53.9) | 0.11 |
| Total blood product (mL/kg)a | 212 | 23.6 [12.0–55.1] | 87 | 18.1 [10.7–31.4] | 77 | 37.2 [16.1–80.3] | 48 | 25.1 [13.2–54.7] | <.01 |
| Any crystalloida, b | 449 | 282 (62.8) | 207 | 110 (53.1) | 153 | 107 (69.9) | 89 | 65 (73.0) | <.01 |
| Crystalloid (mL/kg) | 282 | 25.0 [12.4–41.4] | 110 | 27.1 [14.3–41.5] | 107 | 24.5 [12.0–33.8] | 65 | 25.2 [12.4–55.6] | 0.24 |
| Any colloida, b | 449 | 64 (14.3) | 207 | 8 (3.9) | 153 | 42 (27.5) | 89 | 14 (15.7) | <.01 |
| Any tranexamic acida, c | 449 | 20 (4.5) | 207 | 4 (1.9) | 153 | 15 (9.8) | 89 | 1 (1.1) | <.01 |
| Any aminocaproic acida | 449 | 14 (3.1) | 207 | 0 (0.0) | 153 | 12 (7.8) | 89 | 2 (2.2) | <.01 |
| Any recombinant factor VIIaa | 449 | 10 (2.2) | 207 | 1 (0.5) | 153 | 7 (4.6) | 89 | 2 (2.2) | 0.03 |
Abbreviations M median; N number; P probability; Q quartile.
P < 0.017 trauma v operative
P < 0.017 trauma v medical
P < 0.017 operative v medical
Most children were transfused under MTP activations: trauma (96.1%), operative (69.9%) and medical (82%) (Table 4). The median duration of MT events was 3.5 hours (1.5–6.1). Thirty-six children (8%) were on Extracorporeal Membrane Oxygenation; all had either operative or medical bleeding.
Table 4.
Measures of clinical management during the life-threatening bleeding event
| Overall (N = 449) | Trauma (N = 207) | Operative (N = 153) | Medical (N = 89) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Measure | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | P |
| Massive transfusion protocol activateda, b | 449 | 379 (84.4) | 207 | 199 (96.1) | 153 | 107 (69.9) | 89 | 73 (82.0) | <.01 |
| Location of massive transfusiona, b, c | 449 | 207 | 153 | 89 | <.01 | ||||
| Emergency Department | 160 (35.6) | 139 (67.1) | 3 (2.0) | 18 (20.2) | |||||
| Operative Room | 125 (27.8) | 35 (16.9) | 84 (54.9) | 6 (6.7) | |||||
| Intensive Care Unit | 154 (34.3) | 31 (15.0) | 61 (39.9) | 62 (69.7) | |||||
| Other | 10 (2.2) | 2 (1.0) | 5 (3.3) | 3 (3.4) | |||||
| Extracorporeal membrane oxygenationa, b | 449 | 36 (8.0) | 207 | 0 (0.0) | 153 | 26 (17.0) | 89 | 10 (11.2) | <.01 |
| Massive transfusion duration (hr) | 449 | 3.5 [1.5–6.1] | 207 | 3.0 [1.3–5.7] | 153 | 3.8 [1.9–6.7] | 89 | 4.0 [1.1–6.6] | 0.11 |
| Any red blood cells | 449 | 413 (92.0) | 207 | 191 (92.3) | 153 | 145 (94.8) | 89 | 77 (86.5) | 0.07 |
| Red blood cells (mL/kg)a, c | 449 | 30.2 [15.5–62.3] | 207 | 25.3 [12.5–48.9] | 153 | 48.5 [23.4–87.4] | 89 | 26.4 [10.8–51.8] | <.01 |
| Time to red blood cells (min) | 405 | 8.0 [0.0–42.0] | 187 | 8.0 [0.0–44.0] | 142 | 7.0 [0.0–40.0] | 76 | 11.5 [0.0–45.5] | 0.75 |
| Any plasma | 449 | 359 (80.0) | 207 | 161 (77.8) | 153 | 125 (81.7) | 89 | 73 (82.0) | 0.57 |
| Plasma (mL/kg)a | 449 | 17.1 [4.84–35.6] | 207 | 13.2 [3.57–27.0] | 153 | 22.0 [8.57–48.1] | 89 | 18.1 [5.94–39.5] | <.01 |
| Time to plasma (min) | 347 | 34.0 [15.0–77.0] | 156 | 33.5 [17.5–74.0] | 123 | 39.0 [12.0–83.0] | 68 | 30.0 [15.5–65.0] | 0.86 |
| Any plateletsa, b | 449 | 297 (66.1) | 207 | 120 (58.0) | 153 | 111 (72.5) | 89 | 66 (74.2) | <.01 |
| Platelets (mL/kg)a, b | 449 | 7.30 [0.00–19.8] | 207 | 3.85 [0.00–13.5] | 153 | 13.3 [0.00–33.3] | 89 | 10.4 [0.00–19.8] | <.01 |
| Time to platelets (min) | 293 | 42.0 [15.0–102]. | 118 | 42.0 [20.0–73.0] | 109 | 43.0 [13.0–118]. | 66 | 42.0 [14.0–111]. | 0.96 |
| Any cryoprecipitatea | 449 | 152 (33.9) | 207 | 49 (23.7) | 153 | 71 (46.4) | 89 | 32 (36.0) | <.01 |
| Total blood product (mL/kg)a, c | 449 | 61.3 [29.0–123]. | 207 | 45.0 [21.0–85.3] | 153 | 98.4 [50.6–172]. | 89 | 55.8 [29.0–109]. | <.01 |
| Any crystalloid | 449 | 346 (77.1) | 207 | 163 (78.7) | 153 | 116 (75.8) | 89 | 67 (75.3) | 0.73 |
| Crystalloid (mL/kg)a, b | 449 | 18.2 [0.8–51.0] | 207 | 23.1 [2.0–59.2] | 153 | 10.0 [0.6–43.0] | 89 | 14.6 [0.2–45.6] | 0.06 |
| Any tranexamic acid | 449 | 36 (8.0) | 207 | 19 (9.2) | 153 | 9 (5.9) | 89 | 8 (9.0) | 0.49 |
| Any aminocaproic acida | 449 | 19 (4.2) | 207 | 4 (1.9) | 153 | 11 (7.2) | 89 | 4 (4.5) | <.05 |
| Any recombinant factor VIIaa, b | 447 | 56 (12.5) | 206 | 15 (7.3) | 153 | 26 (17.0) | 88 | 15 (17.0) | <.01 |
Abbreviations M median; N number; P probability; Q quartile.
P < 0.017 trauma v operative
P < 0.017 trauma v medical
P < 0.017 operative v medical
During MT, 92.0% received RBCs, 80.0% plasma and 66.1% platelets. No patients received whole blood. Time to administration of the first product varied by blood product type. Overall median time to first RBCs was 8 minutes (0–42). Median time to plasma administration was 34 minutes (15–77) and to platelet administration 42 minutes (15–102). Total blood volume administered was greater for operative events [98.4 mL/kg (50.6–172)] compared to trauma [45.0 mL/kg (21.0–85.3)] and medical [55.8 mL/kg (29.0–109)]. Volumes of crystalloid administered were higher for trauma [23.1 mL/kg (2.0–59.2)] than operative [10.0 mL/kg (0.63–43.0)] or medical events [14.6 mL/kg (0.19–45.6)]. Hemostatic adjuncts were infrequently used: Tranexamic Acid (8.0%), Aminocaproic Acid (4.2%) and Recombinant Factor VIIa (12.5%).
Post-event, there was stabilization of blood pressure and normalization of hemoglobin, base deficit, platelet count, international normalized ratio (INR), and fibrinogen (Table 5). While lactate concentrations were not routinely measured pre-event, they were mildly elevated post-event. Development of acute respiratory distress syndrome and acute kidney injury were common (20.3% and 18.5% respectively), however acute kidney injury disproportionately occurred in children with operative (28.8%) and medical bleeding (30.3%) as opposed to those with trauma (5.8%). Abdominal compartment syndrome was uncommon, occurring in only 1.8% of children.
Table 5.
Post life-threatening bleeding clinical measures and outcomes
| Overall (N = 449) | Trauma (N = 207) | Operative (N = 153) | Medical (N = 89) | ||||||
|---|---|---|---|---|---|---|---|---|---|
| Measure | Obs | M [Q1–Q3]. n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | Obs | M [Q1–Q3] n (%Obs) | P |
| Hemoglobin (g/dL) | 358 | 11.3 [9.6–13.2] | 161 | 11.3 [9.9–12.8] | 137 | 11.7 [9.8–13.7] | 60 | 10.5 [8.9–12.8] | 0.08 |
| Lactate (mmol/L)c | 253 | 3.5 [1.7–6.9] | 90 | 3.5 [2.0–5.9] | 110 | 3.0 [1.6–6.2] | 53 | 5.8 [2.2–12.5] | 0.02 |
| Base deficit|excess (mmol/L) | 329 | 1.0 [−3.5–5.0]. | 146 | 0.6 [−5.0–5.00]. | 122 | 2.0 [−1.0–4.7]. | 61 | 1.1 [−2.2–5.6]. | 0.27 |
| Fibrinogen (mg/dL) | 247 | 204 [160–272]. | 107 | 197 [157–251]. | 95 | 216 [161–293]. | 45 | 225 [175–327]. | 0.07 |
| Platelet count (x 109/L) | 330 | 142 [99.0–187] | 151 | 147 [111–193]. | 124 | 137 [85.0–187] | 55 | 124 [94.0–171] | 0.09 |
| International normalized ratio | 315 | 1.30 [1.18–1.50] | 145 | 1.30 [1.20–1.41] | 113 | 1.30 [1.10–1.50] | 57 | 1.40 [1.18–1.80] | 0.08 |
| Lowest systolic blood pressure (mmHg)a, c | 354 | 104 [83–123] | 162 | 114 [98–133] | 130 | 89.5 [73–107]. | 62 | 107 [88–126] | <.01 |
| Age-adjusted hypotensiona | 354 | 61 (17.2) | 162 | 18 (11.1) | 130 | 31 (23.8) | 62 | 12 (19.4) | 0.02 |
| Acute respiratory distress syndromea | 449 | 91 (20.3) | 207 | 31 (15.0) | 153 | 39 (25.5) | 89 | 21 (23.6) | 0.04 |
| Sepsisb | 449 | 44 (9.8) | 207 | 15 (7.2) | 153 | 14 (9.2) | 89 | 15 (16.9) | 0.04 |
| Acute kidney injurya, b | 449 | 83 (18.5) | 207 | 12 (5.8) | 153 | 44 (28.8) | 89 | 27 (30.3) | <.01 |
| Abdominal compartment syndrome | 449 | 8 (1.8) | 207 | 1 (0.5) | 153 | 4 (2.6) | 89 | 3 (3.4) | 0.14 |
| Died within 6 hoursa, c | 445 | 69 (15.5) | 205 | 35 (17.1) | 151 | 9 (6.0) | 89 | 25 (28.1) | <.01 |
| during massive transfusion | 69 | 49 (71.0) | 35 | 24 (68.6) | 9 | 6 (66.7) | 25 | 19 (76.0) | 0.78 |
| of hemorrhageb | 69 | 54 (78.3) | 35 | 22 (62.9) | 9 | 9 (100) | 25 | 23 (92.0) | 0.01 |
| of central nervous system injuryb, c | 69 | 13 (18.8) | 35 | 13 (37.1) | 9 | 0 (0.0) | 25 | 0 (0.0) | <.01 |
| Died within 24 hoursa, c | 445 | 99 (22.2) | 205 | 50 (24.4) | 151 | 17 (11.3) | 89 | 32 (36.0) | <.01 |
| during massive transfusion | 99 | 56 (56.6) | 50 | 28 (56.0) | 17 | 8 (47.1) | 32 | 20 (62.5) | 0.58 |
| of hemorrhagea, b | 99 | 71 (71.7) | 50 | 28 (56.0) | 17 | 16 (94.1) | 32 | 27 (84.4) | <.01 |
| of central nervous system injurya, b | 99 | 23 (23.2) | 50 | 21 (42.0) | 17 | 1 (5.9) | 32 | 1 (3.1) | <.01 |
| Died within 28 daysa, b, c | 445 | 168 (37.8) | 205 | 74 (36.1) | 151 | 36 (23.8) | 89 | 58 (65.2) | <.01 |
| during massive transfusion | 168 | 60 (35.7) | 74 | 30 (40.5) | 36 | 8 (22.2) | 58 | 22 (37.9) | 0.16 |
| of hemorrhage | 168 | 82 (48.8) | 74 | 30 (40.5) | 36 | 19 (52.8) | 58 | 33 (56.9) | 0.15 |
| of central nervous system injurya, b | 168 | 58 (34.5) | 74 | 43 (58.1) | 36 | 8 (22.2) | 58 | 7 (12.1) | <.01 |
Abbreviations M median; N number; P probability; Q quartile.
P < 0.017 trauma v operative
P < 0.017 trauma v medical
P < 0.017 operative v medical
Overall, 35.7% of deaths occurred during MT (trauma 40.5%, operative 22.2%, medical 37.9%) (Table 5). Operative and medical 6-hour mortality was almost exclusively from hemorrhage (100% and 92.0% respectively) whereas 62.9% of trauma deaths were due to hemorrhage and 37.1% due to central nervous system (CNS) injury. 24-hour mortality continued to be attributable to hemorrhage for children with operative (94.1%) and medical bleeding (84.4%) compared to children with trauma (hemorrhage 56%; CNS 42%). At 28 days, hemorrhage was the most common cause of death for those with operative (52.8%) and medical bleeding (56.9%). For children with trauma, CNS injury (58.1%) was the most common cause of death. Mortality was greatest among children with medical bleeding (Figure 1).
Discussion
In this multi-center prospective observational study of children with life-threatening bleeding, there was significant heterogeneity in demographics, pre-event clinical status, resuscitation management and outcomes related to the etiology of bleeding. Further, most children with life-threatening bleeding were managed under hospital MTPs, a practice that is grounded in adult trauma literature. Despite this, there were substantial time lags in the administration of blood components essential to hemostasis. Many children did not receive plasma and/or platelets as part of their resuscitation and the median time to both plasma and platelet administration was greater than thirty minutes in those that did. Death from hemorrhage, regardless of etiology, predominantly occurred within 6–24 hours of onset of MT.
Our analysis is unique from other retrospective studies of massive bleeding in children and large multicenter studies in adults.10,13–20 We report blood product and hemostatic agent use during the event as opposed to during a 6 or 24 hour period, or according to location of administration such as the emergency department or operating room.3 We also report blood product volume as mL/kg instead of in units which has more applicability to children than adults. This methodology allows for more precise examination of transfusion practice during actual bleeding events. This study also evaluated surgical and medical etiologies of hemorrhage managed under MTP whereas previous literature on MT in children has been limited to trauma.14,20–25 Clinical status at the onset of life-threatening bleeding and ultimate outcomes varied by etiology.
While trauma was the most common etiology for life-threatening bleeding in our cohort, a comparable proportion of children had surgical bleeding. Children with surgical bleeding were substantially younger than those with bleeding due to trauma and medical causes. This is likely due to the combined surgical risk of young age and the type of procedures performed in young children, in particular surgery for congenital deformities (e.g. craniofacial, heart and liver) and neoplasms (e.g. neuroblastoma, Wilms tumor).26–28 The effect of age-related maturation of the hemostatic system when considering approaches to life-threatening bleeding has not been thoroughly studied. Children less than one year of age are particularly vulnerable and have the most pronounced differences compared to older children and adults.22,29 Thus, infants are an important subpopulation in studies of life-threatening bleeding and including children with surgical bleeding in studies of DCR may be important to accomplishing this goal.
Children with medical bleeding had higher PRISM III scores. Pre-event data for the medical bleeding group are congruent with what we would expect for children who are experiencing primary or secondary failure of their hemostatic system, i.e. increased INR, and lower platelet counts. Many received plasma, platelets and cryoprecipitate prior to the onset of MT. Children with medical bleeding had high rates of end-organ injury and the highest death rates at all time points, with hemorrhage being the leading cause of death. MTPs in these children were likely activated later in their clinical course when goal directed therapy was not keeping pace with the rate of bleeding.30 However, use of MTP is understudied in this population, and ideal ratios of empiric transfusion unestablished.30,31
Consistent with the literature, the mortality among injured children with life-threatening bleeding was approximately 2–3 times that reported in adult populations.10–12 Prior to onset of MT, these children had lower median GCS and nearly 20% had fixed and dilated pupils. Death due to CNS injury was common within the first 24 hours and CNS injury was the leading cause of death at 28 days.32 Children with traumatic injuries were also less likely to receive platelets and plasma prior to MT and received greater volumes of crystalloid and lower volumes of plasma and platelets during the event. Prior research suggested children with life-threatening bleeding and without CNS injury have improved survival when component balanced transfusion is achieved.19 However, timely and balanced transfusion was not often achieved and high volumes of crystalloid were often used. This is surprising since the concepts of reducing crystalloid administration and balanced transfusion during MT were generated from the trauma community. Future analysis of our data will explore the impact of the ratio of blood components transfused, amount of crystalloid received and the presence of co-morbid CNS injury on outcomes for children with traumatic injuries.
The main purposes of MTPs are to expedite the availability of blood products and promote balanced transfusion under emergency circumstances. Despite the use of MTPs, many children in our cohort did not receive plasma (20.0%) or platelets (33.9%) and when they did, there were notable delays (> 30 minutes) from the onset of MTP. Timely balanced transfusion is a guiding principle of DCR and when attained for severely injured adult patients reduces death from hemorrhage 24 hours after injury.11 Evidence in injured adults indicates that for every minute delay there is a 5% increase in mortality.33 Reducing delays in product availability can be accomplished by several measures including bedside emergency blood refrigeration, thawed immediate release plasma, use of shelf stable freeze-dried plasma and rapid plasma thawing devices and procedures.34–43
Another option for assuring timely and balanced transfusion for life-threatening bleeding is the use of low titer group O whole blood (LTOWB). Ttransfusion of LTOWB compared to individual blood components has been independently associated with improved 24-hour and 28-day mortality while requiring lower total blood products in adult trauma populations.44,45 In children, LTOWB compared to blood components has been associated with improved time to normalization of base deficit and reduction in INR.46 It is critical that the use of LTOWB during MT be adequately studied in children before its use is widely adopted.
There are several limitations to this study All participating centers were tertiary children’s hospitals; thus, the patient population and clinical practice may not be reflective of other care settings. MTPs vary by center and it is unknown whether the observed practice was consistent with protocol. There may be patients included in the study who were unnecessarily transfused under MTP. As this was a prospective observational study, we had missing data for variables if a test was not performed as part of routine clinical care or documented in the medical records. Further, tests were not obtained at the same time points pre- and post- event. From the data and analyses presented, we cannot determine causality.
Conclusions
Pre-event clinical status and ultimate outcomes were distinctly different between children with traumatic, operative and medical bleeding. MTPs were used during life-threatening bleeding events, regardless of the cause of hemorrhage. Despite use of MTPs, there were substantial delays in the administration of plasma and platelets. Randomized trials in children are needed for each major etiology of life-threatening bleeding to determine which resuscitation practices reduce morbidity and mortality.
Supplementary Material
Supplemental Figure 1. Cumulative probability of survival differs by cause of bleeding. Date and time of death was reported for children as follows: 58/89 medical, 36/151 operative, and 74/205 trauma. Survival time for all children who did not die was 28 days. The survivor function was estimated using the Kaplan-Meier method.
Funding:
This study was funded by grant R21HL128863 from the National Heart, Lung and Blood Institute.
Copyright Form Disclosure:
Drs. Leonard, Nair, and Spinella’s institutions received funding from the National Heart, Lung, and Blood Institute (NHLBI). Drs. Leonard, Josephson, Davis, Fitzgerald, Muszynski, Steiner, Wills, and Spinella received support for article research from the National Institutes of Health (NIH). Dr. Josephson received funding from Immucor, LLC and Octapharma. Drs. Allen, Hewes, Rouse, and Steiner’s institutions received funding from the NIH grant administered by Washington University in St. Louis. Drs. Davis, Nair, and Steiner’s’ institution received funding from an NIH Exploratory/Developmental Research Grant Award (R21). Dr. Finkelstein received funding from Trauma In Kids Course via the Royal College of Physicians and Surgeons of Canada and the Pediatric Trauma Society from New York Presbyterian Weill Cornell, Carilion Clinic, and Texas Children’s Hospital, Society of Critical Care Medicine as a Faculty Instructor, GIBLIN, COMBS, SCHWARTZ, CUNNINGHAM, & SCARPA, LLC, Aaronson Rappaport Feinstein & Deutsch, LLP, and Ruprecht Hart Weeks & Ricciardulli, LLP; he disclosed that his wife has stock in Pfizer and Proctor & Gamble. Drs. Fitzgerald, Muszynski, and Spinella’s institutions received funding from the NIH. Drs. Johnson and Wills’ institutions institution received funding from Massive Transfusion Epidemiology and Outcomes in Children (MATIC) Study, 5R21HL128863-02 NHLBI. The remaining authors have disclosed that they do not have any potential conflicts of interest.
Footnotes
Affiliation at the time of study enrollment
Reprints: No reprints will be requested
Conflicts of Interest Disclosure: The authors have no conflicts of interest relevant to this article to disclose.
References
- 1.Holcomb JB, Jenkins D, Rhee P, et al. Damage control resuscitation: directly addressing the early coagulopathy of trauma. The Journal of trauma. 2007;62(2):307–310. [DOI] [PubMed] [Google Scholar]
- 2.Gurney JM, Spinella PC. Blood transfusion management in the severely bleeding military patient. Curr Opin Anaesthesiol. 2018;31(2):207–214. [DOI] [PubMed] [Google Scholar]
- 3.Spinella PC, Holcomb JB. Resuscitation and transfusion principles for traumatic hemorrhagic shock. Blood reviews. 2009;23(6):231–240. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Horst J, Leonard JC, Vogel A, Jacobs R, Spinella PC. A survey of US and Canadian hospitals’ paediatric massive transfusion protocol policies. Transfus Med. 2016;26(1):49–56. [DOI] [PubMed] [Google Scholar]
- 5.Camazine MN, Hemmila MR, Leonard JC, et al. Massive transfusion policies at trauma centers participating in the American College of Surgeons Trauma Quality Improvement Program. The journal of trauma and acute care surgery. 2015;78(6 Suppl 1):S48–53. [DOI] [PubMed] [Google Scholar]
- 6.Cure P, Bembea M, Chou S, et al. 2016 proceedings of the National Heart, Lung, and Blood Institute’s scientific priorities in pediatric transfusion medicine. Transfusion. 2017;57(6):1568–1581. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Neff LP, Cannon JW, Morrison JJ, Edwards MJ, Spinella PC, Borgman MA. Clearly defining pediatric massive transfusion: cutting through the fog and friction with combat data. The journal of trauma and acute care surgery. 2015;78(1):22–28; discussion 28–29. [DOI] [PubMed] [Google Scholar]
- 8.Kuczmarski RJ, Ogden CL, Grummer-Strawn LM, et al. CDC growth charts: United States. Adv Data. 2000(314):1–27. [PubMed] [Google Scholar]
- 9.Pollack MM, Patel KM, Ruttimann U, Cuerdon T. Frequency of variable measurement in 16 pediatric intensive care units: influence on accuracy and potential for bias in severity of illness assessment. Crit Care Med. 1996;24(1):74–77. [DOI] [PubMed] [Google Scholar]
- 10.Shroyer MC, Griffin RL, Mortellaro VE, Russell RT. Massive transfusion in pediatric trauma: analysis of the National Trauma Databank. J Surg Res. 2017;208:166–172. [DOI] [PubMed] [Google Scholar]
- 11.Holcomb JB. Pragmatic, Randomized Optimal Platelets and Plasma Ratios (PROPPR). 2013; http://clinicaltrials.gov/ct2/show/NCT01545232?term=PROPPR&rank=1, 2013.
- 12.Holcomb JB, del Junco DJ, Fox EE, et al. The prospective, observational, multicenter, major trauma transfusion (PROMMTT) study: comparative effectiveness of a time-varying treatment with competing risks. JAMA surgery. 2013;148(2):127–136. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Nosanov L, Inaba K, Okoye O, et al. The impact of blood product ratios in massively transfused pediatric trauma patients. American journal of surgery. 2013;206(5):655–660. [DOI] [PubMed] [Google Scholar]
- 14.Butler EK, Mills BM, Arbabi S, et al. Association of Blood Component Ratios With 24-Hour Mortality in Injured Children Receiving Massive Transfusion. Crit Care Med. 2019;47(7):975–983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Borgman MA, Spinella PC, Perkins JG, et al. The ratio of blood products transfused affects mortality in patients receiving massive transfusions at a combat support hospital. The Journal of trauma. 2007;63(4):805–813. [DOI] [PubMed] [Google Scholar]
- 16.O’Keeffe T, Refaai M, Tchorz K, Forestner JE, Sarode R. A massive transfusion protocol to decrease blood component use and costs. Archives of surgery. 2008;143(7):686–690; discussion 690–681. [DOI] [PubMed] [Google Scholar]
- 17.Cotton BA, Au BK, Nunez TC, Gunter OL, Robertson AM, Young PP. Predefined massive transfusion protocols are associated with a reduction in organ failure and postinjury complications. The Journal of trauma. 2009;66(1):41–48; discussion 48–49. [DOI] [PubMed] [Google Scholar]
- 18.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. The Journal of trauma. 2009;66(6):1616–1624. [DOI] [PubMed] [Google Scholar]
- 19.Hwu RS, Keller MS, Spinella PC, Baker D, Tao Y, Leonard JC. Potential effects of high plasma to red blood cell ratio transfusion in pediatric trauma. Trauma. 2017;19(1):21–27. [Google Scholar]
- 20.Cunningham ME, Rosenfeld EH, Zhu H, Naik-Mathuria BJ, Russell RT, Vogel AM. A High Ratio of Plasma: RBC Improves Survival in Massively Transfused Injured Children. J Surg Res. 2019;233:213–220. [DOI] [PubMed] [Google Scholar]
- 21.Murphy CH, Spain DA, Shan H. Coagulopathy and transfusion ratios in pediatric trauma. The journal of trauma and acute care surgery. 2020;88(5):648–653. [DOI] [PubMed] [Google Scholar]
- 22.Toulon P, Berruyer M, Brionne-Francois M, et al. Age dependency for coagulation parameters in paediatric populations. Results of a multicentre study aimed at defining the age-specific reference ranges. Thromb Haemost. 2016;116(1):9–16. [DOI] [PubMed] [Google Scholar]
- 23.Hendrickson JE, Shaz BH, Pereira G, et al. Implementation of a pediatric trauma massive transfusion protocol: one institution’s experience. Transfusion. 2012;52(6):1228–1236. [DOI] [PubMed] [Google Scholar]
- 24.Chidester SJ, Williams N, Wang W, Groner JI. A pediatric massive transfusion protocol. The journal of trauma and acute care surgery. 2012;73(5):1273–1277. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hwu RS, Spinella PC, Keller MS, Baker D, Wallendorf M, Leonard JC. The effect of massive transfusion protocol implementation on pediatric trauma care. Transfusion. 2016;56(11):2712–2719. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Nasr VG, DiNardo JA, Faraoni D. Development of a Pediatric Risk Assessment Score to Predict Perioperative Mortality in Children Undergoing Noncardiac Surgery. Anesth Analg. 2017;124(5):1514–1519. [DOI] [PubMed] [Google Scholar]
- 27.Valencia E, Staffa SJ, Faraoni D, DiNardo JA, Nasr VG. Prospective External Validation of the Pediatric Risk Assessment Score in Predicting Perioperative Mortality in Children Undergoing Noncardiac Surgery. Anesth Analg. 2019;129(4):1014–1020. [DOI] [PubMed] [Google Scholar]
- 28.Nasr VG, Staffa SJ, Zurakowski D, DiNardo JA, Faraoni D. Pediatric Risk Stratification Is Improved by Integrating Both Patient Comorbidities and Intrinsic Surgical Risk. Anesthesiology. 2019;130(6):971–980. [DOI] [PubMed] [Google Scholar]
- 29.Appel IM, Grimminck B, Geerts J, Stigter R, Cnossen MH, Beishuizen A. Age dependency of coagulation parameters during childhood and puberty. Journal of thrombosis and haemostasis : JTH. 2012;10(11):2254–2263. [DOI] [PubMed] [Google Scholar]
- 30.Karam O, Russell RT, Stricker P, et al. Recommendations on RBC Transfusion in Critically Ill Children With Nonlife-Threatening Bleeding or Hemorrhagic Shock From the Pediatric Critical Care Transfusion and Anemia Expertise Initiative. Pediatric critical care medicine : a journal of the Society of Critical Care Medicine and the World Federation of Pediatric Intensive and Critical Care Societies. 2018;19(9S Suppl 1):S127–S132. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Karam O, Tucci M. Massive Transfusion in Children. Transfus Med Rev. 2016;30(4):213–216. [DOI] [PubMed] [Google Scholar]
- 32.Pfeifer R, Tarkin IS, Rocos B, Pape HC. Patterns of mortality and causes of death in polytrauma patients--has anything changed? Injury. 2009;40(9):907–911. [DOI] [PubMed] [Google Scholar]
- 33.Meyer DE, Vincent LE, Fox EE, et al. Every minute counts: Time to delivery of initial massive transfusion cooler and its impact on mortality. The journal of trauma and acute care surgery. 2017;83(1):19–24. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Kuta P, Melling N, Zimmermann R, Achenbach S, Eckstein R, Strobel J. Clotting factor activity in fresh frozen plasma after thawing with a new radio wave thawing device. Transfusion. 2019;59(5):1857–1861. [DOI] [PubMed] [Google Scholar]
- 35.Kuta P, Hauck-Dlimi B, Strobel J, Zimmermann R, Eckstein R. Quality of Clotting Factor Activity in Fresh Frozen Plasma at Thaw with a Microwave System and after Storage at 4 degrees C for 48 Hours. Clin Lab. 2016;62(6):987–991. [DOI] [PubMed] [Google Scholar]
- 36.Martinaud C, Ausset S, Deshayes AV, Cauet A, Demazeau N, Sailliol A. Use of freeze-dried plasma in French intensive care unit in Afghanistan. The Journal of trauma. 2011;71(6):1761–1764; discussion 1764–1765. [DOI] [PubMed] [Google Scholar]
- 37.von Heymann C, Keller MK, Spies C, et al. Activity of clotting factors in fresh-frozen plasma during storage at 4 degrees C over 6 days. Transfusion. 2009;49(5):913–920. [DOI] [PubMed] [Google Scholar]
- 38.Bostrom F, Ekemar L, Olsson D, Egberg N, Lundahl J. Rapid thawing of fresh-frozen plasma with radio wave-based thawing technology and effects on coagulation factors during prolonged storage at 4 degrees C. Vox Sang. 2009;97(1):34–38. [DOI] [PubMed] [Google Scholar]
- 39.Churchill WH, Schmidt B, Lindsey J, Greenberg M, Boudrow S, Brugnara C. Thawing fresh frozen plasma in a microwave oven. A comparison with thawing in a 37 degrees C waterbath. Am J Clin Pathol. 1992;97(2):227–232. [DOI] [PubMed] [Google Scholar]
- 40.Rock G, Tackaberry ES, Dunn JG, Kashyap S. Rapid controlled thawing of fresh-frozen plasma in a modified microwave oven. Transfusion. 1984;24(1):60–65. [DOI] [PubMed] [Google Scholar]
- 41.Westphal RG, Tindle B, Howard PL, Golden EA, Page GA. Rapid thawing of fresh frozen plasma. Am J Clin Pathol. 1982;78(2):220–222. [DOI] [PubMed] [Google Scholar]
- 42.Harris C, Totten M, Davenport D, et al. Experience with uncrossmatched blood refrigerator in emergency department. Trauma Surgery & Acute Care Open. 2018;3:e000184. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Dutton RP, Shih D, Edelman BB, Hess J, Scalea TM. Safety of uncrossmatched type-O red cells for resuscitation from hemorrhagic shock. The Journal of trauma. 2005;59(6):1445–1449. [DOI] [PubMed] [Google Scholar]
- 44.Williams J, Merutka N, Meyer D, et al. Safety profile and impact of low-titer group O whole blood for emergency use in trauma. The journal of trauma and acute care surgery. 2020;88(1):87–93. [DOI] [PubMed] [Google Scholar]
- 45.Shea SM, Staudt AM, Thomas KA, et al. The use of low-titer group O whole blood is independently associated with improved survival compared to component therapy in adults with severe traumatic hemorrhage. Transfusion. 2020;60 Suppl 3:S2–S9. [DOI] [PubMed] [Google Scholar]
- 46.Leeper CM, Yazer MH, Triulzi DJ, Neal MD, Gaines BA. Whole Blood is Superior to Component Transfusion for Injured Children: A Propensity Matched Analysis. Ann Surg. 2020;272(4):590–594. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplemental Figure 1. Cumulative probability of survival differs by cause of bleeding. Date and time of death was reported for children as follows: 58/89 medical, 36/151 operative, and 74/205 trauma. Survival time for all children who did not die was 28 days. The survivor function was estimated using the Kaplan-Meier method.