A Systematic Review of Hospital Trauma Team Activation Criteria for Children

Abstract

Objectives:

Hospital trauma activation criteria are intended to identify children who are likely to require aggressive resuscitation or specific surgical interventions that are time sensitive and require the resources of a trauma team at the bedside. Evidence to support criteria is limited, and no prior publication has provided historical or current perspectives on hospital practices toward informing best practice. This study aimed to describe the published variation in (1) highest level of hospital trauma team activation criteria for pediatric patients and (2) hospital trauma team membership and (3) compare these finding to the current ACS recommendations.

Methods:

Using an Ovid MEDLINE In-Process & Other Non-Indexed Citations search, any published description of hospital trauma team activation criteria for children that used information captured in the prehospital setting was identified. Only studies of children were included. If the study included both adults and children, it was included if the number of children assessed with the criteria was included.

Results:

Eighteen studies spanning 20 years and 13,184 children were included. Hospital trauma team activation and trauma team membership were variable. Nearly all (92%) of the trauma criteria used physiologic factors. Penetrating trauma (83%) was frequently included in the trauma team activation criteria. Mechanisms of injury (52%) were least likely to be included in the highest level of activation. No predictable pattern of criterion adoption was found. Only 2 of the published criteria and 1 of published trauma team membership are consistent with the current American College of Surgeons recommendations.

Conclusions:

Published hospital trauma team activation criteria and trauma team membership for children were variable. Future prospective studies are needed to define the optimal hospital trauma team activation criteria and trauma team membership and assess its impact on improving outcomes for children.

When children are transported to a trauma center, emergency medical services providers communicate the condition of the child to the emergency department (ED) staff before arrival. That information is used to determine whether the trauma team should be activated so that they are present in the ED when the patient arrives. Hospital trauma activation criteria are intended to identify children who are likely to require aggressive resuscitation or specific surgical interventions that are time sensitive and require the resources of a trauma team at the bedside. Presence of the trauma team has been shown to decrease mortality by 25% to 30% for the seriously injured child.¹ Overuse of the trauma team, however, can lead to excess costs and resource use as well as provider fatigue. Indeed activating the trauma team only when a critically injured patient is identified has been shown to reduce costs in adult and pediatric trauma patients.^2,3

The American College of Surgeons (ACS) Committee on Trauma described a tiered trauma team activation that is based on predetermined prehospital criteria.⁴ These criteria are distinct from the prehospital triage criteria that are used to identify patients that need the services of a trauma center. The ACS proposes 6 criteria that serve as the minimum criteria that should be used for the highest level trauma team activation, as follows:

• Age-specific hypotension: systolic blood pressure less than 70 mm Hg + (2 × age in years);

• Gunshot wound to the neck, chest, or abdomen or extremities proximal to the elbow/knee;

• Glasgow Coma Scale less than 9 or deteriorating by 2, with mechanism attributed to trauma;

• Transfer patients from other hospitals receiving blood to maintain vital signs;

• Intubated patients transferred from the scene or patients who have respiratory compromise or are in need of an emergency airway or intubated patients who are transferred from another facility with ongoing respiratory compromise;

• Discretion of the emergency physician;

Trauma centers can expand on these criteria to include any of the prehospital information that is available. This typically includes physiologic, anatomic, and mechanism of injury criteria. There is no referenced evidence to support the ACS recommendations, although they parallel in content the minimum criteria for adults.

Trauma team members that usually respond to the activation are also outlined by the ACS.⁴

• General surgeon

• Emergency physician

• Surgical and emergency residents

• Emergency department nurses

• Laboratory technician

• Radiology technologist

• Critical care nurse

• Anesthesiologist or certified registered nurse anesthetist

• Operating room nurse

• Security officers

• Chaplain or social worker

• Scribe

The rationale for this systematic review is to evaluate the published evidence and compare it with the recommended criteria and team composition for the pediatric trauma patient. This synthesis of data is an incremental step toward understanding current practice, identifying variability, and optimizing hospital trauma activation criteria for children. The objective of this systematic review was to describe for pediatric patients (1) the variation in highest level of hospital trauma team activation criteria and (2) trauma team membership. Lastly, we sought to (3) compare these finding with the current ACS recommendations.

METHODS

Studies were identified with a search of Ovid MEDLINE In-Process & Other Non-Indexed Citations, Ovid MEDLINE Daily, and Ovid MEDLINE using search terms outlined in Figure 1, including only human trials without language restriction. Trauma/emergency service and trauma/emergency team were the terms chosen to search broadly for any publications on the topic, and then, the search was focused to include all children aged 0 to 18 years. A published protocol does not exist for this systematic review. All databases were searched from inception until January 3, 2015. To be included in this review, studies had to describe hospital trauma team activation criteria for pediatric patients that were assessed in the prehospital setting. Studies that included both children and adults were included if trauma activation criteria and number of children evaluated were included separately. Studies that included only prehospital triage criteria that are used to identify patients that need the services of a trauma center were not included. Studies that included only a specific population (ie, winter sports injuries) or mechanism of injury (ie, horse-related injuries) or simply interhospital transfers were excluded. Reference lists from studies included in the analyses were reviewed by hand to identify additional articles meeting criteria, and no additional articles were identified.

FIGURE 1.

Search strategy.

Data were abstracted by a single author (A.D.). Patient characteristics, trauma activation criteria, and trauma team membership were abstracted from each publication. These were reviewed and specific criteria and trauma team members that were shared between publications were used to populate the rows of the tables. The data were then abstracted a second time from each of the publications directly into the tables. Finally, the data in the table were then compared with the published manuscript on a third review to assure accuracy. Patient characteristics included dates of inclusion, patient age, number of trauma patients overall during study enrollment, number transferred to the operating room (variable definitions of emergent operative cases), deaths (including ED deaths, when available), and number of trauma activations. Trauma activation criteria were divided into categories for ease of reporting: physiologic, penetrating, anatomic, mechanism of injury, and other criteria. Criteria were assigned to each of these 5 categories based on their most frequent assignment in the published literature. If the study specifically excluded a population (eg, penetrating injuries or drowning) this was designated as “*” in the table, and the proportion of criteria with that criterion was calculated using only the criteria that included the population. This review is focused on the highest level (most severely injured) of trauma team activation. When the publication included graded or tiered response activation, this was noted in Table 1, but only the highest-level criteria were reported in Table 2. If 2 different criteria were compared, then both criteria were included. The current ACS recommendations were also included for comparison. The membership for the trauma team activated for the highest level activation was also abstracted, and the current ACS recommendations included for comparison.

TABLE 1.

Studies Included in the Systematic Review

Author	Journal	Publication Year	Enrollment Years	Age	Total Trauma Patients	Operating Room	Deaths (ED deaths)	Trauma Activations	Name
Sola, JE	Journal of Pediatric Surgery	1994	1/19/91–2/6/92	0–14		26	15	216	Baltimore "Alpha"
						5	0	736	Baltimore "Bravo"
DeKeyser	Annals of Emergency Medicine	1994	1991					214	Fairfax Blue
DeKeyser	Annals of Emergency Medicine	1994	1991					604	Fairfax Yellow
Quazi	Academic Emergency Medicine	1998	7/12/96–2/28/97	0–14	192			12	Akron Paramedic
Quazi, K	Journal of Trauma	1998	7/1/93–7/31/94	0–14		0	0	194	Akron Mechanism
Vernon, DD	Pediatrics	1999			638	8	7	49	Salt Lake City "Trauma One"
Dowd, MD	Academic Emergency Medicine	2000	1/1/96–12/31/96	0–15	492	32	21	179	Cincinnati Trauma Team Activation
Nuss, KE	Pediatric Emergency Care	2001	1/1/95–12/31/96		542	26	39 (6 ED)	253	Columbus Level 1
						22	1 (0 ED)	289	Columbus Level 2
Chen, LE	Pediatric Emergency Care	2004	1994–1999	0–19	5122	44	(15 ED)	220	St Louis Trauma Stat
						40	(0 ED)	250	St Louis Trauma Minor
Simon, B	Pediatric Emergency Care	2004	1990–2001	0–13	1112	21	6	461	mPTS
Edil, BH	Journal of Pediatric Surgery	2005	1/2000–1/2004	0–13	118	4	18	40	Oklahoma Level 1 Alert
Steele, R	Annals of Emergency Medicine	2006	1/1/97–6/30/04	0–14	2311	8/62/121		87/166/836	Loma Linda Peds
Steele, R	Annals of Emergency Medicine	2007	1/1/97–6/30/04	0–14	2311	8/62/121		534	ACS Criteria 2002
Nasr, A	Journal of Trauma	2007	1993–2003	0–15	3579			628	Toronto mPTS
Bevan, C	Journal of Trauma	2009	4/2004–4/2006	0–15	~4400		6	192	Melbourne #1
							7	82	Melbourne #2
Kouzminova	Journal of Trauma	2009	1998–2007						Santa Clara Major
Kouzminova	Journal of Trauma	2009	1998–2007	*					Santa Clara Minor
Mukherjee, K	Journal of Pediatric Surgery	2010	1 y	0–14	152	18	10	98	Vanderbilt (physiologic focused)
								25	Vanderbilt (standard)
Davis	Injury, International J. Care Injured	2010	2008	*		57	18	468/676	South Wales
Williams, D	Journal of Pediatric Surgery	2011	2006–2008	0–14	4502	2118	48	211	UT Southwestern Stat
							29	1104	UT Southwestern Alert
Bozeman, AP	The American Journal of Surgery	2012	1/1/2009–2/10/2011	0–14	271			176	Arkansas Restraints
Krieger, AR	Journal of Trauma Acute Care Surgery	2012	1/2006–3/2009	0–18				2538	St. Louis Traum Stat MOI
			4/2009–6/2010					1088	St. Louis Trauma Major ABP
Falcone, RA	Journal of Trauma Acute Care Surgery	2012	1 y	0–20	9722	144	72 (24 ED)	656	Multicenter
Nabaweesi, R	Pediatric Surgery International	2014	1/1/2008–12/31/2011	0–14	2235	37	7	369	Arkansas FTTA
								1766	Arkansas PTTA

TABLE 2.

Published Hospital Trauma Activation Criteria for Children

	Author and Pub Year	Sola 1994	Quazi 1998		Vernon 1999	Dowd 2000	Nuss 2001	Chen 2004		Simon 2004			Edil 2005	Steele 2006	Steele 2007	Nasr 2007		Bevan 2009		Mukherjee 2010		Williams 2011	Bozeman 2012	Krieger 2012		Falcone 2012	Nabaweesi 2014	% of Criteria	ACS Criteria
	Criteria type (if multiple)		Physio	Anat						PTS	mPTS	Criteria			ACS 2002		mPTS	1 of 3	single		Physio			mechanism	anat/physio
Physiologic criteria	Cardiopulmonary arrest																					penetrating trauma						28
	Systolic Blood Pressure (SBP)	<80	age-specific		age-specific	<80	hypotension	<80		<90		age-specific			age-specific	<80	<90	hypotension		hypotension	hypotension			hypotension	age-specific	age-specific	age-specific	72
	Heart rate	>130 <50	age-specific				tachcardia	age-specific				age-specific		age-specific						tachcardia	tachcardia			tachcardia		tachcardia	age-specific	44
	Pulse quality							weak pulse		2							2							weak/absent				16
	Perfusion quality					CRF>3 sec	"poor"	cool, clammy, CRF	>2			CRF >2 sec												cool, dammy, CRF	cyanosis			24
	Clinical signs of shock																											32
	Blood transfusion																											12
	IV fluid bolus																									>40 cc/kg		8
	Ongoing blood loss																							or sig blood loss				8
	Resp compromise/ distress									airway integrity	airway integrity						airway integrity	airway compromise		unstable airway	unstable airway			grunt/retract				72
	Intubation							with blood																				52
	Airway foreign body / obstruction																											24
	Stridor																											20
	Respiratory rate	>30, <20	age-specific			>30						age-specific		age-specific		<10>50				tachypnea							age-specific	32
	Glasgow Coma Scale	<9				<8	<9	<9				<14		<9	<8	<8		<8		<8	<8	<8		<13	<13	<8	<9	68
	Suspected spinal cord injury					w paralysis	cervical																					44
	Paralysis																											24
	Focal neuro deficit																											8
	Deteriorating level of consciousness																									GCS by 2		20
	Agitated or confused or altered																											24
	Loss of consciousness							>5 min	<5 min			>5 min								>5 min				>5 min				24
	Any physiologic Criteria																											92
Penetrating	GSW		*	*			Penetrating injury			*	*			Penetrating mechanism			*						*				*	16
	Penetrating injury to extremity		*	*						*	*						*						*				*	27
	Penetrating injury to abdomen		*	*						*	*						*			or pelvis	or pelvix	or pelvis	*			+GSW, BB	*	66
	Penetrating injury to neck		*	*						*	*						*						*			+GSW, BB	*	77
	Penetrating injury to head		*	*				and face		*	*						*						*			+GSW, BB	*	77
	Penetrating injury to torso		*	*						*	*						*						*				*	27
	Penetrating injury to trunk		*	*						*	*						*						*				*	22
	Penetrating injury to the thorax		*	*						*	*						*						*				*	22
	Penetrating injury to groin		*	*						*	*						*						*				*	38
	Penetrating injury to chest		*	*						*	*						*						*			+GSW. BB	*	50
	Any penetrating criteria		*	*						*	*												*				*	83
Anatomic Criteria	Maxlllo/Faclal/Tracheal Injury																											24
	Open or depressed skull fx																											12
	CFS leak																											12
	Flail chest																											24
	Open chest wound																											8
	Pneumothorax																			open				or hemo	or hemo		or hemo	20
	Cardiac/great vessel injury																											4
	Crush of torso																			crush							crush	12
	Crush of chest or abdomen																											8
	Peritonitis																											4
	Burns; total				*	>30% BSA	>30% BSA	>20% BSA				>10% BSA					*			>20% BSA				>20 BSA	>40% TBSA	>30% BSA	*	41
	Burn; 2nd and 3rd degree				*	>15% BSA											*		burns >40%						>15% BSA		*	14
	Suspected airway injury from burn				*												*								*		*	41
	Long bone fractures			2+		2+	2+*	2+				2+							upper/lower	multile				2+				32
	Pelvic fracture/injury																			with tachycardia and hypotension								20
	Skeletal integrety																											12
	Axial skeleton fracture																											8
	Arterial bleeding																											8
	Vascular injury																											4
	Pulseless extremity																											16
	Amputation or degloving																			proximal	proximal	or crush proximal				proximal		56
	2 systems injured																											16
	Blunt abdominal trauma						seatbelt sign												abd injury									8
	Injury above and below diaphragm																											4
	Severe head or neck injury																	midline shift	scalp wound	intracranial injury GCS<14				intracranial bleed				20
	Severe chest or abdominal injury																			with tachycardia and hypotension								12
	subq emphysema
	Open wound																											12
	Any anatomic criteria																											76
Mechanism Criteria	MVC unrestrained																											8
	MVC with ejection																											36
	MVC with prolonged extrication					> 20 min		>20 min								>20 min												12
	MVC with roll-over															>70 mph												20
	MVC with death in same car																											32
	MVC >20 mph					unrestrained																						8
	MVC >40 mph					restrained																						8
	Run-over or dragged by car															or thrown												12
	Major auto deformity >50 cm																											4
	MVC intrusion					>20 in		>12 in								>30 cm							>18 inches					20
	MPC/MBC	thrown 15 ft		10 mph		>20 mph		>20 mph				>20 mph				10 km/hr/30 km/hr			including horse									28
	Drowning				*																				*	*		5
	Fall			>20 ft		>20 ft		>10 ft				3x height				3x height			>3m	>20 ft								28
	Lightening strike / Electric shock																							>200 v	>200 v			20
	Trauma In a hostile environment																											8
	Hanging or strangulation																								*	*		4
	High velocity impact																		explosion									12
	Any mechanism criteria																											52
Other	Multiple simultaneous victims																											4
	Trauma score <12																											4
	Child abuse with viral signs change																								*			4
	Ambulance discretion																											4
	Trauma transfer with criteria																										*	12
	MD discretion																											16

^*Patients with these findings were explicitly excluded.

RESULTS

The search identified 246 articles. Two review authors (A.D. and M.G.) independently screened the titles and abstracts of all citations identified by the search (Fig. 2). Full articles were obtained when they seemed to meet the inclusion criteria or there were insufficient data in the title and abstract to make a clear decision for inclusion. The inclusion of the studies was made after reviewing the full-text articles. Agreement between reviewers was excellent for inclusion of articles, with a Kappa coefficient of 0.91 (0.81, 1.00). Disagreements were resolved between the 2 review authors about study inclusion by discussion and consensus.

FIGURE 2.

Study selection.

Application of the inclusion and exclusion criteria resulted in 18 studies and 25 criteria that were included in our analysis (Fig. 1).^2,5-21 Table 1 shows the included studies and criteria. All but 1 were retrospective cohort studies. The only prospective study empirically derived pediatric trauma activation criteria.¹³ Many (39%) described 2-tiered trauma team activation criteria.^{5,8-10,16,17,21} A number of studies evaluated the accuracy of criteria categories including mechanism^6,8,18,19 and physiologic criteria² as well as simplified criteria including only respiratory compromise and intubation.¹² Two studies evaluated variations of the Pediatric Trauma Score,^11,15 and 2 studies applied the current ACS criteria to the trauma population.^14,20 The included articles spanned 20 years, and 13,184 children meeting the trauma activation criteria were included. Most studies reported the number of children that underwent surgery and number of deaths (Table 1), but this inconsistent reporting limits the ability to compare injury severity.

Physiologic factors reported for trauma team activation generally included assessments for airway integrity, respiratory distress or compromise, signs of shock or resuscitation requirements, and the Glasgow Coma Scale. Nearly all (92%) of the trauma criteria used physiologic factors (Table 2), and most included respiratory compromise (72%), although this was inconsistently defined between studies. Specifically, an intervention for airway compromise like intubation (52%) was reported more often than respiratory rate (32%), but intervention for shock including fluid bolus (8%) and blood transfusion (12%) was less frequent than heart rate (44%) or blood pressure (72%). Cardiopulmonary arrest was the designated criterion for only 28% of trauma activation criteria. Most activation criteria included Glasgow Coma Scale (68%), but the threshold score was variably defined between 8 and 14.

Penetrating trauma (83%) was frequently included in the trauma team activation criteria (Table 2). For penetrating trauma, the site of injury was inconsistent between studies including head (77%), neck (77%), abdomen (66%), chest (50%), groin (38%), and extremity (27%). Anatomic factors (76%) were commonly included, but there was significant variability in the factors included, with the most common criteria being amputation or degloving (56%), burns or suspected airway injury from burns (41%), long bone fractures (32%), facial/tracheal injury (24%), and flail chest (24%). Some published criteria depended heavily on anatomic factors to identify trauma team activation criteria including 14 specific injuries,² but 6 criteria included no anatomic factors at all.^6,7,12-14,18

Mechanisms of injury (52%) were least likely to be included in the highest level of activation (Table 2). The most frequently included criteria were motor vehicle crash with ejection (36%), death in the same car (32%), and falls (28%). A number of unique trauma team activation criteria were included that did not meet the standard categories and are also included in Table 2. One of these, emergency physician's discretion, is now recommended by the ACS but was infrequently included (16%).

Of the studies included in this systematic review, 12 studies reported their trauma team composition (Table 3). The average number of trauma team members was 14 (range, 6–19). All activation criteria included a surgeon, although the level of training required was variable, and most included an ED physician (92%).

TABLE 3.

Trauma Team Composition

Author and Publication Year	Sola 1994	Quazi 1998	Vernon 1999	Dowd 2000	Nuss 2001	Chen 2004	Simon 2004	Nasr 2007	Bevan 2009	Mukheijee 2010	Williams 2011	Nabaweesi 2014	ACS 2014
Surgery attending
Surgery fellow
Surgery Sr resident
Surgery Jr resident
PICU attending
PICU fellow
PICU resident							senior
ED attending					or				or
PEM fellow
EM resident					or		senior
Pediatric resident
Pediatrics
Anesthesiologist
Anesthesiologist resident
Peds Neurosurgery
Orthopeds
RN supervisor
OR Charge RN/RN
OR personnel/team
PICU RN
PICU Charge RN
ER RN		2		2		2	2			2
Float RN
Phlebotomist
Respiratory therapist
Radiology tech
Radiologist
Pharmacist
Distribution
Chaplain / Patoral Care													or
Social Services
Security
Orderly
Blood products / Blood Bank
Administrative supervisor
Trauma nurse coordinator
Patient service attendant
Scribe
Number of participants	9	12	12	18	13	19	7	8	9	18	18 *team members not identified	6	13

Current ACS guidelines recommend only physiologic factors. Only 2 of the published pediatric trauma activation criteria in this review included all elements.^14,20 The currently recommended ACS trauma team member response for highest level activations for a severely injured patient include 12 people.⁴ None of the published team compositions included all of the ACS-recommended members.

DISCUSSION

The published literature about hospital trauma team activation for pediatric patients shows a significant amount of variability in the criteria used and the team that is commenced. This historical perspective provides the opportunity to evaluate the evolution of trauma team activation criteria during the last 20 years, but this evidence shows no predictable pattern of criterion adoption or any data to support the current criteria. Nor any evidence to support the value of these criteria in identifying children who require the activation of a trauma team. Prospective research is needed to evaluate the criteria used for trauma team activation for children.

The initial assessment and evaluation of severely injured trauma patients begins with the prehospital assessment and care. This information is used for both destination decision-making as well as hospital trauma team activation. The cost of missing a severely injured child during destination decision-making can be significant because of hours of delay for the child to be evaluated at a lower level or nontrauma center and then for the child to be transported to a higher level trauma center. In fact, the typical goal is for the transfer to occur in 2 to 3 hours. In contrast, if the hospital trauma team is not activated for a severely injured child being transported to a trauma center, the delay will be much shorter. However, for some critically injured patients, even a short delay will affect morbidity and mortality. This suggests the tradeoff between sensitivity and specificity, for the 2 criteria can be different. For destination decision-making, a goal of 5% undertriage and 30% to 50% overtriage is often quoted, but for hospital trauma activation, it might be reasonable to increase the rate of undertriage to decrease the overtriage rate.²² This would lead to a better use of resources and likely a minimal change in morbidity and mortality.

Trauma centers have published their experiences evaluating potential criteria for trauma team activation because most centers have developed their criteria based on experiences at their center. Historically, most trauma centers included more anatomic and mechanistic factors. This review did not find a reduced number of criteria over time as one might expect with performance review. This systematic review found that pediatric hospital trauma team activation criteria used are consistently inconsistent, not based in empiric evidence, and not consistent with the most recent published guidelines.

The evidence to support each criterion is lacking based on the findings of this systematic review. Only 1 study evaluated each factor (as opposed to full criteria), and the same study was the only prospectively performed study.¹³ Future investigations should include a rigorous evaluation of each factor with the determination of the incremental improvement to the undertriage and overtriage rates for injured children. Most of the included studies did evaluate the undertriage and overtriage rates for the overall trauma activation criteria. However, there was no consistency in the outcome used to define the need for a trauma center, making comparisons impossible in this study. Multidisciplinary performance improvement to refine the undertriage and overtriage rates and the appropriateness of prehospital-based trauma team activations has been recommended. A recent publication of that consensus statement should lead to stronger evidence to support the selection of criteria.²³

This study has limitations. Synthesis of results by combining studies to assess the accuracy of the hospital trauma team activation criteria was not possible because of the inconsistent use of outcomes to assess the criteria. The published literature may not represent clinical practice and may not provide accurate insights into historical practices. However, many institutions rely on the published literature to inform system changes, and this review provides an excellent summary of that information.

In conclusion, published hospital trauma team activation criteria and trauma team membership for children were historically variable. No predictable pattern of criterion adoption was found. The published criteria are also dissimilar from the current ACS-recommended criteria. Future prospective studies are needed to define the optimal hospital trauma team activation criteria and trauma team membership and assess its impact on improving outcomes for children.