Trauma team activation: simplified criteria safely reduces overtriage

The American Journal of Surgery 193 (2007) 630 – 635

Scientific paper

Trauma team activation: simplified criteria safely reduces overtriage Ryan K. Lehmann, D.O.*, Zachary M. Arthurs, M.D., Daniel G. Cuadrado, M.D., Linda E. Casey, R.N., Alec C. Beekley, M.D., Matthew J. Martin, M.D. Department of Surgery, Madigan Army Medical Center, 9040-A Fitzsimmons Ave., Tacoma, WA 98431-1100, USA Manuscript received December 6, 2006; revised manuscript January 21, 2007 Presented at the 93rd Annual Meeting of the North Pacific Surgical Association, Spokane, WA, November 10 –11, 2006

Abstract Background: Our current trauma triage system uses patient and scene variables within a 3-tiered trauma response system. Our purpose was to evaluate the accuracy of the current system and to identify the most reliable variables for trauma triage. Methods: This was a retrospective review at a level II trauma center. Multivariate logistic regression was used to identify independent predictors of the need for any urgent emergency department procedure or operative intervention. The current triage system was analyzed and compared with a proposed simplified system. Results: There were 1495 consecutive trauma admissions identified, the majority (88%) were blunt mechanism. Urgent emergency department interventions were required in 11%, and 4% required emergent surgery. Logistic regression demonstrated that prehospital Glasgow Coma Score ⬍14 (odds ratio [OR] 9.7), hypotension (OR 3.3), altered respiratory effort (OR 4.6), and penetrating truncal injury (OR 10.8) independently predicted the need for urgent intervention (all P ⬍ .01). The current system undertriaged only 1% but overtriaged 51% of patients. A simplified triage system using these 4 variables significantly decreased overtriage and reliably identified patients with severe injury. Conclusions: A simplified triage system using only highly predictive variables can safely decrease the high rate of overtriage of trauma patients. © 2007 Excerpta Medica Inc. All rights reserved. Keywords: Overtriage; Trauma; Triage; Undertriage

Trauma triage systems seek to identify and provide rapid treatment for the most severely injured trauma patients while at the same time identifying less-injured patients in need of only basic care. An ideal system would equally match the severity of injury and resources required for optimal care with the appropriate trauma facility and personnel. However, some degree of imprecision in trauma triage systems is unavoidable, resulting in overtriage and undertriage. As stated in the American College of Surgeons Committee on Trauma guidelines, “ . . . in general, priority has been given to decrease of undertriage, because undertriage may result in preventable mortality or morbidity from

This manuscript represents the opinions of the authors only, and does not represent the views of the Department of Defense, the Department of the Army, or Madigan Army Medical Center. * Corresponding author. Tel.: ⫹1-253-968-2200; fax: ⫹1-253-9685900. E-mail address: [email protected]

delays in definitive care” [1]. Although decrease of undertriage should result in fewer missed injuries or delays in receiving definitive care, the inevitable result has been an increase in overtriage of patients with less-severe or negligible injuries. Although overtriage results in little impact to the patient, it may result in significant strain on hospital and system wide resources and personnel. A wide variety of trauma triage criteria has been proposed and adopted [1–7], but there is no national consensus on the ideal set of variables. Many systems use some variation of the field triage scheme outlined by the American College of Surgeons Committee on Trauma [1], which uses a large number of variables related to patient physiology, anatomic injury, mechanism of injury, age, and presence of comorbid diseases. Although some of these variables have been scientifically validated as predictors of injury [8 –10], the majority have not been rigorously studied as independent predictors of injury severity or resource requirement. The triage system currently used by our facility incorporates field physiologic parameters, mechanism of injury, ana-

0002-9610/07/$ – see front matter © 2007 Excerpta Medica Inc. All rights reserved. doi:10.1016/j.amjsurg.2007.01.017

R.K. Lehmann et al. / The American Journal of Surgery 193 (2007) 630 – 635

Fig. 1. Pierce County (WA) prehospital trauma triage guidelines. This is the algorithm used by emergency department staff at our institution as the guidelines for trauma team activation.

tomic factors, and age to stratify patients into 3 tiers of triage acuity (Fig. 1). The purpose of this study was to evaluate the accuracy of our current triage system for identifying patients in need of urgent interventions and to elucidate the variables that are most predictive of severe injury needing emergent intervention or a higher acuity of care. Methods Madigan Army Medical Center is a military tertiary care referral center and Washington State–verified level II trauma center. Our facility’s trauma triage criteria follow the guidelines set forth in the Pierce County Prehospital Trauma Triage Procedures (Fig. 1). These triage procedures are based on physiologic, anatomic, and mechanistic variables obtained by prehospital personnel that are then used by emergency department staff as an in-hospital triage tool to activate the trauma team using a 3-tiered response system. Full trauma team activation includes the presence of the surgery attending physician on call, surgery chief resident, surgery junior resident, emergency medicine attending physician, emergency medicine senior resident, emergency medicine junior resident, emergency department nurse, anesthesia provider, nursing supervisor, radiology resident, radiology technician, blood bank technician, laboratory officer of the day, chaplain, and trauma nurse coordinator. Our facility maintains a prospectively collected trauma registry, including all patients who meet prespecified Washington State inclusion criteria [11]. The trauma registry throughout the state of Washington has been standardized, and all hospitals use a computerized database (Collector Client Data Management System; Digital Innovations, Forest Hill, Maryland). Responsibility for data input into the program at our facility is equally shared by the trauma nurse coordinator and the trauma registrar. This study was a retrospective analysis of all adult patients (age ⬎16 years) entered into the trauma registry during a 4-year period. Patients whose primary mechanism of injury was a burn and patients transferred from our emergency department to another facility were excluded. The trauma registry was queried for all available prehospital variables to include age, sex, mechanism of injury, extrication, vital signs, consciousness

631

and respiratory assessment, and the subsequent decision for trauma team activation. Data were also collected regarding emergency department vitals, injury severity and diagnoses, hospital procedures, length of stay, and mortality. Patients were then categorized into 2 groups based on whether they required an urgent emergency room or operative intervention. Patients were categorized into the “intervention” group if they were brought directly from the emergency department to the operating room for an urgent surgical procedure (laparotomy, thoracotomy, craniotomy, or neck exploration) or if they required any 1of the following procedures in the emergency department: intubation or surgical airway, tube or needle thoracostomy, thoracotomy, pericardiocentesis, central venous catheter placement, blood transfusion, or cardiopulmonary resuscitation. Overtriage was defined as trauma team activation for a patient who did not require urgent intervention, and undertriage was defined as no trauma team activation for a patient who ultimately required ⱖ1 of the previously mentioned interventions. Overtriage and undertriage rates for our current triage system were analyzed and compared with corresponding rates using multiple permutations and combinations of individual field triage variables. Descriptive and inferential statistics were performed using SPSS 11.0 Statistical Software Package (Mac edition; SPSS, Chicago, Illinois). Parametric variables were subjected to Student t test analysis with central tendencies reported as mean ⫾ SD. Nonparametric variables were subjected to Mann-Whitney U test with central tendencies reported as medians. Nonparametri variables were compared with Chi-square analysis or Fischer exact test where appropriate. Key variables were then subjected to stepwise logistic regression for dichotomous dependent variables. Limits of entry into the regression were set at .05, and at .10 for the variable to be removed. Significance was set at P ⬍ .05. This study was reviewed and approved by the local Institutional Review Board. Results From January 2002 through December 2005, 1782 patients were included in the trauma registry. There were 1495 patients identified who met inclusion criteria and who represented our study population. The demographics of the study group are listed in Table 1. The population was a cohort of predominately men who largely suffered blunt traumatic injury. Only 3% of all patients required emergent operative intervention, whereas 11% of patients required urgent intervention in the emergency room. Overall, the cohort experienced a low incidence of morbidity (9% complication rate) and a low mortality rate (3%). The population was then divided into those patients who required and those who did not require emergent intervention. The univariate comparison of the 2 groups is listed in Table 2. As expected, there were significant differences identified between the 2 groups in demographics, injury severity, and outcomes, all favoring the group that did not require emergent intervention. Similarly, multiple triage variables, including anatomic, mechanistic, and physiologic factors, were significantly different on univariate analysis. The majority of variables included in the current triage system were found to be significant on this phase of analysis.

632

R.K. Lehmann et al. / The American Journal of Surgery 193 (2007) 630 – 635

Table 1 Study population demographics Variable (n ⫽ 1495)

Mean ⫾ SD n (%)

Age Male sex Mechanism of injury Blunt Penetrating Emergent ED procedure Emergent OR intervention ICU admission from ED ISS ICU LOS (d) Hospital LOS (d) Complications Mortality

41 ⫾ 22 1045 (70) 1315 (88) 179 (12) 162 (11) 58 (3) 222 (15) 8.8 ⫾ 9 2.9 ⫾ 4 4.6 ⫾ 6 133 (9) 49 (3)

ED ⫽ Emergency department; ICU ⫽ intensive care unit; ISS ⫽ Injury Severity Score; LOS ⫽ length of stay; OR ⫽ operating room.

We then subjected the key demographic and triage variables to multivariate logistic regression for the end point of requiring an emergent ER or operative intervention. The results are listed in Table 3. Four clinical variables, ie, GCS Table 2 Univariate comparison of patients who required emergent intervention with those who did not require emergent intervention Variable

Emergent intervention (n ⫽ 188)

No emergent intervention (n ⫽ 1163)

P

Age (y) Male sex (%) Mechanism of injury (%) MVC Penetrating Other Penetrating truncal injury (%) Prehospital variables (%) Extrication Consciousness assessment Normal Altered Respiratory effort Normal Labored Absent Bradycardic/tachycardic (%) Hypotensive (SBP ⬍100 mm Hg) (%) ED variables Hypotensive (SBP ⬍90 mm Hg) (%) ICU LOS (d) Injury Severity Score Mortality (%)

41 ⫾ 20 68

42 ⫾ 22 76

.647 ⬍.05 ⬍.01*

37 12 55 11

36 8 59 3

71

80

42 58

75 25

75 14 11 24

98 2 0 17

⬍.05*

10

1

⬍.01*

⬍.01* ⬍.01* ⬍.01* ⬍.01*

21 1 16 12.3

4 1 9 0.7

⬍.01* .786† ⬍.01* ⬍.01*

ED ⫽ emergency department; ICU LOS ⫽ intensive care unit length of stay; MVC ⫽ motor vehicle crash. * Chi-square test. † Mann-Whitney U test and measure of central tendency reported as medians.

Table 3 Prehospital independent predictors of patients requiring emergent intervention (emergent ED intervention, OR procedure, or ICU admission) Variables

OR

Significant GCS ⬍14 Respiratory effort (absent or labored) Hypotension (SBP ⬍100 mm Hg) Penetrating truncal injury Nonsignficant Age (y) Sex Penetrating mechanism Firearms MCC Stab Bradycardia (HR ⬍60 bpm) Tachycardia (HR ⬎110 bpm) Extrication

95% confidence interval

P

9.7

5.5–16.9

⬍.01

4.6 3.3 10.8

1.9–11 1.6–6.8 4.4–26.5

⬍.01 ⬍.01 ⬍.01

1.0 0.79 0.88 1.0 102 105 0.77 1.06 1.46

0.98–1.01 .44–1.43 0.367–2.0 0–10 0–10 0–10 0.225–2.6 0.655–1.74 0.94–2.3

.96 .45 .739 1.0 .677 .676 .675 .791 .100

ED ⫽ Emergency department; GCS ⫽ Glasgow Coma Scale; HR ⫽ heart rate; ICU ⫽ intensive care unit; MCC ⫽ motorcycle crash; OR ⫽ operating room.

⬍14, abnormal respiratory effort (labored or absent), hypotension (systolic blood pressure [SBP] ⬍100 mm Hg), and penetrating truncal injury, were found to be independent predictors for requiring some form of urgent intervention. However, the remainder of the variables included in our current triage system did not independently identify patients who would require urgent attention. Most notably, factors such as age, heart rate, mechanism of injury, prolonged extrication, and mild confusion (GCS ⫽ 14) were not significant independent contributors to appropriate triage of these patients. Although our primary purpose was to analyze prehospital data, we did include several emergency department variables in a multivariate regression model and found that emergency department hypotension (SBP ⬍100 mm Hg), regardless of blood pressure in the field, was a strong independent predictor of the need for urgent intervention (OR 5.22, 95% confidence interval 2.55–10.67, P ⬍ .001). Using the identified independent predictors from the regression analysis, we then compared these 4 variables as a simplified triage system with the current triage system (Table 4). The 2 trauma systems were compared based on the percentage of patients who would have been overtriaged Table 4 Comparison of current triage system with a simplified triage system based on four System

Overtriage (%)

Undertriage (%)

Current triage system Mortality Simplified triage system Mortality

51 0 29 0

1 7 3 0

*Penetrating truncal injury, GCS ⬍14, abnormal respiratory effort, and hypotension.

R.K. Lehmann et al. / The American Journal of Surgery 193 (2007) 630 – 635

and undertriaged. The simplified triage system significantly decreased the overtriage rate from 51% to 29% (P ⬍ .05), and there was no significant difference in undertriage rates between the 2 systems (3% vs. 1%, P ⫽ not significant). The simplified triage criteria performed as well as the more complex current system for identifying severely injured patients (Injury Severity Score ⬎15) and predicting inhospital mortality. Interestingly, there were no deaths among patients who were undertriaged using the simplified criteria compared with a 7% mortality among patients who were undertriaged using our current system. Discussion Optimal care of the traumatically injured patient relies on a rapid and prioritized approach to identifying anatomic and physiologic derangements that require urgent intervention. This process begins with prehospital identification and triage of the injured patient to an appropriate center and personnel able to manage their injuries. Multiple variables and algorithms have been proposed to help guide prehospital and emergency department personnel in making decisions about entering a patient into a trauma system and triggering trauma team activation [1–10]. Although some of these systems have been studied and validated, the majority are anecdotal or of unproven predictive ability. In addition, the complexity of many of these proposed schemes limits their application and reliability by variably trained first responders in the often chaotic prehospital setting. Although undertriage affects the overall safety of a trauma system, overtriage affects its efficiency and can result in increased costs, inappropriate resource use, and provider frustration. This is particularly important in the current climate of limited resident work hours, the overburdening of our nations trauma centers, and physician hesitancy or unwillingness to participate in trauma call [2,12]. Unfortunately, efforts to decrease overtriage rates may come at the expense of increased undertriage, which is unacceptable from a patient safety and ethical standpoint. To safely achieve these seemingly contradictory goals, a system in place that can quickly and accurately identify and separate severely injured trauma patients from uninjured or less severely injured patients. Although many trauma systems have attempted to do this by introducing increasingly complex and “all-inclusive” sets of triage criteria and decision algorithms, the best approach may be to simplify our triage systems and only include variables that have been scientifically validated as significant predictors of injury. Our hypothesis in this current study was that the variables used in our system that are true independent predictors of the need for urgent interventions could be used to triage patients with lower overtriage rates and an acceptable rate of undertriage. Scene variables, such as extrication and significant vehicle damage, should in theory relate well to anatomy of injury and thus severity of injury. However, these were shown to be poorly predictive of severe injury in this study. Although theoretically sound variables can add to the sensitivity of a system and thus decrease undertriage, they may increase overtriage at the same time [2]. Using only statistically proven variables as the basis for a triage system can still provide safe and efficient triage for severely injured trauma patients [2,5].

633

Several physiologic variables have been shown previously to be independent predictors of injury severity and the requirement for emergent intervention. Tinkoff et al reported that SBP ⬍90 mm Hg, endotracheal intubation, and low GCS (⬍8) were associated with increased mortality, need for emergent surgery, and intensive care unit admission [8]. Norwood et al demonstrated that a prehospital GCS ⱕ14 in the setting of blunt trauma was a strong predictor of severe injury and the need for urgent evaluation and hospitalization [10]. The importance of penetrating truncal mechanism as a marker of severe injury has been well demonstrated by Sava et al, who showed that truncal gunshot injuries were independently predictive of severe pathology and the need for urgent intervention, even without the presence of physiologic derangement criteria [9]. Similarly, our data have also shown that hypotension, altered mental status, altered respiratory status, and penetrating truncal injury were valid independent predictors that should be included in any triage system. In addition, we found that the development of hypotension after arrival at the emergency department, even with stable vital signs in the field, was a strong independent predictor of needing some urgent intervention. We also found that the often-used physiologic variable of heart rate was not a good predictor of the need for urgent intervention and did not contribute to accurate triage of our trauma patients. Several other series have also validated that heart rate may not be an accurate variable for identifying severely ill or injured patients [13–16]. This is likely a reflection of multiple other factors that can influence heart rate, such as fever, agitation, drug or alcohol use, medications, age, level of fitness, etc. Altered mental status is another often-used variable to triage patients to a trauma center and to activate a trauma response. Several series have validated the use of the GCS as a triage tool, and a low GCS has been found to reliably predict the need for urgent interventions, injury severity, morbidity, and mortality [3,8,10,17]. Although it is well accepted that low GCS is highly predictive of the need for intubation and other urgent interventions, the exact break point in the GCS score at which it becomes predictive has not been identified. Our findings indicate that including mildly confused or intoxicated patients (GCS 14), who were otherwise physiologically stable, did not have an increased need for urgent interventions, and that a GCS ⬍ 14 would be a better cut point for identifying seriously injured patients. Factors related to the mechanism of injury and not to the patient’s physiologic response to the injury may also have limited value as triage variables. Criteria such as vehicle damage or intrusion, prolonged extrication, and death of another vehicle occupant have produced mixed results when evaluated for predictive value [2,18 –20]. Similarly, the “gut feeling” of the triaging medic or physician has been demonstrated to fare poorly as a triage tool for discriminating injured from uninjured patients. Our results agree with these findings because the majority of mechanistic variables, including prolonged extrication and prolonged scene time, were not found to be independent predictors of injury. It is interesting to note that in the current triage system (Fig. 1), poorly predictive variables, such as tachycardia, activate the highest level of trauma response (step 1), whereas penetrat-

634

R.K. Lehmann et al. / The American Journal of Surgery 193 (2007) 630 – 635

ing truncal injury carries a lesser level of trauma activation (step 2). This study has several limitations that should be noted. It is a retrospective review of trauma registry data and is thus subject to retrospective bias, errors, and incompleteness in data entry, and it is limited to the predefined variables already contained in the trauma registry. Outcome data were limited to the hospital stay only, and we cannot comment on the impact of triage on any longer-term outcome measures. Although this was a large sample size, the numbers of patients requiring interventions and the number of deaths were low and thus could limit the power of the statistical analysis. Similarly, there may be other important triage criteria that did not reach statistical significance in our study. For example, advanced age has been found in several series to both confound standard triage variables, such as heart rate and blood pressure, and to be an independent marker of morbidity and mortality [21,22]. Although we used the need for urgent surgery or intervention as our benchmark for proper triage, there may be patients who do not need any urgent intervention and yet would still benefit from having a full trauma team present. Most importantly, although we retrospectively found that our proposed simplified criteria performed well in terms of undertriage and overtriage, this should be tested and validated before being accepted as a safe practice. To this end, we are currently in the process of analyzing our current and proposed triage criteria in the setting of a prospective observational trial. In summary, a simplified triage system using anatomic and physiologic criteria to determine the need for full trauma team activation can potentially minimize overtriage while maintaining an acceptable rate of undertriage. We found that prehospital altered mental status, hypotension, altered respiratory status, and penetrating truncal injury should prompt full trauma team activation. These factors can be easily and rapidly assessed at the scene of injury by first responders and integrated into a simple system for hospital personnel to use as a guide to activating the full trauma team. In addition, if a patient does not meet any prehospital trauma activation criteria but develops hypotension at any time in the emergency department, immediate trauma team notification and aggressive evaluation for a life-threatening injury should be instituted. References [1] Surgeons Committee on Trauma–American College of Surgeons. Resources for Optimal Care of the Injured Patient. Chicago, IL: American College of Surgeons; 1998. [2] Esposito TJ, Offner PJ, Jurkovich GJ, et al. Do prehospital trauma center triage criteria identify major trauma victims? Arch Surg 1995; 130:171– 6. [3] Cook CH, Muscarella P, Praba AC, et al. Reducing overtriage without compromising outcomes in trauma patients. Arch Surg 2001;136: 752– 6. [4] Long WB, Bachulis BL, Hynes GD. Accuracy and relationship of mechanism of injury, trauma score, and injury severity score in identifying major trauma. Am J Surg 1985;151:581– 4. [5] Koehler JJ, Baer LJ, Malafa SA, et al. Prehospital index: a scoring system for field triage of trauma victims. Ann Emerg Med 1986;15: 178 – 82. [6] Knopp R, Yanagi A, Kalsen G, et al. Mechanism of injury and anatomic injury as criteria for prehospital trauma triage. Ann Emerg Med 1988;17:895–902.

[7] Phillips JA, Buchman TG. Optimizing prehospital triage criteria for trauma team alerts. J Trauma 1993;34:127–32. [8] Tinkoff GH, O’Connor RE. Validation of new trauma triage rules for trauma attending response to the emergency department. J Trauma 2002;52:1153–9. [9] Sava J, Alo K, Velmahos GC, et al. All patients with truncal gunshot wounds deserve trauma team activation. J Trauma 2002;52:276 –9. [10] Norwood SH, McAuley CE, Berne JD, et al. A prehospital Glasgow Coma Scale Score ⱕ 14 accurately predicts the need for full trauma team activation and patient hospitalization after motor vehicle collisions. J Trauma 2002;53:503–7. [11] Washington State Department of Health. Office of Emergency Medical Services and Trauma System Web site. Available at: http://www. doh.wa.gov/hsqa/emstrauma/traumareg.htm. Accessed: December 2006. [12] Abraham T, Freitas M, Frangos S, et al. Are resident work-hour limitations beneficial to the trauma profession? Am Surg 2006;72(1): 35– 41. [13] Victorino GP, Battistella FD, Wisner DH. Does tachycardia correlate with hypotension after trauma? J Am Coll Surg 2003;196:679 – 84. [14] Sander-Jensen K, Secher NH, Bie P, et al. Vagal slowing of the heart during hemorrhage: observations from 200 consecutive hypotensive patients. Br Med J 1986;292:364 – 6. [15] Brown CV, Velmahos GC, Neville AL, et al. Hemodynamically “stable” patients with peritonitis after penetrating abdominal trauma: identifying those who are bleeding. Arch Surg 2005;140(8):767–72. [16] Demetriades D, Chan LS, Bhasin P, et al. Relative bradycardia in patients with traumatic hypotension. J Trauma 1998;45:534 –9. [17] Shatney CH, Sensaki K. Trauma team activation for “mechanism of injury” blunt trauma victims: time for a change? J Trauma 1994;37: 275– 81. [18] Gabbe BJ, Cameron PA, Wolfe R, et al. Prehospital prediction of intensive care unit stay and mortality in blunt trauma patients. J Trauma 2005;59:456 – 63. [19] Lowe DK, Oh GR, Neely KW, et al. Evaluation of injury mechanism as a criterion in trauma triage. Am J Surg 1986;152:6 –10. [20] Henry MC, Hollander JE, Alicandro JM, et al. Incremental benefit of individual American College of Surgeons trauma triage criteria. Acad Emerg Med 1996;11:992–1000. [21] Tornetta P, Mostafavi H, Riina J, et al. Morbidity and mortality in elderly trauma patients. J Trauma 1999;46(2):702– 6. [22] Demetriades D, Sava J, Alo K, et al. Old age as a criterion for trauma team activation. J Trauma 2001;51(4):754 – 6.

Discussion William B. Long III, M.D. (Portland, OR): The results of this retrospective study of trauma system patients seen at Madigan Army Medical Center (Tacoma, Washington), a level II trauma center by Washington State criteria, confirm what we already know: current triage criteria as described by the American College of Surgeons Committee on Trauma and the Washington State Trauma System creates a low under-triage rate, but a significantly higher over-triage rate defined by trauma system patients not needing an urgent intervention, ICU care, or ISS less than 16. The authors note in their discussion, “There may be patients who do not need any urgent intervention, and yet would still benefit from having a full trauma team present.” This statement, while not validated by the data in this paper, is the crux of the controversies over trauma team activations and trauma triage criteria. Is the objective of some trauma centers to activate the trauma team less often or “more appropriately”? Are trauma centers going to reduce costs by doing less diagnostic studies urgently, or are the same diagnostic studies going to occur anyway, but at a slower pace with a smaller team response, or maybe by an emergency physician and a nurse or two?

R.K. Lehmann et al. / The American Journal of Surgery 193 (2007) 630 – 635

If efficiency is the word we are looking for, can an emergency physician work up a trauma patient who does not meet trauma team activation criteria as rapidly and efficiently as a full trauma team? I submit that the answer is no. Annual data from my trauma center shows an average resuscitation time for stable trauma system patients admitted to the emergency department to be 25 ⫾ 5 minutes, before the patient goes to the computed tomography scanner or the operating room. An emergency physician with 1 or 2 nurses for team members takes more than 1 hour before the patient goes to the CT scanner, and maybe longer if an ultrasound is performed. Due to the incompleteness of the work-up for trauma by many emergency physicians, many trauma surgeons are ordering

635

additional diagnostic studies. A “stealth trauma patient” (not entered into the trauma system by prehospital care personnel) is usually in the ED for 3 to 4 hours before the surgeon is called. Both patient populations have eventually the same number of diagnostic tests. Where is the efficiency? The authors affirm the status and accuracy of some of our current trauma triage criteria and highlight the importance of penetrating truncal injury. The subject of this paper is really the necessity for full trauma team activation for most trauma system patients. I hope the authors will address this issue and the importance of rapid, thorough, diagnostic assessment of trauma system patients to rule in or rule out serious injury in a future publication.