TRAUMA in the pediatric age group remains a

Transcription

1 Prehospital Triage in the Injured Pediatric Patient By S.A. Engum, M.K. Mitchell, L.R. Scherer, G. Gomez, L. Jacobson, K. Solotkin, and J.L. Grosfeld Indianapolis, Indiana Background/Purpose: Identifying major trauma patients in the prehospital setting is essential in determining management, destination, and best utilization of emergency department resources. Few methods of trauma triage have been accepted unanimously. This study prospectively evaluates the efficacy of comprehensive field triage using 12 criteria (simplified version of the American College of Surgeon s guidelines) in 1,285 pediatric trauma patients. Methods: Major trauma was defined as occurring in those who died in the emergency room, had major surgery (penetrating injury involving surgery of the head, neck, chest, abdomen, or groin), or were admitted directly to the intensive care unit. The correlation between trauma triage criteria, hospital disposition, and triage accuracy were determined prospectively and compared in the pediatric patients (36 months) with an adult cohort of patients (12 months). Results: A total of 1,285 pediatric trauma patients were evaluated and compared with 1,326 adult trauma patients. The most accurate trauma triage criterion for major injury was a blood pressure 90 mmhg (systolic) with an accuracy of 86%. This was followed by burn greater than 15% total body surface area (79%), Glasgow Coma Scale score 12 (78%), respiratory rate less than 10/min or greater than 29/min (73%), and paralysis (50%). Less accurate criteria included a fall from greater than 20 feet (33%); penetrating injury to head, neck, chest, abdomen, or groin (29%); ejection from vehicle (24%); pedestrian struck at greater than 20 mph (16%); paramedic judgement (12%); rollover (3%); and extrication (0%). The Glasgow Coma Scale score was a more accurate indicator of major injury in children than adults, and paramedic judgement was less accurate in children when compared with adults. Of the 379 major pediatric trauma victims, the Revised Trauma Score and Pediatric Trauma Score missed 36% and 45% of these major trauma victims, respectively. The overtriage rate for children was 71% with a sensitivity of 100% (no missed major trauma patients). Conclusions: Physiological variables, anatomic site, and mechanism of injury provide a sensitive and safe system of triage. Continued education of prehospital personnel regarding pediatric trauma and stratification of the current triage tools are necessary to minimize overtriage in an era of shrinking resources. J Pediatr Surg 35: Copyright 2000 by W.B. Saunders Company. INDEX WORDS: Trauma, triage. TRAUMA in the pediatric age group remains a significant health care problem, accounting for most deaths between ages 1 and 15 years. 1 The systems approach to trauma has resulted in improved survival rates, 2-4 and the concept of the preventable trauma death has led to an evolving regionalization of trauma care in the United States. 2 Regionalized trauma care depends on accurate assessment of the injured patient and triage to an appropriate facility. The increased demand for effective on-the-scene triage led to the development of scoring systems to aid in the triage decisions. 5 Many different scoring systems have been devised for prehospital triage From the James Whitcomb Riley Hospital for Children, Indiana University Regional Trauma Center, Indiana University School of Medicine, Indianapolis, IN. Presented at the 46th Annual International Congress of the British Association of Paediatric Surgeons, Liverpool, England, July 21-24, Address reprint requests to Jay L. Grosfeld, MD, Chairman and Surgeon-in-Chief, James Whitcomb Riley Hospital for Children, 702 Barnhill Dr, Suite 2500, Indianapolis, IN Copyright 2000 by W.B. Saunders Company /00/ $03.00/0 of injured patients (trauma score, 6 trauma index, 7 CRAMS (circulation, respiration, abdomen, motor and sensory), 8 Hopkins (Alpha, Bravo, Delta, and Echo), 9 and the prehospital index 10 ; however, the ideal scoring system has been elusive. Most triage schemes fall into 1 of 4 categories: physiological criteria, mechanism of injury, anatomic site of injury, and the high-risk patient. It is recognized, however, that physiological derangement develops over time, and there is a subset of patients who have lifethreatening injuries despite initial hemodynamic stability and nearly normal physiological scores in the prehospital phase. In addition, anatomic injury criteria and mechanism of injury when used alone as triage tools are inaccurate and often lead to inappropriate triage to higher-level facilities. The purpose of this study was to prospectively (1) assess the usefulness of an adult simplified trauma triage system in pediatric trauma at a level 1 adult and pediatric trauma hospital and (2) determine the accuracy of each criterion, positive predictive value, and areas of under and overtriage when compared with adults in an effort to maximize hospital resources and personnel. 82 Journal of Pediatric Surgery, Vol 35, No 1 (January), 2000: pp 82-87

2 PREHOSPITAL TRIAGE 83 MATERIALS AND METHODS The Indiana University Regional Trauma Center Wishard Hospital Trauma Center is located on the campus of Indiana University Medical Center in Indianapolis and serves as a tertiary referral center for trauma in the Indianapolis metropolitan area. The area encompasses approximately 170 square miles and has a population base of 1.2 million people. The Pediatric Trauma Service serving Wishard and Riley Children s Hospital evaluates all pediatric trauma victims age 0 to 15 years. Protocol A response algorithm was created that relied on patient information routinely reported via radiotelephone by prehospital providers through the Wishard Hospital Emergency Medical Service Systems. Based on 12 Trauma Alert Criteria, patients underwent triage by prehospital and emergency room personnel for evaluation in a trauma resuscitationshock room by a senior surgical resident and activation of the Pediatric Surgery Trauma Team. Table 1 lists the 12 Trauma Alert Criteria modified and simplified from the American College of Surgeon triage decision scheme. 11 Analysis All cases were evaluated by the Pediatric Trauma Service over the last 36 months and were entered into a Trauma Registry that calculated the Pediatric Trauma Score (PTS), 12 Revised Trauma Score (RTS), 13 and Injury Severity Score (ISS) 14,15 for each case. Major trauma patients were defined as those dying in the emergency room, admitted to the pediatric intensive care unit, or requiring a major surgical procedure (craniotomy, neck exploration, thoracotomy, median sternotomy, exploratory laparotomy, and limb-threatening vascular procedures). Minor trauma patients were defined as those discharged from the emergency room, admitted to a hospital ward bed, or requiring a minor surgical procedure. Statistics Statistical formulas used for analysis included the positive predictive value and accuracy of the criterion (true-positives and true-positives plus false-positives), sensitivity (true-positives and true-positives plus false-negatives), overtriage (individuals not meeting major trauma definition), and undertriage (those patients with major trauma missed by the 12 Trauma Alert Criteria). Statistical support for comparison between pediatric and adult patients related to each criterion was based on 2 tests with a P value of less than.05 considered statistically significant. Table 1. Simplified Trauma Alert Criteria Blood pressure 90 mm Hg (systolic) Glasgow Coma Scale score 12 Respiratory rate 10/min or 29/min Penetrating injury to head, neck, chest, abdomen, or groin Second- or third-degree burn involving 15% total body surface area Paralysis Ejection of patient from vehicle Rollover of vehicle Extrication of patient Falls of 20 feet Pedestrian hit at 20 mph by vehicle Emergency medical technician or paramedic judgement Table 2. Simplified Trauma Triage Criteria Distribution (1,285 Patients) No. (%) Emergency medical technician/paramedic judgement 518 (40) Pedestrian struck 20 mph 277 (22) Glasgow Coma Scale score (10) Second or third degree burn involving 15% TBSA 82 (6) Penetrating injury to head, neck, chest, abdomen, or groin 69 (5) Blood pressure 90 mm Hg (systolic) 65 (5) Rollover of vehicle 58 (5) Fall from 20 feet 43 (3) Ejection from vehicle 17 (1) Respiratory rate 10/min or 29/min 11 (1) Extrication of patient from vehicle 10 (1) Paralysis 4 (1) RESULTS Demographics Of the 1,285 injured children, there were 807 boys and 478 girls. The average age was 7.2 years with a range of 1 month to 15 years. Those children identified by the criterion penetrating injury to head, neck, chest, abdomen, or groin had an average age of 13.7 years with a range of 3 to 15 years. Ninety percent of the penetrating injuries were in boys. Those children identified by the criteria of greater than 15% of total body surface area (TBSA) burn had an average age of 4.3 years with a range of 4 months to 15 years. Trauma Triage Criteria Distribution Table 2 presents the distribution of the 1,285 pediatric patients who were distributed throughout all 12 triage criteria. Trauma Triage Criteria Accuracy The most accurate trauma triage criterion was a systolic blood pressure 90 mm Hg, which was 86% accurate for major injury. The next most accurate criterion was a second- or third-degree burn of greater than 15% TBSA (79%), followed by Glasgow Coma Scale score 12 (78%), respiratory rate less than 10 breaths per minute or greater than 29 breaths per minute (73%); paralysis (50%); fall greater than 20 feet (33%); penetrating injury to the head, neck, chest, or abdomen (29%); ejection (24%); pedestrian stuck at greater than 20 mph (16%); paramedic judgement (12%); rollover (3%); and extrication (0%). The accuracy of detecting a major trauma victim for the 12 simplified trauma triage criteria was 29% for pediatric patients and 37% for the adult patients. The Glasgow Coma Scale score was a more accurate indicator of a major trauma victim in pediatric patients when compared with adults (78% v 52%; P.01). In addition, a burn greater than 15% TBSA

3 84 ENGUM ET AL (79% v 91%; P.03), rollover (3% v 18%; P.01), extrication (0% v 38%; P.02), and paramedic judgement (12% v 20%; P.01) were all less accurate in the pediatric patient than adult (Table 3). Comparison With RTS and PTS Of the 1,285 pediatric trauma victims, 379 were determined by the prospective simplified criteria to have major trauma. When the Revised Trauma Score (230 of 379 patients with complete data) and the Pediatric Trauma Score (228 of 379 patients with complete data) were used to identify major pediatric trauma victims (RTS 12, and PTS 8), 36% and 45% of major trauma occurrences were missed, respectively. Elimination of Subjective Criteria When paramedic judgement is eliminated as a triage criteria, accuracy for major injury increased to 41% for the remaining 11 criteria with an overtriage rate of 59%; however, the undertriage rate, or missed major pediatric trauma victims increased to 7% (62 missed major injuries) and the sensitivity fell from 100% to 84%. Mortality The overall mortality rate was 4% (50 of 1,285) with 32 children dying in the emergency room and 18 dying in the intensive care unit. The emergency room death rate is similar to the adult population (3%). DISCUSSION The implementation of Regional Trauma Systems has resulted in improved outcomes for trauma victims. Haller et al showed that system management of lifethreatening injuries in children improves patient outcome. Effective trauma systems require implementation of a mechanism by which decisions are made to transport victims to appropriate hospitals. Criteria that are insufficiently sensitive results in too many severely injured patients not being transported to an appropriate trauma center with the likelihood of increased mortality, whereas a lack of specificity results in too many victims with less severe injuries being transported to the trauma center. Prehospital personnel with varying degrees of training and education must make decisions regarding triage and facility destination rapidly in the field. Important factors to consider in accurate triage assessment include (1) mechanism of injury, (2) the patient s physiological state, (3) the injuries themselves, (4) specialist facilities available in nearby hospitals, and (5) likely transport times. Triage protocols such as the Trauma Score, 19 Revised Trauma Score, 13 CRAMS score, 20 Pre-Hospital Index, 21 Revised Trauma Index, 22 and the Trauma Triage Rule 23,24 have been developed in an attempt to enable paramedics to determine hospital destination. Undertriage rates of 4% to 19% have been reported, 5,25 and it has been suggested that an undertriage rate of 5% may be acceptable to achieve an overtriage rate of 50%. 26 The Injury Severity Score (ISS) is a standard anatomic measure of injury with scores ranging from 0 to 75, with the largest numbers reflecting the most serious injuries. 14 Injury Severity Scores higher than 15 have been used as a proxy for injuries of sufficient magnitude to require trauma center care. However, because the ISS is only an anatomic measure of injury severity and does not measure alterations in physiology or urgency of care, it is inappropriate to use this score as the sole criterion for Table 3. Simplified Trauma Alert Criteria Positive Predictive Value (Accuracy) Criteria No. ER ORMA ICU ORMI WARD DC ACC% BP (133) 28 (30) 7 (47) 21 (27) 1 (13) 7 (6) 1 (10) 86 (78) GCS (152) 1 (5) 13 (19) 88 (55) 2 (12) 18 (18) 9 (43) 78 (52)* RR (4) 1 (0) 0 (2) 7 (0) 1 (90) 2 (1) 0 (1) 73 (50) Penetrating 69 (332) 0 (5) 7 (65) 13 (34) 3 (12) 12 (100) 34 (116) 29 (31) Burn 15% 82 (82) 1 (2) 6 (6) 58 (67) 0 (0) 14 (1) 3 (6) 79 (91) Paralysis 4 (14) 0 (0) 0 (0) 2 (7) 0 (0) 1 (3) 1 (4) 50 (50) Ejection 17 (23) 0 (0) 0 (0) 4 (5) 1 (4) 6 (5) 6 (9) 24 (22) Rollover 58 (55) 0 (0) 1 (4) 1 (6) 2 (2) 5 (6) 49 (37) 3 (18) Extrication 10 (34) 0 (0) 0 (4) 0 (9) 1 (4) 6 (7) 3 (10) 0 (38) Fall 20ft. 43 (34) 0 (0) 0 (0) 14 (9) 0 (3) 12 (7) 17 (15) 33 (26) Ped 20mph 277 (99) 1 (1) 2 (2) 41 (7) 23 (28) 66 (9) 144 (52) 16 (10) Paramedic judgment 518 (364) 0 (0) 4 (14) 58 (60) 28 (46) 182 (49) 246 (195) 12 (20) Total 1285 (1326) 32 (43) 40 (143) 307 (286) 62 (124) 331 (212) 513 (498) 29 (37) NOTE. Values are expressed as pediatric patients with adult patients in parentheses. Abbreviations: ER, died in emergency room; ORMA, major operating room procedure; ICU, admit to intensive care unit; ORMI, minor OR procedure; WARD, admit to ward bed; DC, discharged from emergency room; ACC%, percent accuracy of criterion (ER ORMA ICU)/N. *P.01. P.03. P.01. P.02. P.01.

4 PREHOSPITAL TRIAGE 85 validation of a new triage score. The Pediatric Trauma Score (PTS) is a combination of anatomic and physiological scoring that examines 6 clinical variables in the injured child and assigns each a value of 1, 1, or 2. The resulting score for any one injured child can range from 6 to 12, with the most severe injuries assigned the lowest score. Ramenofsky et al, 27 reported 0% mortality rate for children with a PTS higher than 8 and a mortality rate of 24% for a PTS of 8 or less with a sensitivity of 95.8% and a specificity of 98.6%. Thus, children with a PTS of 8 or less have been appropriately triaged to a trauma center. Eichelberger et al 28 tested the PTS in a triage setting and found it to be effective with an undertriage rate of 15% when ISS scores were 20 or more. Statistical analysis and refinement of the Trauma Score developed the Revised Trauma Score (RTS), which is a physiological score. 19 The score range is from 0 to 12, with the lowest scores representing the most severe injuries. Similarly, patients with a RTS of 11 or less have been identified as a group that requires trauma center care. 29,30 Triage accuracy, primarily related to overtriage, directly impacts resource consumption and health care costs. Overtriage (the proportion of patients transported to a major trauma center who do not require this intensive level of care) is an important resource utilization factor. Undertriage (the proportion of patients who require a major trauma center care but do not receive it) is primarily a problem of inadequate medical care for the individual patient. This may result in treatment delays or otherwise inappropriate care and lead to a worse outcome. To minimize overutilization of resources, a number of considerations need to be addressed. First, the use of an abbreviated trauma team response should be developed for patients at low risk for major trauma. This will maintain high sensitivity in capturing major pediatric trauma patients to a trauma system, but minimize resource overutilization, and allow the abbreviated trauma team to activate the Pediatric Surgical Trauma Team. A higher level response may be activated at any time during the evaluation of those patients whose condition occasionally deteriorates after arrival to a trauma center. This approach will minimize the cry wolf mentality that can develop in an overutilized (abused) trauma system. Secondly, educating paramedic and other hospital personnel about pediatric trauma is essential. This study showed that in our system the majority of personnel responsible for the delivery of healthcare and triage decisions are relatively uncomfortable with the pediatric trauma patient, and this often leads to overutilization of the shock room and the Pediatric Surgical Trauma Team. This observation reinforces the cry wolf mentality that results in treatment delays and possible patient morbidity and mortality. In addition, paramedic questionnaires are useful in determining areas that require educational focus. The pediatric trauma system should periodically reassess and redefine the Trauma Alert Criteria. In a metropolitan system with numerous ambulance services, having one trauma triage system for adults and children is beneficial. This gives the paramedic or prehospital triage staff a simple and efficient guideline to follow. Although the use of adult trauma triage criteria for pediatric patients does result in higher undertriage rates, 31 currently, in the absence of a field triage instrument designed and validated for pediatric trauma patients, it is appropriate to triage injured children to facilities with resources immediately available for expedited care of those severely injured. It may be safer in our environment to make adjustments at the trauma center to maximize resource utilization, rather than impose multiple prehospital triage systems on the prehospital staff creating confusion and the possibility of a missed injury. We have not observed instances of undertriage using our current triage scheme, which further emphasizes the safety of this tool. Results of previous studies in pediatric patients have shown that the Glasgow Coma Scale score, ejection from a vehicle, and penetrating injuries are statistically significant predictors of major trauma. 31 Our data support the accuracy with the Glasgow Coma Scale score as a predictor of severity. Because there is not currently accepted comprehensive measurement for validation of triage scores, we compared our Trauma Triage Criteria with the PTS and RTS as predictors of major pediatric trauma. The PTS has been shown to have severe limitations in detecting spleen and liver injuries in the absence of multiple trauma. 32 We observed a significantly higher missed injury rate using the Pediatric Trauma Score (45%) and the Revised Trauma Score (36%) systems of triage. A revised 2-tiered system (Table 4) would activate the Pediatric Surgery Trauma Team using the following criteria: (1) blood pressure 90 mm Hg systolic, (2) Table 4. Proposed 2-Tiered Triage System Pediatric surgical trauma activation 1. Blood Pressure 90 mm Hg (systolic) 2. Glasgow Coma Scale score Respiratory rate 10/min or 29/min 4. Penetrating injury to head, neck, chest, abdomen, or groin 5. Second- or third-degree burn involving 15% total body surface area 6. Paralysis Senior level surgical resident evaluation and activation of trauma team as necessary 7. Ejection of patient from vehicle 8. Rollover of vehicle 9. Extrication of patient 10. Falls of 20 feet 11. Pedestrian hit at 20 mph by vehicle 12. Emergency medical technician or paramedic judgement

5 86 ENGUM ET AL Glasgow Coma Scale score 12, (3) respiratory rate less than 10 breaths per minute or greater than 29 breaths per minute, (4) penetrating injury, (5) burn greater than 15%, and (6) paralysis. The remaining 6 Trauma Alert Criteria would then be prospectively evaluated with each criterion mandating evaluation by a senior level surgery resident in a shock-trauma room and activation of the Pediatric Surgery Trauma Team when indicated for recognized major trauma. This eliminates much of the subjective data that often are presented such as the distance of a fall or the assessed speed of a vehicle, because few of these injuries are actually witnessed. In addition, with the wide spread use of automobile restraints and child car seats, rollover as a mechanism of injury has a lower accuracy rate for predicting major trauma in the pediatric patient. Our proposed criteria system would ensure upper level surgical evaluation in a cost-effective manor and maintain the ability to activate necessary pediatric surgical involvement and care. This system would maintain 100% sensitivity in our current environment. Paramedic judgement in evaluating the pediatric trauma patient remains a significant problem. Although we would prefer to delete this criteria from the in-field trauma evaluation, we recognize this might increase the undertriage rate and the number of missed injuries (n 62). In addition, other studies have shown that emergency medical technician (EMT) judgement and assessment have performed as well as other scoring systems in the field in identifying patients that have died or require emergency operative intervention. 33 Continued EMT education in identifying risk factors for mortality and the need for trauma center care may be as effective as the implementation of a new scoring system in determining the need for trauma center transport. We believe that continued reevaluation of these criteria will permit fine tuning of a triage system that can be useful in the design and function of regional pediatric trauma centers with a comprehensive emergency medical services system. Use of a second tier triage classification system to further differentiate between potentially severely injured versus moderately injured trauma patients can lead to a more cost-effective delivery of trauma care. At the same time, optimal use of valuable human resources is accomplished. Because a high number of deaths in the pediatric population occur during the attempted resuscitation phase, 34 adult centers must continue to be able to handle pediatric trauma arrests and address the pediatric age groups when planning trauma prevention activities. 1. National Safety Council: Accident Facts. Chicago, IL, National Safety Council, Cales RH, Trunkey DD: Preventable trauma deaths: A review of trauma care systems development. JAMA 254: , Shackford SR, Mackersie RC, Hoyt DB, et al: Impact of a trauma system on outcome of severely injured patients. Arch Surg 122: , West JG, Trunkey DD, Lim RC: Systems of trauma care: A study of two countries. Arch Surg 114: , West JG, Murdock MA, Baldwin LC, et al: A method for evaluating field triage criteria. J Trauma 26: , Morris JA Jr, Auerbach PS, Marshall GA, et al: The trauma score as a triage tool in the prehospital setting. JAMA 256: , Kirkpatrick JR, Youmans RL: Trauma index: An aide in the evaluation of injury victims. J Trauma 11: , Kane G, Engelhardt R, Celentano J, et al: Empirical development and evaluation of prehospital trauma triage instruments. J Trauma 25: , Phillips JA, Buchman TG: Optimizing prehospital triage criteria for trauma team alerts. J Trauma 34: , Koehler JJ, Baer LJ, Malafa SA, et al: Prehospital index: A scoring system for field triage of trauma victims. Ann Emerg Med 15: , Hospital and Prehospital Resources for Optimal Care of the Injured Patient. American College of Surgeons, Tepas JJ, Mollitt DL, Talbert JL, et al: The Pediatric Trauma Score as a predictor of injury severity in the injured child. J Pediatr Surg 22:14-18, Champion HR, Sacco WJ, Copes WS, et al: A revision of the Trauma Score. J Trauma 29: , Baker SP, O Neill B, Haddon W, et al: The injury severity score. REFERENCES A method for describing patients with multiple injuries and evaluating emergency care. J Trauma 14: , Baker SP, O Neill B: The Injury Severity Score: An update. J Trauma 16: , Haller JA, Shorter N, Miller D, et al: Organization and function of a regional pediatric trauma center: Does a system management improve outcome? J Trauma 23: , Haller JA, Buck J: Does a trauma management system improve outcome for children with life-threatening injuries? Can J Surg 28:477, Haller JA, Beaver B: A model: Systems management of lifethreatening injuries in children for the state of Maryland, USA. Int Care Med 15:S53-S56, 1989 (suppl) 19. Champion HR, Sacco WJ, Catnazzo AJ, et al: Trauma Score. Crit Care Med 9: , Gormican SP: CRAMS scale: Field triage of trauma victims. Ann Emerg Med 11: , Koehler JJ, Malafa S, Hillesland J, et al: A multicenter validation of the prehospital index. Ann Emerg Med 16: , Smith JS, Bartholomew MJ: Trauma index revisited: A better triage tool. Crit Care Med 18: , Baxt WG, Jones G, Fortlage D: The trauma triage rule: A new, resource-based approach to the prehospital identification of major trauma victims. Ann Emerg Med 19: , Emerman CL, Shade B, Kubincanek J: Comparative performance of the Baxt trauma triage rule. Am J Emerg Med 10: , Lowe DK, Oh GR, Neely KW, et al: Evaluation of injury mechanism as a criterion in trauma triage. Am J Surg 152:6-10, Committee on Trauma: Field categorization of trauma patients (field triage). Bull Am Coll Surg 71:17-21, Ramenofsky ML, Ramenofsky MB, Jurkovich GJ, et al: The

6 PREHOSPITAL TRIAGE 87 predictive validity of the Pediatric Trauma Score. J Trauma 28: , Eichelberger MR, Gotschall CS, Sacco WJ, et al: A comparison of the Trauma Score, the Revised Trauma Score, and the Pediatric Trauma Score. Ann Emerg Med 18: , Hospital and Prehospital Resources for Optimal Care of the Injured Patient and Appendices A through J. Chicago, IL, American College of Surgeons Committee on Trauma, Champion HR, Sacco WJ: Trauma severity scales, in Maull KI, Cleveland HC, Strauch GO, et al (eds): Advances in Trauma. Chicago, IL, Year Book Medical Publishers, 1986, pp Phillips S, Rond PC III, Kelly SM, et al: The need for pediatric-specific criteria: Results for the Florida Trauma Triage Study. Pediatr Emerg Care 12: , Saladino R, Lund D, Fleisher G: The spectrum of liver and spleen injuries in children: Failure of the Pediatric Trauma Score and clinical signs to predict isolated injuries. Ann Emerg Med 20: , Emerman CL, Shade B, Kubincanek J: A comparison of EMT judgment and prehospital trauma triage instruments. J Trauma 31: , Jubelirer RA, Agarwal NN, Beyer FC, et al: Pediatric trauma triage: Review of 1307 cases. J Trauma 30: , 1990 Discussion D.A. Lloyd (Liverpool, England): Thank you very much for a very clear presentation. It seems that this started off as a look at your prehospital triage system that the paramedics use to decide which centre to take the patient, but you have evolved this now into an in-house triage system for your trauma response team using the refined criteria that you showed us. S.A. Engum (response): We are fortunate that the major ambulance service in our metropolitan area is administratively run by our Regional Trauma Centre. This in-thefield triage system was initiated initially in the adult population brought into our level-1 facility. This system was then extended to children being transported to our Regional Trauma Centre. This trauma triage system was not designed to direct the ambulance paramedics where to transport children, nor has this been mandated. After reviewing the data, it was noted that the majority of patients fell into the paramedic judgement group. It was evident that prehospital triage personnel were not as comfortable with pediatric patient triage, and any injury not triaged by the 11 previous criteria were placed into paramedic judgement with low accuracy and frequent activation of a Pediatric Surgery Trauma Team. This has created a cry wolf mentality and trauma team presentation to trauma resuscitation room delay, which is not safe. These observations have thus stimulated us to re-evaluate our triage system and further trim down the occurrence of trauma team activation in the pediatric patient. D. A. Lloyd (Liverpool, England): In Liverpool, we have developed an in-house 3-tier response system, and we also find that the paramedics tend to overestimate consistently as you say. Once they get to the hospital it is not so bad because it is a children s hospital so people are familiar with children. The Glasgow Coma Scale can be misleading depending on who is doing it. I think there is potential for error there, and we use the AVPU method, which is a lot simpler. I was going to comment on your high overtriage rate, but you have already dealt with that. Do you have any recent assessment of whether you have succeeded in bringing it down or not? S.A. Engum (response): No, we are in the first few months of this prospective evaluation. A couple of things that we have done to maximise the paramedic education issue include our Trauma Director participating in the paramedic morbidity and mortality conference, which is once a month, to bring up some educational issues related to trauma triage. Second, we have instituted, twice a month, a separate paediatric trauma M&M meeting in conjunction with the adult surgeons, residents, and nurses to discuss the triage decisions that were made. Hopefully, these 2 factors will have an effect on minimising overtriage.