Accuracy of Prehospital Diagnosis and Triage of a Swiss Helicopter Emergency Medical Service

ORIGINAL ARTICLE

Accuracy of prehospital diagnosis and triage of a Swiss helicopter emergency medical service

Rebecca M. Hasler, MD, Christian Kehl, MS, Aristomenis K. Exadaktylos, MD, Roland Albrecht, MD, Simon Dubler, MD, Robert Greif, MD, MME, and Natalie Urwyler, MD, Stanford, California

BACKGROUND: Helicopter emergency medical services (HEMSs) have become a standard element of modern prehospital emergency medicine. This study determines the percentage of injured HEMS patients whose injuries were correctly recognized by HEMS physicians. METHODS: A retrospective level III evidence prognostic study using data from the largest Swiss HEMS, REGA (Rettungsflugwacht/Guarde Ae´rienne), on adult patients with trauma transported to a Level I trauma center (January 2006YDecember 2007). National Advisory Committee on Aeronautics (NACA) scores and the Injury Severity Score (ISS) were assessed to identify severely injured patients. Injured body regions diagnosed by REGA physicians were compared with emergency department discharge diagnoses. RESULTS: Four hundred thirty-three patients were analyzed. Median age was 42.1 years (interquartile range, 25.5Y57.9). Three hundred twenty-three (74.6%) were men. Patients were severely injured, with an in-hospital NACA score of 4 or higher in 88.7% of patients and median ISS of 13. REGA physicians correctly recognized injuries to the head in 92.9%, to the femur in 90.5%, and to the tibia/fibula in 83.8% of patients. Injuries to these body regions were overdiagnosed in less than 30%. Abdominal injuries were missed in 56.1%, pelvic injuries in 51.8%, spinal injuries in 40.1%, and chest injuries in 31.2% of patients. CONCLUSION: This study shows that patients are adequately triaged by REGA physicians reflected by a NACA score 4 or higher in 88.7% of patients and a median ISS of 13. However, recognition of injured body regions seems to be challenging in the prehospital setting. Prospective studies on specific training of HEMS physicians for recognition of these injuries (e.g., portable ultrasonography, telemedicine) might help in the future (J Trauma Acute Care Surg. 2012;73: 709Y715. Copyright * 2012 by Lippincott Williams & Wilkins) LEVEL OF EVIDENCE: Prognostic study, level III. KEY WORDS: Helicopter emergency medical services (HEMS); diagnostic accuracy; triage; prehospital emergency medicine.

BACKGROUND with trauma from the scene of injury to a facility where definitive optimal care can be promptly provided remains a core objective Like many innovations in emergency medical services in the management of the severely injured patient.5 (EMSs), the idea of transporting the injured by aircraft has its Three different companies provide HEMSs in Switzer- origins in the military, and the concept of using the aircraft as an land. Two smaller companies, Air Zermatt and Air Glacier, are 1 ambulance is almost as old as powered flight itself. During the located in the alpine region and therefore focus on high- First World War, air ambulances were tested by various military altitude mountain rescue operations.6,7 The third company, organizations and were used regularly for crash rescue by the Swiss Air-Rescue (REGA [Rettungsflugwacht/Guarde Ae´ri- 1 American Army and Navy. The first use of helicopters to enne]), is one of the largest HEMSs in Europe.8 REGA, a non- evacuate combat casualties was by the US Army in Burma during profit organization, organizes 10,000 missions per year and 1 the Second World War. In 1952, the Swiss Air Rescue was operates with 17 helicopters from 13 bases throughout Swit- 2 founded. zerland.8 A REGA helicopter team consists of a pilot, a Today,helicopter EMSs (HEMSs) have become an integral HEMS paramedic, and an emergency physician, usually a se- part of prehospital emergency medicine, with more than 360 nior fellow in anesthesia or intensive care medicine. They 3,4 helicopter bases in Europe. The capacity to transport a patient serve a 6-month full-time rotation at REGA as part of their clinical training at Swiss teaching hospitals.8 For HEMS teams, prehospital triage of patients with trau- 9Y12 Submitted: December 12, 2011, Revised: April 17, 2012, Accepted: April 20, 2012. ma is a difficult task. Transport to trauma centers of patients From the Departments of Anesthesiology and Pain Therapy (R.M.H., R.G., N.U.), and with only minor injuries should be avoided, and injuries to Emergency Medicine (R.M.H., C.K., A.K.E., S.D.), University Hospital Bern, specific body regions should be correctly recognized early in the Bern; and Swiss Air-Rescue (R.A.), REGA (Rettungsflugwacht/GuardeAe´rienne), Switzerland; and Department of Pathology (N.U.), Stanford University School of process to avoid increased morbidity and mortality, facilitate in- 11,13,14 Medicine, Stanford, California. hospital procedures, and avoid increased costs. This work was presented in part at the annual meeting of the European Society of Therefore, this retrospective cohort study aims to assess Emergency Medicine, EuSEM, October 11Y14, 2010, Stockholm, Sweden. Travel to this meeting was supported by a peer-reviewed research grant by Stiftung Stein, the accuracy of prehospital diagnosis by determining the per- Schwyz, Switzerland. centage of patients with trauma transported by REGA in central Address for reprints: Natalie Urwyler, MD, Department of Pathology, L235, Stanford Switzerland who (1) have severe injuries and (2) whose injured School of Medicine, Stanford, CA 94305; email: [email protected]; natalie_ body regions have been correctly recognized in the field. [email protected]. Our hypothesis is that more than 70% of patients trans- DOI: 10.1097/TA.0b013e31825c14b7 ported by REGA have experienced severe injuries, and that J Trauma Acute Care Surg Volume 73, Number 3 709

This study includes all adult patients with trauma trans- TABLE 1. NACA Scores ported by REGA to a university-based Level I trauma center be- NACA Score Patient Status Necessary Intervention tween January 2006 and December 2007. We excluded patients 1 Slight injury or illness No medical intervention younger than16 years because they are directly admitted to the 2 Moderately heavy Ambulatory medical treatment Children’s University Hospital. Furthermore, we excluded injury or illness interhospital transfers because diagnosis in these patients was 3 Heavy but not life-threatening Stationary medical treatment not made by REGA emergency physicians, but by physicians injury or illness of the referring hospital. Finally, we excluded patients trans- 4 Heavy injury or illness, life Emergency medical measures ported by HEMSs other than REGA. In these cases, different threat cannot be excluded injury scoring systems were used, rescue teams were not uni- 5 Acute mortal danger Emergency medical measures formly staffed, and case report forms were noncomparable. 6 Acute cardiac or Emergency resuscitation respiratory arrest 7 Death Outcome Measurements and Data Analysis First, NACA scores15 (Table 1) and the Injury Severity Score (ISS)16 were retrospectively assessed by two board- recognition of injured body regions by REGA emergency phy- certified emergency physicians of our study group. This retro- sicians is correct in more than 70% of patients. spective assessment was based on the final diagnosis made by the consultant emergency physician on call at the emergency department (ED). Imaging and diagnostic procedures, including MATERIALS AND METHODS full-body roentgenograms, computed tomographic scans, mag- netic resonance imaging scans, ultrasonography, and laboratory Inclusion Criteria and Study Population results, are completed in the ED at the University Hospital of This retrospective cohort study uses data from the Swiss Bern as a standard. Therefore, the ED discharge diagnosis for HEMS REGA. REGA emergency physicians triage patients patients with trauma comprises the full extent of injuries. based on the National Advisory Committee for Aeronautics NACA scores based on these ED final diagnoses were (NACA) score, which defines the severity of an injury or illness compared with NACA scores made by the REGA physicians. A (Table 1).15 In addition, REGA physicians classify traumatic cutoff of NACA score of 4 or higher, which classifies patients injuries in 21 prespecified body regions (Table 2). as life threatened, was used to identify seriously injured patients.15 Retrospective data analysis was approved by the relevant Injuries were classified using the Abbreviated Injury Scale for institutional review boards. each anatomic location.16 A cutoff at an AIS score of 3 or

TABLE 2. Comparison of Injured Body Regions (AIS Score Q2): REGA Versus ED Physicians REGA ED Correctly Diagnosed by REGA Missed by REGA Overdiagnosed by REGA Body Region nn n(%*, 95% CI) n (%*, 95% CI) n (%†, 95% CI) Brain/skull 192 169 157 (92.9, 0.88Y0.96) 12 (7.1, 0.04Y0.12) 35 (18.2, 0.13Y0.24) Chest 136 154 106 (68.8, 0.61Y0.76) 48 (31.2, 0.24Y0.39) 30 (22.1, 0.15Y0.30) Spine 126 152 91 (59.9, 0.52Y0.68) 61 (40.1, 0.32Y0.48) 35 (27.8, 0.20Y0.36) Abdomen 76 57 25 (43.9, 0.31Y0.58) 32 (56.1, 0.42Y0.69) 51 (67.1, 0.55Y0.77) Pelvis 53 56 27 (48.2, 0.35Y0.62) 29 (51.8, 0.38Y0.65) 26 (49.1, 0.35Y0.63) Face 46 86 44 (51.2, 0.40Y0.62) 42 (48.8, 0.38Y0.60) 2 (4.4, 0.01Y0.15) Shoulder/clavicle 24 60 19 (31.7, 0.20Y0.45) 41 (68.3, 0.55Y0.80) 5 (20.8, 0.07Y0.42) Femur 54 42 38 (90.5, 0.77Y0.97) 4 (9.5, 0.03Y0.23) 16 (29.6, 0.18Y0.44) Tibia/ﬁbula 38 37 31 (83.8, 0.68Y0.94) 6 (16.2, 0.06Y0.32) 7 (18.4, 0.08Y0.34) Forearm 25 34 23 (67.6, 0.50Y0.83) 11 (32.4, 0.17Y0.51) 2 (8.0, 0.01Y0.26) Ankle/foot 13 23 10 (43.5, 0.23Y0.66) 13 (56.5, 0.34Y0.79) 3 (23.1, 0.05Y0.54) Knee 10 15 5 (33.3, 0.12Y0.62) 10 (66.7, 0.38Y0.88) 5 (50.0, 0.19Y0.81) Wrist/hand 7 12 6 (50.0, 0.21Y0.79) 6 (50.0, 0.21Y0.79) 1 (14.3, 0.00Y0.58) Upper arm 11 12 6 (50.0, 0.21Y0.79) 6 (50.0, 0.21Y0.79) 5 (45.5, 0.17Y0.77) Body regions with G10 patients Hypothermia 9 6 3 (50.0, 0.12Y0.88) 3 (50.0, 0.12Y0.88) 6 (66.7, 0.30Y0.93) Drowning 2 2 2 (100.0, 0.16Y1.00) 0 (0, 0.00Y0.84) 0 (0, 0.00Y0.84) Elbow 4 2 2 (100.0, 0.16Y1.00) 0 (0, 0.00Y0.84) 2 (50.0, 0.07Y0.93)

*Denominator: Number of diagnoses by ED physicians. †Denominator: Number of diagnosis by REGA physicians. No data in the categories burns, frostbites, barotraumas, and suffocation. CI indicates conﬁdence interval.

710 * 2012 Lippincott Williams & Wilkins

Included patients presented with a median age of 42.1 years (IQR, 25.5Y57.9 years). Three hundred twenty-three (74.6%) were men. Twenty-one patients (4.8%) died before leaving the ED.

NACA Scores Forty-nine patients (11.3%) were classified by REGA as NACA score 3, 217 (50.1%) as NACA score 4, 145 (33.5%) as NACA score 5, and 22 (5.1%) as NACA score 6 (Fig. 2). The median in-hospital NACA score was 5 (IQR, 4Y5). Three hundred eighty-four patients (88.7%) were severely injured according to an in-hospital NACA score of 4 or higher. Twenty-one patients died before leaving the ED with a NACA score of 7. Overall, in 283 patients (65.4%), prehospital and in-hospital NACA scores were the same. Thirty-eight patients (8.8%) scored higher in prehospital than in in-hospital assessment, whereas 112 patients (25.9%) presented with lower prehospital NACA scores than in-hospital ones. Figure 1. Flow chart of study patients. AIS Score/ISS higher, which defines serious injury to one body region, was Median ISS was 13 (IQR, 5.3Y22.0). One hundred eighty- used to identify seriously injured patients.16 The AIS scores of eight patients (43.2%) showed an ISS more than 15. Of these the three body regions with the most severe injuries were used patients, 29 (6.7%) showed an ISS more than 42. Two hundred to calculate the ISS.16 seventy-six patients (63.4%) were classified as seriously injured Second, using the REGA classification of different body with an AIS score of 3 or higher in at least one body region. regions for patients with trauma (Table 2), diagnoses made by One hundred sixty-one (37.2%) patients experienced injuries the REGA emergency physicians were compared with the final of an AIS score of 3 or higher to the head, 133 (30.7%) to the ED diagnoses. For this comparison, injuries with a severity of chest, 78 (18%) to the extremities, and 30 (6.9%) to the abdomen. AIS score 1 were excluded because they represent minor in- Injuries of an AIS score of 3 or higher involving the face (n = 13, juries only. 3.0%) or external injuries (e.g., soft tissue injuries, burns, or hy- Results are reported as numbers, percentages, medians, pothermia; n = 5, 1.2%) were rare. One hundred fifty-seven corresponding interquartile ranges (IQRs) (caused by non- patients (36.3%) experienced minor to moderate injuries, with normal data distribution), and 95% confidence intervals, where a maximum of an AIS score of 2 or less in any body region applicable. Certain types of injuries (e.g., pelvic injuries) are (Table 3, Fig. 3). frequently missed or overdiagnosed. This is because such injuries might be missed in a large number of patients who actually sustained a pelvic injury (false-negatives) and diagnosed in many Comparison of Injured Body Regions: REGA other patients who in fact did not sustain this injury (false- Emergency Physicians Versus ED Physicians positives). Injuries missed or overdiagnosed in more than 30% G G of patients were labeled ‘‘frequently missed or overdiagnosed,’’ Rarely Missed ( 30%), Rarely Overdiagnosed ( 30%) whereas injuries missed or overdiagnosed in less than 30% REGA emergency physicians correctly recognized inju- were labeled ‘‘rarely missed or overdiagnosed.’’ This arbitrary ries to the brain/skull in 92.9%, to the femur in 90.5%, and to the cutoff has been introduced to give the reader an easily distin- tibia/fibula in 83.8% of patients. Injuries to these three regions guishable overview of the percentages presented in Table 2. were overdiagnosed in less than 30%.

RESULTS Patients During the study period, 1,196 patients were admitted by various helicopter services. One hundred sixteen patients were excluded because they were admitted by HEMSs other than REGA. Four hundred twenty-four interhospital transfer patients were also excluded. Of the remaining 626 patients directly ﬂown by REGA to the university hospital, 124 experienced unrelated medical conditions and were therefore excluded as well. In addition, we excluded 72 patients younger than 16 years and 27 patients with incomplete data on ﬁnal diagnosis. Figure 2. Comparison of NACA scores by REGA and ED Therefore, a total of 433 patients were analyzed (Fig. 1). physicians.

* 2012 Lippincott Williams & Wilkins 711

TABLE 3. AIS Score of Injured Body Regions AIS Score Head, n (%) Face, n (%) Chest, n (%) Abdomen, n (%) Extremities, n (%) External, n (%) 1 73 (16.9) 27 (6.2) 23 (5.3) 13 (3.0) 59 (13.6) 106 (24.5) 2 35 (8.1) 60 (13.9) 46 (10.6) 75 (17.3) 103 (23.8) 2 (0.5) 3 106 (24.5) 12 (2.8) 106 (24.5) 20 (4.6) 67 (15.5) 0 (0) 4 22 (5.1) 1 (0.2) 10 (2.3) 6 (1.4) 11 (2.5) 1 (0.2) 5 21 (4.8) 0 (0) 16 (3.7) 4 (0.9) 0 (0) 3 (0.7) 6 12 (2.8) 0 (0) 1 (0.2) 0 (0) 0 (0) 1 (0.2)

Frequently Missed (930%), Frequently indicating that HEMS transport was justified. REGA emergen- Overdiagnosed (930%) cy physicians correctly diagnosed moderate-to-severe head, Approximately one half of injuries to the knee (66.7%), to femur, and tibia/fibula injuries in more than 83% of patients. the abdomen (56.1%), to the pelvis (51.8%), and to the upper arm However, they were not that successful in diagnosing abdomi- (50%) were missed by REGA physicians. These injuries were nal, pelvic, spinal, and chest trauma. overdiagnosed in 45.5% to 67.1% of cases. In this retrospective case-control study, we analyzed a notable number of patients rescued by the largest Swiss Frequently Missed (930%), Rarely HEMS, flown to a Level I trauma center. However, our study Overdiagnosed (G30%) has a few limitations. First, in other Swiss regions with differ- A mixed pattern of frequently missed but rarely overdiag- ent emergency systems, the accuracy of prehospital diagnosis nosed injuries was found for the spine (missed 40.1%, over- might be different. Second, we only examined patients directly diagnosed 27.8%), face (missed 48.8%, overdiagnosed 4.4%), transported to a Level I trauma center. The numbers of patients chest (missed 31.2%, overdiagnosed 22.1%), and for seve- with severe injuries transported to Level II or Level III trauma ral extremity injuries (shoulder/clavicle, ankle/foot, forearm, centers are unknown to the authors and, therefore, not included wrist/hand). in the analysis. Consequently, the number of undertriaged patients could not be estimated. Third, this analysis is based Rarely Missed (G30%), Frequently on the final diagnosis of injured body regions and on injury se- 9 verity, whereas emergency medicine triage in the field is based Overdiagnosed ( 30%) on injury mechanisms and vital signs and symptoms. Unfortu- None. nately, physiologic parameters, such as systolic blood pressure, Injuries to the elbow, drowning, and hypothermia injuries heart and respiratory rate, or the Glasgow Coma Scale, which G were rare ( 10 patients each) (Table 2). Therefore, we excluded are part of many triaging scores, are routinely measured but not them from the comparison of body regions above. systematically collected by the HEMS. We were therefore not in a position to use them for further analysis. This may lead to an underestimation of correctly triaged patients in our study. DISCUSSION The NACA score is uniformly used by our REGA physicians, This retrospective cohort study revealed that patients and it is also widely used by many other European HEMSs. admitted by the HEMS REGA are seriously injured, with a However, the score is not validated. Therefore, the NACA median NACA score of 5, a median ISS of 13, and/or an AIS score was only one element in assessing triage accuracy, score of 3 or higher in at least one body region in 63% of along with ISS and injured body regions. patients. Furthermore, in 66% of patients, on-scene judgment Our study found that 43% of HEMS patients showed by REGA emergency physicians was adequate, with prehospi- an ISS of 16 or higher. This is close to the highest figure reported tal NACA scores corresponding to in-hospital NACA scores, in studies from the United States, which observed that 16% to 43% of HEMS patients showed an ISS of 16 or higher.17,18 They concluded that most patients did not need helicopter transfer but could have been transported by ground ambulance. This is not true for the alpine region of Switzerland, where HEMS is used instead of ground ambulance because of specific terrain requirements. Furthermore, we found that 63% of patients showed an AIS score of 3 or higher in at least one body region, and 89% of patients showed a NACA score of 4 or higher. This is in contrast to an earlier Swiss HEMS REGA study of Moeschler et al.19 in 1992 in the region of Lausanne, which found that only 22% of patients were severely injured, with a NACA score of 4 to 6. However, their analysis only included patients rescued Figure 3. AIS scores of different body regions. by a winch in difficult terrains where landing was impossible.

712 * 2012 Lippincott Williams & Wilkins

Regarding the assessment of injured body regions, we CONCLUSIONS observed a range of results. Whereas abdominal, pelvic, chest, and spine injuries seemed difficult to diagnose correctly, with Most patients are adequately transported to a Level I missed diagnosis ranging from 31.2% (chest) to 56.1% (pel- trauma center by REGA emergency physicians based on ISS vis), head, femur, and tibia/fibula injuries were recognized in (median ISS, 13) and NACA scores (median NACA score, 5; more than 83% of patients. Our results are supported by a NACA scores of 89% of the patients, Q4). However, injury to recent study by Muhm et al.11 in 111 German patients with various body regions (abdomen, pelvis, spine, chest) is missed trauma with a mean ISS of 19. They found that injuries to the in 31% (chest) to 56% (pelvis) of patients. Therefore, pro- head, thorax, abdomen, and pelvis are missed or under- spective studies on injury-specific training of HEMS emergen- estimated by prehospital emergency physicians in more than cy physicians for early recognition of abdominal, pelvic, spine, 45% of patients. Our findings are also similar to an Australian and chest trauma, and on the use of technical devices (e.g., por- study, which found that the recognition of severe injuries (AIS table ultrasonography, telemedicine) must be evaluated score, Q3) was worst for abdominal (39%) and best for head further. injuries (58%).20 The authors observed a 98% success rate at recognizing severely injured patients in the prehospital field AUTHORSHIP setting.20 This is similar to our finding, with 89% of patients A.E., R.H., and N.U. conceived the study. A.E., R.H., N.U., and C.K. were having a NACA score of 4 or higher. responsible for conception and design of the study. R.H., N.U., C.K., and The results of our study may have implications for the S.D. did the literature research. R.H., N.U., C.K., and S.D. analyzed the data and interpreted the results in collaboration with R.G., A.E., and R.A. management of prehospital triage of patients with major trauma N.U. and R.A. were responsible for the acquisition of data. R.H. wrote the and for the training of HEMS physicians. A more accurate first draft of the article. All authors critically revised the report for im- prehospital diagnosis of torso injuries is likely to positively in- portant intellectual content and approved the final version of the report. fluence the management of these patients. The ED personnel are ACKNOWLEDGMENTS triggered in a specific direction by the report given by the pre- The authors thank the team of the REGA Base Belp, Switzerland, for its hospital team. According to the prehospital report, the ED invaluable support during the course of this work. The authors also thank physician calls consultants of different surgical disciplines, and Kathrin Dopke and Sabina Utinger for administrative help and Freya anesthesiologists, into the resuscitation room before the patient Shipley for English language editing. arrives to provide the best possible diagnostics, decision making, and treatment immediately (e.g., an experienced anesthesiologist DISCLOSURE is called to manage a difficult airway, an experienced thoracic The authors declare no conflicts of interest. N.U. was supported by a grant from the Swiss National Science Foundation (PBBEP3_131564) surgeon to manage thoracic trauma, etc.). For patients reported to and the Gottfried and Julia Bangerter-Rhyner Foundation, Switzerland. be less severely injured, more time will pass before resources and specialist care become available. This is especially true for those REFERENCES patients admitted during night hours. Out of regular hours, 1. Air Ambulance. Air ambulance service. Available at: http://www. consultants are not based in the hospital (apart from the emer- airambulanceservice.com/history.html. Accessed September 28, 2011. gency and anesthesiology consultant), and it may take up to one 2. REGA Homepage. History of REGA. Available at: http://www.rega.ch/en/ half an hour for them to arrive. Therefore, valuable time may be about-us/history/1946-1959.aspx. Accessed September 28, 2011. lost if prehospital diagnosis of severe injuries is missed. 3. Bledsoe BE, Wesley AK, Eckstein M, Dunn TM, O’Keefe MF. Helicopter scene transport of trauma patients with non-life-threatening injuries: a In Switzerland, HEMS emergency physicians are often meta-analysis. J Trauma. 2006;60:1257Y1266. fellows in their final years in anesthesia or intensive care 4. European HEMS and Air Ambulance Committee (EHAC) Homepage. medicine, rotating to the HEMS as part of their professional Information on EHAC. Available at: http://www.ehac.eu/about.html. training. Classes in initial assessment of patients with trauma Accessed September 28, 2011. and rotations by means of the trauma department might im- 5. Branas CC, MacKenzie EJ, Williams JC, Schwab CW,Teter HM, Flanigan prove physicians’ ability to recognize injuries to specific body MC, Blatt AJ, ReVelle CS. Access to trauma centers in the United States. JAMA. 2005;293:2626Y2633. regions. Furthermore, the use of advanced technical support, 6. Air Zermatt Homepage. Available at: www.air-zermatt.ch. Accessed such as portable ultrasonography devices, may help in diag- September 28, 2011. nosing injuries to the abdomen and chest in the prehospital 7. Air-Glaciers Homepage. Available at: www.airglaciers.ch. Accessed setting.21,22 Portable ultrasonography can be used to image September 28, 2011. patients being transported to the hospital by helicopter, 8. REGA Homepage. Available at: www.rega.ch. Accessed September 28, allowing early assessment of internal abdominal injuries and 2011. fluid in the pericardium.21,22 Focused assessment by sonog- 9. Hasler RM, Schmucker U, Evangelopoulos DS, Hirschberg RE, Zimmermann H, Exadaktylos AK. Improving prehospital trauma management for skiers raphy in trauma protocols, successfully implemented in most and snowboarders: need for on-slope triage? J Trauma Manag Outcomes. trauma units, could be used by HEMS emergency physi- 2011;5:5. cians.23 Furthermore, new handheld devices, using proprietary 10. Kirves H, Handolin L, Niemela¨ M, Pitka¨niemi J, Randell T. Paramedics’ radar technology to noninvasively detect the presence of a and prehospital physicians’ assessments of anatomic injury in trauma pneumothorax, may soon be available to support diagnosis.24 patients: a cohort study. Scand J Trauma Resusc Emerg Med. 2010;18:60. In addition, ED-based telemedicine trauma teams, sup- 11. Muhm M, Danko T, Madler C, Winkler H. [Preclinical prediction of prehospital injury severity by emergency physicians: approach to evaluate porting emergency physicians in the field by telephone and/or validity]. Anaesthesist. 2011;60:534Y540. video portals, might improve prehospital diagnosis and efficient 12. Arntz HR, Klatt S, Stern R, Willich SN, Beneker J. [Are emergency triage.25 physicians’ diagnoses accurate?]. J Anaesthesist. 1996;45:163Y170.

* 2012 Lippincott Williams & Wilkins 713

13. Ku¨hne CA, Ruchholtz S, Sauerland S, Waydhas C, Nast-Kolb D. [Personnel 19. Moeschler O, Refondini S, Ho¨fliger C, Freeman J. Difficult aeromedical and structural requirements for the shock trauma room management rescue situations: experience of a Swiss pre-alpine helicopter base. of multiple trauma. A systematic review of the literature]. Unfallchirurg. J Trauma. 1992;33:754Y759. Y 2004;107:851 861. 20. Mulholland SA, Cameron PA, Gabbe BJ, Williamson OD, YoungK, Smith 14. Hoyt DB, Hollingsworth-Fridlund P, Winchell RJ, Simons RK, Holbrook KL, Bernard SA. Prehospital prediction of the severity of blunt anatomic T, Fortlage D. Analysis of recurrent process errors leading to provider- injury. J Trauma. 2008;64:754Y760. related complications on an organized trauma service: directions for 21. Byhahn C, Bingold TM, Zwissler B, Maier M, Walcher F. Prehospital care improvement. J Trauma. 1994;36:377Y384. ultrasound detects pericardial tamponade in a pregnant victim of stabbing 15. Norum J, Elsbak TM. Air ambulance services in the Arctic 1999Y2009: a assault. Resuscitation. 2008;76:146Y148. Norwegian study. Int J Emerg Med. 2011;4:1. 22. Strode CA, Rubal BJ, Gerhardt RT, Bulgrin JR, Boyd SY. Wireless and 16. Committee on Injury Scaling, Association for the Advancement of satellite transmission of prehospital focused abdominal sonography for Automotive Medicine (AAAM). The Abbreviated Injury Scale (2005 Y revision). Des Plaines, Chicago, IL; 2005. trauma. Prehosp Emerg Care. 2003;7:375 379. 17. Brown JB, Stassen NA, Bankey PE, Sangosanya AT, Cheng JD, Gestring 23. Mazur SM, Pearce A, Alfred S, Goudie A, Sharley P. The F.A.S.T.E.R. ML. Helicopters and the civilian trauma system: national utilization trial. Focused assessment by sonography in trauma during emergency Y patterns demonstrate improved outcomes after traumatic injury. JTrauma. retrieval: a feasibility study. Injury. 2008;39:512 518. 2010;69:1030Y1036. 24. PneumoSonics Homepage. Available at: http://www.pneumosonics.com/. 18. Brathwaite CE, Rosko M, McDowell R, Gallagher J, Proenca J, Spott MA. Accessed September 28, 2011. A critical analysis of on-scene helicopter transport on survival in a state- 25. Latifi R, Keane MG. A review of the role of telemedicine in the accident wide trauma system. J Trauma. 1998;45:140Y146. and emergency department. J Telemed Telecare. 2009;15:132Y134.

714 * 2012 Lippincott Williams & Wilkins

APPENDIX

I) DISTRIBUTION OF ISS BY NUMBER OF PATIENTS.

II) DISCHARGES AND ADMISSIONS Length of Stay, h ISS Discharges and Admissions Patients, n (%) Median (IQR) Median (IQR) Discharged home G4 h 86 (19.9) 1 (1Y6) 3 (1Y6) Discharged to other hospital G24 h 15 (3.5) 14 (10Y21) 13 (9Y19) Admitted for 924 h 298 (68.8) 186 (68Y306) 15 (9Y22) Admitted to intensive care unit 199 (45.7) 166 (50Y316) 22 (13Y27)

III) DISPOSITION AT DISCHARGE: Disposition at Discharge Patients, n (%) Home 214 (49.4) Other hospital 143 (33.0) Rehabilitation 24 (5.5)

IV) IN-HOSPITAL MORTALITY: Mortality Patients, n (%) Overall 52 (12.0) G24 h 34 (7.9) ED mortality 21 (4.8) 924 h 18 (4.2)

* 2012 Lippincott Williams & Wilkins 715