emtras

Emergency trauma score: An instrument for early estimation oftrauma severity*

Marcus R. Raum, PhD; Maarten W. N. Nijsten, PhD; Mathijs Vogelzang, MD; Frank Schuring, MD;Rolf Lefering, PhD; Bertil Bouillon, PhD; Dieter Rixen, PhD; Edmund A. M. Neugebauer, PhD;Henk J. ten Duis, PhD; the Polytrauma Study Group of the German Trauma Society

T rauma scores can be used forevaluation, quality assess-ment, standard definitions,economy, patient selection/

triage, and evaluation of the decision pro-cess. A score that may indicate thechance of mortality shortly after admis-sion can be useful to become aware of

severity of trauma and might influencefurther therapeutic decisions (1). Severalscoring systems for mortality risk havebeen developed for patients with trauma(2). Many scores require knowledge of allanatomical injuries such as the injuryseverity score (ISS), the new ISS (NISS),and the trauma injury severity score (3–5). Complete diagnosis of all relevant in-juries may take many hours after admis-sion to the emergency room (ER) so thatanatomical scores are not available early.The revised trauma score (RTS), availablein two versions, uses physiologic param-eters, so it can in principle be determinedearly, even in the prehospital phase (6).However, calculating the RTS is not easyand the RTS shows a poor correlationwith mortality (7).

We envisioned that a scoring systemfor the ER should use a few clinical pa-rameters that are available commonlyand early, and easy to compute and cor-relate with mortality. To derive, validate,and compare it with existing scoring sys-tems, we used a large database on se-

verely injured patients. We designatedthis score the emergency trauma score(EMTRAS).

PATIENTS AND METHODS

The German Trauma Registry has beenestablished by the Deutsche Gesellschaft furUnfallchirurgie (German trauma society). Asof September 2007, 87 German hospitals, 8Austrian hospitals, 1 hospital in The Nether-lands (Groningen), and 1 hospital in Switzer-land participated. The registry prospectivelyrecords all severely injured patients who wereadmitted alive to the ER, with severe injurydefined as an ISS of at least 16 (3, 8). Thedatabase contains more than 350 items fromthe scene of the accident up to long-termneurologic status.

In patients aged 16 years or older, we eval-uated all parameters that had the potential todescribe the trauma severity of a patient in theER. Later information such as anatomical in-juries was therefore not used. We evaluatedage, sex, and parameters that were recordedprehospital by emergency physicians and para-medics: Glasgow Coma Scale (GCS), heart

*See also p. 2122.From the Department of Traumatology (MRR,

MWNN, MV, FS, HJtD), University Medical Center Gro-ningen, University of Groningen, The Netherlands; In-stitutes for Research in Operative Medicine (RF,EAMN), University of Witten/Herdecke, Germany; andDepartment of Trauma and Orthopaedic Surgery (BB,DR), University of Witten/Herdecke at the Cologne Mer-heim Medical Center, Germany.

Supported by the Deutsche Forschungsgemein-schaft (NE 385/5–1 and 2, Sa 862/1–1) and NovoNordisk, Copenhagen, Denmark.

The authors do not have any potential conflicts ofinterest to disclose.

For information regarding this article, E-mail:[email protected]

Copyright © 2009 by the Society of Critical CareMedicine and Lippincott Williams & Wilkins

DOI: 10.1097/CCM.0b013e31819fe96a

Objectives: Early estimation of the mortality risk of severelyinjured patients is mandatory. To estimate the seriousness of thecondition of patients with trauma, we developed the emergencytrauma score (EMTRAS) for ease of use, with simple parametersthat are available within 30 minutes.

Design: Prospective analysis of the German Trauma Registry ofmultitrauma patients.

Setting: EMTRAS was derived from data from 1993 through2003. Potential parameters that were prognostic for mortality inunivariate analysis were evaluated by multivariate binary logisticregression. Selected parameters were then assigned a subscorethat varied from 0 to 3. The EMTRAS score was a simple additionof these subscores. EMTRAS was compared with other scores’receiver operating characteristic curves. After completion,EMTRAS was validated in patients from 2004 and 2005.

Patients: A total of 11,533 patients were to be used for devel-oping the score and 3314 patients for validating it.

Main Results: The strongest predictors of mortality were age,prehospital Glasgow Coma Scale, base excess (mmol/L), and

prothrombin time (% of reference). These parameters were cate-gorized in subscores of 0 through 3. Age: <40, 40 through 60, 61through 75, and >75 scored 0, 1, 2, and 3, respectively. GlasgowComa Scale: 13 through 15, 10 through 12, 6 through 9, and 3through 5 scored 0, 1, 2, and 3, respectively. Base excess: >�1,�5 through �1, �10 through �5.1, and <�10 scored 0, 1, 2, and3, respectively. Prothrombin time: <80%, 80% through 50%, 49%through 20%, and >20% received a score of 0, 1, 2, and 3,respectively. In the validation dataset, the area under the receiveroperating characteristic curve for EMTRAS was 0.828.

Conclusions: EMTRAS combines four early parameters fromthe emergency room and accurately predicts mortality. Knowl-edge of the anatomical injuries is not necessary. The determina-tion of the EMTRAS will inform caregivers of the seriousness ofpatients with trauma at an early stage. (Crit Care Med 2009; 37:1972–1977)

Key Words: multiple trauma; factor analysis; mortality; prog-nosis; shock; databases

1972 Crit Care Med 2009 Vol. 37, No. 6

rate, systolic blood pressure, oxygen satura-tion, respiratory rate, and the RTS. The fol-lowing parameters from the emergency roomwere evaluated: GCS at the scene of the acci-dent, temperature, heart rate, systolic bloodpressure, respiratory rate, oxygen saturation,base excess (BE), lactate, hemoglobin, plateletcount, prothrombin time (PT; expressed aspercent of the reference value, because inter-national normalized ratio was not available inall centers), and the activated partial throm-boplastin time in the first laboratory after ad-mission on the emergency department. Hos-pital mortality was the outcome measure.Parameters that displayed a highly significantunivariate relation with mortality (p � 0.001)were used for binary logistic regression mul-tivariate analysis. The four parameters thatsubsequently showed the strongest relationwith mortality in multivariate analysis wereused for the Emergency Trauma Score(EMTRAS). After identifying a potential scoreparameter, ranges corresponding to subscores0, 1, 2, or 3 were defined on the basis of theparameters’ frequency distribution, its rela-tion with mortality and with regard to naturalcutoff values to keep the subscores practical.The EMTRAS was defined as the simple arith-metic sum of the integer subscores, thus vary-ing from 0 to 12. The relationship ofEMTRAS to hospital mortality was evaluatedwith a receiver operating characteristiccurves. EMTRAS was compared with estab-lished scores such as the RTS, ISS, NISS,and trauma injury severity scale. A first pa-tient group (1993–2003) was used as themodel generating set according to the pro-cedure described earlier. EMTRAS wastested in a model validation set with patientsfrom 2004 and 2005 only after the modelwas completed. Continuous parameterswere compared with Student’s t test or theMann-Whitney U test in case of skewed pa-rameters. Categorical parameters were com-pared with the chi-square test. All data anal-ysis was performed with SPSS statisticalsoftware (Version 14, SPSS, Chicago, IL).

RESULTS

A total of 11,533 patients entered intothe registry between 1993 and 2003 wereused to generate the model and 3314 pa-tients from 2004 and 2005 were used tovalidate the model. Hospital mortality was21.9% (1993–2003) and 18.9%, respectively(2004–2005; Table 1). Blunt injuries repre-sented the majority (95%) of trauma mech-anisms. Many of the parameters that wereevaluated showed a highly significant uni-variate relation with mortality (Table 2).Unfortunately, the percentage of missingdata varied considerably. Thus, 4808 caseswere included in binary logistic regression,

where four parameters (age, GCS, BE, andPT) emerged that were used for theEMTRAS. Table 3 shows the final parame-ters from the logistic regression model. Thesubscores for these four parameters weredefined as indicated in Table 4. The univar-iate relations of these categories for age,GCS, BE, and PT with mortality are alsoshown in Table 4.

Figure 1 shows the relation of mortal-ity with EMTRAS. Table 5 demonstratesthe areas under the curve for the foursubscores that add up to the EMTRAS,EMTRAS itself, and other trauma scores.EMTRAS had a greater area under thecurve (0.812) than the four subscores,age, GCS, BE, and PT (p � 0.001). Theareas under the receiver operating char-acteristic curves for the existing scores,RTS, ISS, NISS, and trauma injury sever-

ity scale, were 0.730, 0.689, 0.734, and0.810, respectively, compared withEMTRAS in 3295 patients. The area un-der the curve was significantly higher forEMTRAS in comparison with RTS, ISS,and NISS. The second patient group(2004–2005) used to validate the scoringsystem only after the EMTRAS was de-fined and publicly presented (9). Fromthese data sets of 3314 patients, the scorecould be computed in 1292 patients. Theresults were very similar between themodel generating and the model valida-tion sets (Table 5).

For further support of the reliability ofthe score, a goodness of fit test of Hosmerand Lemeshow (10) has been carried out.For the model validation set, chi-squarewas 8.6 with 6 degrees of freedom and a pvalue of 0.20. This nonsignificant p value

Table 1. Model generating patient group and model validating patient group

Model Generating Group1993–2003

Model Validating Group2004–2005 p

N 11,533 3,314Age (yrs) 42.3 (19.2) 44.8 (19.6) �0.001% Male 73.4 (44.2) 72.5 (44.6) NSNew injury severity score 36.4 (14.2) 35.6 (14.8) 0.005Mortality (%) 21.9 18.9 �0.001

NS, not significant.Comparison of patient groups used to generate the emergency trauma score (1993–2003) and to

validate the model (2004–2005). Means � SD are displayed.

Table 2. Univariate analysis

Available Percentage

Mean � SD

pContext Parameter Survived Deceased

GeneralAge (yrs) 100% 40 � 18 50 � 22 �0.001

PrehospitalGlasgow Coma Scale 85 11.0 � 4.5 6.7 � 4.6 �0.001Heart rate (per min) 78 95 � 22 91 � 35 �0.001Systolic blood pressure (mm Hg) 79 118 � 29 103 � 49 �0.001Oxygen saturation (%) 20 93 � 8 88 � 17 �0.001Respiratory rate (per min) 50 15.4 � 6.6 11.7 � 8.2 �0.001Revised trauma score 54 6.7 � 1.5 4.7 � 2.3 �0.001

Emergency roomGlasgow Coma Scale 93 7.4 � 5.4 4.3 � 3.3 �0.001Temperature (°C) 5 36.2 � 1.2 35.4 � 2.1 �0.001Heart rate (per min) 92 90.9 � 19.7 92.9 � 35.3 �0.001Systolic blood pressure (mm Hg) 92 124 � 26 105 � 44 �0.001Oxygen saturation (%) 25 97.3 � 4.4 95.1 � 11.1 �0.001Respiratory rate (per min) 28 14.9 � 5.0 14.8 � 8.0 NSBase excess (mmol/L) 50 �2.5 � 4.5 �5.9 � 7.1 �0.001Lactate (mmol/L) 23 4.7 � 7.5 7.4 � 11.1 �0.001Hemoglobin (g/dL) 90 11.3 � 2.9 9.6 � 3.7 �0.001Platelet count (109/L) 85 198 � 80 162 � 80 �0.001Prothrombin time (%) 81 78 � 21 59 � 27 �0.001Activated partial

thromboplastin time (s)53 54 � 40 35 � 16 �0.001

NS, not significant.Relation of individual parameters in model generating group with outcome.

1973Crit Care Med 2009 Vol. 37, No. 6

points to a very small difference betweenexpected and observed mortality.

Although a considerable number ofpatients had missing data, hospital mor-tality was similar in patients with andwithout missing EMTRAS scores, 21%and 22%, in the 1993–2003 group.

DISCUSSION

The EMTRAS was a feasible tool toestimate the mortality risk of patientswith severe trauma. Despite the fact thatit only uses four items that can be deter-mined early, it compared favorably withother more complex scores.

Existing scores either lack physiologiccomponents (e.g., ISS, NISS) or obscure

clinical insight by cumbersome mathe-matics (2–7, 11–18). Scores such as theISS or trauma injury severity scale thatuse post-ER anatomical information didnot perform better in predicting mortal-ity than the EMTRAS, although EMTRASonly uses physiologic information fromthe ER. For example, serious intestinalinjury, which occurred at the moment ofthe accident may be diagnosed anatomi-cally after many hours, although physio-logically an early sharp decrease in BEmay demonstrate severe metabolic acido-sis. Compiling all anatomical trauma di-agnoses inevitably takes more time thanassessing the patient’s age, the GCSscore, and analyzing samples for bloodgas analysis and coagulation. The RTS,which uses physiologic parameters, is noteasy to calculate. In our patients, the RTScould not be determined in many casesbecause the breathing frequency was notrecorded. In a moving ambulance, mea-suring or observing the respiratory rate ismuch more difficult than measuring theheart rate.

Our philosophy in designing theEMTRAS was that this score should helpto appreciate the trauma severity of apatient at the very moment hospitaltreatment has been initiated, even intrauma centers with moderate facilities.The EMTRAS is further intended to fill thegap between scores that use informationavailable on the ambulance vehicle (e.g.,RTS) and scores that can only be deter-mined long after the patient left the ER.

The strength of the EMTRAS score isprobably related to the fact that it incor-porates four components that each areindependently strongly related to mortal-ity, as Table 4 also underscores. Age isused in many scoring systems and in astudy of the relationship of age to trauma-associated mortality (19), an importantcutoff point was an age of 75 years, whichcorresponds with the highest age sub-score for the EMTRAS. The severity ofcoma after trauma as expressed by theGCS has consistently shown a strong re-lationship to mortality (20, 21). Disad-vantages of the GCS are interobservervariability and early prehospital intuba-tion in 94% of the patients with traumawith severe head injury and 64% of allpatients with an ISS �16 in the Germantrauma registry. Because intubated pa-tients reach the hospital deeply sedated,prehospital GCS as evaluated by emer-gency staff was used in the EMTRAS.

It is not surprising that the two phys-iologic parameters of the EMTRAS (BE

Figure 1. Emergency trauma score and mortality. Relation between emergency trauma score andmortality in the patients from the 1993–2003 group. Low scores such as 0 or 1 are associated with verylow mortality, whereas scores above 8 through 12 are associated with a mortality above 80%.

Table 3. Multivariate analysis of remaining fourparameters selected by univariate analysis thatsubsequently showed the strongest independentrelation in binary logistic regression analysiswith hospital mortality as outcome parameter

Variable Coefficient95% Confidence

Interval

Age (yrs) 0.961 0.957–0.965Glasgow Coma

Scale1.188 1.167–1.209

Base excess 1.066 1.049–1.082Prothrombin

time1.034 1.028–1.036

Constant 0.668

A total of 6339 patients were used in thisanalysis.

Table 4. EMTRAS subscores and mortality

Variable Category Subscore Mortality 1993–2003 Mortality 2004–2005

Age (yrs) �40 0 11.8% 9.6%40–60 1 15.6% 12.3%61–75 2 25.7% 22.7%

�75 3 39.8% 36.1%Glasgow Coma Scale 13–15 0 6.6% 4.7%

10–12 1 14.2% 14.0%6–9 2 17.7% 18.6%3–5 3 42.2% 45.9%

Base excess (mmol/L) ��1 0 10.7% 8.5%�1 to �5 1 14.5% 11.7%

�5.1 to �10 2 28.1% 23.0%��10 3 58.9% 51.5%

Prothrombin time (%) �80 0 7.8% 7.5%80–50 1 16.5% 15.9%49–20 2 41.7% 41.8%�20 3 65.3% 68.4%

EMTRAS, emergency trauma score.The four selected parameters each show a strong univariate association with mortality. EMTRAS

was defined as the sum of the subscores for age, Glascow Coma Scale, base excess, and prothrombintime. Thus, the lowest (best) EMTRAS is 0 and the highest (worst) EMTRAS is 12.


and PT) constitute two of the three com-ponents of the so-called triangle of deaththat describe the interaction among aci-dosis, coagulopathy, and hypothermiaduring hemorrhagic shock (22, 23). Therelationship of metabolic acidosis to mor-tality is better reflected by BE than otheracid–base parameters (23, 24). The rea-son that BE performed better than lactatemay be related to the widespread use oflactated Ringer’s solution in the prehos-pital treatment of patients with trauma inGermany (25). For the second compo-nent of the triangle of death, namelyearly coagulation disturbances, a strongrelationship of PT to outcome has alsobeen demonstrated in patients with andwithout severe brain injuries (26). Sur-prisingly, the third component of the tri-angle of death, namely, body tempera-ture, did not emerge as a mortality-predicting parameter. This may berelated to the very large number of miss-ing measurements, as temperature wasonly recorded from 2004.

A hospital that manages severely in-jured patients probably has the means todetermine BE and PT. If these laboratorymeasurements are rapidly performed, thetrauma team can have the EMTRASwithin 30 minutes after admission to theER. The role of a high EMTRAS (e.g., �8)should be to alert the team that the patienthas a high mortality risk. Such informationmay help in considering damage controlsurgery (27). Whether an improvement ofBE and PT will change the prognosis ofpatient’s outcome is not yet clear and hasnot been tested in this study.

In smaller hospitals, EMTRAS mayprovide earlier information in decidingwhether to transfer the patient for defin-

itive care to a level 1 center. It should bestressed that a score such as EMTRASshould by itself never be the reason tostop treatment.

Several avenues of further researchexist. Now that internal validation of theEMTRAS with respect to the GermanTrauma Registry has been demonstrated,it would be particularly important to per-form external validation of the EMTRASin other trauma populations. As our pa-tients mainly had blunt injury, this wouldbe interesting in-patient groups withsharp injuries. EMTRAS can also be de-termined in existing databases in a retro-spective manner, as long as age, prehos-pital GCS, blood gas analysis, and PT havebeen recorded. In prospective studies,EMTRAS could serve treatment “real-time” and also allow comparison of pa-tient groups in research or for auditingpurposes (2).

Our study has several limitations. Adrawback of many patient registries, in-cluding ours, is missing data. This leadsto different subsets when analyzing thedata. For the logistic regression, only aminority of patients could be used be-cause all parameters had to be available.The similar mortalities in patients withan EMTRAS and those without EMTRASsuggest that the bias introduced by miss-ing data was limited. Also, in both datasets the same areas under the receiveroperating characteristic curves were ob-served and the goodness of fit test showeda high reliability of the EMTRAS in thevalidation data set.

In conclusion, EMTRAS is a simple,fast, and robust tool that adequately pre-dicts mortality compared with existingscores that are not suitable for early use

in trauma management. We hope thatthe EMTRAS will be a useful instrumentto assist in the early treatment of patientswith trauma worldwide on the ER.

ACKNOWLEDGMENTS

The authors thank Dr. Michael Rodg-ers for his support in writing this article.

REFERENCES

1. Gunning K, Rowan K: ABC of intensive care:Outcome data and scoring systems. BMJ1999; 319:241–244

2. Oestern HJ, Kabus K: Comparison of varioustrauma score systems. An overview. Un-fallchirurg 1994; 97:177–184

3. Baker SP, O’Neill B, Haddon Jr W, et al: Theinjury severity score: A method for describ-ing patients with multiple injuries and eval-uating emergency care. J Trauma 1974; 14:187–196

4. Boyd CR, Tolson MA, Copes WS: Evaluatingtrauma care: The TRISS method. Traumascore and the injury severity score. J Trauma1987; 27:370–378

5. Tay SY, Sloan EP, Zun L, et al: Comparisonof the New Injury Severity Score and theInjury Severity Score. J Trauma 2004; 56:162–164

6. Champion HR, Sacco WJ, Copes WS, et al: Arevision of the Trauma Score. J Trauma1989; 29:623–629

7. Moore L, Lavoie A, LeSage N, et al: Statisticalvalidation of the Revised Trauma Score.J Trauma 2006; 60:305–311

8. Tscherne H, Regel G: Tscherne Unfallchiru-rgie Traumamangement. ISBN 3–540-61605–5, Berlin, Germany, Springer, 1997

9. Raum M, Nijsten M, Vogelzang M, et al: Dercombined trauma score zur Prognoseeins-chatzung des Polytraumas, eine analyse desTraumaregisters der DGU an 17.200 Pati-enten. In: Oral Presentation Yearly GermanTrauma Congress 2006, Berlin, 2006

10. Hosmer DW, Lemeshow S: Applied LogisticRegression. Second Edition. New York, NY,John Wiley and Sons, 2000

11. Frankema SP, Steyerberg EW, Edwards MJ,et al: Comparison of current injury scales forsurvival chance estimation: An evaluationcomparing the predictive performance of theISS, NISS, and AP scores in a Dutch localtrauma registration. J Trauma 2005; 58:596–604

12. Kilgo PD, Meredith JW, Osler TM: Incorpo-rating recent advances to make the TRISSapproach universally available. J Trauma2006; 60:1002–1008

13. Lavoie A, Moore L, LeSage N, et al: The NewInjury Severity Score: A more accurate pre-dictor of in-hospital mortality than the In-jury Severity Score. J Trauma 2004; 56:1312–1320

14. Meredith JW, Evans G, Kilgo PD, et al: Acomparison of the abilities of nine scoring

Table 5. Areas under the receiver operating characteristic curves

Variable

1993–2003 2004–2005

AUC95% Confidence

Interval AUC95% Confidence

Interval

Age (yrs) 0.613 0.588–0.638 0.679 0.626–0.733Glasgow Coma Scale 0.702 0.679–0.724 0.735 0.686–0.784Base excess 0.637 0.612–0.661 0.620 0.564–0.677Prothrombin time 0.683 0.659–0.707 0.638 0.584–0.692Emergency trauma score 0.812 0.795–0.829 0.828 0.792–0.865Revised trauma score 0.730 0.708–0.752 0.762 0.717–0.807New injury severity score 0.734 0.712–0.756 0.764 0.716–0.813Injury severity score 0.689 0.667–0.712 0.722 0.675–0.768Trauma revised injury severity score 0.810 0.793–0.827 0.851 0.818–0.885

AUC, area under the curve.The AUCs are shown for both the model generating (1993–2003) and the model validating sets

(2004–2005). The results for the two data sets are quite similar, indicating internal validity of theemergency trauma score.


algorithms in predicting mortality. J Trauma2002; 53:621–628

15. Osler T, Baker SP, Long W: A modification ofthe injury severity score that both improvesaccuracy and simplifies scoring. J Trauma1997; 43:922–925

16. Copes WS, Champion HR, Sacco WJ, et al:The Injury Severity Score revisited.J Trauma 1988; 28:69–77

17. Ruchholtz S, Pehle B, Lewan U, et al: Theemergency room transfusion score (ETS):Prediction of blood transfusion requirementin initial resuscitation after severe trauma.Transfus Med 2006; 16:49–56

18. Yucel N, Lefering R, Maegele M, et al:Trauma Associated Severe Hemorrhage(TASH)-Score: Probability of mass transfu-sion as surrogate for life threatening hemor-rhage after multiple trauma. J Trauma 2006;60:1228–1236

19. Kuhne CA, Ruchholtz S, Kaiser GM, et al:Mortality in severely injured elderly traumapatients-when does age become a risk factor?World J Surg 2005; 29:1476–1482

20. Moore L, Lavoie A, Camden S, et al: Statisti-cal validation of the Glasgow Coma Score.J Trauma 2006; 60:1238–1243

21. Gill MR, Reiley DG, Green SM: Interraterreliability of Glasgow Coma Scale scores inthe emergency department. Ann Emerg Med2004; 43:215–223

22. Ku J, Brasel KJ, Baker CC, et al: Triangle ofdeath: Hypothermia, acidosis and coagulopa-thy. New Horizons 1999; 7:61–72

23. Siegel JH, Rivkind AI, Dalal S, et al: Earlyphysiologic predictors of injury severity anddeath in blunt multiple trauma. Arch Surg1990; 125:498–508

24. Rixen D, Raum M, Bouillon B, et al: Basedeficit development and its prognostic signif-icance in posttrauma critical illness: An anal-ysis by the trauma registry of the DeutscheGesellschaft fur unfallchirurgie. Shock 2001;15:83–89

25. Raum M, Rixen D, Linker R, et al: Influenceof lactate infusion solutions on the plasmalactate concentration. Anasthesiol Inten-sivmed Notfallmed Schmerzther 2002; 37:356–358

26. Raum MR, Bouillon B, Rixen D, et al: Theprognostic value of prothrombin time in pre-dicting survival after major trauma: A pro-spective analysis of 1351 patients from theGerman Trauma registry. Eur J Trauma2001; 3:110–116

27. Taeger G, Ruchholtz S, Waydhas C, et al:Damage control orthopedics in patients withmultiple injuries is effective, time saving,and safe. J Trauma 2005; 59:409–416

APPENDIX

Members of the “PolytraumaStudy Group” of the GermanTrauma Society

H. J. Oestern (Celle), M. Wittke (Celle),O. Steitz (Celle), C. Buschmann (Essen), C.

Kuhne (Essen), C. Waydhas (Essen), D.Nast-Kolb (Essen), S. Ruchholtz (Essen),A. D. Bonk (Frankfurt/Mn.), U. Schweigkofler(Frankfurt/Mn.), F. Walcher (Frankfurt/Mn.), I. Marzi (Frankfurt/Mn.), S. Muller(Frankfurt/Mn.), O. Seibert (Frankfurt/Mn.), B. Hildebrand (Hannover), C. Kret-tek (Hannover), C. Probst (Hannover),H. C. Pape, (Hannover/Pittsburgh), J.Westhoff (Hannover), M. Frink (Han-nover), M. Grotz (Hannover), N. Ising(Heidelberg), R. Simon (Heidelberg), U.Krehmeier (Heidelberg), U. Obertacke(Heidelberg), T. Pohlemann (Homburg),M. Burkhardt (Homburg), H. Joachim(Homburg), A. Seekamp (Kiel), B. Bouil-lon (Koln/Witten-Herdecke), D. Rixen(Koln/Witten-Herdecke), T. Paffrath(Koln/Witten-Herdecke), E. Neuge-bauer (Koln/Witten-Herdecke), H.-M.Kleiner (Koln/Witten-Herdecke), R.Lefering (Koln/Witten-Herdecke), S.Sauerland (Koln/Witten-Herdecke), L.Kumpf (Lippstadt), H. Meyer (Ludwig-shafen), C. Lackner (Munchen), G. Kanz(Munchen), L. M. Qvick (Munchen), S. Hu-ber-Wagner (Munchen), W. Mutschler(Munchen), F. Barth (Munchen), M. R.Raum (Groningen, Netherlands).

Participating Departments ofthe Trauma Registry of theGerman Trauma Society inAlphabetical Order of City

Unfallchirurgie der Universitat Aachen,Klinik fur Unfall- und Wiederherstellung-schirurgie des Zentralklinikums Augsburg,Unfallchirurgie und Orthopadie der KlinikBad Hersfeld, Klinik fur Unfall-und Wieder-herstellungschirurgie der Charite—CampusVirchow-Klinikum Berlin, Unfallchirurgiedes Martin—Luther Krankenhauses Berlin,Chirurgische Klinik des Klinikums Ber-lin—Buch, Berufsgenossenschaftliche Un-fallklinik des Krankenhauses Berlin-Mahr-zahn, Unfallchirurgische Klinik derKrankenanstalten Gilead Bielefeld, Chirur-gische Klinik und Poliklinik BG-Klinik Bo-chum Bergmannsheil, UnfallchirurgischeKlinik des Knappschaftskrankenhauses derRuhr-Universitat Bochum, Klinik fur Un-fallchirurgie der Rheinischen Friedrich-Wilhelms-Universitat Bonn, Klinik fur Un-fall-und Wiederherstellungschirurgie desZentralkrankenhauses Sankt-Jurgen-StraeBremen, Zentrum fur Allgemein-und Un-fallchirurgie am Zentralkrankenhauses OstBremen, Unfallchirurgische Klinik desZentralkrankenhauses Reinekenheide Bre-merhaven, Unfallchirurgie des Kreiskran-kenhauses Burg, Unfallchirurgische Abtei-

lung des Allgemeinen KrankenhausesCelle, Klinik fur Unfall-und Gelenkchirur-gie des Klinikum Chemnitz, Klinik fur Un-fall-und Handchirurgie Dessau, Unfall-undWiederherstellungschirurgie am KlinikumLippe-Detmold, Abteilung fur Chirurgiedes Krankenhauses Dresden-Neustadt,Klinik fur Unfall-und Wiederherstellung-schirurgie der Technischen UniversitatDresden, Klinik fur Unfallchirurgie desStadtischen Klinikums Friedrichstadt,Klinik fur Allgemein-und Unfallchirurgieder Heinrich-Heine-Universitat Dusseldorf,Unfall-, Hand-und Wiederherstellung-schirurgie des Klinikums Erfurt, Un-fallchirurgie des Kreiskrankenhauses Es-chwege, Unfallchirurgische Klinik derUniversitat Essen, UnfallchirurgischeAbteilung des Evangelischen Kranken-hauses Lutherhaus Essen, Berufsgenossen-schaftliche Unfallklinik Frankfurt/Main,Zentrum fur Chirurgie Klinik fur Un-fallchirurgie der Universitat Frankfurt/Main, Chirurgische Klinik des KlinikumsFrankfurt/Oder, Unfallchirurgie, orthopa-dische und wiederherstellende Chirurgiedes Klinikums Furth, Chirurgie des Johan-niter-Krankenhauses Geesthacht, Unfall-und Wiederherstellungschirurgie desStadtischen Klinikums Gorlitz, Klinik furAnasthesiologie, Operative Intensivmedizinund Schmerztherapie der Klinik amEichert Goppingen, Zentrum fur Chirurgieder Georg-August-Universitat Gottingen,Unfallchirurgie des Evangelischen Kran-kenhauses Gottingen-Weende, Universita-tsklinik fur Unfallchirurgie Graz, Un-fallkrankenhaus Graz der AllgemeinenUnfallversicherungsanstalt Graz, Chirurgiedes Kreiskrankenhauses Grevenbroich,Klinik fur Traumatologie der UniversitatGroningen, Unfallchirurgische AbteilungKrankenhaus Gummersbach, Unfall-undWiederherstellungschirurgie am Berufs-genossenschaftliches UnfallkrankenhausHamburg, Abteilung fur Unfall- und Wied-erherstellungschirurgie der Universita-tsklinik Hamburg-Eppendorf, AbteilungUnfallchirurgie Kreiskrankenhaus Hameln,Unfallchirurgische Abteilung des Stadtis-chen Krankenhauses Hannover Nordstadt,Unfallchirurgische Abteilung des Friederik-enstifts Hannover, UnfallchirurgischeKlinik der Medizinischen Hochschule Han-nover, Chirurgische Abteilung Kranken-haus Hattingen, Abteilung Orthopadie I derOrthopadischen Universitatsklinik Heidel-berg, Klinik fur Unfall- und Wiederherstel-lungschirurgie des St. Bernward Kranken-hauses Hildesheim, Unfallabteilung desWaldviertel Klinikums Horn, Abteilung furUnfallchirurgie des LKH Judenburg-Knit-


telfeld, Unfallchirurgische Klinik der Uni-versitat Homburg-Saar, Unfall-, Hand- undWiederherstellungschirurgie des Klini-kums Karlsruhe, Klinik fur Unfallchirurgieder Universitat Kiel, Klinik und Poliklinikfur Unfall-, Hand- und Wiederherstellung-schirurgie Universitat zu Koln, Lehrstuhlfur Unfallchirurgie/Orthopadie der Univer-sitat.

Witten/Herdecke am Klinikum Koln-Merheim, Abteilung fur Unfallchirurgie desAllgemeinen Offentlichen KrankenhausesKrems/Donau, Unfall- und Wiederherstel-lungschirurgie des Stadtischen KlinikumsSt. Georg Leipzig, Klinik fur Unfall- undWiederherstellungschirurgie des Universi-tatsklinikums Leipzig, Chirurgische Abtei-lung des Evangelischen KrankenhausesLengerich, Abteilung fur Unfallchirurgiedes Allgemeinen offentlichen Kranken-hauses Linz, Evangelisches KrankenhausLippstadt, Operatives Zentrum I Klinik furChirurgie Universitatsklinikum Lubeck,Berufsgenossenschaftliche UnfallklinikLudwigshafen, Klinik fur Unfall- und Wied-erherstellungschirurgie des St.-Marien-Hospital Lunen, Abteilung fur Unfallchiru-rgie der Otto-v.-Guericke-Universitat

Magdeburg, Klinik fur Chirurgie des Klini-kums Magdeburg Krankenhaus Altstadt,Klinik fur Unfallchirurgie der UniversitatMainz, Abteilung fur Unfallchirurgie desUniversitatsklinikums Mannheim, Klinikfur Unfallchirurgie der Philipps-UniversitatMarburg, Unfallchirurgie des KlinikumsMinden, Unfallchirurgie des Kranken-hauses Maria Hilf Monchengladbach,Chirurgische Klinik Klinikum Großhadernder Ludwig Maximilian UniversitatMunchen, Unfallchirurgische AbteilungKlinikum Munchen-Harlaching, Chirur-gische Klinik Klinikum Innenstadt derLudwig Maximilian Universitat Munchen,Klinik fur Unfall- und Handchirurgie derWestfalischen Wilhelms-Universitat Mun-ster, Berufsgenossenschaftliche Un-fallklinik Murnau, Klinik fur Unfall- undWiederherstellungschirurgie der Stadtis-chen Kliniken Lukaskrankenhaus Neuss,Unfallchirurgie des Marienhospitals Os-nabruck, Unfallchirurgie des VogtlandKlinikums Plauen, Unfallchirurgie desKrankenhauses Lennep Klinikum Remsc-heid, Klinik fur Unfall-und Wiederherstel-lungschirurgie Klinikum Rosenheim,Chirurgie und Unfallchirurgie Sana-

Krankenhaus Rugen, LAndesklinik furUnfallchirurgie und Sporttraumatolo-gie des Landeskrankenhauses Salzburg,Unfallchirurgie DiakoniekrankenhausSchwabisch-Hall, Chirurgische Klinikdes Kreiskrankenhauses Soltau, Un-fallchirurgie im Johanniter-Kranken-haus der Altmark in Stendal, Unfall-undWiederherstellungschirurgie des Krei-skrankenhauses Traunstein, Berufs-genossenschaftliche-Unfallklinik Tu-bingen, Abteilung fur Anasthesiologieund Intensivmedizin des Bundeswe-hrkrankenhauses Ulm, Abteilung furUnfall-, Hand-und Wiederherstellung-schirurgie der Universitat Ulm, Un-fallchirurgische Klinik des KlinikumsVillingen-Schwenningen, Unfallchirur-gie Klinikum Weiden, Unfallchirurgiedes Asklepios Krankenhauses Weien-fels, Abteilung fur Unfallchirurgie undSporttraumatologie des DonauspitalsWien, Unfallchirurgische Klinik derUniversitat Wurzburg, Klinik fur Un-fall- und Wiederherstellungschirurgiedes Klinikums Wuppertal, Unfallchiru-rgische Klinik der Universitat Zurich,Rettungswache Zusmarshausen.


emtras

Documents