Scoring multitrauma patients: which scoring system?

13

Injury (1993) 24,(I), 13-16 Printed in Greul Brikzin

Scoring multitrauma

patients: which scoring system?

J. A. W. Teijink, B. J. D wars, P. Patka and H. J. Th. M. Haarman Department

of General Surgery, Free University

Hospital, Amsterdam,

Two methods of estimaling the severity of injury and evaluating outcome, the Injury Severity Score (1SS)and the Polytrawna-Schlussel(PTS), were evaluated. 7he records of 37 victim.5 of multiple injuries were assessed retrospectively by nine trauma surgeons using both methods of scoring. The agreement among the uws was calculated by standard deviation. The standard deviation (SD/ among users was swzallerfor the PTS (4.1) than for theES (6.2). Each method of scoring comprises six components contributing to the total score.The three highestscoring components of the ISS were first squared and then summated, Ihe sum being the total ISS score.All the PT.5 categories were summated. Comparison of the SD for each of the components was not possible. However, the category or categories which mostly influenced the standard deviation could be indicated. For the 1.5s these were the circulation and central nervous sytem and for PTS, the limbs. We prefer the Polykaumu-SchlusseI method because this takes account of age, needs revision of only one category andgives more consistent results among users.

Introduction The need for indices of severity of injury for control of injury and evaluation of outcome is widely accepted (Champion et al., 1983; Trunkey et al., 1983). Since early 1970, quantification of the severity of trauma for the victim of multiple injuries has yielded a wide range of methods of scoring. Reliable methods of scoring have several advantages both for the patients and for the organization of their care (McKenzie, 1984; Draaisma, 1987). The acceptance and application of one method of scoring on national and international scales would make its use possible for economic and epidemiological purposes. The numerous methods of scoring make it difficult to decide which should be chosen. For the best use of a method of scoring several requirements have to be met with regard to both the user and reliability (Gibson, 1981; Draaisma, 1987). For the user, the method must be easy to apply and the data contributing to the total score must be easy to collect. Reliability requires good correlation with mortality and/or morbidity (high predictive value), good correlation with the prognosis of the patients, and consistent scoring both by different assessors (general reliability) and by the same assessor (individual reliability). We compared two generally used scoring methods - the Injury Severity Score and the Polytrauma-Schlussel. The 8 1993 Butterworth-Heinemann 0020-1383/93/010013-04

Ltd

The Netherlands

Injury Severity Score (ISS) is derived from the Abbreviated Injury Scale (AIS) and the Hospital Trauma Index (HTI) (Baker et al., 1974). The more recent Polytrauma-Schlussel was introduced in the 1980s(Oestem, 1985). Both methods are easy to handle and are comparable with regard to the prognosis of the patient (Oestem et al., 1985). We assessed the general reliability between the users of each of the two scoring methods and also looked at the applicability of each method.

Material and methods The study was based on the records of 3 7 victims of injuries in traffic accidents, that came to our accident and emergency department from July to December 1988. The data collected from the records included a description of the accident, the blood pressure, pulse rate and Glasgow Coma Score on admission and all the initial major diagnoses made in the department. These variables and the criteria for scoring for the PTS and ISS (Table1 and Table II) were provided for the nine surgeons who participated in the study and who are actively involved in the treatment of polytrauma patients. They were asked to score the injuries of the 37 patients retrospectively, using both PTS and ISS. Both the PTS score and the ISS have six components from which the total score is derived. Both the extremities and the abdomen are included in the PTS and ISS. The remaining four components for the PTS are head, thorax, pelvis and vertebral column. For the ISS these components are respiratory system, cardiovascular system, central nervous system (CNS) and the skin and subcutaneous tissue. Summation of the values in each of the six components of the PTS gives the PTS score, which ranges from 0 to well over 70, which is not consistent with life. The ISS is the sum of the squares of the scores for each of the three most severe injuries. Each component can score from 0 to 5 points which means that the maximum is 75 (three components scoring 5’ points). The two methods of scoring were investigated using analysis of variance (Fleiss, 1981).In this analysis the variation between patients and the variation between assessors were separated. For each of these two components a standard deviation (SD) has been computed. The smaller the SD, the greater the reliability. For the applicability of each method of scoring we estimated the range of SDS for the possible diagnoses in each method.

14

Injury: the British Journal of Accident Surgery (1993) Vol. 24/No.

Table I. Polytraurna-SchIussel (PTS) criteria

1

Table II. Injury Severity Score (1%) criteria Points

Head GCS 13-I 5 points GCS 8-l 2 points GCS 3-7 points Facial fracture Severe facial fracture Thorax Fracture(s) of sternum or ribs (l-3) Several rib fractures Several rib fractures, bilateral Haemothorax and/or pneumothorax Contusion of lung Contusion of both lungs Unstable thorax Ruptured aorta

2 5 10 2 7 9 3 7

Abdomen Rupture of spleen Rupture of spleen and liver Severe rupture of liver Injury of intestines, mesentery, kidneys or pancreas

9 13 13 9

Limbs Fracture-dislocation of acetabulum Fracture of femur Cornminuted fracture of femur Fracture of lower limb Injury of ligaments of knee, fracture of patella, fracture of ankle Fracture of humerus, fracture of shoulder Fracture of elbow, fracture of forearm Vascular trauma above elbow or knee Vascular trauma below elbow or knee Amputation through arm or thigh Forearm or leg amputation Second and third degree open fractures Large wound of soft tissue Pelvis and vertebral column Fracture of pelvis Fracture of unstable pelvis Fracture of pelvis and urogential lesion Fracture of spine Paraplegia or tetraplegia Age in years o-39 40-49 50-54 55-59 60-64 65-69 70-74 75 and older

12 8 12 4 2 4 2 8 4 12 8 4 2 3 9 12 3 3 0 1 2 3 : 13 21

Results of the patients (26 men and II women) ranged from 14 years to 76 years (mean 34 years). The mean scores of PTS and ISS given by the nine surgeons for 37 patients appeared to be almost the same (PTS, 29.2; range 18-54 and ISS, 29.7; range 16-57). Analysis of the total scores showed a higher standard deviation among surgeons who used the ES (6.2) than among those using the PTS score (4.1). Since the mean ES and PTS appeared almost equal, we looked for an arbitrary difference of 7.5 points or more between mean PTS and mean ES for the same patient. This was the case in 15 patients (40 per cent). On seven occasions the mean PTS outnumbered the mean ISS for a given patient; on eight occasions the mean ISS outnumbered the mean PTS. The age

Respiratory No injury Minimal findings Simple fracture of rib or sternum, contusion of chest wall with pleuritic pain; contusion of lung First or multiple rib fractures; haemothorax and/or pneumothorax Open chest wound, paradoxical breathing (unilateral); tension pneumothorax; rupture of simple diaphragm; rupture of trachea Aspiration; bilateral paradoxical breathing; multiple ruptures of diaphragm Cardiovascular No injury < 500 ml blood lost, normal skin perfusion 500-I 000 ml blood lost, diminished skin perfusion; myocardial contusion with normal blood pressure 1000-7 500 ml blood lost with blood pressure < 100 mmHg; myocardial contusion with lowered blood pressure; tamponade with normal blood pressure 1500-2000 ml blood lost with blood pressure < 80 mmHg; tamponade with diminished blood pressure > 2000 ml blood lost with blood pressure < 60 mgHg; cardiac arrest due to loss of blood Central nervous system No injury Head injury without loss of consciousness Head injury with short (< 15 min) loss of consciousness; fracture of skull; one facial fracture; cervical pain without fractures Head injury with loss of consciousness (15-60 min); depressed fracture of skull; fractures of neck with or without slight neurological findings; multiple facial fractures. Head injury with coma (> 60 min) or neurological findings; fracture of neck with paraplegia Head injury with coma (>60 h); cervical fracture with quadriplegia Abdomen No injury Mild pain in abdominal wall, flank or back and tenderness without peritoneal signs Acute discomfort and tenderness in flank, back or abdomen; fractures of a rib (7-l 2); haematuria Injury of liver (minor), small bowel, spleen, kidney, body of pancreas; mesentery, ureter or urethra; rib fractures (7-12) Ruptured liver, bladder, head of pancreas, duodenum, colon or mesentery (large) Crush liver, major vascular bleeding including thoracic and abdominal aorta, caval and/or iliac veins, hepatic veins Limbs No injury Minor sprains and fractures (no long bones) Simple fracture: humerus, clavicle, forearm or leg; single nerve; single ligamentous lesion Multiple moderate fractures, open moderate fracture; fracture of femur; stable fracture of pelvis; stable fracture of lumbar or thoracic spine; major nerve lesion; major dislocation Open fracture of femur; limb crushing or amputation; unstable fracture of pelvis; unstable fracture of lumbar or thoracic spine Open (crush) fracture of pelvis Skin and subcutaneous tissue No injury ~5% burn; abrasion; contusion; wounds 5-l 5% burn; extensive abrasion, contusions and wounds 15-30% burn; avulsion (> 30 x 3Ocm) 30-45% burn; avulsion entire limb > 45% burn

0 1 2 3

4 5 0 1 2

3 4 5 0 1

2

3 4 5 0 1 2 3 4 5 0 1 2

3

4 5 0 1 2 3 4 5

Teijink et al.: Which scoring system for multitrauma patients Table V. Trauma scoring methods

Table III. PTS scores: mean and standard deviation

Method

SD

Head Thorax Abdomen Limbs Pelvis/vertebral column Age

Mean

Patients

Users

7.3 4.7 2.3 8.8 1.3 5.1

3.6 4.9 4.8 7.7 2.5 7.8

0.9 2.0 2.1 3.2 1 .o -

Physiological parameters Prognostic index (Cowley et al., 1974) Acute trauma index (Millholland et al., 1979) Triage index (Champion et al., 1980) Trauma score (Champion et al., 1981) Anatomical parameters Abbreviated Injury Scale (Committee, 1971) Anatomic index (Champion et al., 1980) Physiological and anatomical parameters Trauma index (Kirkpatrick, 1971) Injury severity score (Baker et al., 1974; Baker and O’Neill, 1976) Polytrauma-Schltissel (Oestern et al., 1985)

Table IV. ISS scores: mean and standard deviation SD

Respiration Cardiovascular Central nervous system Limbs Skin and subcutaneous tissues

15

Mean

Patients

Users

1.9 2.0 2.3 2.7 0.3

1.6 1.6 1.0 1 .o 0.4

0.3 0.9 0.7 0.4 0.3

of the scores of the different categories with the PTS, the limbs (3.2, TabfellI), and with the ISS, the cardiovascular and central nervous systems (0.9 and 0.7, TableIV) were the main contributors to the total standard deviation among users. The difference in size of both standard deviations can be explained by the design of the ISS (summation of squares). Analysis

Discussion There are three types of scoring methods, those based on physiological or on anatomical parameters and those based on both. The ISS and PTS are based on both physiological and anatomical parameters (Table V). Oestem and colleagues investigated the ultimate prognosis for three scoring systems (AIS, ISS and PTS) and found almost similar predictive values for PTS (74 per cent) and ISS (73 per cent) (Oestem et al., 1985). For IS patients, a marked difference in the mean PTS and ISS per patient, as assessed by the nine surgeons, showed that each of the two scoring methods has its own characteristics, related to the type of injury. The mean of all ISS and PTS scores appeared identical. The difference of 7.5 points or more between the mean PTS and ISS for one patient was studied in detail. A higher mean PTS was seen for elderly patients, patients with multiple fractures and patients with more than three injured categories. A higher mean ISS was reached for patients who were in hypovolaemic shock, suffered from severe head injury or had severe injury of the thorax. Eight patients who scored in the ISS category abdomen, did score but did not do so for this category in the PTS. The standard deviations for both the PTS and ISS scores were considerable, although the standard deviation between the patients and that between the surgeons for the ISS score (11.6 and 6.2) exceeded the PTS score (9.8 and 4.1). Since the way in which the total score of the ISS is reached is quite different from that for the PTS, the standard deviations of the categories cannot be compared. The category age, obviously no cause for inter-disagreement (SD O.O), is one of the six PTS categories. This important parameter is no part of the ISS.

Variations between the methods are also influenced by the quality and clarity of the criteria for scoring. Comparison of both methods of scoring shows that the ISS scale is far more detailed, with more than 100 diagnoses, the PTS with fewer than 50).

Conclusions The PTS shows a smaller deviation, both between patients and between surgeons. In the PTS, the limbs are mainly responsible for the existing disagreement among the surgeons. For the ISS this is primarily the ‘cardiovascular’ category. In spite of the far more refined ISS scaling dictionary, this does not seem to contribute to a better general reliability. Based on these findings we prefer the PolytraumaSchlussel (PTS) scoring method.

Acknowledgements The authors wish to thank the nine trauma-surgeons at the Free-University Hospital, the AMC University Hospital and the OLVG Hospital in Amsterdam, the Red Cross Hospital in Beverwijk and the MCA Hospital in Alkmaar who spontaneously gave their time to score the 37 patients with both scoring methods. We also thank Drs J. P. Nauta for his statistical remarks.

References Baker S. P. and O’Neill B. (1976) The injury severity

score: an update. 1, Trauma 16, 882. Baker S. P., O’Neill B., Haddon W. et al. (1974) The Injury Severity Score: a method for describing patients with multiple injuries and evaluating trauma care. J. Trauma 14, 187.

Champion H. R., Sacco W. J., Camazzo A. J. et al. (1981) Trauma Score. Crit. Care Med. 9, 672. Champion H. R., Sacco W. J., Hannan D. S. et al. (1980) Assessment of injury severity: the triage index. Crit. Care Med. 8,201. Champion H. R., Sacco W. J. and Hunt T. K. (1983) Trauma severity scoring to predict mortality. World 1. Surg. 7,4. Champion H. R., Sacco W. J., Lepper R. L. et al. (1980) An anatomic index of injury severity. 1. Trauma 20,97. Committee on Medical Aspects of Automotive Safety (1971) Rating the severity of tissue damage: I The Abbreviated Scale. lAiX.44 215,277.

16

Injury: the British Journal

Cowley R. A., Sacco W. J., Gill W. et al. (1974) A prognostic index for severe trauma. J. Trauma14, 1029. Draaisma J. M. Th. (1987) Evaluation of trauma care. With emphasis on hospital trauma care. Dissertation, Catholic University, Nijmegen, The Netherlands. Fleiss J. L. (1988) 77~ Design and Analysis of ClinicalErperimenfs. New York: John Wiley & Sons. Gibson G. (1981) Indices of severity from medical e&ration studies: reliability, validity and data requirements. Infernat. J HealthServ. 4,597. Kirkpatrick J. R. and Youmans R. L. (1971) Trauma Index: an aid in the evaluation of injured victims. J. Trauma2, 711. MacKenzie E. J. (1984) Injury Severity Scales: overview and directions for future research. Am. J. Emerg.Med. 2,537.

ADVERTISE MAKE

of Accident

Surgery

(1993)Vol. 24/No.

Millholland A. V., Cowley R. A. and Sacco W. J. (1979) Development and prospective study of an anatomical index and an acute trauma index. Am. Surg.45, 246. Oestern H. J., Tscheme H., Shum J. et al. (1985) Klassifiziemng der Verletzungsschwere. Unfakhinug. 88, 465. Trunkey D. D., Siegel J., Baker S. P. et al. (1983) Panel: Current status of trauma seventy indices. J. Trauma 23, 185.

Paperaccepted30 May 1992. Requestsfor reprintsshouldbe addressedfo: Professor Dr H. J. Th. M. Haarman, Department of General Surgery, Free University Hospital, PO Box 7057, 1007 MN Amsterdam, The Netherlands.

IN THIS JOURNAL IT WORK FOR YOU

This journal is highly specific and enables companies to advertise their product, service or event to a well defined and attentive audience within a perfectly-tailored editorial environment. Butterworth-Heinemann journals provide: . Precision targeting l

Long ‘shelf life’- advertising goes on working, reinforcing your sales message l

Full details

1

High pass-on readership

of advertisement rates, mechanical data, and copy dates can be obtained MTB Advertising, 1 1 Harts Gardens, Guildford, Telephone: +44 (0) 483 578507 Fax: +44

information on circulation from: Surrey GU2 6QA UK (0) 483 472678