Impact of Advanced Trauma Life Support Training on Early Trauma Management Judith A. Vestrup, MO, A n n e Stormorken, BS, and Victor Wood, MD, Vancouver, British Columbia
Since its introduction in 1979, Advanced Trauma Life Support (ATLS) has become a widely used method of teaching assessment and early management skills for trauma care [•]. As of February 1984, a total of 1,032 provider and 383 instructor courses had been offered, with a total enrollment of 19,686 physicians [2]. Despite the widespread popularity of these courses in the United States and Canada, little is known about the impact of ATLS training on trauma care. In fact, the few published articles on ATLS deal with either the animal laboratory or the influence of the type of medical training on performance in the course itself [3,4]. Ornato et al [5] made reference to ATLS as one of the recent improvements in emergency trauma care, but then went on to conclude that the major factor in reducing trauma mortality was improved prehospital care. In an effort to evaluate the impact of ATLS on early trauma care, a retrospective study of trauma patients admitted to the Vancouver General Hospital, which functions as a Level I trauma center, was undertaken. Time periods before and after ATLS training for key members of the trauma service were selected for comparison. Methods
Charts of patients treated before (April 1983 to March 1984) and after (April 1985 to March 1986) the institution of ATLS training were reviewed. In the intervening year, all general surgeons involved in trauma care completed ATLS instructor courses. ATLS provider courses were also made mandatory for senior general surgeons and orthopedic residents on the trauma team. A majority of full-time emergency room physicians also completed ATLS programs. A scoring chart utilizing the Abbreviated Injury Scale was used to assign injury severity scores (ISS) [6]. All patients with an ISS of 14 or greater were reviewed. Because of inadequacies in our emergency room record-keeping system, there were limitations in our ability to accurately assess the timeliness and appropriateness of airway management and resuscitation from shock in all From the Department of Surgery, University of BritishColumbia,Vancouver, British Columbia. Requests for reprints should be addressed to Judith A. Vestrup, MD, Department of Surgery, Third Floor, Laurel Pavilion, 910 West 10th Avenue, Vancouver, British Columbia, Canada, V5Z 4E3. Presented at the 74th Annual Meeting of the North Pacific Surgical Association,Vancouver, BritishColumbia, November 13 and 14, 1987.
704
patients. Therefore, more reliable endpoints, such as documentation of urinalysis, cystography, and intravenous pyelography (the latter for patients with hematuria) and rectal examinations performed in blunt trauma patients were selected for study. All charts were also reviewed for evidence of delay in diagnosis of intraabdominal injury. These procedures and documentation of delay in diagnosis represent a sampling of performance evaluation rather than an exhaustive study of all indicated behaviors and procedures. A more detailed review of patient deaths was performed by one of us (JAV), looking for evidence of failures in assessment and resuscitation that potentially contributed to death. Autopsies were performed on all patients who died in both study periods. Statistical methods employed included the unpaired Student's t test, chisquare test, and Fisher's exact test. p <0.05 was considered significant. Results
There were 50 patients in the before ATLS group, with a mean age of 41.6 years and an ISS of 29.8 as compared with 71 patients in the after ATLS group, with a mean age of 40.6 years and an ISS of 30.6. Forty-nine before ATLS patients (98 percent) and 62 after ATLS patients (87 percent) sustained blunt trauma. Motor vehicle accidents predominated, accounting for 42 injuries in the before ATLS group and 49 injuries in the after ATLS group. Falls, assaults, and industrial accidents accounted for the remaining cases of blunt trauma. There was one penetrating injury in the before ATLS group and nine in the after ATLS group. Of the 13 deaths before ATLS (26 percent), the mean patient age was 44.8 years and the mean ISS was 36. This was not significantly different from the 14 deaths (20 percent) after ATLS, with a mean patient age of 50.1 years and a mean ISS of 36.6. The data on adequacy of early evaluation for all patients are detailed in Table I. The delay in diagnosis of intraabdominal injuries was not significantly different in the two groups. The delay was an important contribution to outcome, however, resulting in the death of one patient in the before ATLS group and of two of the four patients who died in the after ATLS group. Assessment of the genitourinary system was inadequate in 12 to 29 percent of the patients, with no difference between
The American Journal of Surgery
A d v a n c e d T r a u m a Life Support Training
the before and after ATLS groups. There was, however, a statistically significant increase in the number of rectal examinations performed on patients with blunt trauma after ATLS. In the before ATLS group, four patients died from hemorrhagic shock, with two of these patients without vital signs on arrival. Eight deaths were primarily attributable to central nervous system injury with one of them also related to shock and two to central nervous system injury in combination with sepsis. The remaining patient died from adult respiratory distress syndrome. In the after ATLS group, death was attributable to shock in six patients, one of whom was also hypothermic. An additional six patients died from central nervous system injury. The remaining two patients died from sepsis and multisystem organ failure. With respect to detailed analysis of patient deaths, errors in assessment or management occurred in five patients (38 percent) before ATLS (Table II). In two of these patients (15 percent), death was deemed to be preventable. Both patients had an ISS below the LD~0 for their age group [7]. After ATLS, assessment or management errors occurred in six patients (43 percent) (Table II). In five of these patients (36 percent), inadequate care was considered to have contributed significantly to death; however, two of these patients had an ISS substantially above the LDs0 for their age group; therefore, the actual preventable mortality rate was 21 percent. There was a total of nine assessment or management errors in the before ATLS group compared with 10 errors in the after ATLS group. There were four errors related to the airway and ventilation, all occurring in the before ATLS patients; however, there were four significant missed injuries in the after ATLS patients and only one in the before ATLS patients. Inadequate or delayed resuscitation occurred in three of the before ATLS deaths, and in four of the after ATLS deaths. There was one delayed laparotomy in the before ATLS group and two in the after ATLS group.
Comments Despite the fact that ATLS has been widely taught, little is known about its impact on physician performance in the golden hour of trauma care, for which the course was designed. In one of the landmark studies of quality of trauma care, Moylan et al [8] theorized that "many of the management errors should be preventable with increased physician education, particularly education related to treatment of the cardiovascular and respiratory systerns." In the present study, we failed to show an improvement in overall quality of care after ATLS training was given to key members of the trauma service; however, some behaviors taught in ATLS training did result in improved performance. For example, rectal examination was more consistently
Volume 155, May 1988
TABLE I
Adequacy of E a r l y Evaluation ATLS Training Before After n % n %
Delayed diagnosis of 1 of intraabdominal injuries Blunt trauma without urinalysis 6 of Hematuriawithout cystography 4 o f & intravenous pyelography Blunt trauma without 11 of rectal examination Pelvic fracture without 4 of rectal examination
50
2
4 of 71
6"
49 12 14 29
12 of 62 5 of 26
19" 19"
49
22
4 of 62
61
23
17
1 of 26
4*
* Difference not statistically significant. t Statistically significant difference (p = 0.03). ATLS = Advanced Trauma Life Support.
performed after ATLS training. In patients who died from their injuries, the frequency of delayed or inadequate resuscitation was the same before and after ATLS training. Airway management errors did not occur in the after ATLS deaths, and this may be attributable to ATLS improving airway skills. However, missed injuries occurred more frequently after ATLS. There was no improvement in mortality rates with 26 percent of patients dying in the before ATLS group compared with 20 percent in the after ATLS group. Overall, therefore, no quantifiable impact of ATLS training was observed. It is not surprising that an improvement in early management of patients was not seen after the introduction of ATLS training. A number of studies on knowledge retention rates and performance after cardiac life support training have shown a marked deterioration in proficiency over time [9,10]. For example, only 20 percent of persons trained in basic life support were still proficient i year later [10]. In another study of physicians given basic life support training, both knowledge and skills returned to pretraining levels i year later [11]. Interestingly, Stross [12] found that after physician training in advanced cardiac life support, motor skills were maintained at more than double the rate of other physicians by those teaching cardiopulmonary resuscitation or participating in a number of cardiac arrest cases. As early as 1979, Sims [3] commented that the educational benefit of trauma laboratory training as determined by formal educational testing had yet to be established. Recently, Girotti [13] raised a question as to whether the ATLS course made a difference to trauma care. To our knowledge, we were the first to attempt to address these issues. The impact of ATLS training in the present study appears to be modest. This may be a function of the relatively small trauma volume in our institution, providing insufficient opportunity for adequate maintenance of assessment and resuscitation skills. Dearden and Rutherford [14], in finding a 58 percent manage-
705
Vestrup et al
TABLE II Patient Age (yr)
Trauma Deaths Before and After Advanced Trauma Life Support (ATLS) Training
ISS
Cause
Potentially Preventable Death
Errors Before ATLS
48 87 79 25 67 49 50 65 21 20 30 22 18
29 50 50 32 27 34 59 25 36 29 38 25 34
Shock Shock CNS injury Shock Shock CNS injury, shock CNS injury CNS injury CNS injury CNS injury, meningitis CNS injury ARDS CNS injury, bronchopneumonia
Delayed transfusion
No
Delayed intubation & resuscitation Delayed transfusion & laparotomy
Yes Yes
Delayed hyperventilation ...
No
Delayed intubation
No ...
After ATLS 83 28
34 41
Shock Shock
66 82 47
38 34 54
CNS injury CNS injury Shock
84
41
Shock
58 19
33 41
CNS injury Shock
45 14 44 31 28 72
20 29 29 66 25 27
Shock, hypothermia CNS injury CNS injury CNS injury Sepsis MOSF
Missed hemothorax Delayed resuscitation & missed diaphragm rupture . . . . . . . . . . Exsanguinated from cardiac & hepatic injuries Delayed transfusion, missed ruptured diaphragm, & aortic laceration . . . . . No peritoneal lavage, delayed recognition of liver laceration, & inadequate resuscitation Delayed resuscitation & laparotomy . . . . . . . . . . . . . . . . . . . . . . . . .
Yes Yes . . No Yes . Yes
Yes . . . . .
ARDS = adult respiratory distress syndrome; CNS = central nervous system; ISS = injury severity score; MOSF = multiple organ system failure•
ment error rate in a trauma audit of similar size to ours, also raised concerns about physicians' ability to gain expertise when patient volume is small. Since the ATLS instructor manual [1] specifically states that the "course is targeted at the physician who does not deal with major trauma on a day-today basis," this may still represent a major deficiency in achieving the stated educational goals for the course. The other possibility is that our failure to demonstrate an impact of ATLS training on early trauma care is a result of an inevitable decrement in knowledge and skills after training. Limitations to the present study include its retrospective nature and the inadequacy of chart records in providing full information on resuscitation variables. Since airway management and resuscitation from shock are key elements in the ATLS program, further studies are needed to assess the impact of the course in these areas. To this end, prospective trauma audit [15] and the Major Trauma Outcome
706
Study [16] should prove to be invaluable tools. Further, studies directed at evaluating the retention rates of ATLS education and determining the impact on motor skills are needed. Such studies could potentially lead to strategies for improving knowledge retention and psychomotor skills, which are so necessary in the early management of trauma patients.
Summary To assess the impact of ATLS education on early trauma management, charts of patients with an ISS of 14 or greater were reviewed for a 1 year period before and after ATLS training of emergency room trauma care providers. There were 50 patients in the before ATLS group, with a mean age of 41.6 years and an ISS of 29.8, and 71 patients in the after ATLS group, with a mean age of 40.6 years and an ISS of 30.6. Of those parameters evaluated as measures of early assessment, only rectal examination
The American Journalof Surgery
Advanced Trauma Life Support Training
was found to be performed significantly more frequently after ATLS training. The mortality rates of 26 percent before ATLS and 20 percent after ATLS were not significantly different. In evaluating assessment and management parameters in the patients who died, no airway management errors were found in the after ATLS group; however, there were more missed injuries in this group. We have concluded that ATLS instruction failed to produce a quantifiable improvement in patient assessment or outcome. Further studies directed at assessing the retention rate for ATLS education and determining the impact on clinical performance are needed. References 1. American College of Surgeons' Committee on Trauma. Advanced trauma life support instructor manual. Chicago: American College of Surgeons, 1984. 2. Advanced trauma life support newsletter. Chicago: American College of Surgeons, 1984. 3. Sims JK. Advanced trauma life support laboratory: pilot implementation and evaluation. Ann Emerg Med 1979; 8: 150-3. 4. Aprahamian C, Nelson KT, Thompson BM, Malangoni MA, Schneider TC. The relationship of the level of training and area of medical specialization with registrant performance in an advanced trauma life support course. J Emerg Med 1984; 2: 137-40. 5. Ornato JP, Craren EJ, Nelson NM, Kimball KF. Impact of improved emergency medical services and emergency
Volume 155, May 1988
trauma care on the reduction in mortality from trauma. J Trauma 1985; 25: 575-9. 6. Greenspan L, McLellan BA, Greig H. Abbreviated injury scale and injury severity score: a scoring chart. J Trauma 1985; 25: 60-4. 7. Baker SP, O'Neill B, Haddon W, Long WB. The injury severity Score: a method for describing patients with multiple injuries arid evaluating emergency care. J Trauma 1974; 14: 187-96. 8. Moylan JA, Detmer DE, Rose J, Schulz R. Evaluation of the quality of hospital care for major trauma. J Trauma 1976; 16: 517-23. 9. Sternbach GL, Kiskaddon RT, Fossel M, Eliastam M. The retention of cardiopulmonary resuscitation skills. J Emerg Med 1984; 2: 33-6. 10. Wynne G. ABC of resuscitation. Training and retention of skills. Br Med J 1986; 293: 30-2. 11. Gass DA, Curry L. Physicians' and nurses' retention of knowledge and skill after training in cardiopulmonary resuscitation. Can Med Assoc J 1983; 128: 550-1. 12. Stross JK. Maintaining competency in advanced cardiac life support skills. JAMA 1983; 249: 3339-41. 13. Girotti MJ. Letter. Can J Surg 1987; 30: 228. 14. Dearden CH, RutherfOrd WH. The resuscitation of the severely injured in the accident and emergency department: a medical audit. Injury 1985; 16: 249-52. 15. Shackford SR, Hollingsworth-Fridlund P, McArdle M, Eastman AB. Assessing quality in a trauma system. The medical audit committee: composition, costs and results. J Trauma 1987; 27: 866-73. 16. Champion HR, Frey CF, Sacco WJ. Determination of national normative outcomes for trauma (abstr). J Trauma 1984; 24: 651.
707