s From Podium Presentations To Be Presented At The Trauma Association Of Canada Meeting, Montreal, Canada, September 1999

BACKGROUND Regionalization of trauma care services in our region was initiated in 1993 with the designation of four tertiary trauma centers. The process continued in 1995 with the implementation of patient triage and transfer protocols. Since 1995, the network of trauma care has been expanded with the designation of 33 secondary, 30 primary, and 32 stabilization trauma centers. In addition, during this period emergency medical personnel have been trained to assess and triage trauma victims within minimal prehospital time. The objective of the present study was to evaluate the impact of trauma care regionalization on the mortality of major trauma patients. METHODS This was a prospective study in which patients were entered at the time of injury and were followed to discharge from the acute-care hospital. The patients were identified from the Quebec Trauma Registry, a review of the records of acute-care hospitals that treat trauma, and records of the emergency medical services in the region. The study sample consisted of all patients fulfilling the criteria of a major trauma, defined as death, or Injury Severity Score (ISS) > 12, or Pre-Hospital Index > 3, or two or more injuries with Abbreviated Injury Scale scores > 2, or hospital stay of more than 3 days. Data collection took place between April 1, 1993, and March 31, 1998. During this period, four distinct phases of trauma care regionalization were defined: pre-regionalization (phase 0), initiation (phase I), intermediate (phase II), and advanced (phase III). RESULTS A total of 12,208 patients were entered into the study cohort, and they were approximately evenly distributed over the 6 years of the study. During the study period, there was a decline in the mean age of patients from 54 to 46 years, whereas the male/female ratio remained constant at 2:1. There was also an increase in the mean ISS, from 25.5 to 27.5. The proportion of patients injured in motor vehicle collisions increased from less than 45% to more than 50% (p < 0.001). The mortality rate during the phases of regionalization were: phase 0, 52%; phase I, 32%; phase II, 19%; and phase III, 18%. These differences were clinically important and statistically significant (p < 0.0001). Stratified analysis showed a significant decline in mortality among patients with ISS between 12 and 49. The change in mortality for patients with fatal injuries (ISS > or = 50) was not significant. During the study period, the mean prehospital time decreased significantly, from 62 to 44 minutes. The mean time to admission after arrival at the hospital decreased from 151 to 128 minutes (p < 0.001). The latter decrease was primarily attributable to changes at the tertiary centers. The proportion of patients with ISS between 12 and 24 and between 25 and 49 who were treated at tertiary centers increased from 56 to 82% and from 36 to 84%, respectively (p < 0.001). Compared with the secondary and primary centers, throughout the course of the study the mortality rate in the secondary and tertiary centers showed a consistent decline (p < 0.001). In addition, the mortality rate in the tertiary centers remained consistently lower (p < 0.001). The results of multivariate analyses showed that after adjusting for injury severity and patient age, the primary factors contributing to the reduced mortality were treatment at a tertiary center, reduced prehospital time, and direct transport from the scene to tertiary centers. CONCLUSION This study produced empirical evidence that the integration of trauma care services into a regionalized system reduces mortality. The results showed that tertiary trauma centers and reduced prehospital times are the essential components of an efficient trauma care system.

s Of Papers To Be Presented At The Joint Meeting Of The American Association For The Surgery Of Trauma (Fifty-Eight Annual Meeting) And The Trauma Association Of Canada, September 24-26, 1998, The Renaissance Harborplace Hotel, Baltimore, Maryland

Background The purpose of the study was to compare the outcome of severely injured patients who were transported directly to a Level I, tertiary trauma center with those who were transferred after being first transported to less specialized hospitals. Methods The data were based on all patients treated at three tertiary trauma centers in Quebec between April 1, 1993, and December 31, 1995. There were 1,608 patients (37%) transferred and 2,756 patients (63%) transported directly. Results The mean age of the patients was approximately 45 years, and more than 60% were males. The predominant mechanisms of injury were falls and motor vehicle crashes. The transfer and direct transport groups were similar with respect to age, gender, and mechanism of injury. Body regions injured were also similar with the exception of head or neck injuries (transfer, 56%; direct, 28%; p < 0.0001). The mean Injury Severity Score was 14, the mean Pre-Hospital Index score was 5.5, and the mean Revised Trauma Score was 7.5. The two groups were similar with respect to these injury severity measures. The primary outcome of interest was mortality described as overall death rate, death rate in the emergency room, and death rate after admission. Other outcomes studied were hospital length of stay and duration of treatment in an intensive care unit. When compared with the direct transport group, transferred patients were at increased risk for overall mortality (transfer, 8.9%; direct, 4.8%; odds ratio, 1.96; 95% confidence interval (CI) = 1.53-2.50), emergency room mortality (transfer, 3.4%; direct, 1.2%; odds ratio, 2.96; 95% CI = 1.90-4.6), and mortality after admission (transfer, 5.5%; direct, 3.6%; odds ratio, 1.57; 95% CI = 1.17-2.11). All of these differences were statistically significant (p < 0.003). Stratified and multiple logistic regression analysis did not alter these results and failed to identify a patient subgroup for which transfer was associated with a reduced risk of mortality. After adjusting for patient age, Injury Severity Score, and presence of injuries to the head or neck and extremities, transferred patients stayed significantly longer in the hospital and the intensive care unit as indicated by the mean length of stay (transfer, 16.0 days; direct, 13.2 days; p = 0.02) and the mean intensive care unit stay (transfer, 2.0 days; direct, 0.95 days; p = 0.001). Conclusion The results of this study have shown that transportation of severely injured patients from the scene directly to Level I trauma centers is associated with a reduction in mortality and morbidity. Further studies are required for the evaluation of transport protocols for rural trauma. Economic and cost-effectiveness considerations of patient triage are also essential.

s Of Papers To Be Presented At The Joint Meeting Of The Trauma Association Of Canada And The Royal College Of Physicians And Surgeons Of Canada, September 25-28, 1997, The Pan Pacific Hotel, Vancouver, British Columbia, Canada

s Of Papers To Be Presented At The Joint Meeting Of The Trauma Association Of Canada And The Royal College Of Physicians And Surgeons Of Canada, September 25-28, 1997, The Pan Pacific Hotel, Vancouver, British Columbia, Canada

The study is based on 44 preventable deaths occurring in a cohort of 360 patients with major trauma. These cases were reviewed by a committee of nine experts. The mean Injury Severity Score (ISS) was 28, and most cases had injuries to the head/neck (68%) and chest (64%). The mean (+/-SD) observed prehospital times, and those considered the maximum allowable by the committee, were 40.6 +/- 12.0 minutes for head/neck injuries and 23.9 +/- 12.2 minutes for chest injuries (p < 0.05). Intravenous (IV) lines were started in 38 (86%) of the patients. The committee classified this procedure as harmful for 16 (42%) and neutral for 19 (50%). Among the 18 (46%) that were intubated, this intervention was considered harmful for 17% and neutral for 39%. In two of the three patients for whom a pneumatic antishock garment was applied, this procedure was considered harmful. Of the 34 patients that required direct transport at a level I trauma center, 50% were transferred to such a hospital. These results show significant prehospital delays and high rates of inappropriate IV line initiation and intubation in trauma patients receiving on-site care by physicians. We conclude that prehospital care protocols for trauma patients should emphasize prompt transport and specific on-site care algorithms.

The diagnosis of blunt cardiac injury in traumatized patients is problematic and the implications of such a diagnosis are not clear. Although cardiac selective creatine kinase (CK-MB) assays and electrocardiograms (EKG) are the most widely available laboratory investigations, they often correlate poorly with diagnoses made on clinical grounds, or by other laboratory methods. We therefore retrospectively studied the Montreal General Hospital experience with 342 consecutive blunt trauma patients admitted to our surgical intensive care/trauma unit. Using clinical criteria, cardiac injury was diagnosed in 44 patients (13%). Twenty-seven of these patients (61%) developed arrythmias or cardiogenic hypotension, half of which required treatment. Heart injuries contributed to six of the 12 deaths in this group. Many of the patients maintained normal CK-MB levels and/or had normal admission EKG's despite the clinical diagnosis of cardiac injury. However, using our criteria for CK-MB positivity, there was a strong correlation between CK-MB elevation and the development of cardiac complications, and very high CK-MB levels (greater than 200 mu/L) were associated with a 100% incidence of such complications. Focusing on patients who developed cardiac complications serious enough to require treatment, we found combined CK-MB/EKG positivity in all cases (100% sensitivity). This method also provided a negative predictive value of 100%. We conclude that although blunt cardiac injury is an important source of morbidity and mortality its 'diagnosis' is not the issue. Rather, it is more important to recognize which of these clinically identified 'high-risk' patients will actually develop cardiac complications. We feel our approach will enable clinicians to do this.