Keith Wesley, MD, FACEP Review of:Purtill MA, Benedict K, Hernandez-Boussard T, et al: "Validation of a prehospital trauma triage tool: a 10-year perspective." Journal of Trauma 65(6):1253- 1257, 2008.
The Science
This study from Santa Cruz, Calif. is a rigorous attempt to validate a change in the way this EMS system triaged multiple trauma victims from the scene in rural areas. Their protocol was based on what they term "MAP" for mechanism, anatomy, and physiological changes found on triage. Their triage tool consisted of the following areas:
Mechanism of Injury
___High Impact
Specific High Impact Criteria
___Ejection of Patient
___Roll over
___Fatality in same vehicle
___Intrusion of MV into passenger compartment
___Extrication over 20 min
___Pedestrian hit at 20 mph or more
___Gunshot wound
___Stab Type wound
___Fall > 15 feet
___Submersion Injury (for pediatrics only)
Anatomic Injury (or Injuries)
___Significant Penetrating Injury
___Significant Blunt Injury
___Burns
Specific Burn Criteria
___>10% Body in Children < 1 year
___>15% Body in all others
___Burns to Face/Mouth/Throat
___Singed Nasal Hair
___Respiratory Distress/Cough
___Deep Burns to hands/feet/perineum
___Neuro Injury
Specific Neuro Injury
___Sensory Loss
___Motor Deficit
___Paralysis
Physiologic Criteria
____Glasgow Coma Score <11
___Inadequate Perfusion
___Respiratory Distress
___Unable to Determine Physiology (Pediatric <6)
Other Criteria
___Base Hospital Physician Judgment
___Patient "In extremis"
Prior to the study (control period), based-station medical control was contacted for approval for air medical transport of the patient to the trauma center. During the study period, there was no change in the triage tool. If patients met two or more criteria, the EMTs could summon the helicopter or transport directly to the trauma center without base station approval. If only one criterion was met, they could contact the based-station for approval to transport to the trauma center.
During the control period, there were 6,670 trauma patients. Based-station contact was made 95% of the time, and 7% of the patients were transported to the trauma center. During the study period, 8,414 trauma patients were encountered. Based-station contact was made for 39%, and 10% of the patients were transported to the trauma center.
An analysis of the patients who were transported to the trauma center revealed that 79% of the patients taken to the trauma center during the control period were determined to have multiple trauma as compared with 69% during the study period (p=N.S.). Of those patients not meeting triage criteria in the control period, 75% were later found to have multiple trauma -- compared to 69% during the study period.
This resulted in a triage tool with 93.8% sensitivity and 99.5% specificity with a marked reduction in the need for based-station contact.
The Street
This is a phenomenal study. The issue of prehospital trauma triage is one fraught with great controversy. This system is using a set of criteria that pre-dates the current criteria recently released by the American College of Surgeons (ACS), which focuses more on physiology and less on mechanism and anatomy. It hasn't been validated as this tool has been. The ACS states that an acceptable under-triage rate (one that misses the multiple trauma victim) is 5% while acceptable over-triage is 30%-50%. This system's tool resulted in an over-triage of 31%.
What was unclear was the number of patients who referred to the trauma center with only one criterion after based-station referral. An analysis of these patients would be interesting. Clearly, the elimination of based-station contact resulted in a very efficient system of rapidly moving these rural trauma victims from the scene.
No triage tool is perfect. The goal of any system is to adopt a tool that meets the resources of its region, is consistent with the science as we understand it, and doesn't over burden the trauma center with minor trauma -- all while keeping under-triage to a bare minimum and putting into place a mechanism to address any deficiencies in the system.