Study Objectives: Computed tomography (CT) is a vital adjunct in the evaluation and care of trauma patients. While its usefulness is undisputable, this benefit comes with radiation-related risk given the relatively high doses of ionizing radiation that are used. This concern has generated a debate over the proper role of CT in stable trauma patients. In particular, patients who are awake and alert and have minimal physical complaints after a high energy accident (falling from height or high speed motor vehicle accident) often undergo multiple CT scans “just to make sure” that they don’t have significant injury. While several studies have promoted this approach, to date there has been no well-designed study to exam this practice in this patient population. We addressed the following questions: How closely do physician assessment and CT scan results agree in the alert stable patient who has experienced high energy trauma but has no complaints or physical findings to suggest injury? Can physicians reliably detect severe injuries in this select patient population?
Methods: Design/Setting: This is a prospective cohort study conducted at 3 Level I trauma centers. A convenience sample was enrolled when study personnel were available.
Participants: Patients were included if they met the inclusion criteria: blunt trauma, trauma team activation, Glasgow Coma Score 15, systolic blood pressure on arrival 100, age between 18 and 65. Patients were excluded if they were transferred from another hospital, if they were pregnant or if no CT scan was obtained. Trauma team leaders completed a survey regarding the reliability of the patient and suspicion of any injury and severe injury in various body regions (head, neck, chest, abdomen, pelvis and extremities). The patient’s chart was later abstracted for outcome and injuries detected on x-ray or CT. Major injuries were defined a priori.
Results: 150 patients were enrolled. Mean age was 43 (SD17.6). Mechanisms of injury were: MVA 55%, fall 20%, motorcycle accident 10%, bicycle accident 9%, other 6%. 46% of patients were deemed unreliable mostly because of intoxication or distracting injury. 43% of patients were admitted to the floor and 35% were admitted to the ICU. 18% of patients were discharged home and 2 patients died in the hospital. Among all patients included in the study, there were 89 major injuries, 43 of which were not suspected by the physician. Among the patients that were deemed reliable (n81), there were 11 major injuries that were not detected. These injuries occurred in 5 patients and included sacral fracture, acetabular fracture, C7 facet fracture, T9 fracture and small
intraparenchymal contusions in 2 patients. The calculated negative predictive value of physician assessment ranged from 0.89 to 0.99 (95% CI 0.80 to 0.99). Sensitivity of physician assessment to the presence of any major injury was 0.68 (0.46-0.84). 47 patients were assessed by more than 1 provider. Using the AC1 statistic, there was good inter-observer agreement of patient reliability (0.948 / 0.74) and of major injury assessment (ranging from 0.81 to 1.0 / 0.03 to 0.10).
Conclusion: Even among patients deemed reliable, there were a number of missed major injuries. Physicians should not rely strictly on their assessment to detect major injuries in patients who have experienced high energy trauma. Particular care should be taken among patients who are felt to be unreliable.