Paramedic evaluation of clinical indicators of cervical spinal injury

Purpose. Standard prehospital practice includes frequent immobilization of blunt trauma patients, oftentimes based solely on mechanism. Unnecessary cervical spine (c-spine) immobilization does have disadvantages, including morbidity such as low back pain and splinting, increased scene time and costs, and patient-paramedic conflict. Some emergency physicians (EPs) use clinical criteria to clear trauma patients of c-spine injury. If paramedics were able to apply clinical criteria in the out-of-hospital setting, then unnecessary c-spine immobilization could be safely avoided. The authors designed a prospective, randomized, simulated trial to determine the level of agreement between paramedic and EP assessments of clinical indicators of c-spine injury, hypothesizing that there would be substantial agreement between them. Methods. A convenience sample of ten paramedics and ten attending EPs participated. Ten standardized patients, with various combinations of positive and negative findings, were examined simultaneously by EP-paramedic pairs. Each pair evaluated five randomly assigned patients for six clinical criteria: 1) alteration in consciousness, 2) evidence of intoxication, 3) complaint of neck pain, 4) cervical tenderness, 5) neurologic deficit or complaint, and 6) distracting injury. If any criterion was positive, clinical clearance was considered to have failed, and the simulated patient would have been immobilized. Fifty pairs of examinations were performed. The kappa statistic was utilized to determine level of agreement between the two groups for each criterion and for the immobilization decision. A kappa of 0.40 to 0.75 denotes good agreement and >0.75 denotes excellent agreement. Results. The kappas for the six criteria were: 1) 0.77; 2) 0.68; 3) 0.62; 4) 0.73; 5) 0.68; and 6) 0.62. The kappa statistic for the immobilization decision was 0.90. In only one case did the immobilization decisions differ; the paramedic indicated immobilization, whereas the physician did not. Conclusion. In this model, there was excellent agreement between paramedics and physicians when evaluating simulated patients for possible c-spine injury. No patient requiring immobilization would have been clinically cleared by paramedics. These data support the progression to a prospective field trial evaluating the use of these criteria by paramedics.