National Emergency X-Radiography Utilization Study Criteria Is Inadequate to Rule Out Fracture After Significant Blunt Trauma Compared With Computed Tomography

Abstract

EAST guidelines now recommend computed tomography (CT) to evaluate cervical spine (c-spine) fractures after blunt trauma in patients who do not meet National Emergency X-Radiography Utilization Study criteria (NC), yet no imaging is required in those patients who do meet these criteria. NC are based on patients with both minor and severe (trauma team activation [TTA]) trauma. The purpose of this study was to evaluate the NC using CT as the gold standard in TTA patients. We prospectively evaluated 2,606 blunt TTA patients at our Level I trauma center. NC defined as alertness (Glasgow Coma Scale [GCS] score = 15), evidence of intoxication, clinically distracting injury, midline c-spine tenderness, or neurologic deficits were documented. CT was used to determine the accuracy of these criteria. There were 157 patients with c-spine fractures and 2,449 patients without c-spine fractures. The fracture group was older (age, 43.4 years ± 19.3 years fracture group vs. 37.7 years ± 17.5 years no fracture group, p = 0.0003) with a lower GCS score (fracture group 13.7 ± 4.5 vs. no fracture group 14.4 ± 3.6, p = 0.0001) and initial systolic blood pressure (132.5 mm Hg ± 23.4 mm Hg vs. 139.9 mm Hg vs. 23.5 mm Hg, p = 0.0009). The sensitivity and specificity of clinical examination for all patients were 82.8% (130 of 157) and 45.7% (1,118 of 2,449), respectively. The positive predictive value (PPV) and negative predictive value (NPV) were 8.9% (130 of 1,461) and 97.6% (1,118 of 1,145), respectively. Patients with a GCS score of 15 had a sensitivity of 77%, specificity of 52.3%, PPV of 8.5%, and NPV of 97.5% for clinical examination. In those patients with the GCS score of 15, no intoxication or distracting injury, clinical examination had a sensitivity of 59.4%, specificity of 79.5%, PPV of 12.5%, and NPV of 97.5%. Of 26 patients with missed injuries based on NC, 19 (73.1%) required further intervention (16 collars, 2 OR, 1 Halo). As in our previous trial, NC is inaccurate compared with CT to diagnose c-spine fractures in TTA patients. CT should be used in all blunt TTA patients regardless of whether they meet NC.