Retrospective application of the NEXUS low-risk criteria for cervical spine radiography in Canadian emergency departments

• Published in: Annals of emergency medicine Vol. 43; no. 4; pp. 507 - 514
• Main Authors: Dickinson, Garth, Stiell, Ian G, Schull, Michael, Brison, Robert, Clement, Catherine M, Vandemheen, Katherine L, Cass, Daniel, McKnight, Douglas, Greenberg, Gary, Worthington, James R, Reardon, Mark, Morrison, Laurie, Eisenhauer, Mary A, Dreyer, Jonathan, Wells, George A
• Format: Journal Article
• Language: English
• Published: New York, NY Mosby, Inc 01-04-2004; Elsevier
• Subjects: Adolescent; Adult; Aged; Aged, 80 and over; Anesthesia. Intensive care medicine. Transfusions. Cell therapy and gene therapy; Biological and medical sciences; Canada; Cervical Vertebrae - diagnostic imaging; Cervical Vertebrae - injuries; Emergency Service, Hospital; Female; Humans; Intensive care medicine; Male; Medical sciences; Middle Aged; Radiography; Retrospective Studies; Sensitivity and Specificity; Spinal Cord Injuries - diagnostic imaging; Spinal Injuries - diagnostic imaging; Wounds, Nonpenetrating - diagnostic imaging; United States; North America; America; Human; Intensive care; Prognosis; Telephone; Physician; Trauma; Confidence interval; Radiography; Fixation; Emergency department; Treatment; Cohort study; Risk factor; Computerized axial tomography; Emergency; Reliability; Cervical spine; Resuscitation
• ISSN: 0196-0644; 1097-6760
• DOI: 10.1016/j.annemergmed.2003.10.036

We evaluate the accuracy, reliability, and potential impact of the National Emergency X-Radiography Utilization Study (NEXUS) low-risk criteria for cervical spine radiography, when applied in Canadian emergency departments (EDs).
The Canadian C-Spine Rule derivation study was a prospective cohort study conducted in 10 Canadian EDs that recruited alert and stable adult trauma patients. Physicians completed a 20-item data form for each patient and performed interobserver assessments when feasible. The prospective assessments included the 5 individual NEXUS criteria but not an explicit interpretation of the overall need for radiography according to the criteria. Patients underwent plain radiography, flexion-extension views, and computed tomography at the discretion of the treating physician. Patients who did not have radiography were followed up with a structured outcome assessment by telephone to determine clinically important cervical spine injury, a previously validated outcome measurement. Analyses included sensitivity and specificity with 95% confidence interval (CI), κ coefficient, and potential radiography rates.
Among 8,924 patients, 151 (1.7%) patients had an important cervical spine injury. The combined NEXUS criteria identified important cervical spine injury with a sensitivity of 92.7% (95% CI 87% to 96%) and a specificity of 37.8% (95% CI 37% to 39%). Application of the NEXUS criteria would have potentially reduced cervical spine radiography rates by 6.1% from the actual rate of 68.9% to 62.8%. Of 11 patients with important injuries not identified, 2 were treated with internal fixation and 3 with a halo.
This retrospective validation found the NEXUS low-risk criteria to be less sensitive than previously reported. The NEXUS low-risk criteria should be further explicitly and prospectively evaluated for accuracy and reliability before widespread clinical use outside of the United States.