Abstract & Commentary
Source: The Canadian c-spine rule versus the NEXUS low-risk criteria in patients with trauma. N Engl J Med 2003;349:2510-2518.
In 2000, Hoffman et al published the nexus study delineating five criteria to exclude the need for cervical spine (c-spine) radiographs in a "low-risk" emergency department (ED) patient population.1 A year later, Stiell et al presented a study from Canada that identified different criteria for excluding c-spine injury (and hence the need for radiography) in a low-risk ED patient population.2 These differing criteria are presented in the Table.
Stiell’s purpose for this study, carried out at nine Canadian tertiary care hospitals, was to prospectively compare the NEXUS low-risk criteria (NLC) to the Canadian c-spine rules (CCR) for accuracy, reliability, clinical acceptability, and potential outcomes in patient care and radiography utilization.
In the study, 8283 patients prospectively were evaluated with both the NEXUS criteria and the CCR prior to radiography. There were 169 (2%) clinically significant injuries. Radiography was at the discretion of the examining physician, and approximately 30% of study patients had no radiographs. These patients were accounted for by telephone follow-up at 14 days. Notably, in 845 patients enrolled (10.2%), examining physicians neglected to determine ability to rotate the neck to 45°. These were termed "indeterminate" cases.
When the two criteria were compared and these indeterminate cases were excluded, the CCR was more sensitive than the NLC (99.4% vs 90.7%), and more specific (45.1% vs 36.8%) and would have resulted in lower radiography rates (55.9% vs 66.6%). Stiell performs a secondary analysis in which all indeterminate cases are assumed to be positive, resulting in a sensitivity and specificity of 99.4% and 40.4%, respectively. Finally, he also calculates the same values if all indeterminate cases are assumed to be negative, resulting in a sensitivity and specificity of 95.3% and 50.7%, respectively. The study concludes that the CCR would have missed one clinically significant fracture, and the NLC would have missed 16 clinically significant fractures; hence, the CCR is superior to the NLC for excluding c-spine injury in low-risk patients.
Commentary by Andrew D. Perron, MD, FACEP, FACSM
Decreasing the number of c-spine radiographs obtained on low-risk patients is important for any number of good reasons. Unnecessary irradiation, cost, length of time spent on a board in a collar, and prolonged ED stays are the reasons usually cited (appropriately) for this area of clinical research. As with so many resources in the ED (e.g., plain x-rays of the ankle and knee, computed tomography of the head), we have validated clinical decision rules to help guide us through the diagnostic algorithm.
So, which way should the busy clinician be looking for answers to help guide them through this evaluation: west, to Hoffman’s NLC, or north, to Stiell’s CCR? We know from Hoffman’s study on 34,000 patients that when properly applied and prospectively evaluated, the NLC carry a 99.6% sensitivity. We know from this study that when properly applied, the CCR carry at worst a 95%, and more likely closer to 99% sensitivity. Both criteria markedly reduce unnecessary radiographs, and both result in very few missed injuries.
My conclusion? Pick one, learn it, and stick with it. I think either one gives you the ability to separate the wheat from the chaff with great diagnostic certainty. In my mind, the five NEXUS criteria are easier to remember than the 3 + 5 + 1 criteria for the Canadian rules (one does wonder how more than 10% of the patients had the Canadian rule misapplied [no neck rotation], when that was an integral part of the rule being studied). In the end, memorizing one algorithm over another is a small point, and in this age of the ubiquitous personal digital assistant (PDA), either set of criteria is only as far away as my coat pocket if I can’t remember them.
Dr. Perron, Residency Program Director, Department of Emergency Medicine, Maine Medical Center, Portland, ME, is on the Editorial Board of Emergency Medicine Alert.
1. Hoffman JR, et al. Validity of a set of clinical criteria to rule out injury to the cervical spine in patients with blunt trauma. N Engl J Med 2000;343: 94-99.
2. Stiell IG, et al. The Canadian C-Spine Rule for radiography in alert and stable trauma patients. JAMA 2001;286:1841-1848.