Interrater Reliability of Cervical Spine Injury Criteria in Patients With Blunt Trauma☆,☆☆,★
Introduction
Sporadic case reports of asymptomatic cervical spine injuries1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12 have led to the widespread use of radiographic imaging for virtually all persons who present to the emergency department with blunt trauma.13, 14 Such routine imaging detects injuries in a small minority of cases, and subjects 96% to 98% of those with blunt trauma to the risk and expense of negative cervical spine radiography.11, 12, 13, 14
Concerns about the cost and radiation exposure associated with large numbers of negative radiographs have led to the development of “low-risk criteria” designed to screen patients with blunt trauma for cervical spine injury.15 Ideally, such criteria would reliably identify all patients with cervical spine injury (be maximally sensitive) while excluding injury in a large subset of trauma patients (be relatively specific). Previous studies from our institution have suggested that the risk of cervical spine injury is extremely low among patients with blunt trauma if they meet all of the following criteria: (1) they have a normal neurologic function; (2) they are not intoxicated; (3) they have no posterior midline cervical tenderness; and (4) they have no distracting painful injuries.15 In these studies, application of a decision rule that avoids radiography when all four of these criteria are met would have allowed us to decrease the number of radiographic studies by almost one third, while still identifying every patient with cervical spine fracture or dislocation.
The general applicability of any decision rule, however, is critically dependent on whether its elements can be reliably interpreted by different physicians. No prior study has addressed the question of whether clinicians can agree on either the individual low-risk criteria noted or the overall classification of individual patients based on the combined criteria. Given the absence of any prior information on which to base a precise hypothesis, we designed this study to determine the interrater reliability of these criteria among emergency physicians, both as a whole and for each of the individual components.
Section snippets
Methods
This study was conducted between July 1995 and April 1996 in the EDs of a private university hospital and a public county hospital which share an emergency medicine residency program and are staffed by full-time emergency medicine faculty and residents. On presentation, each patient underwent evaluation by one of the residents or attending emergency physicians normally staffing the ED. A second examination was independently performed by one of five emergency physicians involved in the study.
Results
A total of 122 survey pairs were collected during the study period. There were 56 men and 66 women. Their ages ranged between 4 and 95 years; the median age was 32, and there were 7 patients younger than 18 and 17 patients older than 60 years of age. The time between examinations ranged from 1 to 237 minutes, with a median time of 10 minutes; only three paired examinations were done more than 1 hour apart.
Table 1 lists the κ statistics and standard errors for each of the individual criteria and
Discussion
In our previous work, the risk criteria evaluated in this study exhibited a negative predictive value of 100% (with a lower 95% confidence level of 98.7%). In that series, 353 of 974 blunt trauma victims were classified as “risk-free” and could potentially have been safely evaluated without radiography.15 A multicenter study to validate these criteria and define narrow confidence intervals for their sensitivity is currently underway.
The value of any such criteria would be limited, however, if
References (19)
- et al.
Occult cervical spine fracture in an ambulatory patient
Ann Emerg Med
(1982) - et al.
Clinical presentation of patients with acute cervical spine injury
Ann Emerg Med
(1984) Unstable occult cervical-spine fracture
Ann Emerg Med
(1991)”Asymptomatic” cervical spine injuries: A myth?
Am J Emerg Med
(1985)- et al.
”Asymptomatic” cervical spine injuries: A myth?
Am J Emerg Med
(1985) - et al.
Low-risk criteria for cervical spine radiography in blunt trauma: A prospective study
Ann Emerg Med
(1992) Fractures of the cervical spine with minimal or no symptoms
Med J Malaya
(1972)- et al.
Occult unstable cervical spine injury
J Tenn Med Assoc
(1988) Missed fractures of the cervical spine
Mil Med
(1989)
Cited by (45)
Neuroimaging of the Traumatic Spine
2016, Magnetic Resonance Imaging Clinics of North AmericaCitation Excerpt :These clinical decision rules consider variables from the patient history and examination or simple clinical tests, were derived from clinical research, and are defined as decision-making tools. The most established clinical decision rules for spinal imaging are the CCR6–8 and the National Emergency X-Radiography Utilization Study Low-Risk Criteria (NLC)9–11 (Table 1). By identifying high-risk criteria (Box 1), both clinical decision tools help if radiography is indicated as a screening method on alert (eg, score of 15 on the Glasgow Coma Scale) and stable trauma patients with mild or unspecific symptoms and low risk of spine injury.
Agreement between resident and faculty emergency physicians in the application of nexus criteria for suspected cervical spine injuries
2015, Journal of Emergency MedicineCitation Excerpt :Previous studies assessing the inter-rater reliability of the NEXUS criteria among physicians showed a large range in agreement. One study demonstrated that the evaluation of a focal neurologic deficit had the lowest level of agreement (κ = 0.58) compared to intoxication and the presence of a distracting injury showing “almost perfect” agreement (κ = 0.86 and 0.77, respectively) (9). Another study showed fair to moderate agreement for intoxication and presence of a distracting injury (κ = 0.23 and 0.41, respectively) (1).
Role of cervical spine radiography in the initial evaluation of stable high-energy blunt trauma patients
2011, Journal of Orthopaedic ScienceCanadian C-spine Rule and the National Emergency X-Radiography Utilization Low-Risk Criteria for C-spine radiography in young trauma patients
2009, Journal of Pediatric SurgeryCitation Excerpt :A prospective study where the criteria are assessed at the same time as the physician determines the need for cervical spine imaging may produce different result. Although the RAs were trained to apply CDR criteria, they were not medical doctors [27]. In summary, neither CDR performed at a high enough level to be used with confidence.
Spinal Cord Injury in the Pediatric Patient
2008, Clinical Pediatric Emergency MedicineCitation Excerpt :The target populations for these rules are alert, stable adult patients without neurologic deficit. The National Emergency X-Radiography Utilization Study (NEXUS) criteria were described in 1992 [62,63] and validated in a US study of 34 069 patients, with a sensitivity of 99.6% [64,65]. To meet the NEXUS low-risk profile, the patient must satisfy all of the following 5 criteria: no posterior midline cervical spine tenderness, no evidence of intoxication, a normal level of alertness, no focal neurologic deficit, and the absence of painful or distracting injuries (Table 1).
Cervical spine clearance: A review
2005, Injury
- ☆
From the UCLA Emergency Medicine Center and the Department of Medicine, University of California at Los Angeles School of Medicine, Los Angeles, CA.
- ☆☆
Reprint no. 47/1/87766
- ★
Address for reprints: William R Mower, MD UCLA Emergency Medicine Center 924 Westwood Boulevard, Suite 300 Los Angeles, California 90024 310-825-7209 Fax 310-794-9747 E-mail [email protected]