Comparison of the Canadian CT head rule and the New Orleans criteria in patients with minor head injury in a Spanish hospital.

BACKGROUND AND OBJECTIVE: To compare two scales for assessment of patients with mild head injury. The Canadian CT Head Rule (CCHR) and New Orleans Criteria (NOC) according to their diagnostic accuracy in patients attending an emergency department, and to determine the most important predictive values.

METHOD: Cross-sectional study in a first-level Hospital in the period of January 2011 to January 2013. Patients with mild head injury criteria were included. All the patients underwent a computed tomography (CT) of the head as part of internal protocol and the CCHR and NOC criteria were recorded for each patient. We consider the main variable the presence of traumatic lesions on CT and, as secondary variables, neurosurgical intervention and post-concussion syndrome. Sensitivity, specificity, predictive values and validity index (VI) of the CCHR and the NO criteria in the subgroup of patients with a Glasgow Coma Scale (GCS) score of 15 points were compared.

RESULTS: A total of 217 patients, of whom 197 had a GCS score of 15 points were evaluated. Both rules showed 100% sensitivity when a significant injury was presented in the CT, the CCHR 100% (95% CI: 97.4%-100%) and the NO criteria 100% (95% CI: 97.4%-100%); but the CCHR achieved higher values of specificity 25.3% (95% CI: 18.6%-32%), positive predictive value (PPV) and VI. The two rules showed a 100% sensitivity regarding neurosurgical intervention; however the CCHR with high-risk criteria showed better specificity, PPV and VI 55.2 (95% CI: 8.3%-62.2%) compared to the NO criteria 7.6 (95% CI: 3.8%-11.5%). With regard to post-concussion syndrome criteria NO criteria showed better sensitivity 100% (95% CI: 96.2%-100%) and predictive values, but lower specificity and VI compared with the CCHR 76.9% (95% CI: 50.2%-100%).

CONCLUSIONS: Our study demonstrates the high sensitivity of the CCHR and the NO criteria in patients with mild head injury, both to detect a significant clinical lesion on CT or the need for neurosurgical intervention and better specificity of CCHR compared with NO criteria. The adoption of clinical prediction rules, especially the CCHR, to request a CT scan in patients with mild head injury should be recommended in the emergency department.