In a comparison of clinical decision rules for use of CT in mild head injury, the Canadian CT Head Rule was better able to predict need for neurosurgical intervention than the New Orleans Criteria, according to a study published online Aug. 24 in Annals of Emergency Medicine.
The Canadian CT Head Rule had a higher sensitivity for predicting the need for neurosurgical intervention and identifying patients with intracranial traumatic lesions, wrote authors Wahid Bouida, MD, of Fattouma Bourguiba University Hospital in Monastir, Tunisia, and colleagues. They noted, however, that both decision rules had high negative predictive value for identifying need for neurosurgical interventions.
“A number of studies have been conducted to identify a set of high-risk factors that would clearly indicate which patient with mild head injury should undergo CT scan, yet no consensus has been reached. Current guidelines on the use of cranial CT scanning vary from mandatory scanning for all patients to more selective use based on clinical examination findings,” wrote the authors.
To underscore the importance of having effective guidelines, Bouida and colleagues explained that while mild head injury is one of the most common injuries treated in the emergency department (ED), systematically referring patients for a CT scan would not be cost-effective because life-threatening complications requiring neurosurgical intervention occur in less than 1 percent of cases.
The authors set about comparing two of the leading guidelines by performing an observational cohort study from 2008 to 2011 that included 1,582 patients with mild head injury who were age 10 or older.
Neurosurgical intervention was performed in 2.1 percent of patients and positive CT findings were demonstrated in 13.8 percent, according to the authors. In predicting need for neurosurgical intervention, the Canadian CT Head Rule had a sensitivity and specificity of 100 percent and 60 percent respectively. This compares with a sensitivity and specificity of 82 percent and 26 percent, respectively, for the New Orleans Criteria. Negative predictive values for intervention were 100 percent and 99 percent for the Canadian CT Head Rule and New Orleans Criteria, respectively.
Sensitivity for clinically significant head CT findings was 95 percent for the Canadian CT Head Rule and 86 percent for the New Orleans Criteria, according to the authors.
The researchers speculated that the superior sensitivity of the Canadian CT Head Rule was likely due to the fact that it takes into account a patient’s Glasgow Coma Scale score, while the New Orleans Criteria does not. They suggested that the Canadian CT Head Rule could be further improved by introducing new items such as headache severity to the model.
“Although our findings suggest that the Canadian CT Head Rule outperforms the New Orleans Criteria, it would be interesting to perform the same comparison with other available rules such as CT in Head Injury Patients (known as CHIP) and the Scandinavian or National Institute for Clinical Excellence guidelines,” wrote Bouida and colleagues. “Currently, no study has clearly shown one decision rule to perform significantly better than the others in cost-saving terms.”