Authors (including presenting author) :
Tao YS (1), Ying MT (2), Chan CTP (3)
Affiliation :
(1) Department of Diagnostic Radiology and Imaging, Queen Elizabeth Hospital, (2) Department of Health Technology and Informatics, The Hong Kong Polytechnic University, (3) Department of Health Technology and Informatics, The Hong Kong Polytechnic University
Introduction :
Computed tomography (CT) scanning of the head has become a standard clinical practice for patients with minor head injuries in the accident and emergency department (AED). Nevertheless, most of these patients are classified as having minimal or insignificant head injuries. Only a few of them are considered to have significant head injuries and require further neurosurgical intervention for intracranial hematoma. The study aimed to investigate the feasibility of developing a CT head selective rule for minor head injury patients based on clinical data available in AED. Also, the study verified the reliability of the rule in terms of its effectiveness and accuracy. The CT head selective rule is expected to be used as a quick guide to determine the need for CT head scanning for patients with minor head injuries in AED. This will help avoid unnecessary CT head scans and reduce radiation exposure to patients.
Objectives :
In the phase-one study, the clinical records of minor head injury patients were reviewed. The patient’s clinical data, including their clinical signs and symptoms due to the injury, were recorded. The patient clinical data were classified into different risk factors. Their CT head scan results were also reviewed and were classified into with or without intracranial abnormality (i.e. positive or negative results, respectively). The chi-square test was used to evaluate the correlations between individual risk factors and positive results in CT head scans. In the phase-two study, the significant risk factors determined in the phase-one study were used to prospectively investigate the value of the requested CT head scan in patients with a minor head injury. Multiple logistic regression models were employed to evaluate the correlation between CT head scan results and the risk factors. The risk factors with a high correlation with a positive result in a CT head scan were selected to devise the CT head selective rule.
Methodology :
Of the 2001 minor head injury patients with CT head scan in the phase-one study, 127 (6.3%) patients had a positive result in the CT scan. Results showed four risk factors, including (1)Glasgow Coma Scores lower than 15, X2(1, N=2001) =1163, p<.001. (2)Aged above 65, X2 (1, N=2001) =0.98, p>.05. (3)Loss of consciousness, X2(1, N=2001) =7.52, p>.001, and (4)Episode of vomiting after the head injury, X2(1, N=2001) =14.78, p<.001; were significantly associated with a positive result in CT scan. In the phase-two study, three additional risk factors were included in the study besides the four identified risk factors. They were the patient’s conditions at AED, including alcohol intoxication, an open wound on the head, and the patient’s ambulatory ability. There were 387 patients included in the phase-two study, and 40 (10.3%) had positive results in the CT head scan. All of these 40 patients were presented with one or more risk factors. Among the remaining 347 patients with negative results in CT scans, 126 of them had no risk factor, and the remaining 221 patients had at least one risk factor. Hence, 33% of the CT scans were performed on patients with no risk factors. The sensitivity, specificity, positive predictive value, negative predictive value and overall accuracy were 100%, 33.1%, 14.7%, 100% and 40%, respectively.
Result & Outcome :
The avoidable CT scan would have been possible if the CT head rule had been followed correctly to protect patients from potentially harmful ionizing radiation.
