Use of a Clinical Decision Rule May Reduce Unnecessary Head CT Scans

William T. Basco, Jr., MD, FAAP


September 10, 2008

A Clinical Decision Rule for Cranial Computed Tomography in Minor Pediatric Head Trauma

Atabaki SM, Stiell IG, Bazarian JJ, et al
Arch Pediatr Adolesc Med. 2008;162:439-445

The study authors noted that the use of cranial computed tomography (CT) is increasing in all patients, including children. Studies have quantified "minor head trauma" as some loss of consciousness, with a Glasgow Coma Scale score of 13 or greater; this was the standard used in this study despite some apparent disagreement over this classification.

This study evaluated patients with minor head trauma who presented to 1 of 4 Canadian Level 1 pediatric trauma centers from March 1997 to March 2000. The treating physicians recorded patient clinical characteristics on a data collection instrument before they were able to obtain CT results, and the treating physicians were also asked to predict whether each patient was likely to have an intracranial injury.

The 1000 subjects were all 21 years old or younger; 64.1% were male; and the mean age was 8.9 years. The potential predictor variables were amnesia, dizziness, headache, intoxication, lethargy, having had a seizure, emesis, and altered behavior (by parental assessment). The following physical findings were also used as predictors: scalp hematoma, scalp laceration, palpable skull defect, the presence of a sensory or motor deficit, signs of basilar skull fracture, and measures of loss of consciousness.

The study authors also accounted for mechanism of injury. Overall, 65 patients (6.5%) had an intracranial injury diagnosed by CT, and only 6 (0.6%) required neurosurgical intervention. Forty percent of the subjects with intracranial injury had a subdural hematoma; 27.7% had contusions; 23.1% had subarachnoid hematomas; and 21.5% had multiple injuries. Falls accounted for the cause of injury in 44.4% of injured subjects.

The study authors developed the following decision rule. A patient was unlikely to have an intracranial injury if he or she had none of the following: dizziness, sensory deficit, Glasgow Coma Scale < 15, mental status change, injury due to bicycle, < 2 years old, palpable skull defect, or basilar skull fracture signs on examination. This rule had a sensitivity of 95% for detecting intracranial injury, with a specificity of 48.9% and negative predictive value of 99.3% (the chance that the patient does not have intracranial injury, given that the test is "negative"). By contrast, the treating physicians had a sensitivity rate of 14.8%.

The study authors concluded that they were able to develop a sensitive clinical decision rule that would allow reduction in unnecessary cranial CT scans.

The study authors noted the need to validate this rule prospectively, but they estimated that up to 46% of CT scans in patients with minor head trauma could be eliminated should this approach prove valid. They also raised some interesting questions in their discussion. Would it be important to "catch" the small percentage of children with intracranial injury missed by the decision rule? None of the 6 "missed" in this study had injuries requiring surgical intervention. Does a negative CT result change management of a child with intracranial injury and persistent symptoms at evaluation? Of course, this study can't begin to answer whether any knowledge gained from knowing about the injury in these 6 children is offset by their increased radiation and sedation (for CT) risk. All are very interesting questions currently being actively debated.



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: