Are We Overusing Head CT?
Myles M. Mitsunaga
There seems to be a universal human tendency to use every available tool when evaluating a problem so as not to miss something important.
In "Head CT Scans in the Emergency Department for Syncope and Dizziness," published in the January 2015 issue of AJR, Miles M. Mitsunaga, assistant professor at the John A. Burns School of Medicine, University of Hawaii, looked at the value of using head CT to evaluate such patients.
Is evaluation of patients presenting with dizziness and syncope experiencing what the military calls “mission creep”: using—perhaps unnecessarily—the most sophisticated technology to evaluate and differentiate between benign and life-threatening causes of dizziness and syncope?
Yes. However, the differential diagnosis for these patients presenting to the emergency department (ED) is extremely broad and is often confounded by comorbidities and polypharmacy. Patients often have difficulty accurately communicating symptoms; thus, when the emergency physician’s initial history and physical exam are equivocal, extensive workups with labs and imaging are performed to rapidly rule out serious causes. Our findings suggest that careful attention to the clinical examination and appropriate follow-up may be used as an alternative to expensive cross-sectional imaging for these patients.
How can radiologists assist ED physicians in evaluating such patients?
Radiologists can assist our physician colleagues in the ED and their patients by reminding them that most patients with mild symptoms of dizziness or syncope do not require a head CT. If a careful history and physical examination do not find persistent neurologic signs, then a follow-up clinic visit the next day may be all that is necessary.
In what circumstances is head CT imaging recommended?
According to the American College of Radiology, some of the indications for CT of the brain include acute head trauma, suspected acute intracranial hemorrhage, vascular occlusive disease, mental status change, headache, acute neurologic deficits, suspected intracranial infection, osseous evaluation of the calvarium, and suspected mass or tumor, among others. However, without a complete history and physical examination, too many patients could be suspected of having one of these disease processes.
You analyzed consecutive patients presenting to your ED with dizziness, near-syncope, or syncope over a 6-month period, all of whom underwent head CT. What were your objectives?
We wanted to determine the yield of acutely abnormal findings on head CT scans and determine the potential clinical factors (age, gender, loss of consciousness, acute head trauma, seizure, headache, slurred speech, altered mental status, history of a neurologic deficit, physical exam finding of a focal neurologic deficit, laboratory evidence of drug intoxication or hypoglycemia, and use of anticoagulation medications) that predicted acutely abnormal head CT findings and subsequent hospital admission. Potential positive head CT findings included any intracranial hemorrhage (epidural, subdural, subarachnoid, or intraparenchymal) or intracranial space-occupying lesion with mass effect (brain abscess, tumor, granuloma, etc.).
Did your findings match your expectations?
Yes. We found that head CT has a low diagnostic yield (7.1% for patients with dizziness and 6.4% for patients with near-syncope and syncope). Our study reported a relatively higher diagnostic yield than did previous studies; however, we included patients who had competing indications because we wanted to determine whether any competing indications were predictive of an acutely abnormal head CT. Additionally, although a focal neurologic deficit, age greater than 60 years, and head trauma were statistically significant in predicting an acutely abnormal finding on a head CT, the overall accuracy of that model remained the same (93.3%) as when no factors were applied.
What did you conclude?
The very low prevalence of acute head CT findings in patients presenting with syncope or dizziness suggests that physicians are using head CT as a screening examination, rather than a diagnostic tool. Better clinical algorithms to screen patients will be necessary to avoid unnecessary use of head CT in low-risk patients.