Choosing Wisely, the ABIM Foundation’s widely followed campaign to counter overutilization, urges doctors to exercise great discretion before ordering neuroimaging when the patient’s only symptom is a headache. A new study calls into question the soundness of the recommendation.

Led by Ammar Hawasli, MD, PhD, a Washington University neurosurgeon, researchers looked at 95 patients diagnosed with biopsy-confirmed brain tumors. They found that close to a quarter, 24.2%, had presented only isolated headaches, no symptoms or nonspecific symptoms. Meanwhile isolated headaches were the sole medical complaint from 11 (11.6%) of the patients.

And, if the Choosing Wisely recommendations had been followed, the brain tumors would have gone undiagnosed, at least initially, in three of those 11 patients, the researchers report in the journal Neurosurgery.

“If adopted by governing or funding organizations, these guidelines threaten to negatively affect the care and outcomes of patients with brain tumors, who frequently present with minimal symptoms or isolated headache syndromes,” the authors write. “As physicians grapple with the difficult conflict between evidence-based cost-cutting guidelines and individualized patient-tailored medicine, they must carefully balance the costs and benefits of discretionary services such as neuroimaging for headaches.”

A Choosing Wisely fact sheet endorsed by the American College of Radiology and Consumer Reports states that a standard brain CT scan should cost about $340 without a contrast agent and $840 with, while brain MRIs should ring up bills of $660 without contrast and $970 with.

The fact sheet, aimed at consumer headache sufferers, says most of the time these tests are unnecessary because they rarely help diagnose the problem yet pose risks and can add needless expense.

Nearly half the patients in the Washington University study would have been cleared for imaging by Choosing Wisely, if its recommendations were binding, as 48.4% went to their doctors with a combination of brain-tumor symptoms, such as seizures, cognitive and speech dysfunction or other neurologic abnormalities.

The authors conclude with a statement of support for physicians who exercise good clinical judgment in their ordering of neuroimaging.

“Although we do not recommend routine screening for the general population, we do contend that a substantial number of patients with brain tumors will present with isolated headaches,” they write. “Unvalidated guidelines to prevent neuroimaging in patients with headaches may reduce the perceived global economic burden at the expense of medical errors, delayed diagnoses and inferior outcomes for patients with brain tumor. Ultimately, further research is crucial for the development of validated and tested clinical decision rules on neuroimaging for headaches.”

Neurosurgery has posted the report as a freely available special article