Share this article on:

Incidental Findings on MRI Scans for Primary Headache Are Not Uncommon, Study Finds — How Migraine Specialists Address the Challenge

Fitzgerald, Susan

doi: 10.1097/01.NT.0000520853.12691.3f
Back to Top | Article Outline


Diagnostic imaging for headaches can lead to a plethora of incidental findings, many of which may prompt more worries and needless procedures, according to a new paper. But headache specialists say unnecessary imaging can be avoided if neurologists follow certain diagnostic processes, including a thorough neurologic exam.

It happens often. A magnetic resonance imaging (MRI) scan is ordered for a patient with nagging headaches and the scan turns up “an incidental finding,” which may simply be a benign cyst or a normal anatomical variation. But now the patient has two reasons to be anxious: the headaches and the surprising discovery inside his head.

These are only two of many scenarios that can unfold when MRI is used to evaluate headaches, according to a paper published in May in the journal Headache.

“About 90 percent of all headaches are the primary type where MRI scans of the brain will not reveal any pathology responsible for headaches,” wrote Randolph W. Evans, MD, FAAN, clinical professor of neurology at Baylor College of Medicine in Houston.

Yet in a significant number of cases — upwards of about 30 percent according to one study cited in the article — the scan reveals an incidental finding such as a small pineal cyst or white matter lesions. Such findings may lead to not only worry, but, in some instances, follow-up testing and even surgery.

“Most incidental findings, including a majority (but certainly not all) pineal cysts, arachnoid cysts, saccular aneurysms, primary empty sella, mega cisterna magna, meningiomas, and Rathke's cleft cysts, do not cause headaches,” Dr. Evans told Neurology Today in an email. In a small subset of cases, about 2 to 8 percent of instances, a worrisome incidental finding “such as a malignant neoplasm, larger aneurysms, or an arteriovenous malformation” may be found.

So why do doctors frequently order an MRI for primary headaches? Dr. Evans said that while there are medical indications for doing so, such as focal neurologic findings on a clinical exam or atypical cluster headaches, scans are often done to reassure the patient that there isn't a brain tumor or a blood vessel ready to burst. The doctor may also want to shortcut the diagnostic process or satisfy potential legal concerns.

“Defensive medicine is a common reason for imaging,” Dr. Evans noted. He said while doing imaging studies to rule out any potential problem may add to health care costs, it may also help lower costs if the patient gets reassurance, focuses on treating the headaches, and stops seeking out doctors in search of an explanation.

“I never deny a scan of the brain if someone will pay for it because there is no way to predict whether there will be a significant abnormality, and even following guidelines and denying a scan does not protect you from a malpractice suit,” Dr. Evans said in an interview.

Back to Top | Article Outline


But migraine specialists who were not involved with the analysis questioned the pretext for using imaging for diagnosing migraine or headache. Indeed, as part of the Choosing Wisely Campaign, an initiative of the American Board of Internal Medicine, the American Headache Society pointedly advised against using neuroimaging studies in patients with stable headaches that meet criteria for migraine. They also recommended against performing computed tomography for headache, except in emergency settings when hemorrhage, acute stroke or head trauma are suspected.

Morris Levin, MD, FAAN, professor of neurology and director of the Headache Center at the University of California, San Francisco, said in an email to Neurology Today that several lists have been published to help clinicians sort out headache symptoms and decide when to order a scan.

Dr. Levin, who chairs the AAN Section on Headache and Facial Pain, said his list of “red flags” includes changes in the pattern of headaches; onset in middle age or later; headache that is effort-induced or positional; the presence of a febrile or systemic illness (like AIDS or cancer); changes in personality or cognition; and any focal neurological findings. He uses the mnemonic C-S-F (Change, Sick, and Focal) as a simple way to remember those points.

Christopher Gottschalk, MD, assistant professor of neurology and director of headache medicine at Yale School of Medicine, said he uses a mnemonic, SNOOP, to identify red flags for headache: Systemic symptoms (fever, weight loss) or secondary risk factors (HIV, systemic cancer); Neurologic symptoms or abnormal signs (confusion, impaired alertness or consciousness); Onset (sudden abrupt, or split-second); Older (new onset and progressive headache, especially in middle age>50 years; Previous headache history (first headache or change in attack frequency, severity, or clinical features).

Dr. Gottschalk said that in instances where an MRI is warranted, it may be a good idea for the doctor to prepare the patient for the possibility of an incidental finding, rather than having the news come out of the blue.

Back to Top | Article Outline


Laszlo Mechtler, MD, FAAN, FASN, medical director of Dent Neurologic Institute and professor of neurology and neuro-oncology at the State University of New York at Buffalo, said many patients with headache are first seen by a primary care physician not specifically trained in headache diagnosis and management or an emergency department doctor who may order an imaging scan without a thorough evaluation. The MRI or computed tomography (CT), he said, may then be read by a general radiologist without specific expertise in neuroradiology.

Dr. Mechtler said the resulting report may note an incidental finding without providing needed context as to its potential significance or in association with the patient's headache. The problem may be with the images themselves.

“In neuroimaging, not all MRIs are equal,” Dr. Mechtler said. “In the community, a lot of imaging is done with inadequate sequencing,” or perhaps with an MRI machine that does not utilize the highest strength of magnetic field. That may lead to the ordering of another MRI.

Dr. Mechtler contends more training in neuroimaging is needed for neurology residents so they can better interpret the results of the scans they order. He said it is not enough to simply advise patients “not to worry” about an incidental finding. He has a patient's images in full display at the appointment and walks them through what has been found.

“We have to confront the incidental findings by educating our patients about the meaning of the results,” said Dr. Mechtler. He said physicians also need to provide more consideration to their rationale for ordering an imaging scan for headache. Dr. Mechtler is part of a group of headache specialists working with the American Headache Society to draw up new guidelines on that issue.

The issue is particularly germane to children who present with headache. Amy Gelfand, MD, assistant professor of neurology and director of pediatric headache at UCSF said children may have to be put under general anesthesia to keep still during an MRI or may be frightened by the machine. Computed tomography scans, which are sometimes ordered based on symptoms, expose the child to radiation.

Dr. Gelfand relies on a clinical exam and patient history to pinpoint cases in which imaging seems to be justified. Some red flags she looks for are waking with pain and vomiting in the middle of the night, balance and gait irregularities, focal weakness, double vision, and increasing frequency of headache.

“I am actually a lot less concerned about a brain tumor when the headache has been there for three years than when the headache has been there for three weeks,” Dr. Gelfand said.

Families, however, may be focused on a worst-case scenario such as a brain tumor, and despite reassurance “some families feel they need a scan so they can sleep,” Dr. Gelfand said. She makes a point to raise the possibility that the scan may reveal a finding that is unrelated to the headache and inconsequential, hoping to stem later worry.

Dr. Gelfand uses language children can understand to explain why she isn't worried about an MRI finding. In the case of T2 hyperintensities, “I tell the kids they're like brain freckles. I tell them ‘just like you might have freckles on your skin, sometimes you have little things on your brain when we look very closely.’”

Back to Top | Article Outline


Back to Top | Article Outline


•. Evans RW. Incidental findings and normal anatomical variants on MRI of the brain in adults for primary headaches Headache 2017; 57(5):780–791.
•. Dodick DW. Diagnosing headache: Clinical clues and clinical rules http:// Adv Stud Med 2003; 3(6C): S550–S555.
© 2017 American Academy of Neurology