Neuroimaging for Migraine: The American Headache Society Systematic Review and Evidence-Based Guideline

Randolph W. Evans, MD; Rebecca C. Burch, MD; Benjamin M. Frishberg, MD; Michael J. Marmura, MD; Laszlo L. Mechtler, MD; Stephen D. Silberstein, MD; Dana P. Turner, MSPH, PhD


Headache. 2020;60(2):318-336. 

In This Article


Migraine has a worldwide prevalence of 15%-18%[1] and affects over 40 million people in the United States. When and how to use neuroimaging for migraine is a critical issue which confronts every physician who diagnoses and treats migraine. Evidence from the US population based National Ambulatory Medical Care Survey demonstrated that neuroimaging was ordered during 12% of outpatient headache visits between 2007 and 2010.[2]

There are many reasons why physicians may obtain neuroimaging for suspected migraine, including:

  1. Excluding secondary conditions that mimic migraine.[3]

  2. Discomfort with migraine as a clinical diagnosis, ie, "our stubborn quest for diagnostic certainty."[4]

  3. Cognitive bias.[5]

  4. Busy practice conditions where tests are ordered as a shortcut.

  5. Addressing the expectations, concerns, and anxiety of patients and family which may be reflected in negative online reviews.[6]

  6. Addressing the concerns and expectations of referring clinicians ("better safe than sorry").

  7. Medicolegal issues.[7]

Indiscriminate use of neuroimaging should be avoided. The costs associated with neuroimaging can be significant, and one study estimated nearly $1 billion of annual costs in the United States from neuroimaging.[2] Neuroimaging may also lead to anxiety, further testing, and additional costs from incidental findings which are not clinically significant.[8,9] There are many barriers for obtaining neuroimaging, including cost, as patients may have high deductible insurance plans or lack insurance coverage; lengthy third-party review for payor approval; and insurance companies which consider neuroimaging utilization as a negative in their physician ratings.[10]

Recommendations about the role of neuroimaging in diagnosis of headache vary by specialty. The American Academy of Neurology (AAN) evidence-based review[11] of the role of neuroimaging in non-acute headache patients, published in 2000, recommended: "Neuroimaging is not usually warranted for patients with migraine and normal neurological examination (Grade B). For patients with atypical headache features or patients who do not fulfill the strict definition of migraine (or have some additional risk factor), a lower threshold for neuroimaging may be applied (Grade C)." The AHS' "Choosing Wisely in Headache Medicine" concluded: "Don't perform neuroimaging studies in patients with stable headaches that meet criteria for migraine."[12] The American College of Radiology's "Choosing Wisely" concluded: "Don't do imaging for uncomplicated headache."[13] Neurosurgeons, however, argue against overly restrictive guidelines and for the benefit of neuroimaging of patients with isolated headaches or non-specific symptoms to diagnose brain tumors.[14]

In this systematic review, we aimed to gather evidence about the diagnostic utility (ie, sensitivity, specificity, positive predictive value [PPV] and negative predictive value [NPV]) of neuroimaging (MRI and CT) in adult patients (ages 18 and older) seeking outpatient treatment for episodic migraine, chronic migraine, progressive migraine, migraine with aura, and migraine without aura. Our goal was to answer the question "How often does a CT or MRI of the brain identify potentially symptomatic intracranial abnormalities in this population?" Based on the obtained evidence, we developed a guideline regarding the use of neuroimaging in patients with migraine with a normal neurological examination.