Michael Eller; Peter J Goadsby


Expert Rev Neurother. 2013;13(3):263-273. 

In This Article

Abstract and Introduction


Headaches are described as primary, where no contributing cause is found, or secondary, where a discrete lesion or other condition has triggered the phenotype. Primary headache is a common condition; migraine causes much of the morbidity in this population, at great personal and economic cost. The decision to use MRI is a common dilemma facing clinicians, particularly as primary headache phenotypes can be triggered by secondary causes. Studies demonstrate that there is no appreciable difference in the frequency of pathological and incidental findings in common headache populations compared with the general community. Imaging is therefore not routinely required where a primary headache diagnosis can be made. Clinicians must be aware of the risk of manufacturing morbidity in uncovering incidental and nonsignificant imaging changes. However, patients demonstrating 'red flags' on medical history and examination do require imaging to help exclude a secondary cause of symptoms. Other headache phenotypes, such as the trigeminal autonomic cephalalgias, also generally require MRI.


Most patients with headache do not need an MRI of their brain. Some patients require an MRI to exclude a secondary cause of their headache. Between these two bland statements, palatable to most doctors in most countries, are a host of factors that can make this decision difficult. Significant intracranial pathology can cause nothing more than a mild headache;[1] initially missing the diagnosis of a brain tumor in just such a setting is the most common cause of litigation among neurologists in the USA, and has been for several years.[2] Equally, many typical features of primary headache disorders, such as a feeling of sinus fullness in migraine, may prompt imaging which would be expected to be almost always normal.[3,4]