Classification of migraine and tension-type headache

Headache classification is a complicated business. Despite three subsequent versions of the International Headache Classification there are still mistakes to be discovered. As previous Chairman of the Headache Classification I only recently noticed a flaw in ICHD-3. The classification is hierarchical, and it implies that it should be possible to make a headache diagnoses at all digit levels. In general practice and especially in many low- and middle-income countries headache diagnoses are made only at the first digit level. Furthermore, studies of the costs of headache disorders and most drug trials bring only the first digit diagnosis saying migraine with or without aura. I was therefore surprised to realize that there is no set of diagnostic criteria for 1.0 migraine or for 2.0 tension-type headache. There are only criteria for their sub forms. In ICHD-2 migraine with aura also had no set of diagnostic criteria but it was diagnosed by fulfilling criteria for one of the sub forms. That was amended in ICHD-3, but we did not think about first digit diagnoses for migraine or tension-type headache. This contrasted with cluster headache which is coded at the second digit level. It has a set of diagnostic criteria. I have been thinking a lot about the solution to this problem for migraine and tension-type headache. Migraine without aura requires five attacks and migraine with aura two attacks. In fact, it is primarily the aura that is defined since migraine aura without headache is also common. These features make it almost impossible to design independent diagnostic criteria for 1.0 Migraine. One might consider criteria that just describe the painful migraine attack with or without aura, but that would mean excluding aura without- headache and aura with mild headache. The one-digit diagnosis must, however, include all entities at a lower level. To my knowledge it is not known how many digits are used in different settings or in countries with different income levels. I nevertheless suggest including diagnosis of migraine at the first digit level although it clearly is imperfect.

Proposed diagnostic criteria for 1.0 migraine

At least five attacks fulfilling B

Attacks fulfilling diagnostic criteria for one or more of the diagnoses 1.1–1.5

Not better accounted for by another ICHD-3 diagnosis

For tension-type headache it is easy to develop diagnostic criteria. It can be done by simply removing any mention of the frequency of the headaches. Thus, it would read as follows:

Diagnostic criteria for 1.0 Tension-type headache

At least 10 episodes of headache fulfilling criteria B and C

Headache lasting from 30 minutes to seven days

At least two of the following four headache characteristics 1. Bilateral location 2. Pressing or tightening (non-pulsating quality) 3. Mild or moderate intensity 4. Not aggravated by routine physical activities such as walking or climbing stairs

Both of the following: 1. No nausea or vomiting 2. Not more than one of photophobia or phonophobia

Not better accounted for by another ICHD-3 diagnosis

I hope these considerations will be useful for the next headache classification committee when it starts its work in a few years’ time.