Abstract
The basic problem in diagnosing different types of primary headache is that the diagnosis depends on the typical characteristics of headache pain and on the typical symptoms associated with headache. The information we use for the diagnosis is always subjective, and basically derives from interpretations made by the patient and physician. There is also the problem that a patient may have several different types of headache.
Vomiting, nausea, aura, unilaterality, pulsatility, family history of migraine, photophobia, phonophobia, are some of the features which have been considered to be typical for migraine and have been used as criteria in several previous definitions of migraine and in the International Headache Society (IHS) criteria of migraine. However, none of these typical characteristics is specific to migraine. Migraine and tension-type headache greatly overlap in regard to various features of headache (1 –3). The migrainous characteristics of headache correlate with severe pain (1, 2, 4 –6). To make the classification more accurate it has been suggested that not only the severity of pain but also the intensity of certain other characteristics of pain and associated symptoms should be graded when used as criteria of migraine or tension-type headache. This would perhaps make the borderlines less vague, but it would not make the subjectivity any less.
Recall error is a problem, especially in patients with infrequent headache. In research, we are usually dealing with an infrequent and episodic problem. It can be difficult for the patient to recall symptoms 1 month previously. It can also be difficult for the patient to describe the pain characteristics. In children it is usual for them to say: “I don't know if my headache is pulsatile; I don't remember, or I have not paid attention to it”. It is not just difficult for the patients but sometimes also for the investigator to recognize that the patient has several different types of headache episodes.
These problems are often even more pronounced in children, especially small children. There are also some special problems in diagnosing headache in children.
Special features of childhood headache
There are some special features of childhood headache. Certain types of headache are considered to be less frequent in children than in adults (for example, cluster headache) and may also be tension-type headache. On the other hand, there are disorders like periodic symptoms which are connected with migraine and which are more frequent in children than in adults. The pattern of childhood headache changes with age and puberty. One of the problems is that we know so little about headache other than migraine in children.
Another typical feature of childhood headache is its changing pattern. Some features are considered to be more typical for children than for adults, e.g. aura, bilateral pain, nausea and vomiting. Recall errors can also contribute to these differences. It may be more difficult for children to recognize, for example, aura symptoms than it is for adults. Very little scientific data exist considering differences between childhood and adult headache features.
Children as patients
There are also problems dealing with children as patients. When we are dealing with small children we often get some of the information from the parents. Only certain symptoms can be observed objectively by the parents, e.g. vomiting or the impact of headache. Studies on child-parent discrepancy conclude that there is good agreement on pain intensity and on the disability caused by headache. However, there is disagreement on the characteristics of pain and on the provoking factors between children as patients and their parents (7, 8).
It may be even more difficult for children than for adults to recall their headache, to describe their headache in the given terms, and to identify different types of headache.
Aims of the study
The questions we have tried to answer dealing with the diagnosis of headache in children are as follows:
What happens to children with migraine? How does their headache change from childhood to adulthood?
Are there any particular features which should be taken into account in the criteria of headache for children?
How reliable is the information from children about their headache? Do they recall their symptoms? Do they get a correct diagnosis? How can we make this information better? We compared the information on the characteristics of headache gathered during an interview and in a headache diary.
Population and methods
We have followed-up a 1-year age cohort of children from the catchment area of Turku University Hospital. It started when the children were born, and a total of 5356 children were included. We asked about the occurrence and features of headache of all the available children of the cohort in a postal questionnaire when the children were 8 to 9 years old. At that age, the number of children taking part in the study was 3580, which is 66% of all children originally included in the study. At that age, there were 95 children with migraine (2.7%). Three years later, 84 of these children with migraine plus 67 controls with headache and 67 controls with no headache were interviewed face-to-face. After this interview, these children also kept a headache diary for between 2 and 7 months reporting the characteristics of each of their headache episodes during that time. Again, 4 years later, the children with migraine were contacted by telephone and asked about their present headache.
What happened to the children?
Fifty-three children of those who had had migraine at the age of 8 to 9 still had migraine at the age of 11 to 13 (63%). Four children had ceased to have headache (4.8%) and 27 (32.1%) children had headache episodes that did not fulfil the criteria for migraine (9). Out of the 53 children with migraine, 70% (37) still had migraine at the age of 16 to 17 years.
What happened to the symptoms?
The occurrence of migraine features between ages 11–13 years and 15–16 years was compared in the 37 children who had migraine both in childhood and in adolescence. The gastrointestinal symptoms became less marked but, instead, unilaterality was less typical in adolescents than in children. There were no changes in occurrence of photophobia, phonophobia, and pulsatility. The younger children reported aura symptoms more often. Headache frequency did not change much, but the mean duration became longer.
Special criteria for children?
Children have shorter migraine attacks than adults. As even the IHS limitation of the minimum headache attack duration of 2 h for children has been criticized for being too long, we tried to find out what happened to those children who had short migrainous headache attacks at the age of 8 to 9 years and did not fulfil the IHS criteria of migraine. Did they stop having migraine or did they get a true migraine? In our population, there were 19 children who had short (1–2 h) attacks otherwise fulfilling the IHS migraine criteria at the age of 8 to 9 years. At the age of 15 to 16 years, we could reach 18 of them. Using the IHS criteria, 58% had migraine at the age of 15 to 16 years. This information supports the idea that the limitation of minimum headache attack duration should be lowered for children.
Headache diary
One-hundred-and-forty-seven of the headache diaries were returned (67.4%). Forty-two children reported migraine features in this diary. There were nine “new” migraine patients. These children did not fulfil the criteria of migraine in the interview but reported at least one migraine attack in the diary. The children reported several features of pain and associated symptoms in the diary which they did not report in the interview. When they filled in the diary soon after their headache attack, they had a stronger recollection of the symptoms and were probably more aware of them. We got a better and more accurate image of their headache with the diary (10).
Several different types of headache
Thirty-six out of the 42 children reported not only migraine attacks but also other types of headache in the diary. These were headache associated with head trauma or infections, and migrainous headache which did not quite fulfil the criteria of migraine and tension headache. Fifteen children had both migraine and tension-type headache episodes. Only three children were able to report beforehand in the interview that they had several different types of headache (10).
We can conclude that headache in childhood is not a stable condition. There are some features of headache which are typical for childhood. A headache diary is a useful way of getting more accurate information about the features of headache episodes and about the different headache episodes one child may have. On the other hand, we know too little about the evolution and outcome of different types of headache in children, about the factors contributing to the outcome, and about the co-occurrence of different types of headache. This information would be useful both in research and in the clinic. It could help us decide how to treat the patient and when to intervene so that the headache problem does not turn out to be chronic and lifelong.