Abstract
In connection with a large-scale study of headache epidemiology in the parish of Vågå, Norway (n = 1838), a prevalence of exertional headache (EH) of 12.3% was recently reported (n = 202). The principal aim of the present report is to have a closer look at various clinical features: particular attention has been given to the coexistence of EH and migraine. EH seemed to coexist with migraine in 46% of the cases. In spite of that finding, there was generally no plentitude of ‘migraine features’ in the EH attacks per se. The predominant ‘migraine feature’ in EH attacks was throbbing. EH attacks were generally more short-lasting and seemed to be more lenient than the corresponding migraine attacks. Nausea/vomiting, and unilaterality of pain were generally not inherent features of EH attacks. EH attacks in all probability are not abortive migraine attacks – rather, they seem to be ‘migraine-like’.
Introduction
In a recent communication concerning the Vågå study of headache epidemiology (1), an ‘exertional headache’ (EH) prevalence of 12.3% was reported. Validation of the study was carried out by (1) blinded review of records (n = 100) and (2) repeat, blinded examination of parishioners (n = 41). For each of the repeatability studies there was good conformity between the original results and those obtained at the repeat study.
In contradistinction to what was the case in Rooke's study (where ‘exertion’ largely seemed to equate brief, Valsalva manoeuvre-like movements) (2), ‘effort’ has in the present context been taken to imply sustained, physically exhaustive exercise, just like in the studies of Pascual et al. (3, 4).
As pointed out by Diamond (5) and Massey (6) previously, it was clear also in our study that in addition to the well-known, short-lasting EH (2), there is a long-lasting variety, with attacks lasting for several hours (1). In the present communication, in which we shall elaborate further upon clinical features of EH, these two subgroups as for temporal pattern will be treated together, since the IHS (7) does not discriminate between them. The main body of the present communication will concern the migraine-like features of EH. At a later stage, we shall have a closer look upon the long-lasting variety specifically (8).
Materials and methods
The demography as well as an outline of the design of the Vågå study have been presented previously (9). Among a total of 3907 parishioners, there was a target group of 2075 18–65-year-old-ones, of which 1838 (88.6%) could be examined. Information regarding EH was obtained from 1646 parishioners, i.e. 79.3% of the target group (1). Information could not be obtained for all variables in all parishioners questioned. A total of 202 parishioners exhibited this picture (1).
The parishioners were subjected to a structured interview (carried out by the principal investigator, O.S.), based on an elaborate questionnaire. The parishioners themselves were not even allowed to see the questionnaire. Special, individual traits not infrequently necessitated deviations from the main track of questioning; the term ‘semistructured’ may therefore be an adequate characterization of the interview (1, 9). A short- or long- version clinical neurological examination was carried out, the latter only if it seemed clinically indicated due to special circumstances (9, 10) .
A pulsatile quality of the pain of EH has previously been reported (3, 4). Furthermore, migraine as such is not infrequently present in EH (5). Possibly, the pulsatile quality of the pain could be the only ‘migraine’ characteristic, regularly present in EH: it may be a characteristic of EH as such and not necessarily a trait, symbolizing the presence of a ‘migrainous tendency’. The fact that EH and migraine frequently do coexist (5) could, however, indicate that EH attacks actually are abortive migraine attacks. This conjecture will therefore be given some consideration. This will be done by an intra-individual comparison of ‘migrainous’ symptoms during ordinary migraine attacks and during attacks that followed exhaustion:
In addition to: 1, the throbbing quality of the pain, other migrainous traits (7) will be used in this comparison; 2, duration of attacks (4–72 h); 3, unilaterality of pain; 4, intensity of pain: moderate to severe; 5, accentuation of pain upon minor physical activity; 6, nausea; 7, vomiting; 8, phonophobia; 9, photophobia. The number of migraine characteristics was then summed up, the maximum score therefore being 9. To what extent were these ‘migraine features’ present in the EH attack?
For migraine diagnosis, the IHS criteria (7) will be adhered to. In a few cases, 5 of the aforementioned criteria were considered satisfactory. However, the ‘sieve’ has generally been placed rather high, i.e. at ≥ 6 migraine criteria, in order to secure as far as possible the migraine diagnosis and not to allow questionable cases to any appreciable extent. This study should, accordingly, not be perceived as a pure prevalence study of migraine occurrence in EH. It is rather a study of migraine symptoms in EH when EH coexisted with fairly typical migraine. An exception, as for the level of ‘migrainous’ symptoms and signs, was ‘classic migraine’ (migraine with aura), which might be symptom-poor, but where the classic traits indicated the true nature of the headache. Some of these phenomena may admittedly be hard for the patient to remember, i.e. to separately allocate them to either migraine or EH attacks. This, for instance, would seem to go for aggravation of pain by minor physical exercise (trait 5). This potential shortcoming may be accentuated by the fact that such attacks partly took place decades ago. Most parishioners were rather precise concerning headache duration (trait 2).
A strong, may be the strongest, solitary factor when evaluating whether migraine attacks are being reproduced or not during EH attacks may be laterality of head pain. If EH attacks steadily were bilateral, whilst the migraine attacks were unilateral (and even more so with side-shift), this would probably point strongly in the direction of a dichotomy.
Results
Duration of the entire period of complaints
Approximately 25% of those affected were able to give fairly precise information on this topic. Somewhat more than half of those who provided information had had attacks for less than 10 years (Table 1). It seemed to be rare to have had attacks for > 30 years (approx. 15%). It is worthy of note that many parishioners in this group (approx. 40%) claimed to have had this brand of headache only during a circumscribed period in early life: thus, 5 only during childhood; 12 only as youngsters; and 4 only ‘early in life’. There may also have been repetitive, symptomatic periods within a period of some years, each one seemingly being characterized by special circumstances, e.g. stress or nuchal complaints. Such ‘permissive’ factors could seemingly vary intraindividually.
Total period of complaints
Total number of attacks
Of 202 parishioners with EH, a total of 173 (approx. 86%) provided some information as to the lifetime number of attacks (Table 2). The majority of the parishioners (56%) seemed to have experienced less than 10 episodes. Also, the category ‘many’/‘multiple’ attacks may actually contain parishioners that belong to the category of < 10 attacks. Nevertheless, in a number of cases, there may have been a considerable amount of attacks.
Total number of attacks
Localization and laterality of the pain
Rather definite information as to laterality was obtained from only 24 parishioners. The pain was invariably reported as being ‘global’, and frequently the pain maximum seemed to be in the anterior region, frontally or temporally. No one pointed out a solitarily occipital pain localization or unilateral headache.
Coexistence of EH with migraine
A strikingly high proportion of the EH sufferers also seemed to suffer from migraine (Table 3). Of those with migraine, 31% had the with aura variety, 62% the without aura variety, while 7% seemed to have both varieties. In the migraine + EH combination, there was a certain, but nonsignificant, female preponderance (Table 3).
EH and migraine. Coexistence and sex relationship
∗Sex ratio vs. EH – migraine group (P = 0.16, Fisher's exact test). Percentage of migraine in EH cases(93 of 202): 46%.
Pounding was a typical symptom during the EH attack. Of those with rather definite information, 87% reported throbbing headache during EH attacks. In other words, throbbing did not seem invariably to be present in EH. In a few exceptional cases, the pounding seemed to supersede the headache. But pounding per se does not suffice as a symptom for EH: pain also has to be present. Phono-/photophobia seemed to have been present in 8% during EH attacks and nausea in only 5%. In the occasional case, the symptomatology of the EH attacks with regard to some selected features clearly had some ‘migrainous’ characteristics. It is remarkable, however, that two of the major migraine symptoms and signs, i.e. unilaterality of pain and vomiting, generally seemed to be lacking during EH attacks: Unilaterality of pain was apparently not present in anyone of those with rather adequate information, and vomiting in one single case only.
The duration of attacks (trait 2, Methods) generally seemed to be shorter in EH than in the corresponding migraine attack, probably with 2–3 exceptions. In one case, EH attacks even occasionally seemed to supersede migraine as regards attack duration. The intensity of pain in provoked attacks (trait 4, Methods) generally seemed to be at a lower level than in the corresponding migraine attack, but it occasionally seemed to reach migraine levels.
In a direct comparison between the EH attacks and the corresponding migraine attacks, the mean number of ‘migraine features’ was essentially different in the two headaches (Table 4). As indicated in the Table, there was almost no overlapping between the two ‘ranges’. The difference between the mean values for migraine and EH is impressive (Table 4). However, even statistical results at this level do not rule out systematic errors. In no case did the EH attacks seem to satisfy the IHS criteria for migraine.
Parishioners with a combination of EH and migraine (M) ± aura. Number of migraine criteria during the respective attacks
∗In four cases of M – aura, borderline fulfilment of M criteria of the IHS (number of criteria = 5); in the remainder: ≥6 criteria. In two cases of M + aura: 2 and 4 M criteria, respectively, were present. (It is well known that M + aura may be symptom-poor).
∗∗In two cases, there were 5 M criteria present during EH attacks, the remainder having 4 or less criteria. (In other words, 6 M criteria were not present in any case, see Methods). The mean values are obviously different (P < 10−1 5, paired samples t-test and Wilcoxon Signed Rank test).
Discussion
Exhaustion headache: reproduction of migraine attacks?
A high percentage of EH patients seemed to suffer from migraine (Table 3). The problem whether the EH attacks reflect the migraine is an intricate one. We may be faced with the following questions:
Is EH in fact only a regular migraine attack, in which the effort serves as the precipitating factor?;
or: Is EH a migraine attack in an abortive form?;
or: Is the EH attack after all essentially different from migraine?.
It is, of course, hard to answer these questions in a conclusive way. First, in many parishioners with EH, the episodes occurred a long time ago, and detailed and precise recollection cannot be expected. Second, migraine diagnosis even in modern time rests upon a fragile footing.
The view that EH attacks in reality generally are migraine attacks is hardly in agreement with various of the present observations:
I. Although an astonishingly large fraction of the individuals with EH also seemed to suffer from migraine, not all sufferers seemed to have a coexisting migraine. In 54% of the cases in the present study, there was no migraine (Table 3). Diamond (5) found migraine in 26% of his ‘prolonged’ cases. In the majority of cases therefore EH seems to exist as a separate disorder without any direct link to frank migraine.
II. The finding that various components of the migraine attack were not– or only to a small extent – reproduced during EH attacks to some extent speaks against the assumption that EH are abortive migraine attacks.
None of the EH attacks seemed to have ‘classic’ (‘migraine with aura’) features, such as visual phenomena. Thus, in none of the cases of migraine with aura attacks or a mixture of with/without aura attacks, i.e. in a total of 38% of the cases, could the migraine attacks be entirely reproduced.
Unilaterality of pain is one of the striking properties of the migraine attack. In Ekbom's series (11), unilaterality was present in 95% of the cases (mixture of with/without aura cases, mostly the latter variety); in a migraine (without aura) series from our Department (12) unilaterality was present in 91% of the cases. Unilaterality was not positively reported by any one in the present series as being a feature of the EH attack. Many parishioners, however, gave an incomplete and somewhat vague narration concerning laterality. Admittedly therefore the unilaterality figure in our series of EH might be too low. With these reservations in mind, it, nevertheless, seemed likely that there was a clear discrepancy between migraine and EH with regard to laterality of pain, in intraindividual comparisons.
Nausea/vomiting are other important manifestations of the migraine attack (7) and present in 70% in Ekbom's migraine series (11). In the present series with migraine and EH, a high proportion of the migraine sufferers had nausea during migraine attacks, but only rarely during EH attacks.
Duration of attacks. There were relatively few EH attacks that reached up to the minimum duration of a migraine attack, as stipulated by the IHS (four hours). This will be elaborated upon elsewhere (8).
The pulsatile component of the EH attack seems to be the main reminiscence of a migraine attack. Throbbing is an unspecific quality and per se hardly enough to decisively attach the label ‘migraine’ to EH. In a few cases, there may have been a reproduction of more components of a migraine attacks (Table 4). The most correct description is probably that the EH attack may develop ‘migrainous features’, as used by the IHS (7), or it may be: ‘migraine-like’.
The frequent coexistence of EH and migraine is, nevertheless, striking. A similar coexistence of ‘benign vascular sexual headache’ and migraine was found in the series of Silbert et al. (13): migraine was present in 47% of their cases. Another headache with a partial indomethacin response, the ‘jabs and jolts’ syndrome/idiopathic stabbing headache, has also been claimed to appear particularly frequently together with migraine (14, 15) and then with a clear female preponderance. A certain female preponderance was also observed in the EH + migraine cases (Table 3). Can there after all be some type of interconnection between migraine, jabs, and EH? The answer to this question may have to await the detection of a migraine marker.
The putative ominous nature of EH
Until recently, EH has been viewed in an ominous, threatening light – the threat being a potentially dangerous, underlying, intracranial disease. Rooke (2) warned that until proven otherwise, EH should in the single case be regarded as a potentially serious problem – with possible underlying, intracranial pathology. He claimed that EH on the background of organic disease probably is more frequent than the benign variety. The general attitude towards EH seems to have changed somewhat when evidence for a benign variety started to accumulate. Nevertheless, in the series of Pascual et al., there were 43% symptomatic cases (3), and in their series, we are faced with cases of exhaustion headache, like in the present series and not cases of headache following brief exercise.
In the Vågå study, 202 parishioners, 12.3% of the grown-up population, exhibited EH. In practically speaking every one of them, EH was connected with optimal health otherwise. Only a couple of them had, to the best of our knowledge, ever consulted a physician for this complaint. It is a close to inconceivable thought that most of these individuals should have a basic disorder, implying health hazards. It was, of course, entirely out of the question to do CT/MR brain/upper neck surveys in everyone of the EH sufferers. This may be another example illustrating the difference in the basic composition of clinic patients and grassroots-level individuals.
We would like to propose that the headache dealt with herein should be termed: ‘exhaustion headache’ or: ‘headache following physical exhaustion’.
Footnotes
Acknowledgements
We are indebted to GlaxoSmithKline of Norway, Pharmacia & Upjohn, and the Alf Harborg Foundation, Department of Neurology, St. Olavs. Hospital, Trondheim, Trondheim University Hospital, for generous support during the various phases of the investigation. The authors are also grateful to the personnel at the Vågå Health Centre at Vågåmo for their aid. Last, but not least, we thank the inhabitants of the Vågå commune for their collaboration.