Abstract
According to the study of Obermann et al. in the current issue of Cephalalgia, headache after whiplash injury is a self-limiting condition with a good prognosis. Although 4.6% fulfilled criteria for chronic whiplash headache, i.e. still having headache after 3 months, none had headache lasting more than 1 year. Predictors for headache lasting > 3 months were medication overuse, pre-existing pain conditions, and several psychological factors. Another factor, sore throat after the whiplash, was somewhat speculatively interpreted as a possible result of musculoskeletal damage. The headaches were mostly compatible with tension-type headache (TTH).
Although the study has a positive message, that the headache subsides in less than a year, it also demonstrates some very problematic aspects of the whiplash headache entity. First, the criteria contain no description of the headache itself. It is sufficient that the headache and neck pain occur within 7 days after an event with a possible whiplash mechanism. For the chronic whiplash headache, it is required that the headache persists for > 3 months after the accident, but there is no requirement that the headache occurs with a certain frequency. This is in contrast to, for example, chronic TTH, where the headache must occur on more than half of the days for > 3 months. Quite wisely, therefore, the authors have in their definition of chronic headache added the requirement that the headache shall occur daily or near daily. Nevertheless, to find enough patients with ‘chronic’ headache for the analysis, they also used the criteria of the Quebec task force (QTF), but also these criteria have no definition of chronicity.
An even more problematic aspect of the chronic whiplash headache entity is that, according to the very lax International Classification of Headache Disorders, 2nd edn (ICHD-II) criteria, in principle any headache that occurs in an individual for years after the event can be classified as chronic whiplash headache, provided that there was head and neck pain during the first week. This is so even though this single event leaves no trace in the body that can be demonstrated by objective means, and even if the headache resembles the naturally occurring primary headaches, as demonstrated in this and previous studies (1,2). This issue is even more crucial after the demonstration that even sham whiplash accidents, involving virtually no physical forces but many psychological stressors, could induce head and neck pain in up to 20% of the exposed individuals within the first days after the event (3), hence allowing a diagnosis of whiplash headache according to the current ICHD-II criteria. Thus, it may be that so-called whiplash headache mostly or even exclusively represents primary headache induced by the stress of the situation (2). Theoretically, it is highly unsatisfactory to have a headache category named after each type of precipitant for headache episodes, psychological or other. From a practical viewpoint, it is probably unsound to name the headache after the whiplash event if the headache is a primary headache, particularly since most patients and health professionals will then tend to understand the aetiology in terms of a physical injury to head and neck structures. Many unhealthy practices (e.g. use of a neck collar, hazardous neck operations) that may increase the complaints stem from a traumatic conception of the headache, and the prognosis of headache after rear-end collisions seems to be best in cultures where the traumatic event is ignored (4).
In the present study the incidence of chronic headache is consistent with the prevalence figures of chronic headache seen in population-based studies, and the authors wisely comment that development of chronicity in this whiplash group may have similar pathophysiological or clinical mechanisms to chronic headache in general. In my opinion, this raises the issue of whether whiplash headache is a biologically meaningful entity, i.e. caused by injury to head or neck structures. To solve this problem, descriptive case series based on patients presenting in an emergency room or to a doctor some time after the accident are not very helpful, even if they include a follow-up. Such clinic-based studies have been done again and again for at least 50 years (5), and new studies with the same method add relatively little to the understanding of this very problematic headache entity. One important reason for this is the heavy selection probably involved in the recruitment of patients to clinic-based studies. Accidents with a whiplash mechanism are extremely common. According to an estimate from Norway based on car insurance data in year 2000, > 0.5% of the population were potential whiplash accident victims in the sense that they had been in a motor vehicle that had been hit from the rear by another vehicle during that year (6). Only a tiny minority of these present with complaints to a doctor, even though almost 50% of those involved in such accidents may have acute headache, according to a study with accident-based inclusion, i.e. inclusion based only on involvement in a rear-end collision (7). The selection mechanisms in clinic-based whiplash studies in a country like Germany are unknown, but one may suspect that patients' attitudes and beliefs may be important selection factors, in addition to factors like pre-accident pain, and possibility of secondary gain. Whatever is found in such a population may therefore be highly culturally driven and have relatively little to do with a purported neck injury. To study properly prospective factors, one should use accident-based inclusion, and with a carefully matched control group to study whether headache features, prognosis and prognostic factors are different from those in a population not exposed to a whiplash. In studies where this has been done, no differences in prognosis and features have been found (2).
It has been shown that negative expectations related to common whiplash are high in some western countries (8,9), and this implies that whiplash can be an important nocebo, i.e. a factor that can cause harm by purely psychological mechanisms (for an example in the headache field, see (10)). To counteract such negative influence of the diagnosis, I welcome the positive message of the good prognosis found in the study reported in this issue. However, I also fear that studies like this, which somewhat naively accept the concept of whiplash headache with all its pathophysiological implications and negative connotations, may unwittingly promote the use of a diagnosis with a weak scientific basis, and which can do more harm than good.