Abstract
Acute and chronic whiplash headache are new diagnostic entities in the ICHD-2 (5.3, 5.4). In a prospective cohort study, 210 rear-end collision victims were identified consecutively from police records and asked about head and neck pain in questionnaires after 2 weeks, 3 months and 1 year. The results were compared with those of matched controls who were also followed for 1 year. Of 210 accident victims, 75 developed headache within 7 days. Of these, 37 had also neck pain and complied with the criteria for acute whiplash headache. These 37 had the same headache diagnoses, headache features, accompanying symptoms and long-term prognosis as the 38 without initial neck pain who therefore did not comply with the acute whiplash headache diagnosis. Previous headache was a major risk factor for headache both in the acute and chronic stage. Compared with the non-traumatized controls, headache in the whiplash group had the same prevalence, the same diagnoses and characteristic features, and the same prognosis. Both acute and chronic whiplash headache lack specificity compared with the headache in a control group, and have the same long-term prognosis, indicating that such headaches are primary headaches, probably elicited by the stress of the situation.
Introduction
Acute and chronic whiplash headache are new diagnostic entities in the ICHD-2 (5.3 and 5.4) (1), but the validity of these diagnoses has never been analysed. To fulfil the criteria for these diagnoses, headache (no specific features specified) at the time accompanied by neck pain must occur within 7 days after a whiplash trauma, i.e. an extension of the neck followed by flexion. In the acute form, the pain subsides within 3 months, and in the chronic form it persists after 3 months.
The facts that neck pain usually accompanies the most prevalent primary headaches (2), that the headaches after traumas have been shown to be similar to primary headaches (3), and that a false (‘placebo’) rear-end collision without any extension/flexion of the neck may give rise to head and neck pain (4), all raise the suspicion that headache after whiplash may represent a primary headache induced by the stress of the situation. To investigate this issue further, we used data from a prospective controlled inception cohort study performed in Lithuania from which the main data on headache and neck pain had been published previously (5, 6). In this study we had good data on presence of head and neck pain of accident victims after rear-end collisions, on many headache features allowing us to make more specific headache diagnoses, on the exact timing of the symptoms after the accident, and on the long-term prognosis of the pain. Specificity of symptoms and prognosis could be tested by comparing accident victims whose headache fulfilled the criteria for whiplash headache with those whose post-accident headache did not fulfil these criteria, and by comparing the same properties of the accident group as a whole with a matched non-traumatized control group.
Subjects and methods
In this prospective study (5, 6), 210 accident victims of rear-end car collisions were consecutively identified from the daily records of the traffic police in Kaunas, Lithuania. Two weeks after the accident they received questionnaires inquiring about different symptoms, and also after 2 months and 1 year. In the questionnaire to be filled in after 2 weeks, the victims were asked specifically about head and neck pain before the accident and about the situation after the accident. After 2 and 12 months, they were asked about their current head and neck pain. As controls we used 210 non-traumatized sex- and age-matched individuals taken randomly from the population register of the same geographical area. They received questionnaires shortly after identification and 1 year later, and these questionnaires were similar to those given to victims after 2 and 12 months.
Results
According to the questionnaires, symptoms such as neck pain and headache after the accidents lasted maximally 2–3 weeks. First, we analysed possible differences between accident victims with headache who did or did not fulfil the criteria for whiplash headache. Among the 210 accident victims, 75 reported in the questionnaire after 2 weeks that they had developed a headache within 7 days, which is one criterion of whiplash headache. Among these 75, 37 did report neck pain after the accident, hence fulfilling all the criteria of whiplash headache; 38 had no neck pain, hence not fulfilling the criteria. When we compared the diagnoses and the specific features of the 37 who did and the 38 who did not fulfil the criteria, there were no significant differences after 2 or 12 months. Neither were there differences with regard to the prognosis of headache or the neck pain between these two groups (Fig. 1). With regard to the situation before the accident, a significantly higher proportion of those who fulfilled the criteria for whiplash headache had had neck pain. At 2 and 12 months, however, this difference had disappeared.
Prognosis of headache in accident victims fulfilling or not fulfilling ICHD-2 criteria for whiplash headache.
As we had data on head and neck pain before the accident, we could analyse the effect of pre-accidental headache on the occurrence of acute and chronic pain. Those who reported headache before had a significantly higher risk (6–10 times increased, depending on the frequency of the pre-accidental headache) for headache both in the short and long term after the accident than those without pre-accidental headache.
We then analysed whether there were differences in headache features or prognosis between accident victims and the matched controls. Compared with the non-traumatized controls, headache in the whiplash group had the same prevalence, the same diagnoses and characteristic features, and the same prognosis as the control group at 2 and 12 months (Table 1).
Headache diagnoses, frequency and features in rear-end collision victims before and up to 1 year after collision, and in controls at identification and 1 year later
CGH (cervicogenic headache) features: precipitation of head pain by neck movement + restriction of motion in the neck.
n.d., no data; TTH, tension-type headache.
Chi-squared test.
Discussion
The main finding of this study was that the headache after a rear-end collision lacks specificity both with regard to headache features and with regard to the prognosis. This is true both when accident victims fulfilling the criteria of acute whiplash headache were compared with those who did not fulfil the criteria, and when accident victims as a group were compared with non-traumatized control subjects. This confirms a previous analysis indicating a low nosological validity of these headache diagnoses (7), and it decreases the likelihood that long-lasting headaches should in some way be related to any chronic injury to the neck. All in all, it seems highly probable that the acute headaches after such accidents are occurrences of a primary headache (mostly migraine or TTH), possibly related to the psychological shock or stress of the situation. This also explains why the prognosis of headache is similar to that of a non-traumatized control group.
As would be expected, those who developed headache within 7 days after the accident were mostly those who had had headache previously. This fact illuminates in an interesting way most of the previous studies on whiplash. These studies have included patients who reported some complaints after the accident and therefore presented to a doctor, in an emergency room, to a physiotherapist or in a lawsuit. The present study shows that inclusion in such clinic-based studies will tend to be heavily biased towards the headache-prone part of the population, and any findings in such studies may be explained by this selection bias rather than by any hypothesized neck injury. In contrast, the present accident-based study avoids such selection-bias.
There are some definite advantages with performing such a study in Lithuania, because in this country (i) there has been little awareness of the notion that chronic symptoms may result from rear-end collisions, (ii) accident victims with acute symptoms from rear-end collisions generally view this as a benign injury not requiring any medical attention, and (iii) there is no possibility of economical compensation. A few years after the present study was performed one study documented that there indeed was very little expectation of long-lasting headache after whiplash injury by Lithuanians compared with Canadians. Thirty-two per cent of 179 employees in a Canadian company and only 1% of 171 employees in a Lithuanian company expected they could get long-lasting headache after an imaginary scenario with whiplash after rear-end car collision (8). This makes it possible to study the true natural course of the condition in Lithuanians, whereas in many other countries, the natural course may be influenced by psychological factors and possibly harmful examinations and treatments, and the reporting of problems before and after the accident can be influenced by negative expectations and the possibility for secondary gain.
Our interpretation, that also the acute pain after rear-end collisions of this type very often represents a stress-induced primary headache, has been suggested previously in connection with headache related to other types of accidents (3). This is also a likely interpretation of the results of an experimental study in which a high percentage of volunteers developed head and neck pain for several days after having unknowingly been subjected to ‘placebo’ rear-end collisions with minimal movement and maximal stress and worry (4). In Lithuania, where very few people had any car insurance at the time of our study, the stress and worry connected with car accidents are at least as high as in any western country.
Conclusion
Both acute and chronic whiplash headache lacks specificity compared with the headache in a control group, and it has the same long-term prognosis, indicating that such headaches are primary headaches, probably elicited by the stress of the situation.
Conflicts of interest
The authors have not declared any conflicts of interest.