Abstract
We identified clinical, demographic and psychological predictive factors that may contribute to the development of chronic headache associated with mild to moderate whiplash injury [Quebec Task Force (QTF) ≤ II] and determined the incidence of this chronic pain state. Patients were recruited prospectively from six participating accident and emergency departments. While 4.6% of patients developed chronic headache attributed to whiplash injury according to the International Classification of Headache Disorders, 2nd edn criteria, 15.2% of patients complained about headache lasting > 42 days (QTF criteria). Predictive factors were pre-existing facial pain [odds ratio (OR) 9.7, 95% confidence interval (CI) 2.1, 10.4; P = 0.017], lack of confidence to recover completely (OR 5.5, 95% CI 2.0, 13.2; P = 0.005), sore throat (OR 5.0, 95% CI 1.5, 8.9; P = 0.013), medication overuse (OR 4.2, 95% CI 1.4, 12.3; P = 0.009), high Neck Disability Index (OR 4.0, 95% CI 1.3, 12.6; P = 0.019), hopelessness/anxiety (OR 3.8, 95% CI 1.3, 8.7; P = 0.024), and depression (OR 3.3, 95% CI 1.2, 9.4; P = 0.024). The lack of a control group limits the conclusions that can be drawn from this study. Identified predictors closely resemble those found in chronic primary headache disorders.
Keywords
Introduction
Disability and chronic pain in whiplash-associated disorders following motor vehicle collisions are serious medical and socioeconomic problems. Although great amounts of data have been gathered on the diverse presentations of whiplash-associated disorders, little is known specifically about chronic headache attributed to whiplash injury.
Approximately 50% of people complain about headache following whiplash injury in the acute phase after the accident (1). In most cases acute headache attributed to whiplash injury has a good prognosis and resolves within the first couple of weeks after the accident without specific treatment (2). In a minority of patients, however, headache prevails for > 3 months and is then classified by the International Headache Society as chronic headache attributed to whiplash injury (3). Previous investigations have reported the incidence of chronic headache as between 8% and 15%, but also included patients with mild traumatic brain injury and different causes than whiplash injury (i.e. falls, blows to the head, etc.) (1,4). Mechanisms associated with the development of chronic pain in post whiplash injury patients are highly debated and range from psychosomatic (1) to diffuse axonal brain injury (5). Previous studies have not properly differentiated between headache and other whiplash-associated disorders, while underlying pathophysiological mechanisms might be very different in this regard, depending on the severity and mechanism of the caused damage.
As the therapeutic resources for chronic headache in general are limited and the overall consensus is that the earlier the treatment starts the better the prognosis, it seems reasonable to filter out those patients who are at high risk for the development of chronic headache attributed to whiplash injury in an attempt to prevent or attenuate its development. Therefore, we tried to identify predictive factors for the development of chronic headache attributed to whiplash injury using a semistructured, questionnaire-based personal interview.
Methods
The study was approved by the local ethics committee of the University of Duisburg-Essen. All participants gave their written informed consent according to the Declaration of Helsinki.
Persons with whiplash injury following motor vehicle accident were recruited prospectively from six emergency departments between February 2005 and December 2007. Inclusion criteria were: age > 18 years, onset of acute headache within 7 days after the accident, and Quebec Task Force (QTF) grade I or II (6). Individuals with head or neck injury in the past, brain injury, alcohol or drug abuse, major depression, or chronic headache disorder (i.e. chronic migraine, chronic tension-type headache or medication overuse headache) were excluded.
Questionnaire and instruments
All study participants were personally interviewed by one of the co-authors during the first 14 days after accident. The interview was semistructured and questionnaire based. The questionnaire had four parts: (i) personal data; (ii) medical inquiry including questions about personal and family history of painful disorders (headache, facial pain and back pain) and two established standard questionnaires: Kiel Inventory for the Coping of Pain (KSI) with three independently validated subtests [emotional reaction to pain (ERSS); cognitive reaction to pain (KRSS); coping reaction to pain (CRSS)] (7,8) and Questionnaire for the Assessment of Level of Coping with Pain [FESV (Fragebogen zur Erfassung der Schmerzverarbeitung)] (9); (iii) psychiatric and psychological evaluation including Freiburg Personality Inventory (FPI) (10) and the German version of the Hospital Anxiety and Depression Scale (HADS) (11); and (iv) Neck Disability Index (NDI) (12). In addition, all patients were asked to fill out a headache diary during the following 3 months. The diaries were collected during the follow-up interview 3 months later in case the headache subsided, but were continued in case of chronic headache. One year after the accident patients were interviewed again by phone, where information about the headache frequency (if any) and use of acute headache drugs during the last 3 months was collected. Chronic headache patients were seen again in person.
Statistics
The following variables were considered as possible predictors for the development of chronic headache: age (interval scaled, in years), gender (male vs. female), history of migraine (vs. no history of migraine), chronic facial pain (e.g. trigeminal neuralgia/neuropathy or idiopathic persistent facial pain vs. no facial pain), confidence to completely recover (vs. not confident), sore throat during the first days after accident (vs. not having this symptom), overuse of acute headache medication (vs. no overuse), education (high, i.e. high school or university, vs. low), neck disability index (scored > 25 vs. NDI ≤ 25), helplessness (KRSS score > 30 vs. KRSS score ≤ 30), depression (HADS score > 5 vs. HADS score ≤ 5). The primary outcome variable for the incidence assessment was the cumulative incidence of chronic headache (i.e. proportion of patients who developed chronic headache according to the International Classification of Headache Disorders, 2nd edn (ICHD II; headache that develops within 7 days after whiplash, and persists > 3 months) (3). The secondary outcome variable was the proportion of patients who developed chronic headache according to the definition of the QTF (headache develops after whiplash and persists > 42 days) (6).
We used seven univariate logistic regression analyses to assess the impact of potential predictors on the development of chronic headache according to the definition of the QTF because the number of patients with chronic headache attributed to whiplash injury according to the ICHD-II was too low. P < 0.05 was considered to be statistically significant. All analyses were performed using
Results
Demographics and clinical characteristics
All values ± standard deviation and range in parentheses.
As defined by the International Classification of Headache Disorders, 2nd edn (ICHD-II).
Analgesics only.
QTF, Quebec Task Force; VRS, verbal rating scale.
Six patients (4.6%) developed chronic headache (ICHD-II) 3 months after the accident, complaining about daily or near-daily headache with a mean age of 35.3 ± 13.7 years (66.7% women). None of them still had chronic headache after 1 year. Using the definition of QTF we identified 20 patients with chronic headache (15.2%) with a mean age of 36.4 ± 12.6 years (70% women). No distinct clinical differences between chronic and non-chronic headache associated with whiplash injury patients could be determined (Table 1).
Participants were asked to categorize the quality of their headaches. Thereby, multiple answers were possible. There was a slight preponderance of dull (47.4%) and pressing (51.1%) pain quality; 21.8% of patients described their headache as pulsating, 25.6% as a stabbing pain and 30.1% as drawing. Most whiplash patients experienced bilateral headaches (77.2%). Unilateral pain was unusual (left, 5.3%; right, 2.3%); 13.7% stated that headaches were worse on the left, 11.5% that headaches were worse on the right. The majority of patients (57.9%) localized the pain at the forebrain or temples; 12.8% stated that the whole head hurt. Only 7.5% noted pain at the vertex.
Only 5.3% of whiplash patients did not suffer from additional post-traumatic neck pain initially. In 73.7% of cases, the pain affected the lateral muscles of the neck. Pain in the shoulders was suffered by 69.2% of patients. Regarding the caudal extension of pain radiation, 48.9% suffered from pain in the upper back, and 33.1% reported even lower back pain after whiplash injury. None of these accompanying complaints lasted longer than the headache, and all patients reported that the headache was their major disabling factor. None of the chronic headache patients reported accompanying pains other than headache at 3 months' follow-up.
Predictors for the development of chronic headache associated with whiplash injury (N = 20 fulfilling Quebec Task Force criteria)
CI, confidence interval; HADS, hospital anxiety and depression scale; KRSS, cognitive reaction to pain (KSI subtest).
P < 0.05 was considered statistically significant; multiple univariate regression analyses were used.
Discussion
The data of our study suggest that the incidence of chronic headache attributed to whiplash injury is low (4.6%) and generally has a good prognosis with a recovery rate in our cohort of 100% within the first year after the accident. Therefore, even though headache is one of the most prominent complaints following whiplash injury, it does not seem to play a substantial part in the pain and disability often described in patients suffering from whiplash-associated disorders over the long term (13).
This study aimed at the identification of predictors for the development of chronic headache in this particular patient population. It was not designed to determine what predictors are different and which are specific to headache attributed to whiplash injury compared with chronic primary headache forms. It should be kept in mind that we did not have a proper control group of patients with whiplash injury but without headache or comparable healthy control group without accident from the general population to allow this comparison. Thus, interpretation of the following results remains on a descriptive level and should be regarded as such when drawing conclusions from them.
Predictors for the development of chronic headache attributed to whiplash injury are multifactorial, but resemble previous observations on post-traumatic headache, chronic pain patients in general and primary headache patients (i.e. chronic tension-type headache or medication overuse headache) in particular (1,14–18). Prominent predictors seem to be pre-existing pain conditions, such as facial pain in our cohort. Different studies were able to identify different previously existing pain conditions, such as headache or lower back pain as predictors for the development of chronic headache (19,20). The concept of pre-existing frequent pain facilitating the development of chronic pain even when located elsewhere in the body is very interesting. From a pathophysiological point of view this is easily conceivable, as most chronic pain conditions are associated with central facilitation (21–23). Experimental evidence in this regard has also been provided by Sterling et al., who demonstrated a generalized sensory hypersensitivity to a variety of sensory stimuli that occurred soon after the whiplash injury and that was associated with poor recovery (24). High initial pain as described by other studies (25) did not seem to be an important factor in our cohort, as pain intensity within the first 5 days after the accident was only slightly higher than that after this initial phase and remained rather constant on moderate pain levels (Table 1).
A factor that is recognized more and more in recent literature is medication overuse. Medication overuse headache has increased to being the third most common form of chronic headache after chronic tension-type headache and chronic migraine, with prevalence in the general population between 1% and 3% (26–28). Most (69.3%) of our patients reported the usage of pain medication (analgesics only, ibuprofen the favourite) with a mean duration of 9.3 ± 9.8 days after the accident. Even though the underlying pathophysiological mechanisms are insufficiently understood, it appears that headache patients following whiplash injury should be advised to restrict their medication intake to < 15 doses per month (29).
Orthopaedic and neurosurgery literature naturally has a more mechanistic approach to whiplash-associated disorders and often refers to the cervical facet joint as most common source of chronic neck pain and accompanying headache following whiplash injury (30,31). Pathophysiological considerations include micro trauma to the musculoskeletal systems of the neck (31). This may be represented by our finding that patients who reported a sore throat after the accident were more prone to the development of chronic headache and assuming that a sore throat would indicate additional or more pronounced accident-related minor muscular injury of the throat, neck and larynx. The same assumptions could be drawn from interpreting the NDI score, as it represents the everyday disability patients face after whiplash injury due to movement restrictions of the head and neck, as well as neck pain and post-traumatic headache. This, however, requires further research.
The psychological factors we found are in line with previous research concerning the development of chronic pain (25). Surprisingly, depression is the weakest prognostic factor that we determined, even though it is generally considered to be a serious complication and comorbidity, especially in chronic pain patients (32). Helplessness/anxiety and lack of confidence to recover completely are somewhat interconnected factors, as this helplessness and anxiety may result from lack of confidence that patients' perceived pain will stop eventually. A different aspect of confidence to recover completely could be the wish for legal compensation associated with the accident, as proposed previously (1,33). Unfortunately, our study was not designed to evaluate this phenomenon any further. A very thorough meta-analysis evaluated 50 papers with 29 cohorts on whiplash-associated disorders and identified the following psychological predictors for delayed recovery: a high number of complaints, previous psychological problems, and nervousness. Although the latter could be related to our finding anxiety/helplessness, we were unable to reconfirm the other suggested predictors. The authors of this meta-analysis stated that common, reliable predictive factors are hard to establish due to the heterogeneous definitions and included symptoms of whiplash-associated disorders. Outcome measures were seldom validated and responsive instruments such as the NDI, KSI, FESV, FPI, or HADS were rarely used, making it difficult to compare previously reported results with the general population (25). Therefore, it is useful to consider different aspects of whiplash-associated disorders like headache associated with whiplash injury alone, without confounding factors often attributed to whiplash injury syndrome such as dizziness, difficulty in concentration, nervousness, insomnia, irritability, decreased work ability, depressed mood, etc., and evaluate patients with standardized approved instruments to collect valid data.
The clinical characteristics that we determined in patients with chronic headache attributed to whiplash injury resembled those typically seen in tension-type headache for most patients. This is in line with previous studies. Haas et al. investigated 48 patients with chronic post-traumatic headache and found that 75% resembled tension-type headache, 21% migraine without aura, and 4% were unclassifiable (15). Schrader et al. also reported shared clinical features of primary headaches and chronic headache attributed to whiplash injury (1). They even proposed that this chronic headache was not caused by the accident at all (17).
The incidence of 4.6% of chronic headache associated with whiplash injury is close to findings reported by a large US population-based study on chronic daily headache (CDH) that included 1134 subjects. The reported incidence of CDH was 3% in that cohort (34). A large community-based study on outcomes and predictors of CDH in Taiwan reported a medication overuse headache incidence of 6% and identified depression and anxiety as major predictors for poor outcome (35). This close resemblance of reported incidence rates in different forms of chronic headache may hint at similar pathophysiological or clinical mechanisms associated with the development of chronic headache in general.
Some limitations to our study have to be addressed in addition to the lack of a control group mentioned above. The low participation rate of patients who suffered from acute headache associated with whiplash injury as well as the patient recruitment from emergency departments represents a bias of our data that has to be considered when interpreting results. The true incidence of chronic headache associated with whiplash injury is probably even lower than we have reported here given the assumption that those patients that are most severely affected and might have one of the identified predictors are more likely to participate in a study than mildly affected otherwise healthy patients. We aimed to evaluate the predictors for the development of chronic headache attributed to whiplash injury and not its ‘true’ incidence in the general population. There is a possibility that this particular headache initially, or after an undetermined period of time, is not causally related to the initial whiplash trauma in some chronic headache patients. Some chronic headache patients following whiplash injury could have developed chronic headache despite the accident. After several months it might be very hard to discern which findings are due to the accident itself and which result from all the different influencing factors that the individual patient is confronted with in the mean time. The low power due to low chronic patient numbers may have influenced the identification of predictive factors in the statistical model used and will need reconfirmation in future research. We tried to attenuate this limitation by using less restrictive QTF criteria for analysis.
In summary, clinical characteristics of chronic headache attributed to whiplash injury are quite similar to those described in tension-type headache with generally moderate pain levels and little variation in clinical presentation. It appears to be an infrequent disorder with a good prognosis. Predictive factors are multifactorial and include aspects that have been identified in previous studies on chronic headache disorders. The biological nature and underlying pathophysiological origin of chronic headache attributed to whiplash injury remain unresolved.
Footnotes
Competing interests
None to declare.
Acknowledgments
This work was supported by the German Federal Ministry of Education and Research BMBF 01EM 0513.
