Abstract
Aim
To explore post traumatic headache characteristics and risk factors in compensation claimants by observational retrospective cohort analysis.
Case results
Medicolegal reports on 116 consecutive compensation claimants aged 41.9 ± 15.0 years were reviewed 21 ± 14 months after injury. Eighty eight had suffered head and neck injuries, 21 reported only neck injury and seven had “other injuries”. Ninety four percent of the head injuries were “mild”. The incidence of post traumatic headache following neck injury did not differ from that following head and neck injury, and none of the “other injuries” cases developed post traumatic headache. We anticipated that all head and neck injury claimants would seek compensation for post traumatic headache, but 25% denied developing headache. Post traumatic headache was very strongly correlated with a past history of primary headache (p < 0.0001) but no other risk factors were identified. Post traumatic headache semiology was consistent with “migraine” or “probable migraine” in 90% of cases. Headache resolved in 30% of claimants between 3 and 24 months after injury but 70% continued to suffer headaches at the time of assessment. Forty one percent of claimants had received no treatment for post traumatic headache in primary care.
Conclusions
Our data suggest that post traumatic headache is essentially “migraine” provoked by head or neck concussion. It is not clear why so many post traumatic headache sufferers receive poor or inadequate treatment for this condition.
Introduction
At least one and a half million people attend UK hospitals each year following head injuries. Eighty five percent are classified as “minor” (“concussion”, mild traumatic brain injury (MTBI)), 10% as moderate and 5% as severe (1). Headache is the commonest resulting symptom, and the current International Classification of Headache Disorders (ICHD-3 beta) recognises “acute” and “persistent” post traumatic headache (PTH) as a “secondary” headache (2).
The study of PTH in litigants has some advantages with regard to data capture since full disclosure of medical records is mandatory. Important information that might be forgotten or concealed during routine clinical evaluation may be revealed, such as witness accounts of litigants' behaviour in the immediate aftermath of the injury, previous episodes of head injury and compensation claims, and earlier consultations for headache.
A successful claim for PTH in the UK typically attracts significant financial compensation and we thought it likely that all head and neck injury claimants would complain of headache. Pending litigation is also often cited as a possible contributory factor in the pathogenesis of PTH. We therefore studied a cohort of claimants seeking compensation for accidental injury to identify factors that influence susceptibility to PTH and its prognosis.
Methods
We reviewed a convenience sample of 147 consecutive medicolegal documents written by one of the authors (RL) between April 2012 and February 2016. After exclusion of irrelevant material, 116 reports to the UK court concerning individuals seeking compensation for traumatic injury were available for analysis.
Each claimant had been examined on a single occasion some two years (21 ± 14 months) after injury. We recorded claimant demographics, head injury severity according to the Mayo Clinic classification (3), headache phenotype as defined by ICHD-3 beta criteria, past and family history of headaches, and any pre-accident history of psychiatric or psychological disorder. We also documented the incidence of post concussion syndrome (PCS) as defined under International Statistical Classification of Diseases and Related Health Problems 10th Revision (ICD10), the duration of PTH at the time of assessment and the treatment prescribed for the headaches in primary and secondary care.
Descriptive statistics for continuous variables were calculated using the Microsoft Excel statistical package, and chi-squared tests for ordinal data were performed using 2 × 2 contingency tables (see Appendix). χ2 values giving p-values of < 0.05 were considered significant.
Results
The 116 compensation claimants comprised 53 males and 63 females aged 41.9 ± 15.0 years. The injuries resulted from road traffic accidents (RTAs) in 73 cases, workplace injuries in 23, and other accidents such as accidental “mechanical” falls in 20 cases.
One hundred and nine out of 116 claimants had suffered head and/or neck injury. Of these, 88 had suffered head and neck injury, and 21 claimed neck injury only. The remaining seven cases had suffered injuries to body areas other than the head or neck (leg injury with multiple nerve damage sustained in military action, electrocution of right arm, chronic left-sided meralgia paraesthetica following RTA, relapse of multiple sclerosis following RTA, acute right-sided carpal tunnel syndrome after RTA, right radial digital nerve injury in a workplace accident, and right arm pain of unknown cause following RTA).
Head injuries were classified as minor (“possible traumatic brain injury” under the Mayo system) in 102/109 (94%) of the head and/or neck injury claimants. Only seven had suffered “moderate/severe traumatic brain injury” under this classification.
Incidence of PTH
Eighty two out of 109 head and/or neck injury cases (75%) claimed to have developed headaches or to have suffered worsening of pre-accident headaches. However, 27 claimants (25%) stated that they had not developed headaches following injury. Thirteen out of 21 (61.9%) of the “neck only” injury claimants developed PTH compared to 69/88 (78%) head and neck injury cases. This was not significantly different (χ2 = 2.48, p = 0.12). None of the claimants who had not suffered head or neck injury developed PTH.
Time to onset and duration of PTH
PTH developed immediately after or within an hour of injury in 38/82 cases (46%) and within a day in a further 19 cases (23%). Overall, PTH developed within a week in 90% of cases and in all cases within 6 weeks. PTH had resolved before assessment in 25/82 of cases (30%), with an average time to resolution of 6.96 months (mode 2 months, range 3–24 months). However, the remaining 57 PTH claimants (70%) had ongoing headaches at the time of assessment.
PTH phenotypes
PTH phenotype was consistent with ICHD-3 beta-defined “episodic migraine without aura”, “chronic migraine” or “probable migraine” in 73 of the 82 PTH cases (89%). In 19 cases, the principal phenotype “migraine without aura” was complexed with other ICHD-3 beta phenotypes either concomitantly or as independent events: focal head pain at the site of impact in seven cases, primary stabbing headaches in six, and “neck triggered” (cervicogenic) headaches in six (all being “neck injury only” cases). Seven further cases had attacks of “migraine with aura”, and in two “migraine without aura” cases the headaches were associated with prominent trigeminal autonomic symptoms.
Risk factors for PTH (Table 1)
Risk and prognostic factors for PTH.
LOC: loss of consciousness; PCS: post concussion syndrome; PTSD: post traumatic stress disorder; TBI: traumatic brain injury.
Loss of consciousness at the time of head injury (an indication of impact severity) was recorded in 30/69 head and neck PTH cases and in 10/19 who did not develop PTH. This was not significantly different. There was also no difference between the PTH and non-PTH groups with respect to a family history of headaches or previous psychiatric or psychological disorder (most commonly depression and anxiety). However, a past history of primary headache was very strongly associated with PTH: 40/69 PTH cases vs. 2/19 non-PTH (χ2 = 15.2, p < 0.0001).
Prognostic indicators (Table 1)
Of the 25 PTH cases who reported resolution of headache prior to assessment, half had a past history of primary headache and half did not. A family history of headaches similarly carried no prognostic significance.
There were too few PTH cases with a past history of psychiatric or psychological disorder to determine any influence of such disorder on prognosis. Fifteen out of 82 head and neck injury PTH cases fulfilled criteria for PCS. Four (26%) reported recovery from PTH by the time of assessment while 11 (73%) did not. The difference was not significant (χ2 = 0.13, p = 0.72).
Only 48 of the 82 PTH claimants (59%) had been prescribed treatment for the condition in primary care by the time of assessment: 26 (54%) analgesics (not opiate based) or non steroidal anti-inflammatory drugs, seven triptans and 13 anti-migraine prophylactics (five amitriptyline, two nortriptyline, one propranolol, one topiramate, one an SSRI and three several of these in sequence). Only 14 of the PTH cases had been referred for specialist opinion despite having headaches over months or years. Nine were prescribed anti-migraine prophylactics (three amitriptyline, one nortriptyline, three gabapentin, one an SSRI, and one verapamil). One received a triptan, one underwent a greater occipital nerve block and subsequently botox treatment, and three received no additional treatment.
Of the 25 PTH claimants who reported resolution of headache by the time of assessment, 10 had been prescribed treatment for headache and 15 had not. Of the remaining 57 cases with ongoing PTH, 24 had been prescribed treatment and 33 had not. The claimants prescribed treatment were no more likely to report resolution of PTH than those that had not received treatment (χ2 = 0.0317, p = 0.86).
Discussion
Seventy-five percent of the head and/or neck injury claimants in our cohort developed PTH, consistent with a large prospective study in the United States (4). Onset was within a day in nearly 70%, within a week in 90% and within 6 weeks in all cases. There was no difference in the incidence of PTH following neck injury only compared to head and neck injury, and none of the claimants who had suffered injury to other body areas developed PTH.
The only unequivocal predictive factor identified for PTH was a past history of primary headaches, found in almost 50% of the cases. This is greater that reported in other studies, probably because we had access to claimants' comprehensive medical records, which often disclosed forgotten episodes of primary headache sometimes misdiagnosed as due to ophthalmic disorders or “sinus” headache.
Previous cohort studies of PTH with data on headache phenotype and PTH risk factors.
LOC: loss of consciousness; PTSD: post traumatic stress disorder; TBI: traumatic brain injury; TTH: tension type headache; US: United States of America.
The reported prevalence of PTH a year after head injury was high in all these studies, consistent with our observations. “Migraine” was by far the most common reported PTH phenotype (Table 1) (12,13), as we also found. There is little consistency in these studies with regard to PTH risk factors. In some, an association with a past history of primary headache was noted (4,13) and contrary to our findings, female sex and psychological or psychiatric factors were also identified (10,11). Head injury severity is also sometimes stated to be inversely proportional to PTH risk, but this was not found consistently (4–6,8). A past history of head injury was also noted to predispose to subsequent primary headache (14).
The effect of litigation on the risk of developing PTH has been little studied. De Benedetis and De Santi reported in 1983 that a pending lawsuit had no significant influence on the risk of PTH (5). However, a later study from Serbia (15) noted that while earlier studies had supported the notion of “compensation neurosis”, others showed that favourable financial settlements did not contribute to the resolution of symptoms after head injury. They studied the influence of pending litigation on head injury symptoms, including PTH, in 90 patients and found that individuals with moderate to severe traumatic brain injury (TBI) rarely complained of headache unless they were involved in a lawsuit. Conversely, pending litigation had no significant effect on the risk of PTH following mild traumatic brain injury, in keeping with our findings.
Recent reviews of the pathogenesis of PTH propose a “biopsychosocial” model in which pathology caused by traumatic brain injury – such as inflammation, biochemical changes and axonal and neural network damage – is complexed with adverse emotional and psychological responses. Medication overuse headache, pending litigation and cultural attitudes are also considered to be important influences (16–19). In studies published in 2001, expectations of acute symptoms that might follow head injury were found to be similar among normal young adults in Edmonton, Canada, to those in Kaunas, Lithuania, and Patras, Greece (20,21). However, the respondents in Lithuania and Greece, countries where litigation and insurance claims for injury were extremely uncommon at the time, thought chronic symptoms would be much less likely to occur than did the respondents in Edmonton. In a subsequent study from Kaunas, all 200 people who had suffered concussion with loss of consciousness for more than 15 minutes developed PTH but this had resolved within one month in 96% (22). In contrast, a prospective study of 212 head injury cases in the US found the cumulative incidence of PTH had reached 91% after one year (9).
While these observations suggest that cultural attitudes and litigation play a significant role in the determination of PTH, contrary to expectation a significant minority (25%) of head and/or neck injury claimants in our cohort did not complain of headache despite likely pecuniary advantage. This was also highlighted in a subsequent study from Kaunas, in which it was concluded that headache persisting more than three months after head injury was not the result of the head or neck trauma but due to primary headache (23).
PTH had a poor prognosis in our cohort. Seventy percent of the claimants had ongoing headaches at the time of assessment, many months after injury. We did not identify any clear prognostic indicators. While a pre-accident history of primary headaches was strongly predictive of PTH, it had no influence on the prospect of recovery within the period of observation. Similarly, PCS (in which PTH may be a component symptom) had no independent adverse effect.
The treatment of PTH in both primary and secondary care was either non-existent or inadequate in our cohort. This might have had a bearing on prognosis. In addition, medication overuse headache may have been a factor, but our data is limited as it is not feasible to determine reliably the ad hoc consumption of over the counter analgesics in a single medicolegal assessment. Medication overuse was certainly very common among primary care referrals with chronic headache in our recent study (24).
In conclusion, while the influence of litigation on the pathogenesis and chronicity of PTH cannot be discounted in individual cases, the present study shows that recurrent or chronic headache is not inevitable following head or neck injury even in compensation claimants. PTH is strongly predicted by a past history of primary headaches, and the headache semiology is overwhelmingly consistent with “migraine”. We have argued previously that all primary headaches result from activation of a common process – the “migraine mechanism” – with differing semiologies reflecting the particular neural and neurovascular networks engaged by the process (25). We propose that PTH is triggered specifically by concussive injury to the head or neck and results in activation of the migraine mechanism. We note that a mechanical stimulus to the brain was one of the modalities shown by Leão to provoke spreading depression (26), which many regard as the essential process underlying migraine.
We suggest that PTH should be treated vigorously at an early stage in primary care according to the presenting primary headache phenotype(s), with early referral to expert care in more resistant cases. The reasons why the management of PTH in our litigants was so poor is unclear, but the findings in this study are entirely in keeping with our recent observations on the general management of headaches in primary care in the UK (24). In that study, only a third of a cohort of 200 patients was given a specific headache diagnosis in primary care and the vast majority were prescribed inadequate or inappropriate treatments over many years before referral to a headache expert. It is also possible that the misperception of PTH as somehow “different” from the common primary headache disorders impedes effective intervention.
Clinical implications
PTH is very common following head or neck injury but not inevitable even in compensation claimants. The only significant risk factor for PTH is a past history of primary headache. PTH semiology is typically “migraine” but can be complexed with other primary headache phenotypes. PTH therefore most likely results from provocation of the migraine mechanism by concussive injury. Treatment of PTH in compensation claimants is very often neglected or poorly managed in primary care.
This paper is based on preliminary findings presented at the World Congress of Neurology, Kyoto, 2017.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.