Clinical manifestations

Presentation and course

Headaches have been estimated to occur in 18% to 58% of patients following head injury (25; 94; 56). Headache is the most common but not the only manifestation of the posttraumatic or postconcussion syndrome. Additional symptoms include somatic (dizziness, tinnitus, hyperacusis, photophobia, phonophobia, blurred vision, easy fatigability), cognitive (impaired concentration and memory), and psychological symptoms (depression, anxiety, irritability, apathy, insomnia) (102). Also, posttraumatic headache is an important predictor of overall outcome after concussion.

Headache attributed to head or neck trauma usually begins hours or days after the injury but may be delayed for weeks. The timing of posttraumatic headache onset is crucial because it is the requisite for diagnosis according to the ICHD-3: the onset must be within 7 days after the injury (or regaining of consciousness or discontinuation of medications that impair the ability to sense or report the headache) (47). However, the recognition of headache onset could be hindered by cognitive impairment after the injury or by concomitant orthopedic or multi-organ trauma, in which patients may focus primarily on body pain rather than headache, or by the use of pain control that may alleviate the headache.

The most common clinical phenotypes of posttraumatic headache are migraine-like or tension-type-like headaches (67). However, the percentage of headache phenotypes are different in each study. One Danish study found tension-type-like headache accounts for about 97% of chronic posttraumatic headache patients and only 28% of patients have of migraine-like headache (60). A 5-year prospective study found migraine-like headache (58%) is more common than tension-type-like headache (14%) (111). One study found posttraumatic headache frequently presenting as migraine-like headache but usually without aura (56). One observational study from 100 U.S. soldiers found more than 95% of posttraumatic headache patients fulfilled migrainous headache (24). In contrast to civilian injuries, migraines account for perhaps 58% to 75% of posttraumatic headache in soldiers (119; 113).

Rare case reports have attributed trauma to the onset of cluster headache (116), hemicrania continua (27), short-lasting unilateral neuralgiform headache attacks with conjunctival injection and tearing (SUNCT) (98; 99), short-lasting unilateral neuralgiform headache attacks with cranial autonomic features (SUNA) (52), and chronic paroxysmal hemicrania (76).

Medication overuse is common in chronic posttraumatic headache, and some authorities believe that it is the major cause of headache perpetuation. Warner argues that medication overuse headache is the most frequent cause of chronic posttraumatic headache and that stopping overuse of over-the-counter or prescription analgesics leads to headache resolution in most patients (124; 121). Baandrup reviewed the headache diaries of 53 chronic posttraumatic headache patients from the Danish Headache Center; analgesic overuse was found in 42%, but half were not improved following detoxification therapy (04).

Neck pain is more common in those who have sustained whiplash, a sudden acceleration then deceleration to the neck resulting in pain without an actual blow to the head (109). The major constituents of whiplash syndrome include neck pain and/or tenderness, restriction of neck mobility, and headache. Pain arising from the cervical spine often radiates to the occipital area but may extend to the parietal, temporal, and frontal regions. Range of motion may be limited, and palpation tenderness may be noted. Most whiplash-induced headaches tend to occur close in time with the whiplash injury, resembling tension-type headache or cervicogenic headache in phenotypes, and are often short lasting (109). Cervicogenic headache may occur after head trauma: a site-locked unilateral pain starting in the neck and spreading to the fronto-ocular area (120). Myofascial pain syndrome is also common, featuring the presence of active myofascial trigger points that produce referred pain on palpation (33).

Chronic posttraumatic headache may result in pain in temporomandibular joints (16). Temporomandibular joint dysfunction may present as a decrease in the range of motion of the joints, increased pain with chewing, audible clicking, and tenderness to palpation. However, 1 systematic review found conflicting evidence regarding the effects of whiplash on the development of temporomandibular disorders (30).

Occipital neuralgia can follow whiplash-type injuries. Fracture of the styloid process can produce symptoms resembling Eagle syndrome: unilateral pain in the throat or neck, or referred pain in the shoulder, chest, tongue, eye, cheek, temporomandibular joint, or ear (130).

A continuous hemicranial headache that presented with continuous half-sided headache, cranial autonomic symptoms with exacerbations, and a response to indomethacin were reported after cranial surgery (35).

Other rare case reports, including an orthostatic headache after a bicycle accident that occurred 5 days prior to presentation due to cervical spine fracture owing to cerebral spinal fluid leak, and progressive intracranial hypertension due to superior sagittal sinus thrombosis following mild head trauma, were reported (133).

The rate of posttraumatic stress disorder was higher (97.9%) among U.S. soldiers with comorbid chronic posttraumatic headache attributed to head injury, but only 9.3% among military personnel in Iran (63; 55).

Prognosis and complications

The different types of head injury play an important role for the prognoses. The prognosis is favorable for most patients with mild head injury (70). The children usually recover well with quick resolution of symptoms and few residual deficits. For adults, cognitive deficits and symptoms are common in the acute stage, and the majority would recover within 3 to 12 months (14). Patients with moderate to severe traumatic brain injury may have an extended recovery time. Strong predictors for 6-month outcome after moderate to severe traumatic brain injury include the Glasgow Coma Scale, midline shift on CT scan, subdural hematoma, and pulsatility index on transcranial Doppler (51).

Acute posttraumatic headache occurred within 24 hours after the trauma in 78% and lasted for a median of 3.0 days (70). One month after mild head injury, 30% to 90% of patients have headache; 1 year after injury, headache prevalence ranges from 8% to 35%, based on different definitions and methodologies (107). A prospective study found a 91% 1-year cumulative incidence of new or worsened headaches in patients with mild traumatic brain injury after vehicle-related crash and this study showed that up to 49% of patients met the criteria for migraine or probable migraine, followed by tension-type headache (40%) (72). A systematic review on the prevalence of chronic headache after traumatic brain injury revealed a prevalence of 57.8% (87). The prevalence rate of posttraumatic headache after whiplash-type injury tends to decrease with time (19) and was 15% after 2 years (58). Risk factors associated with chronic post-traumatic headache include female gender, increasing age, mechanical factors such as position of head and neck at the time of trauma, prior headache history, multiple traumatic brain injuries, and premorbid psychological status (71; 42; 129; 70; 126). Preexisting facial pain, lack of confidence to recover completely, sore throat, medication overuse, high Neck Disability Index, hopelessness/anxiety, and depression predict the develop of chronic headache (90).

For unknown reasons, there is inadequate correlation between severity of injury (eg, depth and duration of coma, amnesia) and outcome following traumatic brain injury (128; 18; 87; 21). The advanced functional neuroimaging techniques may be useful for predicting outcome, but further studies are needed (80). Although certain serum proteins have been proposed as potential biomarkers for traumatic brain injury (48; 69), their roles in posttraumatic headache are uncertain.

Clinical vignette

A 33-year-old woman developed chronic daily headache and other postconcussive symptoms following an automobile accident. She was an unrestrained passenger in a midsize car that rear-ended a bus. It was estimated that the car she was traveling in collided with the bus while traveling at approximately 35 mph. There was significant damage restricted to her car’s front bumper. The patient struck her head against the dashboard or windshield. The patient lost consciousness for less than 5 minutes. In the emergency department, she had anterograde amnesia for several hours following the accident and complained of diffuse headache. CT scan of her head was normal. She was observed for 8 hours and discharged.

Over the ensuing 48 hours, she developed dizziness, tinnitus, and headache. The headache was bilateral, located in the occipital-nuchal region and radiated to her temples. It was a dull, constant pressure. She had photophobia and phonophobia, but there were no gastrointestinal or trigeminal autonomic features. Approximately once a week it would intensify and resemble the migraine headaches that she had suffered as an adolescent. She attempted to return to her job as a customer service representative but noted difficulty with distractibility, concentration, and memory. She sought relief through physical therapy, chiropractic manipulation, massage therapy, and naturopathic treatments with only modest, transient benefit. Her symptoms gradually improved over the following months, but her daily headache persisted. Six months post-injury she was evaluated by a neurologist; her examination was unremarkable, and she was educated regarding the good prognosis for her symptoms. She was started on amitriptyline 25 mg and titrated to 50 mg. Additionally, she was referred for biofeedback and relaxation therapy and encouraged to participate in a physical fitness program. At her 2-month follow-up, she was much improved, and headaches were occurring once weekly with minimal associated disability.

Biological basis

Etiology and pathogenesis

Acute posttraumatic headaches result from physical trauma to the head or neck. Trauma may trigger a new headache or may exacerbate existing primary headache disorders. Minor head trauma is an accepted trigger for attacks of migraine; footballer’s migraine (migraines triggered by “heading” the ball in soccer) is an example (77; 125; 96). Another extreme example is developed cerebral edema, coma, and death following trivial head injury (61).

On the other hand, the persistence of posttraumatic symptoms may be related to other factors, including psychogenicity, poor psychosocial support, cultural background and base rate misattribution, expectation, chronic pain, compensation or litigation, and malingering (71; 83; 26). There are some predisposing factors. For example, studies found FHM1- and FHM2-related gene mutation associated with higher neurologic response to minor head trauma (61; 114; 06). Patients with underlying migraine have a higher risk of posttraumatic migraine after concussion (73).

Similar symptom complexes among posttraumatic headache, psychiatric diseases, and posttraumatic stress disorders suggest a common psychogenic origin of these disorders. Pretraumatic neurosis and psychopathology are more common in chronic posttraumatic headache patients than in patients with low back pain or chronic tension-type headache (42). Studies of symptom expectation in normal, healthy subjects show that North Americans are more likely to expect chronic headache following a mild head injury than their counterparts in Greece or Lithuania (where rates of chronic posttraumatic headache are lower) (31; 32). The relationship between compensation and posttraumatic symptoms remains controversial. A meta-analysis of 2523 subjects following closed-head injury concluded that more abnormalities and disability were seen in patients with financial incentives (10). The elimination of compensation for pain and suffering is associated with a decreased incidence and improved prognosis of whiplash injury in Canada (15). Others have argued that compensation is not a significant factor in the perpetuation of posttraumatic headache, as the symptoms typically persist and do not improve following legal settlement (92).

Advocates for a purely neurogenic mechanism point out that patients with mild head injury have more pathologic, radiologic, and electrophysiologic abnormalities than normal controls. However, the majority of patients with persistent symptoms do not necessarily have objective abnormalities on CT, MRI, or SPECT (131). Functional neuroimaging (PET, functional MRI, MR spectroscopy) may demonstrate subtle abnormalities (80). For example, 1 study showed that veterans with mild traumatic brain injury had decreased cerebral metabolic rate of glucose in the cerebellum, vermis, pons, and medial temporal lobe and exhibited subtle impairments in verbal fluency, cognitive processing speed, attention, and working memory (97).

No unitary hypothesis can account for the wide spectrum of headaches that can occur following head injury. Although physical factors likely play the major role in headaches that begin hours or days following head or neck trauma, a purely neurogenic theory is unlikely to account for the range of late-onset persistent headaches that occur following minor nonconcussive injury. Chronification is most likely due to a combination of organic and psychosocial factors (58).

Keidel and Ramadan suggest the following factors may be involved in the pathogenesis of acute posttraumatic headache: (1) referred pain from nociceptive input caused by lesions of musculoskeletal, discoligamentous, and other soft-tissue structures (including vessels and nerves); (2) activation of meningeal nociceptive afferents due to traumatic epidural, subdural, and subarachnoid bleeding; (3) stretching of pain-sensitive intracranial structures from increased intracranial pressure; (4) intracranial hypotension; and (5) activation of the trigeminovascular system by posttraumatic sinus venous thrombosis (58). In cases of military traumatic brain injury or blast injury, different mechanisms may be involved: (1) an under-pressure followed by over-pressure wave during the blast; (2) penetrating injury from flying debris; (3) structural environment collapse or the body being thrown a distance; and (4) small-particle pollutants (89).

Diffuse axonal injury is 1 of the most common and important pathologic features of traumatic brain injury (110). It results from shearing forces between different parts of the brain that stretch, distort, or rupture parenchymal axons and may cause neuronal damage or death. It may be caused by angular acceleration of the head and can arise without increased intracranial pressure or hypoxia. Abnormal cerebral hemodynamics have been demonstrated following traumatic head injury (36). Several serum proteins, such as BB isozyme of creatine kinase, glial fibrillary acidic protein, myelin basic protein, neuron-specific enolase, and S-100B, are elevated during diffuse axonal injury and could be used as the biomarkers (48; 69).

After traumatic head injuries, there is a cascade of biochemical changes that are similar to the migraine biochemical milieu: neuronal depolarization, excessive release of excitatory amino acids and neurotransmitters, abnormalities in serotonin neurotransmission, changes in endogenous opiates, increased nitric oxide activity, alterations in cerebral metabolic activity, and cerebral hemodynamic instability (41; 96; 106).

There is strong evidence showing that calcitonin gene-related peptide (CGRP) plays a primary role in migraine and therapies targeting CGRP showed treatment efficacy in migraine patients. In posttraumatic headaches, animal models showed elevated CGRP level after head trauma; the level was associated hyperalgesia and photosensitivity (22; 20; 117; 11). Pituitary adenylate cyclase-activating polypeptide (PACAP) is another migraine trigger. After severe traumatic brain injury, the PACAP level increased in cerebrospinal fluid and plasma (13). Another interesting finding is that PACAP has some neuroprotective and antiinflammatory effects after traumatic brain injury (74). Other molecules linking posttraumatic headaches included neuropeptide substance P and the growth factor BDNF (22; 17; 86). The transient biochemical abnormalities following acute trauma may cause recurrent or chronic headaches through the development of peripheral and central sensitization. Trauma may trigger the first attack, which could then set the process in motion in predisposed patients. The CGRP release at nerve terminals causes vasodilation and headache, followed by activation of CGRP receptors in the brainstem and central sensitization (09). Another hypothesis is axonal injury, which may lead to dysregulation of brainstem nociceptive pathways. The stress from work-related disability, ongoing litigation, and a reduced capacity to participate in familial and social activities may be an ongoing trigger for the maintenance of chronic headaches (02; 75).

Epidemiology

Annually, as many as 1.7 million individuals in the United States suffer a closed-head injury (28). Of those injuries, 80% are mild, 10% moderate, and 10% severe (58). The common causes of head injury include falls (35%), motor vehicle or traffic accidents (17%), a person being struck by or against something (16%), and assaults (10%) (28).

Headache is the most common symptom following head injury, but the epidemiology of posttraumatic headache is hard to determine due to differences in definitions and difficulties with case ascertainment. The prevalence of acute posttraumatic headache in acute mild head injury was 66% according to the diagnostic criteria of ICHD-2 (70). Estimates of headache after injury to the head or neck (whiplash) range widely from 30% to 90% (93; 58). A population-based survey in Norway reported a prevalence of posttraumatic headache of 0.2% in a population aged 30 to 44 years (01). An investigation of the lifetime prevalence of posttraumatic headache in Denmark found the prevalence was 4.7% in men and 2.37% in women (100). In 1 study of returning U.S. Army soldiers who had 1 or more concussions while serving in combat zones in the Middle East, the prevalence was 37% (113).

Posttraumatic headache usually resolves over a few weeks or months, but some may develop chronic headaches (92; 58). One study in Australia found 15.34% of those who had visited an emergency room for minor head injury continued to complain of persistent posttraumatic headache 3 months later (29). A meta-analysis of 12 studies including 1670 civilians with traumatic brain injury revealed that 57.8% suffered from chronic headache (87). A summary of 4 studies consisting of 1020 military survivors with traumatic brain injury also found that 35.5% had chronic headache (122; 123; 104; 50).

One study in soldiers with mild traumatic brain injury with conscious change found blast injuries were associated a higher risk of persistent headache (127). Migraine and posttraumatic headache have a bidirectional relationship (73; 105). Genetic predisposition may influence the development of posttraumatic headache. One study found adolescents with a family history of migraine were more likely to developed posttraumatic headache with migrainous features after concussion. In the pediatric head injury population, the prevalence rates of chronic posttraumatic headache ranged from 3.2% to 6.8% (85; 59).

Prevention

Increased and proper use of safety helmets for bicyclists and rollerbladers and seat belts and head rests for motorists can reduce the incidence of moderate and severe head injury. However, clinical experience suggests that these measures do not necessarily reduce the incidence of acute and chronic posttraumatic headache.

A single preoperative dose of diclofenac reduces the intensity of acute postcraniotomy headache and decreases analgesic requirements over 5 postoperative days in adults (84).

Differential diagnosis

Headache following head trauma may be indistinguishable from primary headache disorders. Both head injuries and primary headache disorders are more common in individuals between the ages of 20 and 50 years, and the new onset of a headache disorder may be coincident rather than attributable to the craniocervical injury. However, due to the temporal relationship, a new onset headache beginning within 7 days of craniocervical trauma is considered consistent with a posttraumatic headache. Delayed onset (over 6 weeks) suggests a new primary headache disorder, although some authorities assert that patients may develop chronic headaches as long as 24 months after trauma (130). Additionally, head or neck trauma may exacerbate pre-existing primary headache disorders.

Head or neck injury may precipitate subdural or epidural hematomas, carotid or vertebral artery dissection, carotid-cavernous fistula, CSF leaks, cerebral venous sinus thrombosis, skull fracture, temporomandibular joint dysfunction, or cervicogenic headache.

The clinicians should maintain high vigilance to identify such secondary causes, especially if posttraumatic headache becomes persistent. Medication overuse headache should be considered if frequent or daily use of analgesics is noted.

Diagnostic workup

Taking a detailed history and performing a thorough neurologic examination remain pivotal in evaluation of posttraumatic headache patients. The exact mechanism of the trauma, the area and extent of the injury, and the immediate post-injury symptoms should be queried. Determining the timing of headache onset following injury is required for diagnosis. However, the recognition of headache onset could be difficult in patients with moderate to severe traumatic brain injury or major trauma. The characteristics of headache after trauma, functional disability, and psychiatric comorbidities as well as current treatment and the use of symptomatic analgesics all should be assessed. The patients should be queried about the presence and status of any ongoing litigation or work-related disability claims (37). During the neurologic examination, any focal symptoms or signs should be evaluated. A complaint of significant unilateral neck pain may portend dissection. A history of postural headache may signify intracranial hypotension. An orbital bruit and conjunctival injection suggest carotid-cavernous fistula (132).

There are guidelines concerning the indications of neuroimaging for mild traumatic head injury in the acute care setting (53). However, the diagnosis of posttraumatic headache depends on clinical history taking, and routine diagnostic neuroimaging or laboratory examinations have limited diagnostic value (78). CT scan remains the imaging modality of first choice in the acute phase for detecting potential sequelae of intracranial trauma or skull fracture (88), with a sensitivity of 63% to 75% (91; 118). However, an initially normal CT does not preclude brain injury and is of limited prognostic value. A repeated CT scan is warranted in case of neurologic deterioration (Brown et al 2007). Compared with CT, MRI is more sensitive in detecting subtle brain abnormalities (diffuse axonal injury, nonhemorrhagic contusion, small subdural hematoma, etc.), especially diffusion-weighted imaging and diffusion tensor imaging (62). MRI is the technique of choice for patients with persistent symptoms in the subacute phase of head injury and during follow-up, or those with ongoing litigation (80; 37). Some clinical symptoms should guide further neuroimaging procedures: orthostatic headache requires an MRI with gadolinium to look for pachymeningeal enhancement characteristic of intracranial hypotension, and focal neck pain and Horner syndrome necessitate a Doppler ultrasound or MRA to look for carotid artery dissection.

Functional neuroimaging tools (SPECT, PET, functional MRI, MR spectroscopy, and magnetoencephalography) may help to demonstrate subtle abnormalities (07; 68) or correlation with neuropsychological deficits (38), but these tools have their limitations in resolution and availability. Neurophysiologic studies (eg, EEG, evoked potentials) are frequently abnormal, but their diagnostic, prognostic, and medicolegal values are negligible (131).

Neuropsychological testing may be useful to document cognitive deficits, predict the functional outcome, or to distinguish concomitant depression, anxiety or frank malingering (71; 57; 44). Cognitive deficits after mild head injury include alterations in attention, information processing, memory, visuospatial processing, and executive functioning (05). Most objective deficits resolve within 3 months, but subjective complaints may persist and appear to have a pathophysiologic and nonneurologic basis. Abnormal neuropsychological test performance should not automatically be interpreted as proof of neuropsychological impairments (57) because anxiety, depression, physical pain, or medication can impair performance.

Management

Because most patients with posttraumatic headache have varied disability in physical, cognitive, and emotional symptoms, management should be individualized, and multidisciplinary approaches are often necessary--eg, physical therapy, relaxation therapy, biofeedback, coping skills training, individual or family therapy, and appropriate pharmacologic management (02; 43; 65). The manual therapy on the neck could reduce the headache compared to ice packs therapy, but this effect was largely lost by 8 weeks after treatment (54). Cognitive-behavioral therapy and relaxation therapy provide a useful supplement to the treatment of posttraumatic headache (49; 40). However, 1 study showed no benefit for whiplash headache relief with physical therapy compared with education and/or activity (103).

A thorough clinical evaluation and judicious use of neuroimaging studies can exclude serious brain injury and provide reassurance needed to the patient. Patients need realistic expectations regarding the time necessary for full recovery; unrealistic expectations may intensify anxiety and depression. Psychiatric evaluation and treatment are warranted in patients with significant depression, anxiety, or phobias.

There is a lack of high-quality studies in the acute and preventative treatments for posttraumatic headache (66). Therefore, treatment of posttraumatic headache should follow the guidelines for the primary headache disorder that it phenotypically resembles. Like the treatment of other primary headaches, posttraumatic headache management can be divided into acute and preventive treatment.

Acute treatments. The acute treatment options could be divided into acute treatment for “acute posttraumatic headache” and “acute attack of persistent or chronic posttraumatic headache.” For acute posttraumatic headache, an early study found repetitive infusions of intravenous dihydroergotamine with metoclopramide can break the headache cycle and diminish associated symptoms in patients with postconcussion headache (79). Also, intravenous metoclopramide 20 mg plus diphenhydramine 25 mg had been shown effective and 63% of patients reported sustained headache relief for 48 hours (34). For abortive treatment for chronic posttraumatic headache, 1 retrospective study with 100 military personnel found triptans had a higher responder rate (70%) than nontriptans (42%) for headache relief in 2 hours (24). Codeine and butalbital-containing analgesics should be restricted to refractory patients with contraindications to the aforementioned medications. Patients must be questioned directly about their daily and weekly consumption of over-the-counter and prescription analgesics. Use of abortive agents should be restricted to 2 days per week to avoid medication overuse headache. If medication overuse headache is present, withdrawal of the overused substances is recommended, and outpatient or inpatient detoxification may be necessary (108). Nonpharmacological options include exercise, neuromodulation, and cognitive/behavioral approaches. The benefit of exercise may due to the exercise-induced release of BDNF but the best timing of exercise after head injury remains unknown (39).

Preventive treatments. The goal of prophylactic therapy is to convert daily headache into a more manageable episodic form. The tricyclic antidepressants (amitriptyline, nortriptyline, doxepin) are most often used for chronic posttraumatic tension-type headaches for their effect on the associated symptoms of irritability, dizziness, depression, and insomnia. Nortriptyline has fewer sedative and anticholinergic effects and may be preferable for patients who are older or more sensitive to these effects. The value of serotonin-specific reuptake inhibitors is unclear. The first-line prophylactic treatment options for chronic posttraumatic migraine include tricyclic antidepressants, beta-blockers, and anticonvulsants (08; 65). Among anticonvulsants, divalproex sodium showed mild to moderate improvement in 60% of patients with chronic posttraumatic headache (95). Another study found topiramate appears to be an effective headache prophylactic therapy, whereas low doses of tricyclic antidepressants appear to have little efficacy in military troops with chronic posttraumatic headache (24). One retrospective study analyzed the efficacy of preventive pharmacological treatments and found that 72% of patients under amitriptyline and 75% under melatonin achieved 50% or more of a reduction in headache frequency (64).

New options of preventive therapies have been introduced for the treatment of posttraumatic headache. One randomized, placebo-controlled, crossover study showed botulinum toxin type A could effectively reduce headache frequency of posttraumatic headache by 2.24 days per week (43.3%) at the end of 16 weeks, compared with a 1.28 days per week (35.1%) increment of headache frequency (134). Moreover, CGRP-related therapies have been tested in posttraumatic headache. One open-label study of erenumab for persistent posttraumatic headache found 2.8 moderate-to-severe headache days reduction (17.8%) from baseline (15.7 days) to week 9 to 12. However, only 13% patients of patients achieved at least a 50% reduction in headache frequency (03). Future randomized controlled studies analyzing the efficacy of CGRP monoclonal antibodies are undertaken to provide more solid evidence for CGRP-targeted therapies in posttraumatic headache (82).