1. Introduction
While migraine is the second leading cause of disability worldwide, according to the
Global Burden of the Disease 2019 study, neck pain (NP) is also highly prevalent and responsible for 2–6% of the total global years lived with disability
[36][1]. The point prevalence of NP is estimated as 8% and its lifetime prevalence is 48%
[37][2]. Thus, it is not surprising that NP and migraine may present in the same individual just by chance alone. However, in a large American study that included 189,967 adults, NP was self-reported by 38% of individuals with severe headache or migraine, a prevalence much higher than the 11% in individuals without severe headache or migraine
[38][3].
The definition of NP varies between authors and international task forces
[39][4]. The most common definition of NP is a pain located in the anatomical region of the neck with or without radiation to the head, trunk, and upper limbs
[40][5]. NP can arise from many local structures, including muscles, ligaments, facet joints, and visceral structures of the neck.
Initial reports of NP in migraine patients date to more than 3 decades ago, with NP being reported by up to two-thirds of the patients during migraine attacks in the 1980s and 1990s
[41,42][6][7]. Since then, studies explored the link between migraine and NP from every angle, both from a clinical and research perspective. Yet, it is still an on-going debate whether NP contributes to migraine pain as a trigger, or whether it is a consequence, a part of the migraine attack, or simply a symptom of a co-existing neck disorder.
The recognition of NP as a potential migraine feature is of obvious importance. For example, both adult and pediatric population studies show that misdiagnosis is common, with NP frequently attributed to cervical spine pathology, leading to unnecessary investigations and delay of correct diagnoses
[22,43][8][9]. Misdiagnosis is more common in migraine patients with pain onset in the cervical region despite these patients still presenting with typical migrainous features
[43][9].
Research on NP and migraine has several concerns and limitations that are worth noticing. First, the assessment of NP differs substantially between studies. In most studies, the prevalence of NP relies on self-administered questionnaires that are subject to recall bias and interpretation difficulties that may result in misclassification, such as localization of pain. In other studies, NP is assessed based on the clinical observation (trigger points, cervical tension), which also has several issues considering the low accuracy of most cervical examination tests
[44][10]. Second, epidemiological studies exploring NP in migraine patients usually exclude confounders, such as cervicogenic headache (CeH) and tension-type headache (TTH). However, as will be discussed later, the overlap between these disorders is significant, particularly if
wthe researche
rs focus on NP. So, there is potential misdiagnosis bias in the studies. Third, over time, the definitions of concepts, such as premonitory symptoms, have changed as well as diagnostic criteria guidelines, such as ICHD-3
[2,45][11][12]. Thus, results may not be directly comparable. Despite appearing to be very common in migraine, NP is an under investigated topic of research
[46][13].
2. Anatomic and Physiological Considerations and Cervical Musculoskeletal Findings in Migraine
Activation of the trigeminal nerve and its connections is a well-established component of migraine pathophysiology, as discussed above
[47][14]. In addition to the trigeminal nerve, both occipital and cervical afferents converge in the trigeminocervical complex (TCC), which is composed of the C1, C2, and C3 dorsal horns of the cervical spinal cord and the trigeminal nucleus caudalis
[48][15]. Activation of this complex is the mechanism for cervical disease headache as well as for the presence of NP in primary headache disorders, depending on whether it originates in cervical or trigeminal areas
[49][16]. Indeed, there is evidence from experimental studies in healthy individuals that neck muscle induced-pain causes pain in both cervical and trigeminal territories
[50][17].
It is possible that repeated migraine attacks lead to central cervical sensitization and lower pressure pain thresholds associated with increased tension of the neck muscles. The neck muscle stiffness could alternatively be caused directly by alterations in muscles, such as inflammation or trigger points, which may activate sensory neurons and thereby contribute to the development of migraine pain. Another potential mechanism is a dysfunctional descending inhibition in migraine as shown in transcranial magnetic stimulation studies
[51][18]. Further mechanisms include increased stress that may also contribute to prolonged increase of regional muscle tone via the limbic system and the alteration of pain facilitation mechanisms
[52][19]. The well-known interaction between stress and migraine further perpetuates this pain cycle.
Several studies depict cervical musculoskeletal abnormalities and postural changes in different headaches, specifically in migraine
[44,46,53][10][13][20]. One of the most frequent findings in migraine patients is the excessive forward head posture (FHP), which means a reduction of the cranio-cervical angle
[44][10]. FHP is thought to result from repeated tasks in the setting of sub-optimal ergonomics and is usually associated with the shortening of the cervical extensor muscles (sub-occipital paraspinals, scalenes, sternocleidomastoid, levator scapulae, upper trapezius, and pectoralis major and minor)
[40][5]. Indeed, a few MRI studies show hyperintense signals in the trapezius in migraine patients
[54][21]. There is also report of the contraction of the scalp muscles, especially temporalis, during migraine attacks due to patients holding or pressing on the affected site for relief
[41][6]. Another common finding in migraine patients is the increased neck muscle stiffness measured by elastography studies
[46][13].
There is also evidence that the pressure pain thresholds in the neck are reduced in migraine patients even when they are not experiencing an attack
[46][13]. Furthermore, one meta-analysis found that pressure pain thresholds measured by quantitative sensory testing (QST) are lower in patients with migraines for the head and neck region but not for the rest of the body
[49][16].
Despite all these data, evidence for cervical musculoskeletal dysfunction in migraine is debatable. Some authors point out that mild impairment in cervical function may be due to variations of the normal, and that differences between healthy controls and migraine patients are only minor
[53,55][20][22]. It is not clear whether migraine has an associated overall pattern of cervical musculoskeletal dysfunction nor if those changes have clinical meaning and might respond to different therapeutic strategies.
3. Neck Pain as a Migraine Feature
As stated previously, NP is not part of the ICHD-3 criteria for migraine
[45][12]. However, NP not only seems to be one of the most reported symptoms in migraine, but in some studies NP is more prevalent than classical migraine symptoms, such as nausea
[56][23]. One recent meta-analysis assessing the prevalence of NP in patients with migraine found a pooled relative frequency of 77% in migraine vs. 23.2% in a control group
[57][24]. However, as highlighted by the authors, there is significant heterogeneity among the studies that
wthe
researchers must consider in the interpretation of the results.
In most cases, NP is reported before or during a migraine attack (ictal)
[37,44,53,56,58][2][10][20][23][25]. However, it has been reported after the headache stops (postdromal) and also interictally
[41][6]. In clinical practice, it is often difficult to distinguish between true premonitory and ictal symptoms. Perhaps a more relevant fact is that patients often have difficulties in recognizing premonitory symptoms and mistake them as triggers
[59][26]. Among a cohort of 50 patients with migraine experiencing neck pain, most (89.1%) reported NP only during the migraine pain phase, while 10.9% had NP-triggered migraine attacks
[60][27].
Pharmacological studies with nitroglycerin showed that almost all patients with migraine had at least one premonitory symptom, while 83% went on to experience a full migraine attack
[61][28]. NP was one of those premonitory symptoms showing good agreement for spontaneous and triggered attacks
[61][28]. Furthermore, in an adult population involving 1010 migraine patients, premonitory symptoms (onset of 2 or more hours prior to the headache) were present in 38.9%, the most frequent being a neck tension, phonophobia, and difficulty concentrating
[59][26], which relates very closely to spontaneous complaints in clinical practice.
Migraine sufferers may report NP as the initial site of pain that later radiates forward and reaches criteria for migraine; as a concomitant site of pain during a migraine attack; or as a site of pain following the acute migraine episode. Up to 40% of the migraine patients report that pain starts and/or concentrates in the occipital and neck areas
[62][29]. Regarding gender differences, there are only a few reports showing that women with migraine more frequently report NP during a migraine episode
[63][30].
The presence of NP in migraine has been associated with several clinical and prognostic measures. The prevalence of NP correlates with headache frequency
[56][23]; NP is associated with delayed acute treatment of migraine up to 30 min
[64][31]; NP is associated with treatment resistance and greater disability
[56,65][23][32]; and NP is more prevalent in chronic migraine than episodic migraine, up to twice the prevalence
[57][24].
Further evidence for NP being part of the migraine cycle comes from studies assessing cervical musculoskeletal function in migraine suffers with NP. First, more than half of those patients have normal neck function
[60][27]. Second, among the patients with cervical dysfunction, cervical dysfunction itself was not associated with pain hypersensitivity, in contrast to NP that was associated with allodynia
[60][27]. Also, neck disability evaluated by The Neck Disability Index seems to be more related to migraine factors including allodynia than to local neck dysfunction
[66][33]. Additional evidence is the interictal reproduction of the typical headache in migraine patients by cervical manipulation
[67][34].
4. Neck Pain as a Migraine Trigger
Prospective clinical epidemiological studies have reported NP as a trigger in up to 38.4% of migraine patients
[68][35]. In these studies, NP was considered when it was specifically not associated with the migraine pain. Furthermore, NP as a trigger was more common in chronic migraine
[68][35]. In one study, the pressure pain thresholds were similar in migraine patients with and without ictal NP, suggesting a peripheral cause for neck stiffness
[46][13].
Regarding the endurance of the neck muscles, one prospective study showed that migraine patients were more likely to report NP and migraine attacks up to 24 h after the assessment
[69][36]. The authors discuss that the neck soreness and increased nociceptive input from cervical routes triggers the migraine attack. One interesting finding was that not all patients with migraine developed an acute attack after cervical tests, which points out once again to different pain thresholds
[69][36].
Part of the evidence for the neck as a migraine trigger comes from clinical experience, which suggests that reducing neck muscle tension through exercise, physiotherapy, or acupuncture may reduce the frequency of migraine attacks
[70][37]. While the data is scarce, there is evidence of the benefit of massage therapy and physiotherapy on migraine
[70][37]. Addressing postural control impairments is of noted importance as well as the manual treatment of trigger points
[70][37]. Physiotherapy is not necessarily uniform as a preventive drug and should be individualized.
5. Neck Pain in Other Primary and Secondary Headaches and Differential Diagnosis
The differential diagnosis of NP is extensive and can be challenging. Tension-type headache (TTH) and cervicogenic headache (CeH) are the two main types of headaches associated with the cervical spine and NP. Yet, as the weresearchers discussed NP is common in migraine, so it is relevant to discuss the differential diagnosis of these disorders.
TTH is usually easy to recognize as a different headache from migraine even when NP is present. In TTH, the pain is not particularly intense and, although relatively common, nausea and photo or phonophobia are less marked than in migraine
[45,55][12][22]. There are important differences regarding the findings of musculoskeletal dysfunction in these patients. One meta-analysis studied specific cervical musculoskeletal impairments in migraine and TTH patients, such as FHP and cervical range of motion (ROM)
[55][22]. Patients with TTH had more cervical impairments than migraine and healthy controls, while migraine patients, even though reporting NP, had no differences in most assessed cervical outcomes compared to controls
[55][22]. TTH is also usually distinguished from CeH except in the rare cases of bilateral CeH
[71][38]. Supporting evidence for cervicogenic headache includes mechanical precipitation of an attack
[45][12].
Mechanical precipitation of headache is characteristic of CeH, but it is not specific and can happen in migraine
[71][38]. Also, migraine may coexist with CeH
[72][39] and the co-occurrence may increase the number of migraine episodes and analgesic use
[73][40]. Pain location itself may not be enough to distinguish between migraine and CeH as migraine patients can have pain concentrating on occipital and neck regions, and up to 50–72% of patients with CeH report pain in the frontotemporal area
[74][41]. Differences between CeH and migraine include epidemiological–in CeH women are not clearly more affected than men and CeH starts later in life; and clinical findings–unilateral pain is usually side-locked and does not change sides between episodes
[71][38]. One meta-analysis showed that flexion-rotation test/ROM and neck flexion strength may support the diagnosis of cervicogenic headache over migraine
[71][38].
Finally, there is evidence that greater occipital nerve block can improve the treatment response in migraine. However, using nerve blocks alone is not sufficient for the differential diagnosis of these disorders, sine both TTH and CeH may respond to these treatments
[75][42].
There are other headaches that may occasionally present with NP. Occipital neuralgia is a paroxysmal lancinating pain over the greater occipital nerve distribution area that is associated with hypoesthesia or dysesthesias and tenderness on palpation
[45][12]. The pain is expected to disappear following a local anesthetic block. Cluster headache patients may also report cervical pain, however, the diagnosis is usually confirmed by the other clinical findings
[76][43].
6. Neck Pain among Children with Migraine
Similar to autonomic symptoms, NP appears to be a particular common migraine feature in the pediatric population
[38][3]. As for the adult population, there is data for NP as both part of the migraine and as a trigger. In a prospective study that included 170 pediatric patients with migraine who were referred to a tertiary hospital for neurological assessment, NP or neck stiffness was reported as a premonitory symptom in 41.5% patients
[77][44]. Moreover, almost half of these patients were first referred to another specialty, such as orthopedics or neurosurgery. Interestingly, NP was more frequent among patients with CAS
[77][44]. This is shown by several other studies that suggest that NP is one of the most common premonitory symptoms, as well as fatigue and mood changes.