Practice Points
- Individuals with multiple sclerosis (MS) have a higher rate of migraine than the general population. Both conditions are associated with psychiatric comorbidity and cognitive symptoms that should be monitored given the potential for additive burden when MS and migraine co-occur.
- Initial migraine assessment should include a headache diary to track migraine symptoms, triggers, frequency, and treatment response. Acute medications should be prescribed, and if migraines occur for more than 4 days per month, preventive treatments should also be considered.
- With stress acting as a migraine trigger, psychological interventions should be offered as a nonpharmacological adjunct to migraine management medications.
The co-occurrence of migraine and multiple sclerosis (MS) has become increasingly recognized. Both are disorders of the central nervous system (CNS) and are thought to be the product of a complex interaction between environmental triggers and genetic predisposition.1,2 Both conditions have a higher prevalence among women, a relatively young age of onset, and are associated with higher rates of cognitive dysfunction, depression, anxiety, pain, and fatigue compared to those of the general population.2,3 Given the typical age of onset, these disorders tend to affect individuals in the prime of their career development and active family lives.
MS is a chronic autoimmune disease affecting an estimated 2.8 million people worldwide.4 In MS, the T cells and B cells of the immune system are abnormally activated and, once within the CNS, cause inflammation and surround the axons of neurons, resulting in demyelination.1 This disrupts the transmission of signals between neurons to, from, and within the brain, giving rise to neurological and neurocognitive symptoms. There are relapsing and progressive forms of MS. Relapsing-remitting MS (RRMS) constitutes 85% of MS cases.4
Migraine is the second-highest cause of disability worldwide.5 The International Classification of Headache Disorders (2018) defines a migraine as a headache attack lasting 4 to 72 hours accompanied by at least 2 of the following: unilateral location, a pulsating quality, moderate to severe pain intensity, and/or aggravated by or avoidance of physical activity.6 There must also be either (1) nausea and/or vomiting, or (2) photophobia and phonophobia.6
In this review, we aimed to describe the pathophysiological links between MS and migraine, discuss the need to consider neuropsychological symptoms in people with MS and migraine, and outline the treatment of migraine in those with comorbid MS. By building awareness of this co-occurrence and its impacts on mood, cognition, and disability, clinicians can better recognize and address the health needs of this vulnerable patient group.
Methods
This review is based on literature searches of PubMed and Google Scholar for articles that examined the pathology and disability outcomes in each disorder and the overlapping biological mechanisms and additive symptom burden in MS and migraine. This was followed by targeted reviews of pharmacological and nonpharmacological treatment options for migraine. Literature searches included articles published up to June 2023 and were conducted by the first author (P.T.). Key search terms included multiple sclerosis and/or migraine, in combination with pathology, mood, cognition, and treatment. As research emerged in 2024, articles were identified and reviewed for relevance based on the authors’ clinical experience and knowledge.
Migraine Pathophysiology
The pathophysiology of migraine remains incomplete, with several biological mechanisms proposed. Researchers implicate the activation of the trigeminovascular system (TGVS) as a key mechanism in precipitating migraine.7 The cause of its atypical activation and sensitization is unknown. It is hypothesized that the TGVS pathway is activated by pain signals from trigeminal nociceptive afferents.7 In the theory of peripheral sensitization, these afferent nociceptors are thought to have reduced threshold for and/or increased responsiveness to potentially damaging stimuli.8 In contrast, central sensitization describes the heightened excitability and sensitivity of neurons in the CNS (in particular, the trigeminocervical complex) due to the sensory input received from the nociceptors at the site of inflammation.8 Ultimately, this dysfunctional activation and lack of inhibition of nociceptive input manifests as increased sensitivity to pain responses (hyperalgesia) as well as activation by nonthreatening stimuli (allodynia).9 Migraine is therefore thought to occur when the pain system of the brain is maladaptively alarmed. The sensitization resulting from repeated attacks involves both central and peripheral systems. Whether a migraine attack is triggered peripherally or centrally remains a matter of debate.
The loss of inhibitory pain modulation from brainstem nuclei has also been implicated in migraine.10 The periaqueductal gray is a key region mediating pain modulation in the midbrain. It allows communication between the cortical and brainstem regions, regulating the balance between protective pain responses when needed and inhibition of incoming nociceptive information.11,12 Imaging studies have demonstrated persistent brainstem activation in the periaqueductal gray, locus coeruleus, and dorsal raphe nucleus using PET scanning during migraine attacks.12 Additionally, the release of the neuropeptide calcitonin gene-related peptide (CGRP) has been strongly linked to migraine initiation. Neurons containing this neuropeptide are found in the trigeminal nucleus caudalis, locus coeruleus, and thalamus.13 Study results show that people with migraine have a higher level of CGRP during a migraine attack,14 as well as a change in the activity of the trigeminal nucleus caudalis on functional MRI in the hours and days preceding a migraine.15 People with migraine have higher interictal levels of CGRP than healthy controls, and those with chronic migraine have a higher level of CGRP than those with episodic migraine.16 Edvinsson and colleagues proposed that continued activation of the TGVS leads to the sustained release of neurotransmitters and results in neuroinflammation, which contributes to migraine chronification rather than the initiation of attacks.17 It is posited that the failure in pain modulation may result from functional changes in brainstem regions that facilitate TGVS activation and neuropeptide release.10,11 Both the abnormally activated TGVS and the brainstem dysfunction in pain modulation and sensory processing are thought to play a role in migraine. While these processes have been identified in the precipitation of migraine attacks, much remains unknown about migraine pathophysiology.
Occurrence of Migraine in MS
In 1952, Compston and McAlpine were the first to document migraine in MS. They noted that 2% of their patients with MS reported migraine within 3 months of their initial relapse.18 In the first systematic study of this occurrence, Watkins and Espir found that a significantly greater proportion of people with MS had migraine (27%) relative to healthy controls (12%).19 More recent studies estimate that migraine occurs in 20% to 45% of the MS population.20-23 This range in prevalence is likely due to variations among studies in diagnostic criteria, participant selection (eg, specialist clinics), and geographic location. In RRMS specifically, 49.8% to 50.9% of individuals had co-occurring migraine across studies.22,23 For people with MS, migraine is more common in women and 2.5 times more likely in those with a family history of migraine.23,24 Relative to periods of remission, during relapses, those with RRMS experienced worsened migraine pain and more frequent migraines.25 Given its prevalence and burden, it is crucial that migraine be managed effectively within MS.
Proposed Underlying Mechanisms
Researchers have proposed several hypotheses on the mechanisms that cause the MS and migraine co-occurrence (Figure). First, recurring inflammation of the CNS may make people with MS more susceptible to migraine, particularly early in the disease course when inflammation is prominent.20 In particular, MS lesions in the brainstem may increase the risk of migraine22; when MS lesions were observed in the midbrain, there was a fourfold increase in the likelihood of migraine.12 Additionally, people with MS and migraine had a higher number of lesions in the substantia nigra and red nucleus than people with MS without migraine.26 While these findings may demonstrate that migraine is triggered by MS lesions that disrupt pain-modulating circuitry, this does not account for migraines occurring before or early in the MS disease course. It is also possible that migraines occur in the absence of brainstem lesions in MS, as this has not yet been investigated.
Second, neuronal activity in migraine leads to inflammatory reactions in the central and peripheral TGVS (ie, neuroinflammation), with repeated neuroinflammation linked to migraine chronification.17 Thus, migraine may contribute to the neuroinflammation observed in MS. A longitudinal survey of 116,000 female nurses by Kister and colleagues supported this theory of migraine as a risk factor for early manifestation of MS.27 They found that participants with migraine were 1.39 times more likely to develop MS compared to those without migraine.That said, the absolute risk of developing MS over a 15-year period in women with migraine remained low (0.47%) relative to the rate in women without migraine (0.32%).27 In this patient group, however, migraines tend to precede MS onset.22 The typical age of presentation of migraine is younger than that of onset of MS.28,29 As such, the fact that migraine occurs prior to MS onset may simply be in accordance with the epidemiology of each distinct disorder. This is consistent with Watkins and Espir’s finding that the age of MS onset did not differ between individuals who had preexisting migraine, those who went on to develop migraine, and those with no headache disorder.19
A third hypothesized mechanism is that migraine pathophysiology interferes with the integrity of the blood-brain barrier, acting as a trigger for MS. Cortical spreading depression—a transient wave of neuronal depolarization across the cortex—has been linked to migraine with aura.2 This could impact the permeability of the blood-brain barrier. It was thought that neuroinflammation linked to migraine may also alter the integrity of the blood-brain barrier, allowing immune cells and leukocytes to enter the brain. Given the lack of evidence of macroscopic disruption in imaging studies of migraine, this is unlikely to be the case. Additionally, studies have found no link between migraine and the opening or increased permeability of the blood-brain barrier.30,31
Shared Risk Factors
It is notable that both migraine and RRMS share a host of common factors. Both are chronic relapsing-remitting disorders of the CNS, which have a higher prevalence among young women.21 Hormonal changes can affect the manifestation of symptoms in each disorder. Reduced MS activity and migraine frequency are observed during pregnancy, and menses can trigger migraine and worsen MS symptoms for some women.21,32 Given the elevated psychological burden in both disorders, it has been suggested that stress and anxiety associated with MS may trigger and account for migraines occurring at a higher rate in those with MS.19,21 While the cause of their co-occurrence remains unclear, shared risk factors (eg, female sex, childbearing age, hormone changes) may play a similar role in their etiology.21 Accordingly, rates of MS would be expected to be higher among those with migraine and vice versa.
The Effect of MS Treatments on Migraine
Disease-modifying treatments (DMTs) used to stabilize MS, such as interferon beta, natalizumab, and fingolimod, can worsen preexisting migraine and provoke de novo headache.23,24,33 Interferon beta has been associated with the onset or worsening of headache at a higher incidence than other DMTs, at a rate of 58.7%.24,33 Pöllmann and colleagues found that 30% of patients with preexisting migraine reported that their migraines worsened in frequency and duration after interferon beta treatment.34 Kister and colleagues showed that people with MS and migraine have a significantly lower tolerance of interferon beta, with 34% experiencing a headache “always or almost always” after treatment compared to people with MS who do not have preexisting migraine.35 Similarly, Villani and colleagues examined people with MS and migraine who took interferon beta and then continued interferon beta treatment or switched to natalizumab.36 People who switched to natalizumab had a reduction in migraine frequency, while those who continued on interferon beta had no such change. Research results suggest that interferon beta may generate an upregulation of proinflammatory cytokines36,37 or promote neocortical hyperexcitability,38 which gives rise to migraine; however, no studies to date have examined the biological underpinnings of this link. In clinical trials of fingolimod, headache and migraine were reported as common adverse events. Headache was reported by 26% of patients taking fingolimod, and there was a slightly higher relative risk of migraine when compared with those who received a placebo.39 Since the co-occurrence of migraine and MS was identified prior to the introduction of DMTs, the use of DMTs likely perpetuates the occurrence of migraine in MS but does not fully account for it.
Thus, the debate about the link between MS and migraine continues. Their co-occurrence may reflect migraine as an early symptom of MS, a risk factor for MS, or they may be distinct disorders that occur as comorbidities. This distinction is yet to be confirmed. Among people with both RRMS and migraine, 68.3% reported moderate to severe levels of migraine-related disability.24 Research on the nature and degree of cognitive issues, mood symptoms, and functional difficulties in this group is scarce, despite evidence of dysfunction in these areas in people with MS alone or migraine alone. The potential for an improvement in these symptoms via effective treatment of migraine also requires further investigation.
Burden of Migraine in MS
People with MS and migraine experience a greater burden of neurological and psychological symptoms than people with MS without migraine. Kister and colleagues compared people with MS and migraine to people with MS without a headache disorder and they found that participants with both MS and migraine had significantly greater fatigue and sleep disturbance, sensory symptoms (eg, spasms, allodynia), brainstem symptoms (eg, facial weakness, vertigo), and pain symptoms (eg, trigeminal and occipital neuralgia) relative to the headache-free group.35 Although it is beyond the scope of this review, these shared symptoms of pain, fatigue, and sleep disturbance warrant consideration in clinical care.
Psychiatric Comorbidity
There is a clinically significant increase in the prevalence of psychiatric disorders in MS. Bipolar disorder and psychosis are more prevalent in MS than in the general population.40,41 In MS, psychiatric comorbidity is associated with faster disease progression and poorer quality of life.42 Relative to the general population, MS is associated with elevated levels of depression and anxiety.40 Depression has been identified in 35% of people with MS, and anxiety has been found in 55% of patients.40 The lifetime prevalence of depression in MS is 50%.43 The lack of definitive cause and unpredictability of the course of MS can lead to uncertainty and exacerbate depression and anxiety symptoms.
Migraine is similarly associated with increased prevalence of psychopathology, including depression (28%), anxiety (20%), and panic disorder (23%), and these are particularly high for individuals reporting severe interictal disability.44,45 Migraine chronification describes the process of episodic migraine transforming into chronic migraine, with migraine increasing in frequency. Depressive symptoms and stressful events were identified as risk factors for migraine chronification.23,46,47 Depression and anxiety symptoms were also associated with higher general pain and migraine-related pain levels.46,47 Much like MS, the unpredictable onset and course of migraine, along with its burdensome consequences, can be distressing for people with the condition.
It is, however, difficult to disentangle the effects of mood on the precipitation and worsening of migraine and MS symptoms, as these relationships are likely bidirectional. Kister and colleagues found significantly higher levels of depression and anxiety symptoms in those with co-occurring migraine and MS relative to people with MS without migraine, highlighting the elevated burden of psychopathology when these conditions co-occur.35
Cognitive Impairment
It is estimated that 43% to 70% of people with MS experience some form of cognitive impairment.3 This is typically characterized by impairments in processing speed, high-level attention, working memory, episodic memory, and executive functioning.3,48 As would be expected, RRMS is associated with less frequent and less severe deficits than progressive forms of MS, although the broad domains affected remain consistent.49 Importantly, even mild cognitive dysfunction in MS can impact social and occupational functioning, activities of daily life, and quality of life.3,50
People with migraine rank cognitive impairment as their second most disabling symptom after pain intensity.51 Commonly reported symptoms include distractibility, impaired reasoning, and slowed thinking during a migraine attack.52 A few cognitive studies have found objective deficits in the interictal period, including poorer attention, working memory, processing speed, executive functioning, and verbal learning and memory when compared to healthy controls.53,54 However, processing speed, attention, and verbal learning and memory are significantly poorer during a migraine attack relative to the interictal period,55 suggesting the possible reversibility or partial improvement of cognitive deficits in the interictal period. Researchers postulate that this is due to disturbances in the prefrontal and temporal cortices and frontal subcortical white matter, given the attention and executive deficits characterizing migraine.55,56 As expected, gray matter density was significantly reduced in people with migraine relative to healthy controls, particularly in the frontal and temporal cortices.57 This also correlated with T2-visible lesion load, age, and disease duration.To date, there has been no research investigating the cognitive phenotype of people with both MS and migraine. While these conditions share common cognitive deficits, the degree, nature, and impacts of cognitive symptoms when migraine co-occurs with MS remain unclear but are highly clinically and functionally relevant.
Current Treatment Options
Pharmacological Treatments
Treatment pathways for migraine in MS include acute and preventive medications and nonpharmacological approaches, akin to those given to individuals with migraine alone. Most people treat migraine acutely with paracetamol, triptans, and nonsteroidal anti-inflammatory drugs.24 According to the recently published International Headache Society Global Practice Recommendations,58 acute treatments should provide pain freedom within 2 hours of administration and should be taken as early as possible in the headache phase. Commonly prescribed nonsteroidal anti-inflammatory drugs include ibuprofen, aspirin, and diclofenac potassium. Triptans are typically used for headaches of moderate to severe intensity. Oral sumatriptan, for example, was found to provide freedom from moderate to severe pain within 2 hours of usage for 32% of individuals compared with 11% using placebo.59On the other hand, overuse of analgesics can precipitate medication-overuse headache and should prompt a review of preventive medication efficacy. Triptans, opioids, and barbiturates pose a higher risk of causing medication-overuse headache.60 Appropriate management of nonmigraine pain in MS is therefore also important to minimize regular use of analgesics. Gepants (ie, small molecule CGRP antagonists) and lasmiditan (ie, a selective agonist of the 5-HT1F receptor) are more recently developed acute treatments for migraine. While evidence is currently limited, gepants have not been associated with medication-overuse headache.58 With further research, gepants may become the preferred treatment for those at higher risk of medication overuse.58 It is of note that access to gepants varies globally, as approval of types of gepants differs among countries. For comprehensive guidance on selecting and monitoring migraine medications, refer tothe global practice recommendations for both acute58 and preventive61 treatment of migraine. These recommendations include guidance on medications for menstrual migraine, refractory migraine, and migraine during pregnancy and breastfeeding. Table 1 presents the key considerations and uses of acute migraine medications and nonpharmacological interventions for people with MS and migraine.
Initial assessment should include monitoring with a headache diary. This has multiple benefits of providing (1) information on the presence of migraine symptoms (eg, aura, nausea, vomiting), (2) an understanding of the number of migraine days per month for treatment decisions, (3) identification of patterns of triggers to be targeted (eg, stress, physical activity), and (4) assessment of treatment response. Preventive treatments for migraine aim to reduce the frequency, severity, or duration of migraines. Prophylactics are typically prescribed when there are more than 4 migraine days per month and considered for those with over 2 migraine days per month if refractory to acute treatments or attacks are very disabling.60,61 The choice of preventive therapy is guided by consideration of its adverse effects and tolerability along with efficacy, cost, patient preference, medical comorbidities, and MS symptoms60 (Table 2). All oral agents should be uptitrated gradually to improve tolerability and maintained at an effective dose for at least 8 to 12 weeks to determine efficacy.
Due to MS-related adverse effects and potential benefits to MS-related symptoms, clinicians need to consider preventive medication use carefully. For example, for people with MS and migraine, an antidepressant (eg, amitriptyline) may benefit those with comorbid depression and sleep disturbance, while topiramate would not be recommended for people with prominent fatigue and cognitive deficits because it is known to have adverse effects in these areas.60 OnabotulinumtoxinA delivered through pericranial injection has also proven efficacious in preventing chronic migraine, with no systemic adverse events reported.63 The use of monoclonal antibodies that act against CGRP or its receptor has been shown to be effective in preventing episodic and chronic migraine and is generally well tolerated.60 However, Ray and colleagues noted the potential for inflammatory and immunological complications following treatment with CGRP monoclonal antibodies64 so there may be some concerns about use in people with MS. A recent evaluation of its use in 27 people with MS found that treatment with CGRP monoclonal antibodies was associated with a reduction in migraine frequency, and mild adverse events occurred for 11% of users (eg, muscle spasms, constipation, headache), but there was no worsening of MS symptoms with concurrent DMTs.65 While the emerging evidence on the safety and tolerability of CGRP monoclonal antibodies in people with MS is reassuring, larger studies are needed. If used, it is recommended that MS disease activity be closely monitored and medication reviewed if there is any MS symptom deterioration. Choice of prophylactic treatment also depends on availability and access. Access to CGRP monoclonal antibodies and onabotulinumtoxinA differs internationally depending on national funding and prescribing guidelines. For example, qualifiers may include appropriate management of medication-overuse headache and prior failure of 3 oral prophylactics. In addition, in Australia, administration of onabotulinumtoxinA requires a neurologist familiar with the administration of the PREEMPT protocol.
For people with MS and migraine, preventive medications are underutilized. Only 10.8% of people with MS and migraine have ever taken a preventive medication,23 despite the finding that 21% have greater than 15 days of migraine per month.24 This underscores the significant undertreatment and limited recognition of migraine in MS.
Suboptimal treatment of migraine is a risk factor for its chronification.68 As would be expected, nonadherence to prescribed treatments hinders effective management of migraine. Gallagher and Kunkel reported that 71% of their patients with migraine delayed or avoided the use of prescribed medications.69 The reasons for nonadherence were mainly the high cost or adverse effects of medications, as well as medication misuse (eg, incorrect use, under-use) and forgetfulness. The undertreatment of migraine, as well as issues with patient adherence to and appropriate use of medications, are ongoing challenges in establishing effective migraine management for people with MS.
Nonpharmacological Approaches
Importantly, a combination of pharmacological and nonpharmacological approaches has been shown to be more effective in reducing the frequency of migraine than delivery of either treatment arm alone.70 Individuals with migraine may lack the coping skills to manage stressors, which can precipitate and exacerbate migraine.71 As a result, a multidisciplinary approach has been recommended. Psychological interventions for migraine aim to reduce its frequency, severity, fatigue, pain, and psychological symptoms, and to improve psychosocial outcomes and minimize reliance on acute medications.44,71 These interventions often include a combination of education, cognitive-behavioral components, biofeedback, relaxation techniques, and stress management training.44,62 Clinicians work with the patient to promote an understanding of how maladaptive thoughts and behaviors precipitate and exacerbate stress or migraine, thereby affording some self-efficacy in symptom management.44 Hunter and Ross were the first researchers to identify that a migraine attack can be exacerbated by worry, distress, frustration, and rumination.72 Since then, evidence from trials and meta-analyses has demonstrated reductions between 30% and 50% in the number of migraine days, pain intensity, and duration after cognitive behavior therapy and/or relaxation training relative to control conditions (ie, migraine education, wait-list, or intervention placebo71,73). These reductions were maintained at 1-year follow-up.74
Conclusions
For individuals with co-occurring MS and migraine, the combination of cognitive compromise, comorbid psychological or psychiatric illness, fatigue, sleep disturbance, social and occupational dysfunction, and the undertreatment of migraine creates an area of significant unmet need and a gap in clinical care. Through appropriate identification and management, both pharmacological and nonpharmacological treatments for migraine can converge and may provide an opportunity to address these comorbid symptoms. In this way, a multidisciplinary approach to migraine management can be established within an MS framework. Given the high prevalence and level of disability of migraine in MS, randomized controlled trials are needed to investigate the efficacy of treatments in reducing migraine frequency and in improving secondary outcomes (eg, relief of fatigue). Such research is likely to improve clinical care and the quality of life of people with MS and alleviate the associated burden on health care systems.
In contrast to this body of literature, the systematic review of 21 trials and meta-analyses by Sharpe and colleagues found no effect of psychological intervention on migraine intensity, mood, quality of life, migraine-related disability, and medication usage compared to control groups.75 The authors did identify several methodological limitations plaguing the studies, including a lack of randomization of participants, a lack of blinding for outcome measures, small sample sizes, low treatment integrity (eg, lack of treatment manual and reporting on therapist adherence), and overrecruitment from tertiary headache centers.75 Thus, the authors felt that the studies included were of very low quality due to these significant design limitations and, consequently, they had little confidence in the null findings. They concluded that the findings are likely to change if studies with valid and reliable research methods are conducted.
