One of the most relevant psychiatric comorbidities in individuals with migraine is major depression. Several studies [11, 58, 129] suggested that those with migraine were 2.5 times more likely to be depressed when compared with those without. Antonaci and colleagues [2] reported the existence of a high variability (from 8.6 to 47.9 %) of major depression in individuals with migraine according to their meta-analytic study. The comorbidity with major depression has been reported especially among individuals with migraine with aura and chronic migraine [3]. In addition, individuals with migraine and depression are more likely to be refractory to commonly available migraine treatments more frequently developing medication overuse and disability [78]. Based on existing evidence, a bidirectional relationship between migraine and major depression has been suggested [11, 84, 85].

Although there is no evidence, to our knowledge, showing that an improvement of major depression may reflect an improvement of migraine symptoms, some of the mechanisms associated with the efficacy of antidepressant medications in the prevention of migraine seem to be related to the shared causative mechanisms underlying both depression and migraine [12]. Specific pathophysiological hypotheses have been postulated in order to explain these shared causative mechanisms: serotonergic dysfunctions, periventricular white matter abnormalities, hormonal influences, and sensitization of the sensory/emotional neural networks [5, 103, 104]. These mechanisms were frequently mentioned in the effort to explain the commonly observed comorbidity between migraine and most of psychiatric conditions in clinical populations.

Given that more studies about psychiatric comorbidity in migraine focused on unipolar depressive symptomatology, the exact prevalence of bipolar disorder spectrum in chronic headache patients is controversial. Based on available evidence, it has been suggested that bipolar disorder rates in chronic headache as well as in other headache disorders are largely underestimated. According to the findings of Robbins [94] who conducted their study in a sample of 287 total chronic headache subjects, bipolar disorder rates were 6.6 %, but a lifetime bipolar disorder rate in migraine was 8.6 % in the overall sample, 6.4 % in episodic chronic migraine patients, and 6.8 % in those experiencing chronic cluster headache were reported.

As for major depression, a bidirectional relationship has been also postulated between migraine and bipolar disorder. Several reports [8, 67, 68] found that the likelihood of being diagnosed with a bipolar disorder was three times higher in migraine patients with aura compared to the general population. It is also important to note that approximately 30 % of bipolar patients suffer from migraine. Recently, Fornaro and Stubbs [25] reported that those with a bipolar disorder type II were more likely to report migraine than those with bipolar disorder type I (54.17 % vs. 32.7 %). According to recent evidence [29, 31], subjects who suffered from a comorbidity with migraine had a type of bipolar disorder which is characterized by an earlier age of onset, a rapid cycling course, more frequent comorbid panic attacks, and increased prevalence in females than those without this migraine comorbidity. Dilsaver and colleagues [20] reported the existence of many similarities such as the episodic course, increased vulnerability to stress, family history of affective disorders, and a better response to valproate among those with bipolar disorder and comorbid migraine. Similarly to other psychiatric comorbidities, it has been hypothesized that even migraine and bipolar disorder may share common abnormalities (e.g., dysfunction of calcium channels) that play a fundamental role in the pathophysiology of both these conditions [31]. These conditions seem to also have a profound link with sleep disturbances as they share relevant neuroendocrine dysfunctions together with a significant response to lithium therapy as suggested by Costa et al. [18]. The importance of making the appropriate diagnosis may have a significant impact on headache care and suggests the utilization of more appropriate mood stabilizers conversely to what was commonly observed in clinical practice (e.g., bipolar disorder is often misdiagnosed, especially on initial presentation).

Patients with migraine were also more likely (in more than 50 % of cases) to develop lifetime anxiety disorders. The prevalence of anxiety disorders is nearly 2–5 times higher in subjects with migraine when compared to the general population and significantly higher in patients with chronic than episodic migraine [13]. Baskin and colleagues [4] noted that comorbidity between depression, anxiety disorders, and migraine is common. Migraine is also commonly associated with generalized anxiety disorder, obsessive compulsive disorder, and panic disorder [4].

As suggested by McWilliams et al. [66] and Smitherman et al. [107], a bidirectional association was found between migraine and anxiety disorders as the risk of experiencing migraine was 3.86 times higher in subjects with generalized anxiety disorder compared to those without. Moreover, individuals with migraine were 3.13 times more likely to develop generalized anxiety disorder than those without. According to recent reports [32], subjects with obsessive compulsive disorder were 4.57 times more likely to experience migraine with a frequency of 2–6 times per week. Moreover, Smitherman and colleagues [107] reported that patients with migraine, especially those affected by migraine with aura, were 3.76 times more at risk to have a panic disorder. Individuals with panic disorder usually experience a higher migraine frequency, an elevated psychosocial impairment and disability as well as an increased risk of medication overuse and chronicity [107]. There are many similarities such as the fear associated with illness episodes, autonomic and gastrointestinal symptoms, functional impairment, and overall burden of disease [107] between these groups. The strong relationship between migraine and panic disorder is also indirectly supported by the increased self-reported use of non-pharmacological treatments for headache in those with a family history of anxiety [69]. Baskin and colleagues [4] suggested that the correct management of anxiety in patients with migraine usually reflected the improvement of quality of life, a higher adherence and effectiveness of both pharmacological and non-pharmacological treatments for migraine.

Interestingly, Gonda and colleagues [30] reported that both subjects with anxiety and migraine showed a common genetic vulnerability to develop these conditions (e.g., higher frequency of the s-allele in a study about the 5HTTLPR polymorphism of the serotonin transporter gene). Several abnormalities have been postulated to play a role in the association between anxiety and migraine: serotonergic and hypothalamic-pituitary-adrenal (HPA) axis dysfunctions, hormonal abnormalities, pain-related cognition, and avoidance learning [59, 107].

Impulsivity and aggression may be commonly observed in patients with chronic headache since individuals with migraine often experience restlessness or agitation during headache episode [35]. Importantly, up of 90 % of patients with migraine demonstrated a variety of aggressive behaviors and impulsivity [117]. According to functional neuroimaging studies [41], the posterior hypothalamus is undoubtedly implicated in determining the occurrence of aggressive behavior throughout the chronic headache attacks.

It has been suggested that self-aggressive and depressive cognitions in chronic cluster headache correlated with depressive symptoms and impairment [60]. The authors evaluated 26 patients with chronic, 25 with active episodic, and 22 with episodic cluster headache outside the active period that have subsequently been compared to 24 migraine patients and 31 headache-free volunteers. Luerding and colleagues [60] reported that in chronic cluster headache and active episodic cluster headache patients, high levels of “self-aggression/depression” were all significantly associated with higher depressive symptoms and impairment measured at both an emotional and functional level. The authors stressed the notion that depression levels are associated with elevated self-aggression. Overall, aggressiveness and impulsivity represent complex phenomena that may be a frequent correlate of disabling conditions such as pain and major depression [1, 124].

One of the most relevant migraine triggers may be identified with stressors. Subjects with chronic migraine usually presented with higher levels of stress [22] as well as more disabling life events in the year prior to their migraine chronicity when compared to those with episodic migraine [100]. These data have been also replicated by the longitudinal Brazilian study that demonstrated the association between higher migraine frequency and negative life events [98]. Several studies [36, 39, 40, 87, 88] showed the efficacy of relaxation therapies, biofeedback, and cognitive behavioral therapies in managing stress as well as in migraine prophylaxis. Both structural and functional impairments and even repeated migraines may be related to chronic stress exposure [7, 65]. It has been demonstrated that negative changes significantly influence the onset and maintenance of pain processing and central sensitivity, in conjunction with the same pain experience in migraine patients [7, 88]. fMRI studies also revealed that migraine patients exerted an increased activation of the perigenual cortex (associated with allostatic alterations in rats; [113]) relative to patients without migraine after a painful heat stimulus.

Important abnormalities have been also found in patients who suffered from analgesic or other headache medication abuse. For example, abnormal ACTH and cortisol release have been reported in individuals with chronic migraine and medication overuse [89]. A positive correlation between duration of chronic migraine and endocrine abnormalities has been observed [89]. Moreover, the fact that the decrease of stress usually reflects a subsequent higher risk of migraine attacks further enhances the existence of a biological link between migraine and stress [57]. It is important to note that both neuroendocrine and autonomic nervous systems are usually enhanced by stress with subsequent elevated production of glucocorticoids. The early exposure to stressful traumatic experiences might enhance the likelihood of developing major psychiatric disorders and is associated with the decrease in the expression of glucocorticoid receptors (GR) influencing gene function and response to future stressors [16].

Tietjen and Peterlin [116] reported that those with a history of emotional, physical, and sexual abuse as well as physical neglect were more likely to develop migraines than the general population. Moreover, those who have been emotionally abused were more likely to develop chronic and more invalidating migraines as well as an earlier migraine onset relative to those who did not [114]. Tietjen and colleagues [115] observed that this correlation was independent of comorbidity with depression and anxiety.

Migraine frequency has been also linked to the number of childhood traumatic experiences [116]. A higher risk of migraine diagnosis rather than tension headaches has been reported in those with headache and a history of emotional abuse, emotional neglect, and sexual abuse [115]. Recent studies focused on cortisol dysfunctions and enhanced stress reactivity presumably related to past traumatic experiences in order to explain the association between abuse and migraine [53, 101]. Recently, it has been suggested that post-traumatic stress disorder, being a valid predictor of migraine, is able to mediate the association between migraine and childhood abuse [107]. Post-traumatic stress disorder among veterans, in particular men, is associated with enhanced headache drug utilization [102]. Post-traumatic stress disorder may also moderate the association between chronic pain and childhood abuse [92]. Raphael and Widom [93] hypothesized that the comorbidity with post-traumatic stress disorder is more relevant in these patients than the same burden related to childhood abuse in patients with chronic pain.

There are some psychological similarities between patients with post-traumatic stress disorder and those with chronic pain including fear avoidance and anxiety sensitivity [75, 105]. Xie and colleagues [126] stressed the presence in subjects with migraine and comorbid post-traumatic stress disorder of serotonin dysfunctions and 5HTTLPR polymorphisms of the serotonin transporter gene, whereas other reports [5, 34, 127] focused on the relevance of estrogen and the HPA axis in order to explain the comorbidity between migraine and post-traumatic stress disorder. As suggested by the study of Peterlin and colleagues [80], 4.5 % of patients without headache, 12.6 % of those with episodic non-migraine headache, 21.5 % of those with episodic migraine, and 19.2 % of those with chronic daily headache suffered from a lifetime post-traumatic stress disorder, respectively. It is also reasonable that major depressive disorder and post-traumatic stress disorder are able to reliably explain the more commonly observed medication abuse in headache individuals [79–82]. The comorbidity between major depressive disorder and post-traumatic stress disorder is undoubtedly more frequent in chronic daily headache than in episodic migraine [79, 82]. Patients with post-traumatic stress disorder were also more likely to show comorbid depressive and anxiety disorders [50], and thus post-traumatic stress disorder seems also able to account at least for the occurrence of some psychiatric comorbidities related to both episodic and chronic migraine. Importantly, targeting the multiple psychological implications related to post-traumatic stress disorder and depression may help to reduce the illness burden and attenuate the disability associated with migraine comorbidities.

Migraine subjects with post-traumatic stress disorder may be considered at greater risk of disability and experience more lost work days as they more frequently suffer from physical, mental health, and substance problems together with interpersonal difficulties when compared to migraine individuals without post-traumatic stress disorder [79, 82, 91]. Muse [72] and other recent authors have suggested that the disability of migraine individuals who suffered from a comorbid post-traumatic stress disorder may be significantly reduced using cognitive-behavioral therapy.

Overall, given the same frequency and prevalence exposure to stressful traumatic events, post-traumatic stress disorder is undoubtedly more common in those with chronic migraine relative to those with episodic migraine [51, 81, 99]. In summary, as suggested by Peterlin and colleagues [80], migraine subjects showed an increased lifetime and 1-year risk to develop post-traumatic stress disorder than the general population. But, relevantly, migraine subjects more commonly reported maladaptive stress responses that could enhance their vulnerability to develop a post-traumatic stress disorder [80]. Several neurobiological hypotheses have been formulated to explain the correlation between migraine and post-traumatic stress disorder such as serotonergic dysfunctions, HPA overactivity, and autonomic abnormalities. Since these abnormalities may be also found in subjects who were diagnosed with affective and anxiety disorders, this could be an interesting research area for the future in order to provide further explanations about the frequent comorbidity between post-traumatic stress disorder and migraine [47].

Rains and Poceta [90] suggested that patients with migraine may commonly suffer from sleep disorders when compared to the general population. Similarly, a higher risk of disabling sleep disorders in patients with migraine has been reported by Odegard and colleagues [74]. Kelman and Rains [49] found that more than 50 % of migraine subjects showed occasional sleep disturbances but more than 30 % of them reported common and invalidating sleep disorders together with sleep duration of less than 6 h per night. More severe and disabling headaches may be associated with chronic short sleep [49], but patients with migraine usually reported a variety of sleep disorders including insomnia, restless leg syndrome, parasomnias, narcolepsy, and periodic limb movements during sleep [17].

Insomnia is a frequent comorbid condition in patients with migraine, depression, anxiety, and post-traumatic stress disorder. However, migraine and psychiatric conditions negatively influence sleep quality; however, poor sleep quality significantly predisposes one to migraine. Thus, sleep disturbances may be a very interesting indicator of psychiatric comorbidity in migraine patients helping to identify specific subgroups at risk for negative outcomes.

When compared to a placebo intervention, behavioral sleep modification is also able to both ameliorate headache frequency/severity and attenuate the illness burden related to both chronic and episodic migraine based on a small randomized placebo-controlled trial [15]. Chronic migraine patients also reported significant benefits after managing sleep apnea and other sleep disturbances [90].

Similar to other psychiatric conditions, a bidirectional relationship has been postulated to explain the comorbidity between sleep disturbances and migraine: importantly, sleep disorders may be considered relevant risk (triggering) factors for migraine, but even migraine may significantly influence patients’ quality of sleep [23]. Dodick and colleagues [21] suggested the existence of similar neurobiological mechanisms related to the occurrence of both sleep disturbances and migraine. Furthermore, major depressive and anxiety disorders are usually comorbid with both sleep disturbances and migraine.

Overall, the interaction between headache/migraine and sleep disturbances has been confirmed by several studies both in populations of adults and children. A detailed anamnesis for the presence of sleep disturbances should be conducted in order to carefully assess and manage either the disability or functional impairment of these conditions on migraine subjects [23].

The association between substance abuse/dependence and migraine has been commonly reported [55, 97]. In 1982, Tennant and Rawson [112] demonstrated that more than 15 % of dependent patients first used opioids due to their headache, although not all subsequent studies confirmed these data [80, 111]. In particular, Peterlin and coworkers [80] did not find any association between substance abuse and migraine after controlling for depression and post-traumatic stress disorder.

Alcohol abuse/dependence does not seem to be correlated with migraine and chronic headache. According to the study of Jette and colleagues [45], a 12-month prevalence of alcohol dependence was not reported among those who suffered from migraine and healthy controls. Some researchers [76, 128] reported that the absence of a correlation between alcohol abuse/dependence and migraine may be explained by the fact that alcohol is an important triggering factor for the occurrence of headache/migraine attacks. Migraine subjects commonly reported that foods, alcohol, atmospheric changes, exposure to light, sounds, or odors are triggering or aggravating factors related to the occurrence of migraine attacks [38].

According to a pathophysiologic hypothesis, some specific environmental stimuli may affect the vulnerable brain leading to complex cellular and molecular changes, which have been shown to be significantly involved in the emergence of migraine attacks. Panconesi [76] suggested that the relationship between alcohol and headaches could be mediated by the 5-hydroxytryptamine system. Given this framework, some anticipatory symptoms, such as food craving, and abnormalities of the sleep-wake cycle may be correctly interpreted as relevant warning signs in preceding (at least 72 h before) the pain phase of the same migraine attack.

Suicidality includes suicidal ideation, attempts, and death. The association between migraine aura and suicidality has been commonly described, and the shared pathophysiology between migraine, major depression, and suicidal behavior has been frequently reported [8, 77, 86]. Suicide attempts have been more commonly observed among patients with migraine relative to the general population, in particular among females and in those who suffered from migraine with aura [37]. The association between migraine and suicidal ideation has been investigated by Wang et al. [119] in 3963 adolescents where suicidal ideation has been reported in 8.5 % of the sample. Individuals with migraine, in particular patients affected by migraine with aura, exerted a higher suicidal ideation relative to healthy subjects. Only migraine with aura is correlated with the higher headache frequency after controlling for major depression scores and sociodemographic features. Only migraine with aura, but not headache frequency or the diagnosis of migraine per se, was significantly associated with suicide attempts even after adjusting for the presence of psychiatric conditions [8] in a sample of young adults with migraine.

Cluster headache has been renamed by clinicians as “suicide headache,” but recent studies that focused on the occurrence of suicidal behavior in chronic headache patients criticized this terminology [63]. Rozen’s survey found that 55 % of chronic headache subjects reported lifetime suicidal ideation and 2 % suicide attempts with no relevant differences between males and females [96]. When addressing impairment in chronic headache patients, Jürgens and colleagues [48] reported that suicide thoughts and suicidal tendencies were experienced by 22 % of chronic cluster headache patients, 15 % of episodic cluster headache patients in the active phase, 14 % of episodic cluster headache patients outside the active period, 4 % of migraine patients, and 3 % of healthy controls, respectively. Conversely, Robbins and colleagues [95] found that 6.3 % of subjects with episodic chronic headache patients and 5.9 % of those with chronic cluster headache patients had lower rates of active suicidal ideation, respectively.

Possible differences between migraine and non-migraine headache in suicide have been reported by Ilgen and colleagues [42]. The authors demonstrated in a 3-year follow-up study that suicide rates were increased in both migraine patients (OR = 1.68) and headache/tension headache (OR = 1.38). Based on their findings, migraine was the only significant predictor of suicide (OR = 1.34) even after controlling for age, sex, a medical comorbidity score, and psychiatric disorders. Chronic daily headache subtypes, headache frequencies, and medication overuse were not found to be correlated with suicidal behavior [120], whereas only headache severity, but not headache diagnosis, seems to predict the emergence of suicide attempts in a 2-year follow-up community-based study [10]. Other larger community-based studies such as the Epidemiological Catchment Area Study [111] and National Comorbidity Survey-Replication [43] showed the association between self-reported headache and suicidality.

The relationship between migraine with aura and major depression seems to be bidirectional [11], and common pathophysiological mechanism such as serotonergic dysfunctions and pain have been reported underlying both suicidal behavior and migraine [24, 119]. It is important to note that pain related to headache is a potential independent risk factor for suicide, especially for those with chronic headache or multiple pain locations [43]. Early and systematic suicide evaluation in the clinical practice should be carried out in migraine patients who suffer from chronic pain.

As suggested by Maizels [62], an appropriate screening for comorbid psychiatric disorders in subjects with migraine and chronic headache should be performed. Many psychometric instruments are available to screen patients with psychiatric symptoms and specific disorders such as major depressive, anxiety, and bipolar disorders. The presence of comorbidity may significantly modify the illness outcome, worsen the quality of life and contribute to the psychosocial impairment of migraine patients; thus, the early and careful screening for psychiatric comorbidity is recommended [61, 62]. There are reports indicating the possibility to conduct verbal screening with unstructured questions [123], other studies supporting the appropriateness of four depression and anxiety questions included in the Patient Health Questionnaire [52, 125] and others assessing comorbidity using more articulated and detailed instruments which are also frequently used in psychiatric populations. As suggested by Smitherman and colleagues [108], a more accurate evaluation for the presence of psychiatric comorbidity seems to be necessary subgroups of patients who are unsuccessful with conventional treatment and/or show a partial response to the commonly available migraine medications.