14.1 Introduction

Cluster headache (CH) is among the most severe and disabling primary headache disorders. A bout of CH results in excruciating unilateral pain occurring several times daily, with each episode lasting up to 3 hours and being accompanied by prominent ipsilateral cranial autonomic features [1]. Extreme restlessness, explosive anger, and even self-injury often occur during full-blown attacks, something rarely if ever seen in other acute pain conditions. Although first coined by Horton in 1952 as the “suicide headache” [2], it may be more appropriately characterized as the “aggressive headache,” as violent behaviors are much more commonly reported than actual suicide attempts or completions [3]. Dr. Lee Kudrow, a seminal investigator on CH, early on provided the following personal account of a cluster headache attack: “I am stuck with the additional fear that the pain will never end, but I dismiss it as impossible. Even if it were the case, I would surely kill myself” [4]. No matter what descriptor is used, it is clear the pain experienced during these attacks is most often unbearable.

Between 2000 and 2017, more than 200 literature reviews addressing CH were published: the most cited—and oldest—ones dealt with diagnostic and pathophysiological issues, as well as with treatment options [5–8], and the same is true for reviews published in the last 2 years [9–12]. That said, the potential role of behavioral and psychological factors in this disorder remains uncertain, as psychopathological assessment results have been variable, due in part to the complexity of this disorder as well as methodological shortcomings (i.e., uncontrolled clinical interviews, small sample sizes, absence of appropriate control or comparison groups, non-standardized assessment procedures or measures, etc. [13]). Given the widely acknowledged pronounced impact of CH, it is surprising that this aspect has been relatively ignored in the literature. During CH attacks, patients are typically unable to function as desired, which in turn leads to adverse psychosocial sequelae, marked impairments in daily activities, and reduced quality of life (QoL) also outside the CH bouts. In this chapter, we examine what is currently known about psychological and behavioral factors related to CH, how CH impacts disability and QoL, and, in brief, how what is known may help inform management of CH.

14.2 Psychological and Behavioral Aspects

Unlike other pain disorders, CH is more prevalent in men than women, although this ratio appears to have decreased from a high of 6.2:1 in the 1960s to 2.1:1 in the 1990s [14]. Early characterizations of psychological profiles of individuals with cluster headache (such as those described by Friedman and Graham in the 1950s–1970s [15, 16]) were “hypermasculinized” and framed as ambitious, hard-working, rugged in appearance, heavy smokers and drinkers, yet internally passive. Much of the literature produced during this time was driven by psychoanalytic perspectives and lacked methodological rigor [17].

Since then, more rigorous attempts, incorporating objective, validated measures, with a focus on multiple headache disorders as well as matched non-headache samples, have been pursued, but, again, with limited convergence. In one of the earliest controlled investigations, 12 patients with episodic cluster were compared to 26 migraineurs, 39 tension-type (TTH; then termed “muscle contraction”), 22 combined migraine and TTH, and 30 non-headache controls (who were asked to assist in recruiting a friend or relative of the same sex, approximate age and marital status, and presumably similar socioeconomic status) on a comprehensive battery of psychometrically validated measures: Minnesota Multiphasic Personality Inventory (MMPI), modified Hostility Scale derived from the MMPI, Beck Depression Inventory, State-Trait Anxiety Inventory, Autonomic Perception Questionnaire, Rathus Assertiveness Schedule, Social Readjustment Rating Scale, Psychosomatic Symptom Checklist, Schalling-Sifneos Scale, Need for Achievement, and Hostile Press [18]. On the MMPI scales of most interest, patients with CH were rarely significantly different from controls (and incidentally most similar to the profiles for migraineurs). The only scale where differences emerged between controls and CH patients was scale 3, which reflects tendencies toward somatization during periods of stress. It is important to point out that the slight differences in elevations did not reach the level reflecting clinical significance. The remaining tests in the battery yielded 16 separate scores, none of which revealed significant differences between controls and CH patients (although two scales showed some slight, but again clinically nonsignificant elevations). Prior to this more comprehensive investigation, Kudrow and Sutkus [19] had researched whether the main clinical scales of the MMPI alone could reliably distinguish six different headache types: migraine, TTH (then termed chronic scalp muscle contraction), combined migraine and TTH, cluster, post-traumatic cephalalgia, and conversion cephalalgia. For males and females alike, CH and migraine shared a similar profile, with no clinically elevated scales (with the other groupings showing increased scores). These findings for CH and migraine were soon thereafter replicated with the MMPI for males and females [13]. A subsequent investigation incorporating the MMPI, with larger sample sizes (160 migraineurs, 95 TTH, 149 migraine combined with TTH, and 30 CH) yielded similar findings, showing no major differences between different types of headache [20]. Although slight elevations were noted overall, none fell into the clinically significant range, and, as non-headache controls were not included, it is not possible to determine if the slight elevations would have exceeded those for individuals absent of headache. A more recent investigation of 120 patients with CH, which included a like number of age- and gender-matched controls (case-control study), revealed no significant elevations or differences in MMPI scores between the two groups [21]. Additionally, no visible differences in appearance were found as well when neurologists performed a series of blind ratings of pictures for a subset of both groups.

Yet even more recently, however, investigations using different broad-spectrum personality measures have produced markedly different findings. In a small sample of 26 CH “in-ward” patients, 92% were reported as evidencing pathological levels of personality disorders, particularly obsessive-compulsive and histrionic, when assessed by the Millon Clinical Multiaxial Inventory-III questionnaire [22]. How the obtained profiles might have compared to other headache types on the inpatient unit or to non-headache controls is unknown. A larger study, comparing outpatients with CH (n = 80) to migraineurs (n = 164), reported a higher percentage of negative personality traits among patients with CH compared to migraineurs [23]. Specifically, paranoid and schizoid traits were significantly more prevalent among CH patients (30% and 42.5%, respectively) than migraineurs (11.6% and 25.6%, respectively), as assessed by the Salamanca screening test. Again, comparisons to non-headache controls were lacking.

All studies here-to-date have failed to consider the role that having a pain condition, such as CH, may play in how individuals respond to and are evaluated by personality tests and psychological measures. Nearly all such tests are aimed at capturing traits that are presumed to be relatively stable throughout the lifetime, and many have not been normed among medical populations. With this in mind, it could well be the case that what appears to be psychopathological traits or disorders are more aptly viewed as a reflection of pain state. Indeed, in a broad review of 32 studies on personality characteristics evidenced before and after treatment of chronic pain, 90% of studies using the MMPI showed an improvement in MMPI scores after treatment [24]. One study, specifically focused on headache and craniofacial pain, showed MMPI scores improved after treatment for women, but not men [25]. Overall, these traits were not predictive of treatment outcome.

So, what can be made of this? Although some research advances have been made over the years, significant methodological problems remain—differences in sampling strategies, sample sizes, and settings where data are collected; quality and type of measures used; adequacy of comparison groups selected, with non-headache controls often omitted and other recurring pain conditions rarely included; and, perhaps of most importance, is the failure to take into account the headache status of patients at the time of assessment, distinguish episodic (ECH) from chronic forms (CCH) of CH, and track changes over time. Hence, debates surrounding the “cluster headache personality” continue [3, 26]. Although individual personality characteristics may be informative for how patients may respond to treatment, the conflicting evidence steers us away from using personality as a diagnostic indicator of headache type.

Even though a typical personality profile does not seem to exist in general for CH patients, some findings point to other possible reliable behavioral correlates. Pain disorders in general are largely known to have high co-occurrence with psychological symptoms, particularly anxiety and depression, especially when the condition is chronic [27–29]. Although evidence bearing on this topic is more limited for CH than for migraine, available findings to date suggest the patterns are similar. One particular study of interest was able to examine US insurance claims for individuals diagnosed with CH (both ECH and CCH) and compare them to patients without a headache diagnosis [30]. Claims due to depressive disorders, sleep disturbance, anxiety disorders, and suicidal ideation were around twice as likely in the CH group than controls. Using an insurance claims database allows for large sample sizes (7,589 CH patients and 30,341 controls in this instance), but it is only representative of select individuals with insurance coverage that perceived their condition so burdensome to require some kind of compensation. Additionally, claims and diagnoses could not be cross-verified; thus, results should be interpreted with caution. A more well-controlled study included patients with CCH and ECH (both in and out of active cluster bouts), migraine patients, and healthy controls who were screened for psychiatric co-occurrence using the Mini-DIPS, a validated structured clinical interview [31]. These researchers found that CCH patients fared the worst, with over half the group endorsing depressive symptoms compared to 27% of active ECH patients, 36% of inactive ECH patients, 29% of migraine patients, and 19% of controls. Suicidal tendencies and symptoms of agoraphobia were also higher in this group. It is important to note that these findings are based only on screener items and that few symptoms were endorsed across groups overall, such that only descriptive statistics could be provided. It could be the case that the chronicity, rather than CH itself, is the more salient linkage to these elevated symptoms (consistent with a point made earlier).

Taken altogether, it appears that those who have CH do not experience much psychological distress during times of remission, suggesting the experience of pain itself leads to any psychological sequelae noted. This is consistent with the findings of Liang and colleagues [32] who were able to analyze 673 patients with CH residing within a Taiwan National Health Insurance database over an extensive time period (2005–2009) and compare findings from this group to two large age-, sex-, and comorbidity-matched cohort comparison groups (2,692 patients with migraine only versus 2,692 patients absent migraine or CH; comorbidity matching was based on scores derived from the Charlson Comorbidity Index). Over an extensive follow-up period (median of 2.5 years), the patients with CH were 5.6 times more likely to develop depression when compared to pure controls. However, rates of depression were similar when compared to the migraineurs.

The early descriptor based on observations of the CH patient as a heavy smoker may not be far off, however. Govare’s review [33] found a higher prevalence of reported licit and illicit drug use in CH compared to the general population, particularly for tobacco and cannabis. In the more recent study conducted by Choong and colleagues [30] using an insurance claim database, those diagnosed with CH were three times more likely to have a tobacco use disorder than controls. What is not clear is the nature of this relation. Does it represent a linked predisposition to addiction and CH pathologies, or might substance use contribute to CH onset? An interesting note to consider is that as the gender gap for CH prevalence has closed, so too has the gap for smoking habits, lowering from nearly 9:1 (male to female) in the 1960s to around 2:1 in the 1990s [14], alongside a general decrease in smoking habits in the USA [34] and EU [35]. More longitudinal research in this area may help to further inform our understanding of behavioral precipitants to CH.

Sleep, too, is closely tied to CH pathology [36]. Nocturnal sleeping is often a trigger for cluster headaches, and patients with CH show much poorer sleep compared to controls. Attacks often occur during rapid eye movement (REM) sleep, though they are not limited to this phase. Heightened activity in emotional arousal brain centers, namely, amygdala and medial prefrontal cortex, suggests these areas are key to regulating REM sleep [37]. Interestingly, one study showed abnormal metabolism in the amygdala in ECH during an active phase [38]. This could cause a derangement in the crosstalk between the amygdala and hypothalamus, with a putative role to generate a permissive state of the brain leading to the activation of the cluster circuit [39]. As with substance use and abuse, the full picture of the relationship between sleep dysregulation and CH is not yet clear.

14.3 Quality of Life, Disability, and Impact

Although psychological and behavioral factors related to the onset and maintenance of CH can be informative, what is equally important to consider is how CH affects the lives of these individuals. The socioeconomic burden of CH patients was evaluated in 2007 in a Danish survey using a telephone interview [40]. Data showed that 78% reported restrictions in daily living (13% also outside of cluster periods) and 25% reported a major decrease in their ability to participate in social activities, family life, and housework. Furthermore, the absenteeism rate was 30%, which was significantly higher than 12% among the general population.

14.3.1 Evaluation of QoL

QoL was evaluated in six studies relying on generic tools: in five studies, the Medical Outcome Study Short Form (MOS-SF-36) was used [41–45] and in one case with the 12-item version [41], while three studies used the EuroQoL-VAS [41, 42, 46]. Parallel to this, the Migraine Specific Quality of Life Questionnaire (MSQ) was used in only one study [44]. Table 14.2 summarizes cross-sectional and longitudinal data evaluating QoL. On average, the Physical Composite Summary (PCS) of MOS-SF scored 37.5, the Mental Composite Summary (MCS) scored 35.2, and the EuroQol-VAS scored around 50. The SF-36 and the EuroQoL are generic instruments and, thus, not intended to specifically measure QoL in headache populations. The SF-36 [47] comprises eight scales that can be summed to yield two composite scores: the first addresses physical health issues (e.g., pain, general health, or ability to undertake physical activities), while the second addresses mental health issues (e.g., vitality, emotional state, and mental health). The EuroQoL [48] covers five domains (mobility, self-care, usual activities, pain/discomfort, and depression/anxiety) that comprise a single composite score and also includes a visual analog scale addressing “current health state.” The MSQ [49] is a migraine-specific QoL tool, which has not been formally validated in patients with CH. For all of the three instruments, higher scores indicate higher QoL.

Table 14.2

Studies addressing quality of life

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

and Clinical Features of Cluster Headache Drugs of Cluster Headache and Other Trigeminal Autonomic Cephalalgias Neuralgia: Channels, Pathophysiology, and Therapeutic Challenges Models for Trigeminal Autonomic Cephalalgias

Stay updated, free articles. Join our Telegram channel

Join