Abstract
Morning headache is regarded as a ‘red flag’ alerting clinicians to a potential underlying sleep disorder. Morning headache has been most frequently associated with obstructive sleep apnea and ‘sleep apnea headache’ is the only headache secondary to a sleep disorder recognized by the diagnostic criteria of the International Classification of Headache Disorders 2nd Edition (ICHD-2). However, the sensitivity and specificity of morning headache to predict sleep apnea is uncertain. Insomnia and a wide variety of other sleep disorders have been associated with morning headache as well, while psychiatric disorders are comorbid with both headache and sleep disorders. The report by Chen and colleagues (1) advances our understanding with a well-designed prospective study of morning headache in habitual snorers considering predictors and functional impairment of morning headache.
Morning headache (also known as awakening headache) is the most common headache descriptor in clinical and epidemiological studies of sleep-related headache, generally referring to any form of headache that is proximally related to sleep. Morning headache emerges during sleep or upon awakening, and differs from chronic daily headache in that morning headache tends to remit after an interval of being awake rather than being unremitting. Morning headache patterns are often considered to be potential markers for a sleep disorder. Unfortunately, most studies of sleep related headache have not followed ICHD-2 criteria for sleep apnea headache or other standardized headache diagnostic criteria. Thus, the reliability of diagnosis is unknown. Inconsistencies in headache diagnosis limit generalizability across studies and probably account for at least some of the conflicting outcomes reported in the literature.
Previous studies have noted a prevalence of morning headache in 4–6% of the general population, 18–60% of sleep apneics, 18% of insomniacs, and 21% of depressed individuals across studies (2). Chen et al. (1) found that 23.5% of a sample of 268 habitual snorers presenting for evaluation to a sleep disorder center had morning headache. Interestingly the 23.5% of snorers with morning headache showed greater functional impairment across all domains of social, physical, role and emotional impairment after controlling for sleep disorders, including obstructive sleep apnea, age, gender, psychological distress, BMI, migraine diagnosis, and smoking. Of total sample of 268 snorers, 43% (n = 116) had insomnia and 72% (n = 193) had elevated scale scores suggesting psychological distress. Thus, morning headache appears to be a significant symptom in its own right and should alert one to potential functional impairment.
Headache has been associated with a wide range of sleep disorders in epidemiological and case-control studies reviewed elsewhere, including obstructive sleep apnea, insomnia, circadian rhythm disorders, movement disorders (e.g. restless legs syndrome, periodic limb movements), and parasomnias (2,3). Likewise, obesity and snoring as well as sleep disturbance have been identified as risk factors for progression from episodic migraine to chronic daily headache (4,5).
Chen et al. (1) observed that habitual snorers with sleep apnea were indeed significantly more likely to have morning headache (27.2%) than snorers without sleep apnea (15.5%). Headache characteristics did not differ between morning headache patients with and without sleep apnea. Although sleep apnea is generally the sleep disorder most closely associated with morning headache, the adjusted odds ratio (adjusted for age, gender, smoking habits, and BMI) for sleep apnea (AOR = 2.6) was actually less in the case of sleep apnea than for other predictors including migraine (6.5), insomnia (4.2), and psychological distress (3.9). Interestingly, BMI was not a significant predictor of morning headache (1.6). Data indicate that sleep apnea is neither necessary nor sufficient for morning headache, even among a sample of habitual snorers. In fact, the data suggest that migraine was the most important predictor of morning headache. Psychiatric symptoms and insomnia were also found to be important contributors. This is consistent with findings of earlier studies. Insomnia typically occurs in half to two-thirds of migraineurs presenting to neurology or specialty headache practices (2,3). For example, Kelman and Rains (6) reported sleep complaints from a sample of 1283 migraineurs: 71% reported morning headaches and over half had difficulty initiating or maintaining sleep.
Psychiatric disorders are comorbid with both headache and sleep disorders, particularly insomnia. Chen and colleagues (1) identified a compelling relationship between morning headache and insomnia: 43% of snorers were diagnosed with insomnia and insomnia was a strong predictor of morning headache. A powerful relationship between insomnia and headache frequency has been noted in earlier literature. For example, in a UK cross-sectional study, Boardman et al. (7) identified a dose-response relationship between headache severity and sleep complaints (trouble falling asleep, waking up several times, trouble staying asleep, or waking after usual amount of sleep feeling tired or worn out). Among 2662 respondents, headache frequency was associated with slight (age/gender adjusted OR = 2.4 (95% confidence interval 1.7–3.2)), moderate (OR = 3.6 (2.6–5.0)), and severe (OR = 7.5 (4.2–13.4)) sleep complaints. The study also identified an association with anxiety.
Morning headache was present in 27.2% of apneics and insomnia and psychiatric disorders were at least as closely associated with morning headache as sleep apnea (1). Among other studies of sleep apnea, the proportion is found to vary widely. Provini and colleagues (8) reviewed literature examining headache in sleep-related breathing disorders; across 12 studies, the prevalence of headache varied widely ranging from 18% to 60% of obstructive sleep apneics, including five studies that further specified ‘morning headache’ in 11% to 58% of apneics. Although ‘sleep apnea headache’ is the only formal diagnosis for headache secondary to a sleep disorder recognized by ICHD-2, this diagnosis warrants more precise research methods and diagnostic precision to understand the phenomenon and validate the diagnosis.
ICHD-2 diagnostic criteria classifies sleep apnea headache as secondary headache under ‘headache attributed to hypoxia or hypercapnia’ that emerges de novo or represents exacerbation of a preexisting headache such as migraine or cluster headache. Recurrent headache is present upon awakening in an individual who has demonstrated sleep apnea by overnight polysomnography (Respiratory Disturbance Index ≥ 5) and at least one of the three following characteristics: it occurs on >15 days per month; it has a bilateral, pressing quality and not accompanied by nausea, photophobia, or phonophobia; and each headache resolves within 30 minutes. The final criterion states that headache resolves within 72 hours and does not recur after effective treatment of sleep apnea. Thus, the sleep apnea headache can be definitively diagnosed only after treatment of the sleep apnea is successful.
The ICHD-2 diagnostic criteria were introduced in 2004 and criteria have yet to be validated. There is some indication that a sizeable proportion of apneic patients with awakening headache would not fulfill the new criteria. Alberti et al. (9) observed that sleep apnea headache may present as: migraine, tension, cluster, or unclassifiable; bilateral (53%) or unilateral (47%); located frontal (33%), frontotemporal (28%), or temporal (16%); with pressing/tightening pain in the majority of patients (79%); with intensity mild (47%), moderate (37%) or severe (16%). Headaches remitted within 30 minutes of waking in only 40% of cases.
Chen and colleagues examined the sleep apnea headache diagnostic features in a subsample of 50 patients with confirmed sleep apnea and morning headache (1). Headache characteristics did not differ between morning headache patients with and without apnea. Sixty-two percent (n = 31/50) of sleep apnea patients fulfilled at least one of three proposed headache symptoms. In only 32% of cases, headaches remitted within 30 minutes. Interestingly, pain characteristics were bilateral, pressing quality in only 40% of cases. Patients with a history of migraine were more likely to report morning headaches that were pulsatile.
Although the authors of the ICHD-2 have acknowledged that the mechanisms and specificity of apnea headache remain uncertain, the pathogenic basis of morning headache was initially presumed to be a consequence of abnormal respiration (e.g. hypoxemia or hypercapnia). This hypothesis was supported by earlier polysomnographic research showing a dose–response relationship between the severity of sleep apnea (e.g. the number of apneic events, the severity of nocturnal oxygen desaturation) and severity or morning headache (10) and a higher incidence of morning headache in apneics than in similarly sleep-disturbed insomniacs (9). Some indirect support is suggested by a handful of studies showing improvement in headache following treatment of sleep apnea with non-invasive positive pressure ventilation treatment or surgical modification of the upper airway to improve breathing, but there are no controlled outcome studies.
Other studies, by contrast, have found no differences in hypoxemia or other respiratory measures comparing obstructive sleep apnea patients with and without morning headache (11,12). Likewise, Chen et al.’s (1) data would generally contradict the theory that headache is secondary to hypoxemia or hypercapnia. Only a diagnosis of sleep apnea predicted morning headache. Other respiratory (e.g. mean apnea-hypopnea index, mean arterial oxygen saturation, lowest arterial oxygen saturation, desaturation index, and snoring index) and sleep (percentage stage 1, stage 2, stage 3, or REM sleep, and arousal index) parameters did not differ between habitual snorers with and without morning headache.
The clinical and research implications of Chen et al. (1) are notable. As noted by a variety of earlier studies, morning headache is not constrained to sleep apnea. Further development of diagnostic criteria for sleep apnea headache and other sleep related headaches are needed. Insomnia is probably the most common sleep disorder related to morning headache. Greater diagnostic precision and reporting in studies of sleep-related headache are necessary to facilitate comparison between sleep and headache literatures and to foster accumulation of findings. Improved research methodology would increase the yield of future studies, particularly improved sampling methods that facilitate generalization of results, use of standardized questionnaires, objective measures for sleep, and uniform reporting of results. Morning headache, although not a sufficient diagnosis, is an important clinical symptom in its own right beyond serving as a potential indicator for a sleep disorder. Morning headache was associated with functional impairment and psychological distress and clinicians and researchers should consider headache-related quality of life including disability assessment as an outcome variable in evaluation and treatment of sleep-related headache. Future research may identify factors mediating the sleep and headache relationship.