Abstract
Background
Hypnic Headache, also known as “alarm clock headache”, is a rare primary headache disorder. It is characterized by frequently recurring headache attacks, which only develop during sleep, especially nighttime sleep.
Objective
This article gives a narrative review on the current knowledge about Hypnic Headache with a focus on secondary Hypnic Headache.
Methods
Based on literature research, using Pubmed and Google Scholar, latest case reports, studies, and systematic reviews about Hypnic Headache were analyzed and summarized focusing on therapeutic options and causes of secondary Hypnic Headache.
Conclusion
Hypnic Headache mainly occurs in elderly patients. However, younger patients and children may also suffer from Hypnic Headache. Many different causes of secondary Hypnic Headache are described in the literature and ought to be ruled out before diagnosing primary Hypnic Headache. The pathophysiology of primary Hypnic Headache remains unclear, but a dysfunction of the hypothalamus seems to play a key role.
Introduction and epidemiology
Hypnic Headache (HH), formerly named “alarm clock headache”, is a rare primary headache disorder, first described in 1988 by Raskin (1). In 2004 the first diagnostic criteria were published in the International Classification of Headache Disorders-2 (ICHD-2) (2). Since 1988 more than 350 cases have been described in literature (3). HH attacks develop exclusively during sleep and cause awakening of the patient, mostly at the same time at night. According to the diagnostic criteria of the ICHD-3 the duration of a HH attack lasts from 15 minutes up to four hours and there are no cranial autonomic symptoms or restlessness accompanying the headache (4). Secondary HH is defined as headache occurring as a symptom of an underlying medical or neurological condition while meeting the diagnostic criteria of primary HH. There is also a huge variety of causes for secondary HH. Considering these might not only be important for diagnostics, but also for future investigations on pathophysiology.
To date the exact prevalence of HH is still unknown. Some older data indicate a prevalence of 0.07% up to 1.4%, whereas current investigations show a prevalence of 0.22% up to 0.3% in the western population (5–7). Although the first five patients described by Raskin were all male, current data suggest that HH it is more common in women than men with a ratio of 2:1 (3,6). HH is mainly described in elderly patients over the age of 50, however, some case reports about children and adolescents do exist (8).
Clinical characteristics
According to the ICHD-3 criteria HH is described as “Frequently recurring headache attacks, developing only during sleep, causing wakening and lasting for up to four hours, without characteristic associated symptoms and not attributed to other pathology” (4). The quality of pain is described as dull in most cases, but also a throbbing, pulsatile or stabbing character has been observed in some cases. The intensity is described as moderate in most cases (3,9). The localization is mainly bilateral, but can also be unilateral (10). In contrast to migraineurs, HH patients report that they get out of bed when they awake with headache and perform some kind of motor activity, e.g. taking a shower, eating, reading, or going for a walk. Additionally, HH patients do not present with the kind of restlessness that can be observed in cluster headache patients (7,11). Although HH occurs episodically in some cases, most patients have a chronic course of disease (10,12). Current diagnostic criteria are shown in Table 1.
Current Diagnostic Criteria of HH According to the ICHD-3 (4).
Former ICHD-2 diagnostic criteria included dull quality of headache, no obligatory lateralization, occurrence during sleep and awakening of patients, high frequency but no upper limit for duration of attacks, occurrence in aging patients, absence of autonomic symptoms, and relative rarity of nausea, photophobia, or phonophobia (11). Current case series and reviews show a greater sensitivity of the ICHD-3 criteria for HH compared to the ICHD-2 criteria (6). However, the absence of autonomic symptoms may still be a point of conflict. As in case series and reviews 8 to 15% of the patients do have mild autonomic symptoms and do not fulfill the criteria for another primary headache type (6,7,13). For the next revision especially this criterium may be reevaluated.
HH in children
HH in children is even more rare than in adults. To date there are in total five published cases about HH in children (8,14,15). Patients were between seven and 11 years old with an average age of 9 ± 1.6 years (3,16). Three out of five patients were girls, two were boys. HH was diagnosed 15.8 ± 25.0 months after the headache started, with a range of 1 to 60 months (17). The headache attacks lasted for up to 30 min in four of five patients (80.0%), in one patient HH attacks lasted between 30 and 60 min. The pain was bilaterally localized in four of five children (80%) and described with severe intensity in four of five children (80%). In four of five patients (80%) the frequency was less than 15 days per month. All patients denied the occurrence of autonomic symptoms. two patients were treated with melatonin in a dose of 3–4 mg and both had a great benefit (8,17). In one case a spontaneous remission of HH attacks has been described (14). Due to the frequency of HH attacks, only one child meets the diagnostic criteria of ICHD-3 for HH, the other four children only fulfill the diagnostic criteria for a “probable HH”. However, differences in characteristics of pediatric and adult patients with migraine have been known for many years (18) and have been described in the ICHD-3. As migraine attacks in children tend to be of shorter duration than in adults, the same might apply to HH in children. Though extremely rare, one should consider that HH can occur in children and adolescents as well.
Current knowledge about pathophysiology of HH
Little is known about the pathophysiology of HH so far. There are many existing theories and approaches mainly based on the clinical picture of the disease. The strictly sleep dependent headache attacks, occurring mostly at the same time at night, indicate a hypothalamic dysfunction as pathophysiological correlate as one possible cause of HH. The hypothalamus as circadian pacemaker regulates sleep and wakefulness and is involved in pain control (19). MRI-based measurements showed a gray matter volume decrease in the posterior hypothalamus of patients suffering from HH, which supports this hypothesis of a pathological circadian pacemaker (20). The pathophysiologic relevance of the observed hypothalamic volume decrease in HH patients remains unclear, even whether it is cause or consequence of HH (21).
Although HH was thought to be REM-phase-related (22), serial polysomnographic investigations showed a proportion of 20-50% REM-related HH attacks while 50-70% of HH attacks occurred during a non-REM period. Additionally, REM- and non-REM-related HH attacks could be detected in the same patient within the same night. There seems to be no connection between sleep stages and onset of HH attacks (23–27). Additionally, the serum melatonin levels were investigated in HH patients, showing no significant differences compared to healthy controls (28).
Further investigations are needed in order to understand more about the pathophysiology of HH and cases of symptomatic HH may give some further hints.
Secondary HH syndrome
HH and brain tumors
For secondary HH, differently located brain tumors have been reported: hemangioblastoma of the cerebellum, nonfunctioning pituitary macroadenoma, growth hormone secreting pituitary tumor and posterior fossa meningioma. Surgical removing of the tumor led to a complete remission of HH attacks in all patients (29–32). Although secondary HH caused by a tumor does not give a specific hint for general causes of HH, it shows that cerebral imaging is important for ruling out a brain tumor as one possible cause of symptomatic HH.
HH due to vascular pathologies
In one case HH occurred after subarachnoid hemorrhage and clipping and coiling of the distal left internal carotid artery aneurysm (33). Cases with different vascular abnormalities such as arterial dolichoectasias and ischemic stroke in the pontine reticular formation have been described (34–36). Idiopathic cyclic edema is a rarely described disease that involves vascular hyperpermeability associated with edema due to interstitial retention of fluid. A 56-year-old patient was reported to have HH attacks associated with the edema. After treatment with aminaphthone, which interferes with the capillary permeability, the generalized edema and headache disappeared completely (37).
Obstructive sleep apnea syndrome and HH
Obstructive sleep apnea syndrome (OSAS) is an often-described comorbidity of HH, but no difference has been seen in the prevalence of OSAS in HH-patients and non-HH patients of the same age (21,38). Moreover, polysomnography showed oxygen desaturations during sleep, that are not in temporal connection to HH attacks (22–24,27). Literature shows morning headache in a proportion of 20 to 40% of patients with OSAS, but no association with HH (39). However, in singular case reports of HH-patients suffering from severe OSAS using a continuous positive airway pressure or mandibular advancement device led to a complete remission of HH (38,40). OSAS may be a cause of secondary HH and should be ruled out through polysomnography.
HH due to low glucose levels
A recent case report described episodes of hypoglycemia, during a 72-hour glucose monitoring period, related to the onset of HH attacks. Rescheduling the patient’s mealtime and increasing the calories at night resulted in a complete remission during a follow-up for 12 months (41).
HH after medication withdrawal
As lithium is one of the first line prophylactic therapeutics for HH, also one case of transient secondary HH after lithium withdrawal is reported. Due to a bipolar disorder the patient was receiving lithium for six years. One month after withdrawal she developed HH attacks, that showed a spontaneous remission after another month. Lithium may have modified the cyclic regulation of melatonin- and serotonin-levels, which may have led to a transient HH (42).
HH due to nocturnal arterial hypertension
Arterial hypertension during sleep has also been described as cause of HH attacks in two case reports. One patient became pain free two weeks after treatment with beta-blockers was started and the other one’s HH attacks stopped immediately after starting treatment with ramipril (43,44). Another patient developed HH attacks after pausing enalapril for arterial hypertension. Again, the attacks stopped immediately after restart of angiotensin converting enzyme treatment (45). Arterial hypertension is a common comorbidity in elderly patients, though nocturnal episodes of hypertension might cause symptomatic HH attacks. Since angiotensin-converting-enzyme (ACE) inhibitors also have a prophylactic effect in patients suffering from migraine, there may also be other mechanisms preventing from HH (45,46).
HH associated with medication overuse headache (MOH)
In one case HH was associated with MOH. The patient was also reported to suffer from migraine and overused ergotamine. After washout therapy with amitriptyline in a daily dose of 75 mg the patient had a complete remission of HH attacks. After discontinuing amitriptyline, the HH attacks returned and stopped after restart of amitriptyline intake. Whether the remission of HH is due to the medication washout or because of the daily intake of amitriptyline remains unclear. This may be a case of primary HH associated with of MOH with and good response to amitriptyline (47).
In general, HH should be only diagnosed after other secondary causes of headache have been excluded. There might be an even wider variety than cases reported in the literature.
Diagnostics in patients with HH
Referring to the previous section, secondary causes must be ruled out when diagnosing primary HH. After anamnesis and clinical examination an MRI should be conducted to rule out vascular pathologies and a tumor. Furthermore, therapy of severe OSAS, diabetes, hypertension and other comorbidities should be optimized. HH begins mostly after the age of 50, but may also occur in children and adolescents (48). Other primary headache disorders according to the ICHD-3 must be ruled out. Current diagnostic criteria of HH are shown in Table 1.
Acute and prophylactic therapeutic options
So far, no randomized controlled studies were published about medication in HH. Consequently, recommendations for acute and prophylactic treatment are based on single case reports, larger case series and systematic reviews. Table 2 and 3 give an overview of the most frequently used acute and prophylactic agents for HH.
Prophylactic agents with best effect on HH.
Acute agents with best effect on HH.
The most common acute and prophylactic medication is caffeine. In larger case reports it was effective in approximately 70% of all patients, used as an acute treatment (3,9,49,50). For acute therapy, a cup of strong coffee or caffeine-containing analgetics can be taken when awaking with headache (10,49,51). One should keep in mind the risk of mixed analgetics for development of MOH and gastrointestinal bleeding, because most HH patients suffer from HH attacks >15 days per month. Therefore, intake of mixed analgesics should be avoided and intake of caffeine alone is recommended (19). For prophylaxis a cup of strong coffee can be consumed before going to bed (10,16). The response rate to caffeine as prophylactic agent is 54% due to a recent systematic review (3). Its analgetic effect seems to be based on an antagonism of adenosine A1, A2A and A2B receptors (52). Sleep problems should be considered as side effects of this therapy, although they occur far less than expected (10). Because of the good tolerance and effectiveness, it is one first-line agent for HH.
Lithium carbonate is another first-line prophylactic agent for HH. Reviews and long-term follow-up data showed a response rate of 73% (3,50). Commonly, lithium doses between 150 mg and 600 mg were used (intended serum level 0.5-1.0 mmol/l) (12). It can also be taken in combination with caffeine before bedtime (1,12,53). Lithium is presumed to have effects on melatonin levels, downregulate serotonin receptors in the hippocampus, and to increase serotonin release (54). Thus, lithium is effective, but one should keep an eye on common side effects like drowsiness, tremor, lack of coordination and weight gain. These often result in incompliance. Furthermore, monitoring renal and thyroid function as well as electrocardiograms (ECGs), since cardiovascular adverse effects can occur, is necessary while the patient receives lithium (55).
Indomethacin is a non-steroidal anti-inflammatory drug (NSAID) and has potent inhibitory effects on the synthesis of prostaglandins (21). Indomethacin, used as prophylactic treatment in a daily dose of 25 to 150 mg, has a response rate of 52%. In comparison, other NSAIDs showed a response rate of 13%, when used as an acute treatment (3). On the one hand indomethacin crosses the blood brain barrier better than naproxen and ibuprofen, on the other hand it inhibits nitric oxide (NO) induced dural vasodilation specifically. This might explain the greater impact of indomethacin in HH and the existence of other indomethacin-responsive headaches (21,56).
Melatonin has been successfully applied for treatment of children suffering from HH in two cases (8). However, in elderly patients it showed a response rate of approximately 50% with doses between 3 and 5 mg (3,50,57). Melatonin is a neurohormone synthesized and secreted by the pineal gland mainly during the night. Melatonin secretion is mainly regulated by the hypothalamus as circadian pacemaker. A lack of melatonin- and cortisol-rhythmicity has also been seen in patients with cluster headache but there is still no data for HH (9,58). Although lithium leads to an increase of melatonin-levels, the intake of melatonin does not seem to be as effective as lithium. In one patient melatonin intake led to aggravation of symptoms (9).
For acute treatment ergotamine derivates, opiates, 100% oxygen inhalation, triptans, paracetamol and NSAIDs showed a response rate below 25% and are consequently not effective as HH treatment (3,9,49). For prophylactic treatment, topiramate and flunarizine had response rates of approximately 40% and may be effective (3,59,60). Lamotrigine was tried in very few cases, but with good response (61,62). As well, the anti-serotonin agent oxetorone showed good response in eight cases (7).
The application of botulinum toxin type A also had a good effect in one patient suffering from HH (63). Recently, a first successful treatment of HH using greater occipital nerve block was described (64).
Conclusion
HH is a rare primary headache disorder. Due to a low level of awareness, it might often be misdiagnosed or overlooked. HH is mainly described in elderly patients over the age of 50, some case reports about children and adolescents do exist (8).
There is little known about the pathophysiology of HH so far. Nevertheless, the hypothalamus as circadian pacemaker, regulating sleep and wakefulness and being involved in pain control seems to play a key role. Supporting this hypothesis, a gray matter volume decrease in the posterior hypothalamus of patients suffering from HH has been seen in MRI-based measurements (20). Some smaller polysomnographic studies and case reports could not show any link of HH to a sleep state (24). A concrete pathomechanism is still unclear.
A huge variety of secondary causes for HH, such as brain tumors (29,30,32) or metabolic disorders have been reported (41). Those must be examined in all patients suffering from HH before diagnosing primary HH. Secondary causes of HH may also give a hint for further investigations on pathophysiology. Future studies on HH should therefore include endocrinological diagnostics and imaging of brain vessels.
Until today, therapeutic options are only evaluated in case reports, case series and reviews. There is no randomized controlled study on a medication for HH. Still, caffeine, lithium, indomethacin and melatonin show promising effects in treatment of HH (3). New treatment attempts include botulinum toxin A and greater occipital nerve block (63,64), but these need further investigation and greater patient cohorts.
Clinical implications
HH is a rare primary headache disorder, but probably often misdiagnosed or overlooked. HH is mainly described in elderly patients over the age of 50, some case reports about children and adolescents do exist. Dysfunction of the hypothalamus may play an important role in the pathophysiology of HH. Secondary headache causes have to be ruled out. Therapeutic options include caffeine, lithium, indomethacin, and melatonin.
Footnotes
Declaration of conflicting interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.