Headache Horizons: Recognition, Diagnosis, and Treatment of Hypnic Headache

Clinical Presentation

CB, age 71, who has no previous headache or migraine history, reports having experienced headaches at night for 5 months. CB describes being awoken on most nights between 2 am and 4 am, typically with a bilateral headache of moderate intensity without associated features. The headaches can last up to a couple of hours. Taking an over-the-counter aspirin/acetaminophen/caffeine combination upon awakening provides some relief. CB has experienced some similar headache attacks during the day, often just before going to bed. Medical history includes treated hypertension.

Clinical Features

Hypnic headache, sometimes referred to as alarm clock headache, is an uncommon and relatively recently described syndrome (1988)¹ of nocturnal, short-lived, mostly bilateral headache. By definition, the headache begins during sleep, awakening the individual. The diagnosis remains clinical, without corroborative testing. The rarity of hypnic headache precludes an analysis of epidemiologic data on prevalence or natural history. Only 2 of 921 participants answering a headache questionnaire in a study from Iceland had probable hypnic headache.² This rare headache type would be expected to account for fewer than 1% of specialist headache medicine clinic referrals.

Nocturnal headache does not in itself indicate a diagnosis of hypnic headache. Headache at night can be attributable to a variety of causes. Nocturnal headache may be triggered by sleep itself, prone body orientation, time of day, withdrawal from or ingestion of medications, or severe hypertension, among other causes. Practitioners must assess for the differential diagnosis of headache exclusively or predominantly at night. More common nocturnal headache diagnoses include cluster headache in those with unilateral headache, migraine in younger individuals, cervicogenic headache in the older population, and substance-related headaches (eg, analgesic withdrawal, alcohol-triggered headache).

Hypnic headache, like cluster headache, may be triggered by naps during the day. Daytime headache does not exclude the diagnosis of hypnic headache, but headache attacks while awake should not ordinarily be referred to as hypnic headache attacks. Exploration of attacks while asleep in a chair at night or asleep during the day may help reveal triggering factors in unclear cases.

Careful history-taking and attention to detail help lead to the correct diagnosis. One particular diagnostic challenge for clinicians is misdiagnosing cluster headache in a minority of individuals with hypnic headache who are experiencing unilateral headache, especially those with some unilateral autonomic features. Differentiating hypnic headache from cluster headache may be difficult, owing to phenotypic overlap. These hybrid cases characteristically result in attacks that are not severe in pain intensity (as is the case with cluster headache), suggesting that a form of hypnic headache may be the best classification for them. Although the length of hypnic headache attacks is typically a few hours, longer attacks up to 8 hours have been described.

Among 250 individuals with hypnic headache, Liang and Wang³ reported lacrimation in 6.3%, ptosis in 2.4%, and rhinorrhea or nasal congestion in 7.7%. Spontaneous remission can occur. Because of the rarity of hypnic headache, information on natural history is limited.

There is no consistent relationship between hypnic headache and sleep cycle. A study by Pinessi et al.⁴ suggested that hypnic headache attacks occur during REM sleep; however, a later study of 37 individuals showed that the majority experienced attacks during non-REM sleep.⁵

The International Classification of Headache Disorders, third edition (ICHD-3)⁶ defines hypnic headache as frequently recurring headache attacks developing only during sleep, causing wakening and lasting for up to 4 hours, without characteristic associated symptoms and not attributed to other pathology. Criteria for ICHD-3 hypnic headache diagnosis are as follows:⁶

A Recurrent headache attacks fulfilling criteria B through E

B Developing only during sleep, and causing wakening

C Occurring on ≥10 days/month for >3 months

D Lasting from 15 minutes to up to 4 hours after waking

E No cranial autonomic symptoms or restlessness

F Not better accounted for by another ICHD-3 diagnosis

ICHD-3 also has provision for probable hypnic headache when not all criteria are fulfilled.⁶

Experience over the past 30 years has revealed the following features of hypnic headache:

• More common in adults older than 50 (although children with hypnic headache have been described⁷)
• Higher prevalence in women (in one study of 25 individuals, 80% were female; in a meta-analysis of published cases [343 individuals], 69% were female^8,9)
• Usually only 1 attack per night
• Attacks last 15 to 240 minutes
• Intensity is mild or moderate in two-thirds of attacks; severe attacks are less frequent
• Autonomic features in fewer than 10%

The most challenging differential diagnosis is between unilateral hypnic headache (especially unilateral hypnic headache with autonomic features, which is best understood as a distinct subset of hypnic headache) and cluster headache.

Differentiating Unilateral Hypnic Headache from Cluster Headache

• Hypnic headache is less severe; severe attacks, if present, are a minority, and the preponderance of attacks have moderate intensity.
• Hypnic headache tends to be of longer duration (64% last more than 2 hours).
• Hypnic headache attacks occur later at night (more than 50% occur from 4 am to 6 am).
• Hypnic headache may be bilateral in some attacks.
• A smaller number of attacks occur per night with hypnic headache (94% of individuals experience 1 or 2 attacks per night).
• Hypnic headache starts at an older age.
• Most cluster headache is episodic and disappears for months, whereas hypnic headache generally is persistent.

Treatment

Because of the rarity of hypnic headache, evidence-based data to guide treatment are lacking. Case reports and case series are needed to generate hypotheses on what treatments might be useful. People who are awoken at night with hypnic headache often report taking over-the-counter analgesics, with partial effect. The original 1988 report¹ on hypnic headache described 6 individuals (5 men), age 65 or older, who experienced single episodes of bilateral headache at night without autonomic features, recovering in 30 to 60 minutes without sequalae. All individuals responded to lithium, up to 600 mg nightly. Use of amitriptyline and propranolol provided no relief in any of the individuals.¹

A publication bias exists for cases that improve with treatment, compared with those that do not exhibit treatment response. Therefore, such case reports may not translate well into clinical practice, but nevertheless must serve as a basis for treatment approach. Silva-Néto and colleagues⁸ published a review of 348 cases of hypnic headache, including an assessment of medication response. Outcomes were classified as no, partial, or good response. There was a 73.4% response rate for lithium, 54% for nighttime caffeine, 51% for indomethacin (which also has been reported to be effective in a case report of an adolescent with hypnic headache¹⁰), 50% for melatonin, 41% for topiramate, and 40% for flunarizine. Other medications had such a small number of reports as to not warrant comment.

A 2016 study⁹ from a headache clinic reported on 40 cases of hypnic headache, noting that 80% of individuals were female and 40% had unilateral attacks. Autonomic features were found in only 2 individuals and were bilateral. The most effective treatments were lithium and caffeine. Of 10 individuals who tried lithium, 7 found it to be completely effective, 2 moderately effective, and 1 partially effective. Of the 21 participants who tried caffeine, 6 found it to be completely effective, 9 moderately effective, and 4 partially effective. Six participants tried melatonin, with 2 finding it to be completely effective, 1 moderately effective, and 1 partially effective.

Treatment should begin with melatonin and caffeine at night before bed. If this treatment is not effective, lithium may be attempted. Lithium necessitates more oversight, often leading to initial reluctance; however, given the potential usefulness of lithium as well as the paucity of options, it should be strongly considered.

Pathophysiology

The pathophysiologic underpinnings of hypnic headache remain unknown. Hypnic headache is a brain disorder, with some evidence for involvement of the hypothalamus.¹¹ Hypnic headache shares a periodicity feature with cluster headache, which has been associated with hypothalamic abnormalities.

Summary

Hypnic headache is a rare disorder diagnosed on the basis of clinical features of nighttime headache attacks predominantly in older individuals. The diagnosis is made after careful review for secondary headache disorders, such as cervicogenic headache, migraine, and analgesic withdrawal. Hypnic headache tends to be bilateral and of moderate severity. A subset of hypnic headache with cluster headache characteristics, such as unilateral pain and unilateral autonomic features, exists; however, attack intensity remains mild to moderate for the majority of attacks. Evidence-based data to help guide therapy are lacking. Treatment should start with melatonin and caffeine taken at night. Despite the oversight required for the use of lithium, a lithium trial may be attempted if melatonin and caffeine fail. Unfamiliarity with the diagnosis leads to diagnostic delay.