Hypnic headache in children

Abstract

Background: Hypnic headache (HH) is a rare, short-lasting headache occurring exclusively during sleep and usually affecting the elderly population. According to the ICHD-II diagnostic criteria, HH is characterized exclusively by sleep-related dull headache attacks, either lateralized or bilateral, a recurrence of >15 times per month and a persistence of pain >15 minutes after waking. No autonomic symptoms and no more than nausea, photophobia, or phonophobia are present.

Cases: We report three children between 7 and 11 years old with HH features. The characteristics of our patient’s headache, with particular reference to the nocturnal pattern, the short duration and the absence of autonomic symptoms, lead us to consider the diagnosis of HH.

Conclusions: Considering the very few cases of HH reported in paediatric age, our cases may expand the clinical spectrum of this disorder, suggesting a possible revision of the diagnostic criteria, with particular regard to the developmental age.

Introduction

Headache is a common presenting somatic complaint in the paediatric age. Paediatric tension-type headache and migraine are being recognized increasingly. Other primary headaches appear to be extremely rare and often remain under-recognized (1), and poorly diagnosed.

Hypnic headache (HH) is a rare sleep-associated primary headache, usually affecting the elderly population (2,3). First described by Raskin in 1988 (4), this primary headache disorder was included in the 2004 International Classification of Headache Disorders, 2nd edition (ICHD-II). According to the ICHD-II diagnostic criteria, HH is defined by attacks of dull headache that typically occur after the age of 50 years, more than 15 times per month, and occur exclusively during sleep. No autonomic symptoms are present and no more than one case of nausea, photophobia or phonophobia (5). The attacks wake the patient during sleep and usually last more than 15 minutes after waking (Table 1). Because of its typical onset after 50 years of age, HH has been extensively studied and is well defined in the adult population (2–4). However in recent years, Scagni et al. and Grosberg et al. described two children complaining of sleep-related dull headache attacks but the attack duration and headache frequency didn’t fully meet the ICHD-II criteria (6,7). We report three children, followed in two different Italian Headache Centres, with typical HH features.

Table 1.

ICHD-II diagnostic criteria for hypnic headache

A. Dull headache fulfilling criteria B–D

B. Develops only during sleep, and awakens patients

C. At least two of the following characteristics:

1. occurs > 15 times per month

2. lasts ≥ 15 min after waking

3. first occurs after age of 50 years

D. No autonomic symptoms and no more than one of nausea or photo/phonophobia

E. Not attributed to another disorder

Case 1

A 7-year-old right-handed boy was referred to our Headache Centre because he complained of headache attacks occurring only during sleep for 6 months (twice a month, each episode lasting from 15 to 30 minutes).

Psychomotor development and neurological examinations did not suggest any neurological damage. The boy had a positive family history of migraine without aura (mother). Headache episodes occurred only during nocturnal sleep and usually began 1–3 hours after he fell asleep, awakening him. The pain was described as ‘tension-type’ headache. Pain was bilateral, localized to the frontal-temporal regions, never unilateral, and the intensity was moderate. No vomiting, photophobia, phonophobia or nausea were associated with attacks. There were no cranial autonomic symptoms (lacrimation, nasal congestion, miosis, ptosis or conjunctival injection). Attacks were short-lasting (ranging from 20 to 30 minutes) and they occurred twice per month. He did not take any medication and after these sporadic nocturnal attacks, the patient was able to return to sleep. Brain MRI and EEG were normal, as well as blood tests (biochemical, haematological, liver and kidney routine tests and metabolic investigations). Considering the headache characteristics and the rare and very short nocturnal episodes described, no prophylactic treatment was started. Up to now, headache severity and frequency have showed no changes or modifications for 12 months.

Case 2

An 11-year-old right-handed boy affected by thalassemia major was referred to our Headache Centre because he complained of headache attacks occurring during sleep for 1 month. Six months before the patient had undergone a bone marrow transplantation. His drug therapy included cyclosporine, methylprednisolone, amoxicillin, fluconazole and acyclovir. The headaches occurred almost every night, sometimes with numerous attacks in the same night. Usually the dull attacks began 1 hour after he fell asleep, awakening him from sleep. Subsequent awakenings for the headache occurred between 01.00 and 02.00 hours, 04.00 and 05.00 hours. The pain was described as a pressing headache (“pushing as a stone on my head”). Pain was bilateral, localized to the frontal-temporal regions, rarely unilateral, and the intensity was moderate to severe. No vomiting, photophobia or phonophobia were associated with attacks, but he occasionally experienced nausea. There were no cranial autonomic symptoms. Attacks were short-lasting (ranging from 10 to 20 minutes). He had not taken any medication and after each attack, the patient was able to return to sleep. Psychomotor development and neurological examinations were normal. He had only a mild hypotrophy of the calf and leg muscles. The boy had a negative family history of headache. Brain MRI and EEG were normal. Blood tests (biochemical, haematological, liver and kidney tests and metabolic investigations) showed normal results but GPT 104 UI/l (5–31), blood iron 144 mg/dl (35–130) and ferritin 1928 ng/ml (13–150). A prophylactic treatment with melatonin was started, at an initial dose of 2 mg at bedtime. Headache severity decreased from severe to moderate and the attack frequency immediately showed a reduction (from 2–3 attacks/night to 1 attack/night and from 25–30 attacks/month to 10–15 attacks/month). The dose was increased to 4 mg, which rendered him headache-free over a 6-month period.

Case 3

A 10-year-old right-handed girl with negative family history for primary and secondary headaches was referred to our Headache Centre because she complained of headache attacks occurring during sleep for 10 months before admission. Psychomotor development and neurological examinations were normal. At onset, the attack frequency was “sporadic”, showing a severe increase in the previous 5 months (from 10 to 15 nocturnal episodes/month) usually beginning 1–2 hours after she fell asleep. The pain was described as a “pulsating headache”, localized bilaterally to the frontal-temporal regions, never unilateral, and the intensity was moderate to severe. No vomiting, photophobia or phonophobia were associated with attacks, but she occasionally experienced nausea. There were no cranial autonomic symptoms. Attacks were short-lasting (ranging from 10 to 30 minutes) and they occurred 10–15 times per month and, rarely, 2–3 times per night. Before being admitted to our Child Neurology Outpatient Service she had not taken any medication and after each of the attacks, the patient was able to return to sleep. Brain MRI and EEG were normal. Routine blood tests (biochemical, haematological, liver, kidney and metabolic investigations) yielded normal results.

Considering the headache characteristics and the clinical history, a prophylactic treatment with melatonin was started, at an initial dose of 3 mg at bedtime. Headache severity decreased from severe to moderate and the attack frequency immediately showed a reduction, decreasing from 12–15 attacks per month to 1–2 attacks per month. Recently, during the last two months, she reached a headache-free condition.

Discussion

Our study describes three cases of HH in paediatric age. The nocturnal pattern, the short duration of attacks without autonomic symptoms and the absence of nausea, photophobia and phonophobia are similar to those found in typical adult HH (2,5). The main differential diagnosis of HH includes secondary headache and primary headache, such as trigeminal autonomic cephalalgias (TACs), migraine and tension-type headache. In our patients, the bilateral pain localized to the frontal or frontal-temporal regions without signs of autonomic involvement excludes a diagnosis of TACs. No nausea, photophobia or phonophobia were present, so we could exclude a migraine diagnosis. No trigger factors were detected. Though the diagnosis of tension-type headache cannot be excluded (bilateral pain location), the close temporal relation of attacks to nocturnal sleep and the short duration make this disorder most unlikely in our patients. Though rare, the possibility of a secondary cause of HH must also be considered but a normal neurological examination and imaging findings of our children also exclude this diagnosis. Although HH is commonly considered as a headache disorder of the elderly, the recent original descriptions suggest that the age spectrum of HH could be wider than thought to date (6,7). In line with the previous paediatric cases of HH reported, the patients in this study showed some atypical features, not fulfilling the ICHD-II criteria. The attack duration and the headache frequency were variable in all patients (Table 2). In the developmental age, it is not uncommon that younger children with primary headache show some atypical features that do not fully meet the ICHD-II criteria. Despite attack duration and frequency (not always fitting with current criteria for HH), in our children the exclusively sleep-related dull headache attacks, bilateral severe pain and absence of autonomic symptoms were strongly suggestive of HH diagnosis. In the patient transplanted for thalassemia major, it is unlikely that the headache represents a neurological complication induced by cyclosporine. Cyclosporine-related neurotoxicity in bone marrow transplantation patients usually appears within 1 month of treatment onset, as a prodromal sign that precedes the severe cyclosporine-related neurotoxicity, such as coma and seizures. Also brain MRI often shows abnormalities (8). Our patient began to complain of headache 5 months after transplantation, so the clinical history of our patient does not imply any of these clinical features.

Table 2.

Clinical characteristics of previously described children and our paediatric patients with hypnic headache

Case report	Scagni	Grosberg	Case 1	Case 2	Case 3
Age (years)/gender	8/F*	9/F*	7/M*	11/M*	10/F*
Sleep-related headache	+	+	+	+	+
Attacks/month	1	8–12	2	20–25 (2–3/night)	10–15 (2–3/night)
Attack duration (min)	30–60	30	15–30	15–20	10–30
Autonomic symptoms	–	–	–	–	–
Nausea	–	–	–	+/−	+
Photophobia	–	–	–	–	–
Phonophobia	–	–	–	–	–

*

M = male; F = female.

Little is known about the natural history of HH in the paediatric population (9). The history of Scagni’s patient was characterized by persistent headache over 5 years, in contrast with a spontaneous 2-month remission reported by Grosberg et al. Patients in our study experienced a chronic headache course ranging from 2 to 12 months.

Headache episodes are moderately responsive to acetylsalicylic acid. Lithium has a good efficacy as a prophylactic agent (10). Caffeine appears to be a well-tolerated treatment option to consider because it does not disrupt nocturnal sleep (11). Cases 2 and 3 benefit from melatonin, rendering them quickly headache free.

The pathophysiology of HH is still unknown. The close temporal relation of attacks to nocturnal sleep, mainly during REM phases, suggests that HH could be a primary headache disorder of chronobiological origin (4,12,). In conclusion, the characteristics of our patients’ headaches, with particular reference to the exclusive relationship between sleep and headache, lead us to consider the diagnosis of HH. The attack duration and frequency of our patients did not completely meet the ICHD-II criteria, as observed in the previous paediatric cases reported.

Another interesting speculative consideration could be the lack of reported cases between paediatric age and individuals over 50; in this respect, we have to ask ourselves whether our described cases are truly cases of paediatric HH (especially considering the ‘clinical picture’ atypicality) and, moreover, whether we should carefully search for misdiagnosed HH cases among adults between 18 and 50 years of age.

In any case, new paediatric cases in the future could expand the clinical spectrum of HH, stimulating a possible revision of diagnostic criteria with particular regard to paediatric age.