Errata: Erratum Volume 52, Issue 8, 784, Article first published online: 8 July 2010
Acknowledgements I would like to acknowledge the late Dr Stuart H Green MA FRCP, Consultant Paediatric Neurologist and Senior Lecturer in Paediatrics and Child Health, Birmingham Children’s Hospital, for his valuable advice in the preparation of this manuscript.
To my knowledge trigeminal autonomic cephalgias (TACs) have not previously been reported in infancy. The diagnosis is dependent on an accurate history, including parents noting any physical signs at the time of the episode. Obtaining a clear history can be challenging when such symptoms occur in preverbal children. Similarly, physical signs, being transient, may have resolved by the time the parents take the child to a doctor. In addition, the investigations may also be normal. In such circumstances, taking a photograph during an episode can confirm the diagnosis. I describe a case of probable trigeminal autonomic cephalgia starting in a 3-month-old male infant who presented with screaming episodes associated with characteristic changes seen on his face. Investigations, including cranial magnetic resonance imaging, electroencephalography, and urinary catecholamines, were normal. The diagnosis was confirmed from a photograph taken by the parents at the time of the attack. As the condition is very rare in young children, there is little information available in the literature on using treatment for prophylaxis or for aborting acute episodes in this age group.
A 3-month-old healthy, male, Caucasian infant developed sudden episodes of screaming during which the parents noted reddening of half of the face accompanied by closing and lacrimation in the eye and watering of the nostril on the same side. The episodes were usually brief and lasted about from 5 to 10 minutes, followed by complete spontaneous recovery; rarely, the duration was up to 20 minutes. The episodes occurred about once a month, usually on the same side of the face (rarely on the opposite side), but never on both sides during the same episode. Occasionally, there were two episodes about 12 hours apart and, rarely, a third one some hours later. On these occasions the infant was irritable on the day between the episodes. The infant was completely well between episodes, as was the case when his mother took him to the general practitioner and when he was seen at two local hospitals. There was no identified trigger for these episodes. The parents had given him paracetamol, which did not provide relief. He was subsequently referred for a paediatric neurology opinion at the age of 15 months. He was born at term by normal delivery. There were no neonatal problems, he was feeding well, and his height, weight, and head circumference were just above the 97th centile. Physical examination, including neurological examination and specialist ophthalmological examination, was normal. His parents and two siblings were all well.
Investigations, including cranial magnetic resonance imaging (MRI), electroencephalography, and urinary catecholamines, were normal. The parents were asked to take a photograph of the infant during an attack (Fig. 1), which showed right-sided flushing of the face, conjunctival redness, drooping of the eyelid, and discharge from the nostril: typical manifestations of trigeminal autonomic cephalgias. By the time the diagnosis was made the child was 17 months old. The attacks spontaneously became less frequent and severe at 2 years and ceased at the age of 3 years.
Consent for publication of this case report has been obtained from the parent of the child.
The trigeminal autonomic cephalgias (TACs) are a group of syndromes defined by the International Classification of Headache Disorders, 2nd edition (ICHD-II).1 They share the clinical features of headache and prominent cranial trigeminal parasympathetic autonomic features. They are rare2 in childhood. To meet the diagnostic criteria for TACs as specified in the ICHD-II, in addition to having the characteristic symptoms and signs of TACs, the history, neurological examination, and diagnostic tests must not suggest another classifiable disorder. The conditions included under TACs are (1) cluster headache, (2) paroxysmal hemicrania, (3) short-lasting unilateral neuralgiform headache with conjunctival injection and lacrimation, and (4) probable TAC (when all the diagnostic criteria of the first three are not present).
Cluster headache1 is described as severe unilateral orbital, supraorbital, or temporal pain lasting between 15 and 180 minutes if untreated. The attacks are associated with one or more of the following, all of which are unilateral: conjunctival injection, lacrimation, nasal congestion, rhinorrhea, forehead and facial sweating, meiosis, ptosis, and eyelid oedema. Most individuals are restless or agitated during an attack. Typically, the age at onset of cluster headache is 20 to 40 years. The prevalence is 3 to 4 times higher in males. There are a few reports of cluster headache-like disorders3 in older children, and the diagnosis is often delayed. It is important to perform cranial MRI to exclude serious intracranial pathology, including tumour or arteriovenous malformation pressing on the trigeminal nerve.
The signs and symptoms of paroxysmal hemicrania1,4,5 are similar to those of cluster headache, but the attacks are more frequent and are of shorter duration. They may occur in an episodic or chronic pattern. The non-steroidal anti-inflammatory drug indomethacin produces relief of symptoms.
Short-lasting unilateral neuralgiform headache with conjunctival injection and lacrimation1,5,6 most commonly presents in older males and is characterized by 3 to 200 short-lasting attacks of unilateral pain in a day, of much briefer duration than occurring in any of the other TACs, lasting from 5 seconds to 4 minutes, and very often accompanied by marked autonomic features.
The screaming attacks in this infant would seem to suggest pain and were clearly associated with trigeminal parasympathetic autonomic features, as illustrated in the photograph (Fig. 1). It is reasonable to speculate that this infant was suffering from either cluster headaches or paroxysmal hemicrania, as the attacks were too long for short-lasting unilateral neuralgiform headache with conjunctival injection and lacrimation. According to the clinical features described in the ICHD-II, this case seems to fit under the subclassification of probable TAC (ICHD-II subheading 3.4). The infant was too young to be able to verbalize the site and duration of pain to fulfil all the criteria for a more accurate diagnosis. The attacks did not truly cluster, as in cluster headache, and did not strictly follow an episodic or chronic pattern. In addition, no seasonal variation was observed.
The autonomic features in cluster headache are thought to result from an abnormality in the region of the posterior inferior part of the hypothalamus. Although the symptoms are not life threatening, they can be extremely debilitating. The use of medications in children with this condition is empirical, and therefore making a recommendation for this age group is difficult.
Oral medications taken after the headache starts are not useful for brief infrequent episodes because the headache peaks before the medication takes effect. Inhaled or injectable medications used to provide pain relief7 include 100% oxygen, sumatriptan, and lidocaine. Prophylactic drugs, including calcium-channel blockers, lithium, and corticosteroids, are sometimes used in adults, and they appear to help to reduce the frequency and severity of attacks in a proportion of individuals. It is difficult to draw conclusions on their effectiveness in children from the few cases reported.3 In a 3-year-old child8 with chronic paroxysmal hemicrania, indomethacin used prophylactically was effective. In adults with intractable chronic cluster headaches unresponsive to medical treatment, deep brain stimulation of the hypothalamus9 has been used with success. In this child, the options for using home oxygen and prophylactic drugs were discussed with the parents. However, treatment was not instituted because the attacks were brief and they spontaneously became less frequent over time.
There was some delay in making a diagnosis in this child because the age at onset was very early and the attacks were infrequent and of short duration, which means that the physical signs were not present when seen in clinical settings. Although the parents very astutely noted the associated features during the screaming attacks, there was reluctance to make the diagnosis of probable TAC at such a young age without independently verifying parental observations. Taking a photograph during an attack was invaluable in making a diagnosis. The ease with which this can be done, for example with a mobile phone camera, provides parents and clinicians with an important diagnostic tool for visual signs that are episodic and should be encouraged.