Clinical Reasoning: A child with pulsatile headache and vomiting
Laurence Morin, MD,
Assia Smail, MD,
Jean-Christophe Mercier, MD, PhD and
Luigi Titomanlio, MD, PhD
From the Departments of Pediatric Emergency Care (L.M., A.S., J.-C.M., L.T.) and Pediatric Neurology (L.T.), APHP-University Hospital R. Debré, Paris, France.
Address correspondence and reprint requests to Dr. Luigi Titomanlio, Pediatric Emergency Dept., Robert Debré University Hospital, 48, Bld. Sérurier, 75019 Paris, France luigi.titomanlio{at}rdb.aphp.fr.
A child was born to nonconsanguineous, healthy parents. Pregnancyand delivery were uneventful, and psychomotor development wasnormal. At the age of 6 years and 10 months, he was admittedto a local hospital because of vomiting and nonfebrile unilateralheadache. Neurologic examination had normal results. Blood tests(complete blood count, C-reactive protein, electrolytes, bloodurea nitrogen, creatinine, glucose, serum bicarbonate and pH,anion gap, transaminase, and urine culture) were within normallimits. Abdominal x-ray and abdominal ultrasound imaging hadnormal results. Based on these results and on clinical observation,common medical and surgical causes (viral illness, gastroenteritis,diabetes, intestinal obstruction) were ruled out. Head CT scanhad normal results. The EEG showed some irregular activity inthe occipital regions, with rare sharp waves, more prevalenton the right side. One week later, a further awake EEG was performedand had normal results. A presumptive diagnosis of migrainewith aura was made after 2 months by a pediatric neurologistbecause of several episodes of unilateral pulsatile headacheand vomiting (one to two episodes per week). The episodes werepreceded by a sensation of sickness, and lasted about 5–10minutes each. Pallor, poorly defined abnormal ocular movements,and transitory unresponsiveness were also reported by his parents.After the episode, the child asked to sleep. Acetaminophen andibuprofen were prescribed to control symptoms.
Five months later, the patient was brought to the EmergencyDepartment of our hospital because of recurrent and long-lastingepisodes of headache beginning the same day. He had four episodesof nausea, vomiting, pallor, and unilateral (right-sided orleft-sided) pulsatile headache, each one lasting from 5 to morethan 30 minutes. The prescribed treatment was ineffective, andthe child was considered to be in a migraine aura status byhis pediatrician.
A critical episode was observed during clinical examination:the child reported a sudden feeling of sickness and a severeunilateral pulsatile headache, followed by nausea. Left eyelidmyoclonus followed, and the child described a short-lastingsensation of blindness. Then his head turned toward the rightand he became unresponsive for about 20 seconds. Soon after,he vomited and became bradycardic (sinus rhythm, 35–40bpm).
Questions for consideration:
What is the differential diagnosis?
What features of the historyhelp make certain entities moreor less likely?
The differential diagnosis in children presenting with pulsatileheadache and vomiting, sensation of sickness, pallor, or otherautonomic symptoms includes emergent etiologies such as intracranialmass (tumor, bleed, infection) and encephalitis, and non-emergentdiseases such as migraine (mainly basilar migraine), gastroenteritis,vagal syncope, cyclic vomiting syndrome, intoxication, and partialseizures (occipital or temporal lobe epilepsy). Other rare etiologiesto consider are vascular syndromes (Klippel-Trenaunay-Weber,arteriovenous malformations of the brain), familial dysautonomia(e.g., Riley-Day syndrome), breath-holding spells of early infancyprogressing to isolated syncope, postural orthostatic tachycardiasyndrome (POTS), and metabolic diseases. This child showed prolongedand severe autonomic symptoms (nausea, vomiting, pallor, bradycardia)that are mainly due to acute cerebral insults, but can alsobe diagnosed as status migrainosus or autonomic status epilepticus.In his personal history, we can identify shorter but similarepisodes, suggesting that the two latter hypotheses are mostlikely correct. Migraine and epilepsy are highly comorbid conditionsthat may share the same pathophysiology, but the nature of theirassociation is unclear. Our case does not fulfill the diagnosticcriteria for migraine with aura of the International Classificationof Headache Disorders, 2nd edition (ICHD-II).1 The aura, whichcan be visual, sensory, or dysphasic, is the consequence offocal cerebral dysfunction that immediately precedes or coincideswith the headache onset. In our patient, autonomic symptomscould be related to a basilar-type migraine rather than to anaura. Differential diagnosis between seizure and migraine couldbe complicated by the presence of headache in both. Seizurescan be followed by postictal headache (headache attributed toseizure, ICHD-II 7.6) and can also occur during or within 1hour of a typical migraine aura attack (migraine-triggered seizure,ICHD-II 1.5.5). In migraineurs, interictal EEG findings areusually normal, although various abnormalities, including mainlydiffuse slowing, have been reported. Sharp waves, observed inour patient, are not seen in migraine.
A clinical diagnosis of autonomic status epilepticus was made,and a rectal dose of 0.5 mg/kg of diazepam was administered,stopping the episode. The diagnosis of autonomic seizures issuggested by the episodic recurrence of unexplained vomitingor abdominal pain, migraine, or other autonomic symptoms, withEEG showing focal seizure activity.
The child fulfills the clinical and likely the EEG criteriafor Panayiotopoulos syndrome (PS), a form of benign focal epilepsyof early childhood: several nonfebrile occipital seizures, occipitalspike-wave activity at EEG (clinical history), absence of knownetiologic factors, normal psychomotor development, and benignclinical evolution under treatment (when prescribed). PS isa common, benign, and idiopathic childhood autonomic epilepsythat has recently been incorporated into the international classificationof epileptic syndromes.2 Of children aged 1 to 15 years whohave had one or more nonfebrile seizures, PS affects approximately6%, and 13% of children aged 3 to 6 years. Age at onset is between1 and 14 years with a peak between 4 and 5 years. Crises arefocal, initially characterized by a complaint from the childof not feeling well, followed by autonomic signs or symptomsfrequently characterized by emetic symptoms (nausea, retching,vomiting), paleness (or, less often, cyanosis or facial blushing),mydriasis (or, less often, miosis), coughing, hypersalivation,urinary and fecal incontinence, and cardiorespiratory and thermoregulatoryalterations.3 In nearly all critical episodes, consciousnessis initially intact. During seizure evolution, the child canbecome flaccid and unresponsive in 20% of cases (ictal syncope),with tonic eye and head deviation. Headache is often concurrentwith other autonomic symptoms. Speech arrest, visual hallucinations,oropharyngolaryngeal movements, and behavioral disturbancesoccur less frequently. Autonomic seizures in PS are frequentlyprolonged, more than 30 minutes in nearly half of cases (autonomicstatus epilepticus).4
In PS, the neuroanatomic and neurophysiologic pathways involvedin the generation of autonomic signs are unknown. Usually, autonomicmanifestations are generated by activation or inhibition ofparts of the central autonomic network that involves the insularcortex, medial prefrontal cortex, amygdala, hypothalamus, andventrolateral medulla. In PS, the epileptogenic zone is wideand bilateral. Therefore, ictal discharges may easily activatethe lower threshold autonomic centers. In children with PS,autonomic manifestations also may be attributed to a maturation-relatedsusceptibility of the central autonomic network.5 Seizures remainpurely autonomic if ictal neuronal activation of non-autonomiccortical areas fails to reach the threshold to produce othersymptoms; otherwise autonomic and localization-related corticalsymptoms occur.
The child had a complete recovery and was kept under medicalsupervision for 1 day. No more episodes occurred.
Questions for consideration:
What testing would you obtain at this point to confirm the diagnosis?
What is the prognosis for this patient?
Would you prescribea treatment, and, if yes, which one?
Interictal EEG testing was repeated to confirm the EEG criteriafor PS, and it showed independent bilateral occipital spike-wavecomplexes. Brain MRI had normal results. Antiepileptic therapywas started (valproic acid, 20 mg/kg/day). At 8 years and 4months of age, he remained symptom-free.
An awake and asleep EEG is the only investigation that commonlyshows abnormal results (90% of cases). The epileptiform activityis characterized by spikes or spike-wave complexes of greatamplitude, with multifocal localization predominating in theposterior regions.6 Interictal EEG findings show intraindividualvariability.5 High-resolution brain MRI has normal results.The overall prognosis of PS is excellent, with remission usuallyoccurring within 1 or 2 years after onset. Children have normalphysical and neuropsychological development and the risk ofepilepsy in adult life appears no higher than in the generalpopulation. Treatment might not be necessary because in one-thirdof cases there is only a single seizure, but benzodiazepines,administered by IV, rectal, or buccal preparations, are commonlyused to terminate autonomic status epilepticus. Whereas PS isbenign in terms of long-term evolution, autonomic seizures canbe associated with potentially life-threatening cardiorespiratoryarrest and death.7 To date, therapeutic management of PS andautonomic status epilepticus is based on consensus. There isno evidence of the superiority of any antiepileptic drug, andcarbamazepine and valproic acid are both widely used.8 If antiepileptictreatment is started, it is suggested to consider its withdrawalwithin 2 years. Because of autonomic status epilepticus andbradycardia in our patient, valproic acid therapy was startedand symptoms resolved completely.
Diagnosis of autonomic status epilepticus can be difficult,especially if this possibility is not considered by the clinicianin an emergency setting. Most general practitioners and pediatriciansare not familiar with the notion that prominent autonomic symptomsand signs may occur as epileptic seizure manifestations of occipitalorigin. As a consequence, this diagnosis can be easily missedand have potentially life-threatening sequelae.9 Detecting keysymptoms usually associated with PS may prevent erroneous diagnoses,shorten the length of clinical observation, and prevent unnecessaryinvestigations. Early recognition of PS can also provide rapidreassurance to families in situations that can be very alarming.
Cardiovascular changes in PS have received the most attention,probably because of their possible contribution to sudden deathin these patients.7 The association between seizures and heartrate changes has already been documented in several studies,tachyarrhythmias being more common than bradyarrhythmias. Ictalbradycardia is seen primarily in association with focal seizures,particularly involving the temporal and limbic lobes.10 Conversely,there are very few cases of ictal cardiorespiratory arrest reportedin PS7; therefore, its best management is unclear. In our case,an intrarectal dose of diazepam was rapidly effective in normalizingheart rate, possibly preventing a cardiorespiratory arrest,with all its consequences.
PS often eludes diagnosis as it presents with manifestationsthat could be overlooked and symptoms that are frequently mistakenfor more common childhood disorders (migraine in the presentcase). The risk of cardiorespiratory arrest in PS should beknown by practitioners in clinical emergency medicine. Somechildren have been reported to be resuscitated, intubated, andmechanically ventilated as a consequence of PS attacks. Ourcase illustrates the efficacy of an intrarectal dose of diazepamin case of ictal bradycardia, possibly preventing a cardiorespiratoryarrest. Although more studies are needed on the subject, supportivefamily management should also include specific education aboutautonomic status epilepticus symptoms.
International Classification of Headache Disorders, 2nd edition. Cephalalgia 2004;24 suppl 1:9–160.[Medline]
Covanis A, Panayiotopoulos CP. Improving the diagnostic yield in Panayiotopoulos syndrome. Eur J Neurol 2008;15:317–319.[Medline]
Panayiotopoulos CP. Autonomic seizures and autonomic status epilepticus peculiar to childhood: diagnosis and management. Epilepsy Behav 2004;5:286–295.[Medline]
Dura-Trave T, Yoldi-Petri ME, Gallinas-Victoriano F. Panayiotopoulos syndrome: epidemiological and clinical characteristics and outcome. Eur J Neurol 2008;15:336–341.[Medline]
Panayiotopoulos CP, Michael M, Sanders S, et al. Benign childhood focal epilepsies: assessment of established and newly recognized syndromes. Brain 2008;131:2264–2286.[Abstract/Free Full Text]
Ferrie C, Caraballo R, Covanis A, et al. Panayiotopoulos syndrome: a consensus view. Dev Med Child Neurol 2006;48:236–240.[Medline]
Verrotti A, Salladini C, Trotta D, et al. Ictal cardiorespiratory arrest in Panayiotopoulos syndrome. Neurology 2005;64:1816–1817.[Free Full Text]
Caraballo R, Cersosimo R, Medina C, Fejerman N. Panayiotopoulos-type benign childhood occipital epilepsy: a prospective study. Neurology 2000;55:1096–1100.[Abstract/Free Full Text]
Covanis A. Panayiotopoulos syndrome: a benign childhood autonomic epilepsy frequently imitating encephalitis, syncope, migraine, sleep disorder, or gastroenteritis. Pediatrics 2006;118:e1237–1243.[Abstract/Free Full Text]
Schuele SU, Bermeo AC, Alexopoulos AV, et al. Video-electrographic and clinical features in patients with ictal asystole. Neurology 2007;69:434–441.[Abstract/Free Full Text]
Top.
SECTION 1
SECTION 2
