Article Text

Alternating hemiplegia of childhood or Hashimoto’s encephalopathy?
  1. Institute of Clinical Paediatrics, University of Siena, Siena, Italy
  1. Dr Paolo Balestri, Institute of Clinical Paediatrics, Viale Bracci, Le Scotte, 53100 Siena, Italy. Telehone 0039 577 586522; fax 0039 577 586143; email balestri{at}

Statistics from

A healthy 14 year old boy had an episode of fever (38.5°C) for 1 day followed, 2 days later, by a progressively worsening confusional state, with slurring of speech, dizziness, and unsteady gait lasting 5 hours. Three months later he had a tonic attack followed by a prolonged postcritical coma which lasted 3 days. On both occasions haematological and biochemical blood tests, brain CT and MRI, CSF examination, and EEG were normal. A few days later, he had another tonic attack of 1 minute, with a confusional state, nominal dysphasia, and left hemiplegia for 8 hours followed by stupor for 3 days. The patient was therefore admitted to our hospital where nominal dysphasia, dystonia, dysmetria, tremulousness, increased deep tendon reflexes, ankle clonus, nystagmus and an enlarged thyroid gland were noted. During his stay in hospital, the patient had a paroxysmal horizontal nystagmus with bilateral mydriasis followed by a tonic attack, involving the right side of the body, with subsequent drowsiness and right hemiplegia for about 10 hours. The next morning he seemed recovered after the nocturnal rest. During that episode several examinations were carried out. EEG recording during wakefulness and spontaneous sleep showed irregular slow waves of 2 Hz–3 Hz on the left frontal region. Brain CT and MRI showed a mild reduction of volume of the left nucleus caudatus, MR angiography was normal. Ictal brain SPECT showed left hemispheric hypoperfusion. Biochemical evaluation encompassing lactate, pyruvate, ketone bodies, amino acids, ammonia, platelets, and protein C and S plasma concentrations performed ictally as well as interictally were normal. Postictal SPECT and EEG performed the next day were normal. A week later, the patient experienced a further tonic attack involving the left side of the body with subsequent drowsiness and left hemiplegia lasting 8 hours, which disappeared after nocturnal rest. EEG recording performed during that episode showed irregular slow waves of 2 Hz–3 Hz on the right hemisphere. Because of the alternating hemiplegic episodes associated with transient hemispheric hypoperfusion, flunarizine (10 mg/day) was administered.

Thyroid function investigation displayed a Hashimoto’s thyroiditis, as serum TSH concentration ws 4.8 mUI/l (normal range 0.3–3.1 mUI/l), antithyroglobulin antibodies 1/100 (normal range <1/100), thyroid microsomal antibodies 1:25.600 (normal range <1/100), and TSH receptor antibodies were absent. Serum T3, free T3, T4, and freeT4 concentrations were normal. Flunarizine treatment was maintained for 28 months at the dose of 10 mg/day, and no further clinical relapses occurred during the follow up period. After 8 months, L-thyroxine (50 μg/day) was given in addition to flunarizine because of mild hypothyroidism. Neurological examination and quantitative neuropsychological tests were still normal. The thyroid microsomal antibody concentrations were unchanged at 1/25 600.

Alternating hemiplegia is a main feature of alternating hemiplegia of childhood (AHC), in which multiple paroxysmal manifestations, especially tonic-dystonic attacks, oculomotor disturbances, and the consistent restorative effect of sleep can occur. All these features were present in our patient and AHC was considered as a possible diagnosis. However, we cannot definitely state that the normalisation of the clinical findings after nocturnal rest, which occurred on two occasions in our patient, was related to the restorative effect of sleep or simply appeared during sleep by chance. The appearance of symptoms at the age of 14 years and the absence of mental deterioration were not consistent with AHC.

Intraoperative neurophysiological recording. Two types of cell discharges were recorded. (A) Tonic neuron: A1, raw data as recorded in the operating theatre of a subthalamic cell, it discharges in a regular pattern (tonic) at high frequency (mean=85 Hz): discharges are subsequently changed into events to be analysed (A2–A3); A2, interval histogram, a symmetric frequency distribution can be seen with the highest peak at 10.3 ms, the insert shows only one action potential of the same cell (negative downwards); A3, autocorrelation histogram made with every action potential to show the type of cell’s activity. The flat outline indicates a great regularity in the firing. (B) Phasic neuron. B1, raw data, as recorded in the operating theatre, of a different type of subthalamic cell, it discharges irregularly at high frequency (mean=87 Hz) with bursts formed by several action potentials subsequently changed into events for statistical analysis (B2-B3); B2, interval histogram. It displays a different outline distribution than the tonic cell with the highest peak at 5.3 ms corresponding to the time interval found when discharging in burst mode, the insert illustrates only one action potential of the same cell with a total duration of 1.4 ms; B3, autocorrelation histogram. The cell shows a tendency to discharge in bursts as seen by the waves (arrows) consisting in periodic increments and decrements in the discharge rate reaching a rhythmical activity of 6–7 Hz.

On the other hand, acute and relapsing neurological symptoms can also occur in patients with Hashimoto’s encephalopathy. In our patient the presence of Hashimoto’s thyroiditis and a history of a febrile illness just before the onset of the clinical picture manifested by confusional state and coma, are compatible with Hashimoto’s encephalopathy. Transient hemiparesis has also been reported in patients with Hashimoto’s encephalopathy,1 however, to our knowledge clear cut transient, recurrent, and alternating hemiplegia episodes restored after sleep have never been reported. EEG findings such as left frontal and right hemispheric slow waves, appearing in our patient during the right and left hemiplegic episodes respectively, and the interictal normalisation are consistent with AHC. However Hashimoto’s encephalopathy cannot be ruled out as generalised or focal slow waves on the EEG have been reported.2 Brain SPECT, performed in our patient during a right hemiplegic attack, showed a left hemispheric hypoperfusion, which was completely resolved interictally. These findings are characteristic of AHC. On the other hand, brain SPECT has been reported in only a few cases of Hashimoto’s encephalopathy, showing global decreased perfusion restored during clinical improvement in a patient,3 and left temporal hypoperfusion in another patient.4 Therefore, as proposed by Forchetti et al 3 in patients with Hashimoto’s encephalopathy, we might hypothesise that a possible autoimmune mechanism causes an alteration in the vascular reactivity of the cerebral microvasculature inducing a reduction of blood flow that in our patient was prevalent in one or in the other hemisphere alternatively. With respect to the treatment, flunarizine is the elective drug in AHC, even if it is only able to reduce the long lasting nature and the severity of the attacks; it does not influence their frequency. A rapid control of symptoms, as in our patient, was therefore unexpected. To our knowledge, there are no reports on Hashimoto’s encephalopathy treated with flunarizine, even if a favourable effect of the drug on decreased brain blood flow might be expected, considering the positive results obtained in migraine and peripheral vascular disorders. In our patient, L-thyroxine was administered 8 months after flunarizine monotherapy, therefore it did not influence the neurological picture. Assuming that our patient is really affected by Hashimoto’s encephalopathy, we have to admit that flunarizine is effective in this condition. In patients with Hashimoto’s encephalopathy corticosteroids are considered the elective drugs. However, reoccurrence of symptoms when the corticosteroids are withdrawn or even while taking them and their inability to prevent mental deterioration in some cases have been reported.5For these reasons, the possibility of using another effective drug such as flunarizine in Hashimoto’s encephalopathy would be of paramount importance, because it has fewer side effects and can be administered continuously.

In conclusion we think that our patient has an unusual form of Hashimoto’s encephalopathy, even if an atypical AHC associated with Hashimoto’s thyroiditis by chance cannot be theoretically ruled out. Therefore, we should consider the effectiveness of flunarizine therapy, that will have to be validated further in patients with Hashimoto’s encephalopathy in whom a decreased brain perfusion is documented.


View Abstract

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.