A J Sommerfield, R Stimson, IW Campbell
Department of Diabetes and Endocrinology, Victoria Hospital, Hayfield Road, Kirkcaldy
Correspondence to: Professor I.W. Campbell, Victoria Hospital, Hayfield Road, Kirkcaldy, Scotland
e-mail: Jackie.Wallace@faht.scot.nhs.uk
SMJ 2004 49(4): 155-156
Abstract: Three cases are described of a reversible encephalopathy, all presenting with marked neurological disturbance. In all three, the diagnosis was not clear at the time of presentation but eventually it was felt all of the cases were consistent with Hashimoto’s encephalopathy. The diagnosis of Hashimoto’s encephalopathy should be considered in any case of unexplained encephalopathy. Common features are high anti-thyroid peroxidase antibody titres, an abnormal EEG and an elevated CSF protein concentration. The encephalopathy is independent of thyroid hormonal status. Treatment with corticosteroids leads to a prompt resolution of symptoms and long term low dose steroid therapy prevents further neurological recurrence Key words: Hashimoto’s, acute, emergency, encephalopathy, medical
Neurological disturbance is a well-recognised complication of hypothyroidism. Dementia, psychosis, ataxia and mood disturbance can all occur. However, symptoms invariably resolve when thyroxine replacement is commenced. Hashimoto’s encephalopathy (HE) has been postulated to be a separate clinical entity. It is independent of hormonal thyroid function, the majority of patients being euthyroid, is associated with high titres of anti-thyroid autoantibodies, and responds well to treatment with corticosteroids. Here we report of three patients suffering from a reversible encephalopathy associated with high concentrations of antithyroid autoantibodies. All three cases had symptoms for several months and were initially misdiagnosed.
Case 1
In February 1999, a 38 year-old woman was admitted with severe headache and vomiting. She had a past history of type 1 diabetes of 16 years duration and had been diagnosed with primary hypothyroidism at the age of 34. Her only medication was insulin and thyroxine. Physical examination was normal, and routine baseline investigations (FBC, U&Es, glucose, TFTs) were also normal. Her symptoms resolved with simple analgesia and she was discharged home the following day with a putative diagnosis of migraine.
Six months later she was readmitted having been found unresponsive in the street. Her conscious level was impaired (GCS 9), she was flailing her limbs and making unresponsive sounds. She was mildly pyrexial (37.4oC), but was haemodynamically stable and there were no focal neurological signs. Hypoglycaemia was initially suspected, but venous blood glucose measured 6.2 mmol/l. Routine bloods, including thyroid function, were again normal. A CT scan, and subsequently an MRI scan, of the brain were also normal. A lumbar puncture revealed elevated CSF protein at 1.1g/l, but there were no cells, and no organisms were identified on gram stain or culture. An EEG showed non-specific abnormalities. She was initially treated with intravenous antibiotics and acyclovir. Over the next seven days her clinical condition gradually improved and she was discharged home.
The following day she was readmitted following a seizure. Once more, routine blood tests and neuroimaging were normal. The CSF protein remained elevated, and the EEG was again non-specifically abnormal. Her antithyroid peroxidase antibody titres were measured, which were significantly raised (1/6400). She was commenced on 40mg prednisolone, and the dose gradually reduced over the ensuing six months. She continues on 5mg of prednisolone daily and has had no further episodes of neurological disturbance.
Case 2
In April 1999, a 25 year-old woman was admitted with sudden onset of right arm weakness, expressive dysphasia, and loss of vision in her right eye. She complained of frontal headache and photophobia, and had vomited. Her only past medical history was of primary hypothyroidism, for which she was taking thyroxine. On examination, she was mildly confused, with an AMT 8/10, and had an expressive dysphasia. She was apyrexial and neurological examination was otherwise normal. FBC, U&Es, glucose and TFTs were normal, as was a CT scan of the brain. Her confusional state, headache and dysphasia improved spontaneously and she was discharged home with no definite diagnosis made. She was readmitted in October 1999 with headache, seizures and confusion. Examination was unremarkable and routine bloods were normal. CT and MRI scans of the brain were also normal. Lumbar puncture showed an isolated elevated CSF protein at 0.9g/l and an EEG was reported as showing encephalopathic changes. She was commenced on anticonvulsants and treated empirically with acyclovir. Her symptoms resolved within one week.
One year later she was readmitted with a three-day history of vomiting, abdominal pain, dizziness, headache, myoclonus and seizures. She was extensively investigated for a cause of encephalopathy. CSF protein remained high at 1.07g/l, but serum electrophoresis, LDH, CK, porphyria screen, hepatitis serology, lead, copper and caeruloplasmin, and duodenal biopsy were all normal. Autoantibody levels were normal apart from antithyroid aperoxidase antibodies, which were elevated at 1/6400. She was commenced on 40mg of oral prednisolone. Her symptoms took three weeks to resolve. Her dose of steroids was gradually reduced over the following 12 months, and then discontinued altogether. In August 2002 she re-presented with severe generalised headache. At this time, she had been off the steroids for six months. Her steroids were reintroduced and her symptoms quickly settled. She continues on 5mg daily of oral prednisolone and has been free of neurological symptoms since.
Case 3
In January 1999, a 43 year-old woman was admitted following a collapse. She had no significant past medical history and was taking no medication. On nerological examination, she was agitated, GCS 10/15, with generalised increased tone and brisk reflexes, and choreiform movements. FBC, U&Es, and glucose were normal. CT scan of the brain was also normal. CSF protein was elevated at 0.74g/l. She was treated with acyclovir, her symptoms resolved and she was discharged 10 days later. The following month she was readmitted with a similar story. CSF protein remained high at 0.76g/l. MRI scan of the brain and EEG were normal. TFTs showed normal thyroxine at 15pmol/l, but the TSH was raised at 25mmol/l. Antithyroid peroxidase antibodies were high at 1/ 6400. She was commenced on 40mg of prednisolone and 100mg of thyroxine and her symptoms resolved completely within days. She is maintained on 5mg daily of prednisolone and remains well on this.
Hashimoto’s encephalopathy was first described in 1966 by Brain et al.1 They reported a case of a 40 year-old man who presented with episodic encephalopathy. He had a goitre, but thyroid function was normal. Biopsy of the thyroid gland showed changes typical of Hashimoto’s thyroiditis. Antithyroid autoantibody titres were elevated and lumbar puncture revealed elevated CSF protein. There was no pleocytosis or evidence of infection. An EEG showed non-specific abnormalities.
Since that time there have been more than 100 published cases of Hashimoto’s encephalopathy. A recent extensive literature review analysed 85 cases of non-infectious encephalopathy associated with high serum antithyroid autoantibody levels.2 Eighty-one percent of cases were female, with mean age of onset of symptoms of 44 years (range 9-78 years). Thyroid status varied greatly. The most common abnormality was subclinical hypothyroidism (35% cases), while the majority had normal serum thyoxine levels (67%). Antyithyroid autoantibodies were elevated in 100% cases, antithyroid peroxidase antibody being the most commonly identified. The most common abnormality in CSF was an elevated protein concentration (78%). In 76% cases there were small numbers (0-3) of mononuclear cells within the CSF, and in only 4% was the cell count >100. MRI scan of the brain was abnormal in 49%. Changes were predominantly non-specific. Ninety-eight percent cases had an abnormal EEG. This usually showed non-specific changes of diffuse slowing. Histology of brain tissue was available in only four cases, but has shown evidence of a lymphocytic infiltrate.3
Treatment of HE is with immunosuppressive therapy and, in general, the prognosis is good with improvement of symptoms within 4-6 weeks. Of the 85 patients reviewed, 45 were treated with steroids alone. Symptoms improved in 98% of these cases. Another 24 patients were treated with a combination of steroids and thyroxine; 92% of these patients improved. The dose and duration of steroid therapy has varied considerably.
Two subtypes of the disorder have been proposed – a vasculitic type, characterised by multiple stroke-like episodes, and a diffuse progressive type, characterised by dementia and psychiatric symptoms.4 Coma, tremor, seizures or myoclonus may occur in either form. There is no difference in the biochemical or radiological findings between the two forms, nor in the response to treatment.4 The three cases that are described in this paper exhibit features typical of HE. They also demonstrate the difficulty in diagnosing the condition. Consequently, as in these cases, many patients with HE are initially misdiagnosed. This in part reflects the lack of a specific diagnostic test for HE. The most common pathological findings include elevated antithyroid autoantibody titres, elevated CSF protein and an abnormal EEG. Increased awareness of the condition is required, and will aid the diagnosis of this potentially treatable form of encephalopathy.
The existence of HE is still debated.5 Antithyroid autoantibodies have been found in up to 10% of the general population,6 and it is possible that HE represents an idiopathic encephalopathy coincidentally associated with high titres of antithyroid autoantibodies. However, the combination of encephalopathy, high serum antithyroid antibody concentration, and responsiveness to corticosteroids seems unlikely to be due to chance. The pathogenesis of HE is, as yet, unclear. Exact histopathological data is lacking. However, it appears that the encephalopathy is not caused by hypothyroidism per se. Overt hypothyroidism can cause cerebral dysfunction,7 including changes in mood, cognition, conscious level and psychosis. It can also cause high CSF protein concentrations and an abnormal EEG.7 All of these changes improve with the administration of thyroxine.7 Patients with HE are often euthyroid, and subclinical hypothyroidism has less, if any, effect on cerebral function, the EEG or CSF.8 Possible pathogenic mechanisms that have been suggested include excess TRH, localised cerebral oedema, cerebral hypoperfusion or autoimmune-mediated cerebral vasculitis. The latter seems particularly likely. An autoimmune mechanism is suggested by the presence of an inflammatory infiltrate on brain biopsy and the improvement of symptoms in response to steroids.
HE is probably an underestimated condition due to overlap of the symptoms with more common disorders, such as depression or seizures. A high level of suspicion, and increased awareness of the condition, is necessary to establish the diagnosis. This treatable form of encephalopathy should be considered, and antithyroid autoantibody levels checked, in any case of unexplained encephalopathy after other causes of metabolic, inflammatory, infective and paraneoplastic encephalopathy have been excluded.
ACKNOWLEDGEMENTS: The authors thank their colleagues in the Department of Clinical Neurosciences, Western General Hospital, Edinburgh for their help in the investigation of the three cases.
REFERENCES
1 Brain L, Jellinek EH, Ball K. Hashimoto’s disease and encepahalopathy. Lancet 1966; 2: 512-514.
2 Chong JY, Rowland LP, Utiger RP. Hashimoto encephalopathy. Syndrome or myth? Arch. Neurol. 2003; 60: 164-171.
3 Nolte KW, Unbehaun A, Sieker H, Kloss TM, Paulus W. Hashimoto encephalopathy: a brainstem vasculitis? Neurology 2000; 54: 769-770.
4 Kothbauer-Margreiter I, Sturzenegger M, Komor J, Baumgartner R, Hess CW. Encephalopathy associated with Hashimoto thyroiditis: diagnosis and treatment. J. Neurol. 1996; 243: 585-593.
5 Sunil GS, Mariash C. Hashimoto’s encephalitis. J. Clin. Endocrinol. Metab. 2001; 86: 947.
6 Vanderpump MP, Tunbridge WM, French JM et al. The incidence of thyroid disorders in the community: a twenty-year follow-up of the Whickham survey. Clin. Endocrinol. 1995; 43: 55-68.
7 Whybrow PC. Behavioral and psychiatric aspects of hypothyroidism. In: Braverman LE, Utiger RD, eds. The Thyroid: A Fundamental and Clinical Text. 8th ed. Philadelphia, PA: Lippincott Williams & Wilkins; 2000: 837-842.
8 Nystrom E, Hamberger A, Lindstedt G, Lundquist C, Wikkelso C. Cerebrospinal fluid proteins in subclinical and overt hypothyroidism. Acta Neurol. Scand. 1997; 95: 311-314.