New-Onset Psychosis: Differentiating Encephalopathy From Psychopathology

A 57-year-old African-American man with no known past medical or psychiatric history presented to the psychiatric unit under an involuntary psychiatric commitment petitioned by law enforcement after he was found at a gas station stating that he was being followed. On initial presentation, he constantly looked around the examination room and appeared to be paranoid as he reported several different accounts of being followed, including by a friend wanting to murder him because he "knew too much," as well as by assassins trying to kill him. He stated that the only other people who knew about these attempts to kill him had been incarcerated.

Collateral information was obtained from the patient’s wife. She confirmed that he had no past psychiatric history and no known family psychiatric history. In the past month, he had become paranoid and began hiding knives at home, reporting that he felt that he was being followed and that someone wanted to kill him. He then went missing and was later found to have unexpectedly driven himself out of state.

On admission, the patient scored 18/30 on the Montreal Cognitive Assessment, with significant deficits observed in executive function domains. He was oriented to person and time but not to place. During his hospital course, he denied any physical symptoms; however, he had slow responses to questioning, and on physical examination, he exhibited bradykinesia in the form of walking with short slow steps. On mental status examination, he demonstrated guarded and anxious behavior; he reported "good" mood, but his affect was irritable. He was isolative on the psychiatric unit, exhibited confabulation and delusional thought content, and demonstrated poor insight into his situation. Because he refused psychiatric treatment, the medical team obtained consent for treatment from his next of kin (wife) in accordance with state law. Subsequently, he was started on risperidone (oral disintegrating tablet, 0.5 mg twice daily) and treated for 16 days without any improvement of symptoms.

The differential diagnoses included unspecified schizophrenia spectrum disorder, meningoencephalitis, delirium, Lewy body dementia, dementia with behavioral disturbance, frontotemporal dementia, Parkinson’s disease, and anti-N-methyl-d-aspartate (NMDA) receptor encephalitis. Despite the patient’s apparent psychiatric presentation, given his age and lack of family and personal psychiatric history, we pursued a neurological consultation and clinical workup. Results of his neurological and physical examinations were within normal limits. Diagnostic tests included brain CT and MRI, EEG, rapid plasma reagin, NMDA antibodies, and lumbar puncture serology/immunology testing (see box). Imaging and EEG were unremarkable. A urine drug screen was negative, and CBC was unremarkable. NMDA testing was negative. Rheumatoid factor, ANA pattern, c-ANCA, p-ANCA (anti-granulocyte antibodies), anti-proteinase 3, and anti-myeloperoxidase were within normal limits. Tests for syphilis, HIV, and tuberculosis were all negative. Lumbar puncture CSF alpha 1-globulin, alpha 2-globulin, beta globulin, gamma globulin, beta 2-microglob, serum protein electrophoresis M spike, Venereal Disease Research Laboratory Test, Lyme immunoglobulin G/M (IgG/IgM), coccidiosis IgG, cryptococcus Ag, herpes DNA, toxoplasma IgM/IgG, and varicella-zoster virus DNA were also all unremarkable. However, the patient’s lumbar puncture revealed elevated CSF IgG, increased IgG synthesis rate, and elevated total protein. Thyroid-stimulating hormone (TSH) was 1.340 uIU/mL (normal range, 0.358–3.740), and free T4 was 1.04 ng/dl (reference range, 0.76–1.46). Thyroid peroxidase antibody was elevated at 39.4 (reference range, 0.0–35.0). Thyroglobulin antibody was <20.0 (reference range, 0.0–40.0). Serology showed elevated Epstein-Barr virus (EBV) capsid Ag IgG and EBV nuclear antigen antibodies. C-reactive protein was elevated at 0.40 (reference range, 0.00–0.29), and erythrocyte sedimentation rate was elevated at 14 (reference range, 0–10).

On the basis of the presentation and elevated thyroid peroxidase antibodies, the patient was diagnosed as having Hashimoto’s encephalopathy. He was transferred to the medical unit and started on methylprednisolone sodium succinate (1,000 mg intravenous infusion, over 24 hours for 5 days). Treatment with steroids led to resolution of the patient’s delusions and paranoia. His mental status and cognitive function returned to his baseline, and the patient was discharged home.

Etiologies of autoimmune encephalopathy are currently divided into two groups: those of rheumatic conditions with neuropsychiatric symptoms and antibody-associated autoimmune encephalitis (3). The latter include Hashimoto’s encephalopathy, which has a prevalence of 2.1 per 100,000 and mean age at onset of 44–46 years (4). Clinical features include acute confusion or diffuse progressive pattern of cognitive impairment, seizures, tremors, hyperreflexia, or psychosis (1). It is not uncommon for patients to be euthyroid at the time of presentation, and a previous report on the topic indicated that over 40% of patients were euthyroid at the time of diagnosis (1). For patients with Hashimoto’s encephalopathy, the results of thyroid function tests may be normal or may suggest a range of thyroid pathologies, including subclinical hypothyroidism, hypothyroidism, subclinical hyperthyroidism, or hyperthyroidism (2).

Two primary types of antibodies play a role in the pathogenesis of autoimmune thyroid disease: antibodies against TSH receptors, and antibodies against the thyroid gland (5). These are antiperoxidase antibodies, antithyroglobulin antibody, and antisodium-iodine symporter antibodies, which are the most important factors to the pathogenesis of the disease. About 70% of patients diagnosed as having Hashimoto’s thyroiditis will have positive antithyroglobulin antibodies (3). In addition, up to 85% of patients will be positive for antithyroid peroxidase antibodies (3); however, this antibody can also be detected in about 10% of the healthy population. In the case reported above, antithyroid peroxidase antibodies were found, but no antithyroglobulin antibodies were found. The presence of either of these antibodies can help establish the diagnosis of Hashimoto’s encephalopathy (6). Some evidence suggests that thyroid autoimmune factors are directly involved in the presentation of encephalopathy, although no significant correlation has been detected between the level of antibodies and the severity of the disease (7). In addition, although excessive secretion of TSH may cause damage to brain tissues in multiple ways, no significant correlation has been found between nervous system impairment and thyroid gland dysfunction (8, 9). Furthermore, there is no evidence to suggest that antithyroid antibodies directly contribute to the damage of neurons. Thus, it is unknown at this time whether altered hormonal feedback related to thyroid gland dysfunction or autoimmune response is responsible for the associated encephalopathy.

Treatment of choice includes high-dose prednisone (1–2 mg/kg) for up to 4–6 weeks, followed by a slow taper to avoid recurrence (5). Serum thyroid antibody levels do not correlate with response to treatment and can remain elevated even after treatment. Clinical improvement is best monitored by resolution of symptoms, which tend to improve with high-dose steroids (9).

Psychiatrists must maintain a broad differential diagnosis for nonpsychiatric causes of acute psychosis that do not fit the normal clinical picture. Although patients with Hashimoto’s encephalopathy may exhibit psychotic symptoms similar to those seen in other primary psychiatric disorders, the treatment of autoimmune encephalopathy is vastly different. Both early recognition and an understanding that normal thyroid studies do not automatically rule out the diagnosis of Hashimoto’s encephalopathy is important to the treatment of this condition. In cases of psychosis in which thyroid studies are equivocal, psychiatrists must maintain a high index of suspicion for Hashimoto’s encephalopathy. Additional research is needed to further identify the pathophysiology of the syndrome.

The authors have confirmed that details of the case have been disguised to protect patient privacy.