FDG-PET changes in LGI1-encephalitis

Case presentation: We present the case of a 63-year-old female who was brought to the emergency department due to seizures and progressive memory deficits. She had a prior hospitalization 3 months earlier after presenting with a new-onset generalized tonic-clonic seizure, following a 1-month period of insomnia and forgetfulness. Initially, she was diagnosed with a symptomatic seizure due to severe hyponatremia (116mEq/L), thought to be secondary to multiple oral nootropic drugs. She had no comorbidities, and both brain MRI and prolonged EEG monitoring were normal. She was discharged with normal blood sodium levels and minor improvement in cognitive status, with the recommendation to withdraw oral drugs. During the following 3 months, however, she experienced progressive cognitive deterioration. An FDG-PET was ordered and showed hypermetabolism in the right temporal lobe, and her MoCA score was 18. At the second admission, she had three new short and self-limited episodes of generalized tonic-clonic seizures. Her sodium level was again 116mEq/L. After an extensive search for the cause of hyponatremia, a diagnosis of probable SIADH was made without an evident clinical cause. A new MRI showed no significant findings, and continuous EEG suggested cortical dysfunction with occasional right TIRDAs and FIRDAs. Lumbar puncture results were normal, without pleocytosis and oligoclonal bands. Oral levetiracetam was started, and hyponatremia was slowly corrected; the patient remained seizure-free. Serum anti-LGI1 was ordered, and the patient was discharged with close outpatient follow-up. A month later, results came back positive for anti-LGI1, and the patient underwent corticotherapy with IV 1g methylprednisolone for 5 days followed by oral 80mg prednisone. She showed marked cognitive improvement in the first week after therapy, with MoCA score improving to 26, with minor visuospatial memory deficits. Discussion: We presented a case of anti-LGI1 encephalitis with significant improvement following corticotherapy, as consistently shown in previously reported cases. Clinical criteria for limbic encephalitis were all met except for the absence of MRI changes. Most cases present with typical bi-temporal changes on MRI; however, they can show multiple patterns of impairment. Metabolic imaging could reveal more subtle temporal changes and act as a more precise tool for diagnosing limbic encephalitis. There is, however, a lack of definite criteria for analyzing FDG-PET imaging, as anti-LGI1 encephalitis can present as either hypo or hypermetabolism in the temporal lobes or basal ganglia. The presence of anti-LGI1 antibodies in such cases is indispensable. Final comments: Anti-LGI1 encephalitis is considered the second most common autoimmune encephalitis and tends to show good outcomes. It may be beneficial to have a low threshold for considering it even when not all criteria for limbic encephalitis are met, especially if paraclinical signs such as hyponatremia are present.