Diagnosis of rare dementia syndromes: an algorithmic approach

The etiology of dementia can be diagnosed in most patients using a standard clinical approach consisting of physical, neurologic, and mental status examinations, and laboratory testing, lumbar puncture, and neuroimaging. In some cases, however, the clinical presentation or historical data are unusual, or the results of the workup are inconclusive or atypical. A rare cause of dementia may then be present and a complicated evaluation may be necessary to identify the specific disease process. A potentially useful approach to the diagnosis of rare dementing disorders consists of a series of diagnostic algorithms. This approach utilizes results of neuroimaging studies to guide the evaluation through additional diagnostic steps such as specific enzymatic or immunologic assays or biopsy of extraneural tissues. The disorders potentially detected by these algorithms typically have unusual clinical features such as early age of onset, abnormal neurologic signs and symptoms early in the clinical course, early personality and mood changes, extrapyramidal or cerebellar signs and symptoms, seizures, peripheral neuropathy or myopathy, and extraneural abnormalities involving the dermatologic, cardiovascular, musculoskeletal, or ocular systems. Accurate diagnosis of these rare causes of dementia is important for medical and psychiatric management, prognosis, and genetic counseling.