Sarcoidosis Presenting As Late‐Onset Dementia: Are Cerebrospinal Fluid Biomarkers Analyses Helpful?

Abstract

To the Editor: Neurosarcoidosis is uncommon in elderly adults1 and is strongly associated with dementia.2 Biopsy of the central nervous system (CNS) is usually required to confirm the diagnosis.3 The case of a 71-year-old man with rapidly progressive onset of memory deficit and finally with signs of dementia syndrome, apathetic state, and abrupt weight loss (20 kg) is reported. His medical history was positive for high blood pressure and depression. Initial physical examination demonstrated bilateral Babinski signs, and initial cognitive assessment revealed a Mini-Mental State Examination (MMSE) score of 23 that rapidly declined to 9. Executive disorder and storage memory deficit were also initially noticed. Magnetic resonance imaging (MRI) of the brain showed minor hyperintense white matter lesions at T2, with an irregular aspect of the cerebral gyri and microbleeding next to the frontal horn of the lateral ventricle. Ethylcysteinate dimer enhanced single-photon emission computed tomography of the brain showed hypoperfusion in frontal lobes. With subsequent imaging and biological findings (Table 1) indicating systemic sarcoidosis, bronchial biopsy revealed an epithelial and giant cell inflammatory granuloma (EGG) without caseous necrosis.1 Rapid improvement of cognitive function and general health status (weight gain and partial regression of the apathetic state) was observed with cortiosteroid theraphy (1 mg/kg per day). The cortiosteroid theraphy dose was reduced to 0.5 mg/kg per day and maintained. The diagnosis of neurosarcoidosis was made based on the above-mentioned findings, which matched the diagnostic criteria for probable neurosarcoidosis.3 Thus, clinical findings, cognitive profile, laboratory test results supporting CNS inflammation (high cerebrospinal fluid (CSF) protein levels, presence of oligoclonal band), and pathology-confirmed systemic sarcoidosis (EGG), were compatible with the diagnosis of neurosarcoidosis. With no leptomeningeal or cerebral parenchyma damage, which would have been biopsied, there was no indication for brain biopsy. Unspecific brain imaging results do not exclude a diagnosis of neurosarcoidosis.4 Abnormalities of the white matter detected using computed tomography and degenerative changes affecting the perivascular spaces or the brain parenchyma in the form of small granulomas have been described in biopsy material.5 The microbleeding might also have been related to neurosarcoidosis in the present case.6 Additional support for the diagnosis of neurosarcoidosis was discovered. Over the following 44 months, the situation continued to improve while cortiosteroid theraphy was progressively reduced and finally stopped. Although he returned to independence in activities of daily living with stimulation (but not in instrumental activities of daily living), his cognitive deficit persisted (MMSE score 15). He was able to perform a description task from the Boston Diagnostic Aphasia Examination 7 and to read a text and the time on a clock, but his verbal fluency was poor, as well as his results on the Frontal Assessment Battery (5/18). In addition, some verbal (utterances repeated multiple times such as inappropriate laughter) and gestural (rubbing hands together; rubbing the face) stereotypies appeared and become almost permanent. He no longer exhibited disinhibited or aggressive behavior, but apathy persisted. The combination of a dementia syndrome of degenerative origin such as frontotemporal dementia (FTD) and pulmonary sarcoidosis could not be excluded.8 Normal CSF cytology does not exclude neurosarcoidosis, and positive oligoclonal bands are seen in at least 19% of neurosarcoidosis cases.9 Although FTD, which is a neurodegenerative disease with different behavioral subtypes, might easily explain the apathy and hypoperfusion in the frontal lobes, the inflammation and continuing improvement with corticotherapy are not consistent with this diagnosis. Moreover, during follow-up, as shown in Table 1, repeated measurements of CSF biomarkers at Months 3, 17, 21, and 27 showed variations in Tau and PhosphoTau-181 protein levels with subsequent normalization, whereas Aβ1–40 and Aβ1–42 levels remained abnormal. Although these CSF biomarker levels demonstrated acute cerebral stress and alteration of β-amyloid clearance, normalization of Tau and PhosphoTau-181 were not indicative of Alzheimer's disease and even contributed to the exclusion of this diagnosis10 and of FTD.11 Variations in CSF biomarkers were relevant for determining the differential diagnosis between FTD with pulmonary sarcoidosis and sequelae of neurosarcoidosis with symptoms mimicking, which was the case here. In the present clinical case, the variation in CSF biomarkers levels probably indicates brain tissue damage associated with the inflammatory process. If this raises interest in measuring CSF biomarkers for the diagnosis of rapidly progressing non-Alzheimer's dementia syndromes, further studies are necessary to better interpret value variations in Aβ1–40, Aβ1–42, Tau and PhosphoTau-181 proteins, although the measurement of CSF biomarkers would never replace pathological analysis of the brain tissue to confirm the origin of cognitive disorders. Conflict of Interest: None of the authors has any conflict of interest to declare. Author Contributions: MK, POL, RC, OB, TV, and GK wrote the manuscript. Sponsor's Role: There was no sponsor for this manuscript.