A visual rating scale for cingulate island sign on 18F-FDG-PET to differentiate dementia with Lewy bodies and Alzheimer's disease

Le Gjerum*, Kristian Steen Frederiksen, Otto Mølby Henriksen, Ian Law, Lasse Anderberg, Birgitte Bo Andersen, Eva Bjerregaard, Anne Mette Hejl, Peter Høgh, Steen Gregers Hasselbalch

*Corresponding author af dette arbejde

Publikation: Bidrag til tidsskriftTidsskriftartikelForskningpeer review

19 Citationer (Scopus)
110 Downloads (Pure)

Abstract

Valid diagnosis of dementia with Lewy bodies (DLB) is essential to establish appropriate treatment and care. However, the diagnostic accuracy is complicated by clinical and pathological overlap with Alzheimer's disease (AD). Cingulate island sign (CIS), defined as sparing of posterior cingulate cortex (PCC) relative to precuneus and cuneus on 18F-fluoro-deoxy-glucose positron emission tomography (18F-FDG-PET), is included in the revised diagnostic DLB criteria. There are no guidelines for the visual grading of CIS, although visual rating is a fast-applicable method in a clinical setting. The objective was to develop a robust visual CIS scale and evaluate the performance in differentiating DLB with and without amyloid beta pathology (Aβ+/−), and AD. 18F-FDG-PET scans from 35 DLB patients, 36 AD patients, and 23 healthy controls were rated according to a visual CIS scale based on specific reading criteria. The visual CIS scale was validated against a quantitative CIS ratio derived from a region of interest analysis of PCC, precuneus, and cuneus. DLB patients had a significantly higher visual CIS score compared to AD patients, and controls. A cut-off visual CIS score of 4 significantly differentiated DLB Aβ− patients from DLB Aβ+ patients. In conclusion, the visual CIS scale is clinically useful to differentiate DLB from AD. The degree of CIS may be related to Aβ pathology in DLB patients.

OriginalsprogEngelsk
Artikelnummer116645
TidsskriftJournal of the Neurological Sciences
Vol/bind410
Antal sider10
ISSN0022-510X
DOI
StatusUdgivet - mar. 2020

Citationsformater