Autopsy validation of 123I-FP-CIT dopaminergic neuroimaging for the diagnosis of DLB

Objective: To conduct a validation study of ¹²³I-N-fluoropropyl-2b-carbomethoxy-3b-(4-iodophenyl) nortropane (¹²³I-FP-CIT) SPECT dopaminergic imaging in the clinical diagnosis of dementia with Lewy bodies (DLB) with autopsy as the gold standard.

Methods: Patients >60 years of age with dementia who had undergone ¹²³I-FP-CIT imaging in research studies and who had donated their brain tissue to the Newcastle Brain Tissue Resource were included. All had structured clinical research assessments, and clinical diagnoses were applied by consensus panels using international diagnostic criteria. All underwent ¹²³I-FP-CIT imaging at baseline, and scans were rated as normal or abnormal by blinded raters. Patients were reviewed in prospective studies and after death underwent detailed autopsy assessment, and neuropathologic diagnoses were applied with the use of standard international criteria.

Results: Fifty-five patients (33 with DLB and 22 with Alzheimer disease) were included. Against autopsy diagnosis, ¹²³I-FP-CIT had a balanced diagnostic accuracy of 86% (sensitivity 80%, specificity 92%) compared with clinical diagnosis, which had an accuracy of 79% (sensitivity 87%, specificity 72%). Among patients with DLB, 10% (3 patients) met pathologic criteria for Lewy body disease but had normal ¹²³I-FP-CIT imaging.

Conclusions: This large autopsy analysis of ¹²³I-FP-CIT imaging in dementia demonstrates that it is a valid and accurate biomarker for DLB, and the high specificity compared with clinical diagnosis (20% higher) is clinically important. The results need to be replicated with patients recruited from a wider range of settings, including movement disorder clinics and general practice. While an abnormal ¹²³I-FP-CIT scan strongly supports Lewy body disease, a normal scan does not exclude DLB with minimal brainstem involvement.

Classification of evidence: This study provides Class I evidence that ¹²³I-FP-CIT dopaminergic neuroimaging accurately identifies patients with DLB.

Neurology 2016