Transcranial sonography and [18 F]fluorodeoxyglucose positron emission tomography for the differential diagnosis of parkinsonism: a head-to-head comparison

Strong support for transcranial sonography in this paper. I must say my own experience performing TCS to date has been a positive one. The merits are clear to see and if atypical PD can be distinguished from idiopathic PD that would be a huge bonus! Alastair Noyce

Eur J Neurol. 2014 Mar 6. doi: 10.1111/ene.12394. [Epub ahead of print]

Hellwig S, Reinhard M, Amtage F, Guschlbauer B, Buchert R, Tüscher O, Weiller C, Niesen WD, Meyer PT.

Author information

Department of Neurology, University Hospital Freiburg, Freiburg, Germany; Department of Psychiatry and Psychotherapy, University Hospital Freiburg, Freiburg, Germany.

Abstract

BACKGROUND AND PURPOSE:

Brain imaging with positron emission tomography using [18 F]fluorodeoxyglucose (FDG-PET) and transcranial B-mode sonography (TCS) improves the differential diagnosis of parkinsonism. The diagnostic merits of these approaches in identifying and differentiating atypical parkinsonian syndromes (APS) are compared.

METHODS:

Data were included from 36 patients with clinically suspected APS who underwent PET and TCS. FDG-PET scans were analyzed by visual assessment (including voxel-based statistical maps) of a priori defined disease-specific metabolic patterns. Sonographers achieved diagnoses according to pre-defined criteria for echogenicities of the substantia nigra and lenticular nucleus, and third ventricle diameter. Patients with APS were identified and allocated to the subgroups multiple system atrophy (MSA), progressive supranuclear palsy (PSP) or corticobasal degeneration (CBD).

RESULTS:

After a median follow-up period of 9 months, the final clinical diagnoses (reference standard) were Parkinson's disease, n = 15; MSA, n = 9; PSP, n = 7; and CBD, n = 5 (n = 21 APS in total). Six patients (4 APS) showed an insufficient bone window for TCS. In the remaining 30 patients, sensitivity/specificity for diagnosing APS were 82%/100% and 82%/85% for FDG-PET and TCS, respectively. Diagnostic accuracies did not differ between FDG-PET (90%) and TCS (83%; P = 0.69). Likewise, overall accuracy of subgroup classification (non-APS, MSA, PSP and CBD) did not differ between modalities (FDG-PET 87% and TCS 83%; P = 1.00).

CONCLUSIONS:

FDG-PET and TCS show comparable accuracies for differential diagnosis of neurodegenerative parkinsonism. This preliminary study supports the use of TCS and warrants further prospective validation.