Is the time ripe to adopt semiquantitative analysis of SPECT evaluation in movement disorders as a standard?

Abstract

Dear Sir, In a very interesting review article recently published in the European Journal of Nuclear Medicine and Molecular Imaging on SPECT imaging evaluation in movement disorders [1], Badiavas et al. also evaluated the performance of the most popular software packages adopted worldwide for “semiquantitative” or “voxel-by-voxel” analysis of uptake by the basal ganglia. Quantitation has been the main diagnostic goal of nuclear medicine and always will be, since the possibility of measuring a process or quantifying a disease reflects the ability of a diagnostic tool to provide a detailed understanding of the mechanism of a physiological process. Visual interpretation or measurement are completely different processes and they represent the difference between modern evidence-based medicine and the medicine of past years based on physicians’ subjectivity rather than objective interpretation of the data. As correctly underlined by the authors, differentiating among possible aetiologies of movement disorders is crucial because of the different therapeutic strategies and prognoses: incorrect information about a reduction in uptake by the basal ganglia may contribute to clinical confusion. However, borderline clinical features and results of instrumental techniques can be found in clinical practice. Particularly under these conditions, visual analysis suffers from interobserver discrepancy, as also underlined by many other studies [2, 3], whereas an automatic method should be less affected by this drawback. The results derived from regions of interest, especially if anatomically shaped for accuracy, can add value in the clinical setting providing a continuous variable for the assessment of disease progression and monitoring of neuroprotective treatments, rather than a categorical variable, since a continuous variable is more likely to express the wide clinical variability and is more suitable for correlative analysis with other disease markers, such as those expressing the severity of motor impairment [4]. The main limits for adopting a semiquantitative analysis in clinical routine are physicians’ inexperience in using this relatively new type of software and the lack of ageand gender-matched databases of healthy control subjects. In fact, the ratio of specific to nonspecific activity can be affected by several factors such as patient age and sex [5], imaging equipment and data processing [6]. As a consequence, each centre should ideally build its own database of normal values before interpreting semiquantitative measurements. A voxel-by-voxel analysis (i.e. standard parametric mapping, SPM) suffers from the same limitation, but it has been used in other studies with reasonable findings [7]. However, as underlined by Calvini et al. [4], even if standard templates represent an approximation, the Talairach atlas is currently considered by the neurological community as the standard reference for locating structures and functional areas; therefore its use could be implemented in clinical practice. In recent years many software packages have been introduced to overcome limitations and many attempts to provide standardized reference values have been made and, as stated by the authors, although in some cases the results F. Caobelli (*) : R. Giubbini Chair of Nuclear Medicine, University of Brescia, P.le Spedali Civili 1, 25133 Brescia, Italy e-mail: fedefournier@libero.it