European Journal of Hybrid Imaging (May 2018)

Interpreting 123I–ioflupane dopamine transporter scans using hybrid scores

  • Kenneth J. Nichols,
  • Brandon Chen,
  • Maria B. Tomas,
  • Christopher J. Palestro

DOI
https://doi.org/10.1186/s41824-018-0028-0
Journal volume & issue
Vol. 2, no. 1
pp. 1 – 13

Abstract

Read online

Abstract Background Dopamine transporter (DaT) 123I–FP-CIT scans most commonly are interpreted visually. Alternatively, absolute quantitation of radiopharmaceutical uptake may improve scan accuracy. However, neither approach accomodates dependence of striatal uptake on age and gender. We investigated whether demographic indexing of visual and numerical variables improve discrimination of patients with essential tremor (ET), Parkinson’s disease (PD), and dementia with Lewy bodies (DLB). Methods Data of 132 consecutive patients undergoing DaT SPECT scans were reviewed retrospectively. The clinical impression in the latest neurology note was utilized as the final clinical diagnosis. Caudate and putamen specific binding ratio (PSBR) were computed. 123I calibration phantoms were constructed to enable absolute quantitation of putamen radiopharmaceutical uptake. A single experienced nuclear medicine physician graded visual certainty on a 3-level scale. Demographic indexing normalized metrics to published normal PSBR values. Methods were compared by simultaneous ROC analyses to identify the technique of maximal accuracy. Results Thirty-four patients (26%) were diagnosed with ET, 85 (64%) with PD, 6 (5%) with multiple system atrophy, and 7 (5%) with DLB. For discriminating DLB from PD, visual analysis was significantly less specific and accurate than the other techniques. However, indexing significantly improved specificity and accuracy of visual scores, such that indexed visual scores were statistically equivalent to all other methods. Indexed PSBR yielded essentially the same results as non-indexed PSBR, for which highest overall test efficacy was achieved. Conclusions Our results in this small series of patients with DLB suggest that if 123I–FP-CIT visual scores are to be used to discriminate DLB from other neurologic disorders, demographic indexing should be applied. However, best results overall are obtained using quantified parameters, regardless of whether or not demographic indexing is applied to these values.

Keywords