Impact of the cerebrospinal fluid-mask algorithm on the diagnostic performance of 123I-Ioflupane SPECT: an investigation of parkinsonian syndromes

Background A cerebrospinal fluid (CSF)-mask algorithm has been developed to reduce the adverse influence of CSF-low-counts on the diagnostic utility of the specific binding ratio (SBR) index calculated with Southampton method. We assessed the effect of the CSF-mask algorithm on the diagnostic performance of the SBR index for parkinsonian syndromes (PS), including Parkinson's disease, and the influence of cerebral ventricle dilatation on the CSF-mask algorithm. Methods We enrolled 163 and 158 patients with and without PS, respectively. Both the conventional SBR (non-CSF-mask) and SBR corrected with the CSF-mask algorithm (CSF-mask) were calculated from I-123-Ioflupane single-photon emission computed tomography (SPECT) images of these patients. We compared the diagnostic performance of the corresponding indices and evaluated whether the effect of the CSF-mask algorithm varied according to the extent of ventricle dilatation, as assessed with the Evans index (EI). A receiver-operating characteristics (ROC) analysis was used for statistical analyses. Results ROC analyses demonstrated that the CSF-mask algorithm performed better than the non-CSF-mask (no correction, area under the curve [AUC] = 0.917 [95% confidence interval (CI) 0.887-0.947] vs. 0.895 [95% CI 0.861-0.929], p < 0.001; attenuation correction, AUC = 0.930 [95% CI 0.902-0.957] vs. 0.903 [95% CI 0.870-0.936], p < 0.001). When not corrected for attenuation, no significant difference in the AUC was observed in the low EI group between the non-CSF-mask and CSF-mask algorithms (0.927 [95% CI 0.877-0.978] vs. 0.942 [95% CI 0.898-0.986], p = 0.11); in the middle and high EI groups, the CSF-mask algorithm performed better than the non-CSF-mask algorithm (middle EI group, AUC = 0.894 [95% CI 0.825-0.963] vs. 0.872 [95% CI 0.798-0.947], p < 0.05; high EI group, AUC = 0.931 [95% CI 0.883-0.978] vs. 0.900 [95% CI 0.840-0.961], p < 0.01). When corrected for attenuation, significant differences in the AUC were observed in all three EI groups (low EI group, AUC = 0.961 [95% CI 0.924-0.998] vs. 0.942 [95% CI 0.895-0.988], p < 0.05; middle EI group, AUC = 0.905 [95% CI 0.843-0.968] vs. 0.872 [95% CI 0.800-0.944], p < 0.005; high EI group, AUC = 0.954 [95% CI 0.917-0.991] vs. 0.917 [95% CI 0.862-0.973], p < 0.005). Conclusion The CSF-mask algorithm improved the performance of the SBR index in informing the diagnosis of PS, especially in cases with ventricle dilatation.