Objective: To evaluate visual assessment of [(11)C]PIB and [(18)F]FDDNP PET images as potential supportive diagnostic markers for Alzheimer's disease (AD).
Methods: Twenty-one AD patients and 20 controls were included. Parametric [(11)C]PIB and [(18)F]FDDNP global binding potential (BP(ND)) images were visually rated as 'AD' or 'normal.' Data were compared with ratings of [(18)F]FDG PET images and MRI-derived medial temporal lobe atrophy (MTA) scores. Inter-rater agreement and agreement with clinical diagnosis were assessed for all imaging modalities. In addition, cut-off values for quantitative global [(11)C]PIB and [(18)F]FDDNP BP(ND) were determined. Visual ratings were compared with dichotomised quantitative values.
Results: Agreement between readers was excellent for [(11)C]PIB, [(18)F]FDDNP and MTA (Cohen kappa kappa> or =0.85) and moderate for [(18)F]FDG (kappa=0.56). The highest sensitivity was found for [(11)C]PIB and [(18)F]FDG (both 1.0). The highest specificity was found for MTA (0.90) and [(11)C]PIB (0.85). [(18)F]FDDNP had the lowest sensitivity and specificity (0.67 and 0.53, respectively). The cut-off for quantitative [(11)C]PIB BP(ND) was 0.54 (sensitivity and specificity both 0.95) and for [(18)F]FDDNP BP(ND) 0.07 (sensitivity 0.80, specificity 0.73). Agreement between quantitative analyses and visual ratings was excellent for [(11)C]PIB (kappa=0.85) and fair for [(18)F]FDDNP (kappa=0.40).
Conclusion: Visual assessment of [(11)C]PIB images was straightforward and accurate, showing promise as a supportive diagnostic marker for AD. Moreover, [(11)C]PIB showed the best combination of sensitivity and specificity. Visual assessment of [(18)F]FDDNP images was insufficient. The quantitative analysis of [(18)F]FDDNP data showed a considerably higher diagnostic value than the visual analysis.