Molecular imaging in the diagnosis of Alzheimer’s disease: visual assessment of [C]PIB and [F]FDDNP PET images

Objective: To evaluate visual assessment of [C]PIB and [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 [C]PIB and [F]FDDNP BPND images were visually rated as ‘AD’ or ‘normal’. Data were compared with ratings of [F]FDG PET images and MRI derived medial temporal lobe atrophy (MTA) scores. Inter-rater agreement and agreement with clinical diagnosis was assessed for all imaging modalities. In addition, cut-off values for quantitative global [C]PIB and [F]FDDNP BPND were determined. Visual ratings were compared with dichotomised quantitative values. Results: Agreement between readers was excellent for [C]PIB, [F]FDDNP and MTA (Cohen’s kappa κ≥0.85) and moderate for [F]FDG (κ=0.56). Highest sensitivity was found for [C]PIB and [F]FDG (both 1.0). Highest specificity was found for MTA (0.90) and [C]PIB (0.85). [F]FDDNP had lowest sensitivity and specificity (0.67 and 0.53, respectively). Cut-off for quantitative [C]PIB BPND was 0.54 (sensitivity and specificity both 0.95) and for [F]FDDNP BPND 0.07 (sensitivity 0.80, specificity 0.73). Agreement between quantitative analyses and visual ratings was excellent for [C]PIB (κ=0.85) and fair for [F]FDDNP (κ=0.40). Conclusion: Visual assessment of [C]PIB images was straightforward and accurate, showing promise as a supportive diagnostic marker for AD. Moreover, [C]PIB showed the best combination of sensitivity and specificity. Visual assessment of [F]FDDNP images was insufficient. Quantitative analysis of [F]FDDNP data showed considerable higher diagnostic value than visual analysis. pe er -0 05 57 42 4, v er si on 1 19 J an 2 01