Coincidence detection FDG PET in the management of oncological patients: attenuation correction versus non-attenuation correction.

The aim of this study was to determine if attenuation correction (AC) in a dual-head, coincidence, positron emission tomography imaging system (Co-PET) improved image quality, lesion detection, patient staging and management of various malignant neoplasms, compared to non-attenuation corrected (NAC) images. Thirty patients with known or suspected malignant neoplasms underwent fluorodeoxyglucose (FDG) Co-PET, which was correlated with histopathology, computed tomography (CT) and other conventional imaging modalities and clinical follow-up. The number and location of FDG avid lesions detected on the AC images and NAC Co-PET images were blindly assessed by two independent observers. Semi-quantitative grading of image clarity and lesion-to-background quality was performed. This revealed markedly improved image clarity and lesion-to-background quality in the AC versus NAC Co-PET images. AC Co-PET was statistically superior to NAC Co-PET in relation to lesion detection (P<0.01) and tumour staging (P<0.01). NAC Co-PET demonstrated 51 of the 65 lesions (78%) detected by AC Co-PET. AC Co-PET altered tumour staging in five additional patients (16%) compared to NAC Co-PET. Management was altered in two of these five patients.