Clinical feasibility and impact of fully automated multiparametric PET imaging using direct Patlak reconstruction: evaluation of 103 dynamic whole-body ¹⁸F-FDG PET/CT scans.

Abstract

Functional imaging by standard whole-body (WB) ¹⁸F-flurodeoxyglucose (FDG) positron emission tomography (PET) is an integrated part of disease diagnostics. Recently, a clinical dynamic whole-body (D-WB) FDG PET/CT scanning protocols has been developed allowing for quantitative imaging of tissue metabolic rate of FDG (MR_FDG). It was the purpose of this retrospective study to evaluate whether MR_FDG imaging is feasible in a clinical setting and whether it improves lesion detectability. One hundred nine patients representing a broad range of referral indications for FDG PET/CT were invited to undergo a D-WB FDG PET/CT scan. Two sets of images were produced: parametric images and standard static SUV images. Both sets of images were reviewed visually, and 310 individual lesions were quantitatively analysed using the target-to-background (TBR) and contrast-to-noise (CNR) metrics. One hundred three out of 109 patients completed the D-WB FDG PET/CT scan. There was no difference in the number of pathological lesions identified visually on the MR_FDG and the SUV images, whereas MR_FDG images yielded 4 fewer false positives than the SUV images. Quantitatively, MR_FDG TBR was significantly higher than SUV TBR in 299/310 lesions, and better MR_FDG CNR was found to facilitate the challenging reading of lesions with low SUV TBR. D-WB FDG PET/CT is feasible in a clinical setting and produces MR_FDG images of good visual quality and superior lesion contrast. In addition, MR_FDG images complement the standard SUV images providing better quantification and enhanced image reading. However, although MR_FDG also reduced the number of false-positive findings, no additional malignant lesions were identified. The technique therefore appears to be best suited for select patient groups or possibly treatment response evaluation.