Abstract
1522
Purpose: The skin is the largest organ in the human body and has the function of protection and sensation as well as synthesis and excretion. Glucose metabolism in the skin could be evaluated by F-18 fluorodeoxyglucose (FDG) positron emission tomography (PET); however, limitation in the spatial resolution had hampered this usage. Our hypothesis was that recently developed PET detector with silicon photomultiplier (SiPM) could yield accurate measurement of FDG uptake in the thin structure, and its combination with simultaneously acquired magnetic resonance imaging (MRI) enabled accurate segmentation of the skin in the whole body. Thus, the purpose of this study was to assess the repeatability of the quantitatively assessed physiological FDG uptake of the skin on PET/MRI and explore its regional distribution and relationship with gender and age. MATERIALS AND METHODS: Two hundred and twenty-five patients (112 men and 113 women; mean age, 63.2±16.3 y.o.; range, 0.5 to 94 y.o.) who underwent whole body FDG PET/MRI (Signa PET/MR, GE Healthcare) were retrospectively enrolled in this study. Repeated examinations within one year without medical interventions were performed in 32 out of 112 subjects, and were evaluated for the reproducibility assessment. The skin segmentation on PET was performed by three steps as follows: First, body surface contouring on MRI based attenuation correction (AC) map in vivo (without bed and coils) generated by the off-line PET reconstruction software (PET toolbox, GE healthcare) using a 2-point Dixon MR sequence; second, body surface extracting in three voxels thick in the MRAC on the commercially available workstation to create a region of interest (ROI) of the skin; and third, copying the ROI to the corresponding PET to evaluate the skin standardized uptake value (SUV) in the following six regions: overall, forehead, scalp, chest, back, and abdomen. To assess the repeatability of FDG uptake of the skin, the Bland and Altman plot was created for the mean SUV of the skin in the overall region and was compared with that of the liver. To evaluate the relationship between the skin maximum SUV (SUVmax) and regional distribution, as well as age and gender, a one-way ANOVA was performed.
Results: The coefficient of repeatability for the mean SUV of the overall skin region (0.052; 95% CI, 0.042 to 0.067) was better than that of the liver (0.6415; 95%CI, 0.5230 to 0.8299) with statistically significant difference (p<0.0001). The highest mean SUVmax was observed in the forehead (3.09±1.04) and was followed by the scalp (2.07±0.53) region. The mean SUVmax in the forehead in men (3.45±1.48) was significantly higher than that in women (2.88±1.12) (p=0.0019). The forehead mean SUVmax in men with less than the age of twenty years (1.76±0.93) and that in women with less than ten years (0.98±0.38) were significantly lower than those with the ages and over (3.94±1.23 and 3.00±1.04, respectively) (p<0.001). In women with the age group of 70 years and above, the SUVmax in the forehead (2.32±0.71) was significantly lower than those in their thirties (3.83±0.82) (p<0.05).
Conclusions: PET/MRI with SiPM enables quantitative analysis of FDG uptake of the human skin with a high repeatability. The degree of physiological FDG uptake of the skin is highest in the forehead and varies among gender and age.