Is imaging the extremities with PEM feasible? A novel application for a high-resolution positron emission scanner.

INTRODUCTION: Positron emission mammography (PEM) has better spatial resolution than positron emission tomography/computed tomography (PET/CT), or PET/CT. We evaluated the feasibility of extremity imaging with PEM using PET as a standard. METHODS/MATERIALS: Fourteen patients underwent sequential PET/CT and PEM. RESULTS/DISCUSSION: PEM visualized with equal or improved resolution all of the lesions identified on PET/CT. It often provided additional information such improved uptake localization and also visualized activity in an adjacent structures that was not seen on PET/CT or magnetic resonance imaging. We believe PEM can image the extremities in diseases like melanoma, arthritis and osteomyelitis and patients with metallic hardware.

Correlation of PUV and SUV in the extremities while using PEM as a high-resolution positron emission scanner.

OBJECTIVE: Owing to its unique configuration of two adjustable plate detectors positron emission mammography, or PEM, could theoretically also function as a high-resolution positron emission scanner for the extremities or neck. PEM quantitates its activity via a "PEM uptake value," or PUV, and although its relationship to the standardized uptake value, or SUV, has been demonstrated in the breasts, to our knowledge there are no studies validating PUV in other sites such as the extremities. MATERIALS AND METHODS: This was a retrospective chart review of two separate protocols of a total of 15 patients. The patients all had hypermetabolic lesions in the extremities or neck on imaging with PET/CT and were sent after their PET/CT to PEM for further imaging. Owing to the sequential nature of these examinations no additional radiotracer was administered. RESULTS: Spearman's rank order correlation was calculated between the PUVmax obtained from PEM images, and the SUVmax for all. Spearman's rank order correlation for all sites was 0.42, which is not significantly different from 0 (p = 0.13). When neck lesions were excluded from the group, there was a strong and statistically significant correlation between PUVmax and SUVmax, with Spearman's rank correlation of 0.73, and significantly different from 0 (p = 0.0068). DISCUSSION: The correlation of PUV and SUV in the extremities indicates the potential use of PEM as a semiquantitative, high-resolution positron emission scanner and warrants further investigation, especially in the realms of disease processes that often present in the extremities, such as melanoma, osteomyelitis, and arthritis, as well as playing a role in the imaging of patients with metallic hardware post-limb salvage surgery.