The reproducibility of deep-inspiration breath-hold (18)F-FDG PET/CT technique in diagnosing various cancers affected by respiratory motion.

ABSTRACT

BACKGROUND: The deep-inspiration breath-hold PET/CT (DIBH PET/CT) technique improves the limitations of diagnosing a lesion located in an area influenced by respiratory motion that results in spatial misregistrations caused by respiration between PET and CT. However, its reproducibility with regard to calculating the maximal standardized uptake value (SUV(max)) and metabolic volume (MV) in DIBH PET/CT has not been elucidated. OBJECTIVE: The purpose of this study was to investigate the reproducibility of the DIBH PET/CT technique including calculating the SUV(max) and the MV. METHODS: Sixty patients with various cancers were enrolled. The subjects had 47 abdominal lesions and 28 chest lesions. All patients demonstrated a misregistered image in the early whole-body image taken 50 min after FDG intravenous infusions. We added the delayed spot images 40 min after the first image. On the delayed image, we performed both conventional techniques with non-breath-hold (NBH) and the DIBH technique. In the four times DIBH technique, we obtained the coefficient of variance (CV) in calculating these indices for evaluating reproducibility. RESULTS: The SUV(max) value with DIBH showed an increase of 16.1-60.1% compared with that measured by NBH. The mean value of CV was 5.5 in thoracic lesions and 6.3 in abdominal lesions. The values of MV with DIBH showed a decrease of 14.0-20.1% compared with those measured by NBH. Regarding reproducibility, mean value of CV was 7.1 in thoracic lesions and 11.9 in abdominal lesions. CONCLUSION: The DIBH technique improves the inaccurate quantification of both SUV(max) and MV. Although the CV value of SUV(max) with DIBH technique is better in thoracic lesions compared with that in abdominal lesions, the reproducibility was acceptable.