Multi-bed-position acquisition technique for deep inspiration breath-hold PET/CT: a preliminary result for pulmonary lesions.

OBJECTIVE

The deep inspiration breath-hold (DIBH) technique for positron emission tomography/computed tomography (PET/CT) is under investigation for its contribution to the accurate diagnosis of pulmonary lesions. "Step and shoot" or multi-bed-position image acquisition is necessary to improve the accuracy of whole-lung evaluation. The purpose of this study was to describe the method and preliminary results of evaluating pulmonary lesions using multi-bed-position deep inspiration breath-hold (MDIBH) PET/CT.

METHODS

Thirteen patients with a total of 32 metastatic pulmonary lesions underwent both whole-body free-breath (FB) and whole-lung MDIBH-PET/CT sessions with suitable axial slice overlap. The self-breath holding technique was used for reproducibility of the DIBH condition. The standard FB-PET/CT was performed under the FB condition, which was followed by the MDIBH-PET/CT performed under the DIBH condition. Accuracy of alignment between CT and PET images and CT image quality were evaluated independently on dependent density, motion artifact, and heterogeneity attenuation. Mean standardized uptake value (SUV) in normal lung [Background (BG)-SUV(mean)], maximum SUV (SUV(max)) of lesion (lesion-SUV(max)), tumor background ratios (TBRs), and uptake volumes (UVs) were evaluated quantitatively.

RESULTS

Improvements in artifacts were statistically significant using MDIBH-PET/CT. Improvements in alignment were statistically significant with the MDIBH-PET/CT for the diaphragm, heart and lung apices. CT image quality was statistically significantly higher with the MDIBH-PET/CT than with the FB-PET/CT in all indices. The decreases in BG-SUV(mean) were statistically significant (in all patients) with an average of -37%. Lesion-SUV(max) was increased in 7 of 32 (22%) lesions although average lesion-SUV(max) showed no statistical difference between the FB- and the MDIBH-PET/CT images. The increase in TBRs was statistically significant in 31 of 32 lesions (97%) in the MDIBH-PET/CT with an average of 57%. UVs were lower in 23 of 32 lesions (72%) in the MDIBH-PET/CT by -12% on average, although no statistical difference was confirmed between the techniques.

CONCLUSIONS

The MDIBH-PET/CT can provide better-aligned fused images, featuring superior image quality, in both PET and CT images. The PET images showed low BG, non-blurring and high TBRs, and the CT images provided diagnostic capability of detecting small pulmonary lesions with negligible radiation exposure.