Detection Rate, Location, and Size of Pulmonary Nodules in Trimodality PET/CT-MR: Comparison of Low-Dose CT and Dixon-Based MR Imaging

OBJECTIVE: The objective of this study was to prospectively compare the detection rate, the location, and the size of pulmonary nodules in low-dose computed tomography (CT) and in magnetic resonance (MR) imaging with a 3-dimensional (3D) dual-echo gradient-echo (GRE) pulse sequence using a trimodality positron emission tomography (PET)/CT-MR setup. METHODS: Forty consecutive patients (25 men and 15 women; mean [SD] age 64 [12] years) referred for staging of malignancy were prospectively included in this single-center, Institutional Review Board-approved study. Imaging using trimodality PET/CT-MR setup (full ring, time-of-flight PET/CT and 3-T whole-body MR imager) comprised PET, low-dose CT for anatomic referencing/attenuation correction of PET, and MR imaging with 3D dual-echo GRE pulse sequence, allowing the reconstruction of water-only (WO) and in-phase (IP) images. Two blinded and independent readers assessed all images randomly for the presence, the location, and the size of pulmonary nodules. Detection rates, defined as the proportion of screened participants with at least 1 pulmonary nodule, were compared between low-dose CT and MR imaging including both WO and IP images. RESULTS: Inter-reader agreements were high regarding the location (k = 0.93-0.98) and the size of pulmonary nodules (intraclass correlation analysis = 0.94-0.98) in CT and in MR imaging. Computed tomographic scans revealed 66 pulmonary nodules in 34 of the 40 patients (85%), whereas WO and IP images showed 56 and 58 pulmonary nodules in 33 of the 40 patients (83%), respectively. The detection rates of CT and MR imaging were similar (P's > 0.05) regarding all nodules, 18F-Fluordesoxyglucose-positive pulmonary nodules, and 18F-Fluordesoxyglucose-negative pulmonary nodules. The size of pulmonary nodules was significantly smaller on WO (P < 0.05; mean difference, 3 mm; 95% confidence interval, - 13 to 18 mm) and IP images (P < 0.001; mean difference, 4 mm; 95% confidence interval, -5 to 12 mm) compared with in CT. CONCLUSIONS: Our study indicates that a 3D Dixon-based, dual-echo GRE pulse sequence might be suitable for lung imaging in clinical whole-body PET/MR examinations. Although the detection rates were lower, there was no statistically significant difference on a patient-based evaluation concerning detection rates of pulmonary nodules compared with low-dose CT. Assessment of nodule location can be performed equally well with MR imaging