Diffusion-weighted Whole-body MRI at 3 Tesla for the Detection and Discrimination of Pulmonary Tumors

Abstract

Objective: To evaluate diffusion-weighted whole-body MRI with background body signal suppression (DWIBS) at 3.0 T for pulmonary lesion detection and characterization. Materials and Methods: 19 patients with 25 pulmonary lesions were examined with DWIBS using 2 b-values (b = 0 and 1000 s/mm²) and partly additionally with DWIBS using 3 b-values (b = 0, 50, 1000 s/mm²). DWIBS was compared to FDG PET. For characterization of hyperintense lesions by DWIBS, Lesion-to-Spinal cord Ratio (LSR) and apparent diffusion coefficient ADC were analyzed. From repeated measurements, the Coefficient of Variation (CV) was calculated. From 3-b-value data, the ADC(0,1000) and ADC(50,1000) values were compared in order to assess perfusion influences. Results: Sensitivity and specificity of detecting malignant lesions were comparable for DWIBS and FDG PET. Malignant compared to benign lesions had lower ADC(0,1000) and higher LSR values. CV of LSR was more than a factor of 8 higher than CV of ADC(0,1000) (23.9% vs 2.9%, P = 0.012). Perfusion effects were largest for metastases, medium for adenocarcinoma and benign lesions, and lowest for squamous cell carcinoma. Conclusion: DWIBS at 3.0 T is appropriate for lesion detection and characterization. ADC analysis is superior to signal intensity ratio determination with respect to repeatability. An analysis of perfusion influences provides additional information.