Indexed on: 27 Sep '07Published on: 27 Sep '07Published in: Journal of Magnetic Resonance Imaging
To conduct a prospective comparison of the accuracy of short inversion time (TI) inversion-recovery (STIR) turbo spin-echo (SE) imaging and coregistered 2-[fluorine-18] fluoro-2-deoxy-D-glucose (FDG)-positron emission tomography (PET) with computed tomography (CT) (coregistered FDG-PET/CT) to assess the N-stage in non-small-cell lung cancer (NSCLC) patients.A total of 115 consecutive NSCLC patients prospectively underwent CT, STIR turbo SE imaging, and FDG-PET, as well as surgical and pathological examinations. All STIR turbo SE images were obtained with a 0.9% saline phantom, which was placed alongside the chest wall of each patient, and coregistered FDG-PET/CTs were reconstructed using commercially available software. For quantitative assessments, the ratio of signal intensity (SI) of each lymph node to that of 0.9% saline phantom (lymph node-saline ratio [LSR]) and maximal standardized uptake value (SUV(max)) of each lymph node were calculated. Feasible threshold values were determined by using the receiver operating characteristic (ROC) curve-based positive test, and diagnostic capabilities of N-stage were compared by McNemar's test on a per patient basis.When feasible, threshold values were adopted, quantitative sensitivity (90.1%) and accuracy (92.2%) of STIR turbo SE imaging were significantly higher than those of quantitative and qualitative sensitivities (76.7% and 74.4%) and accuracies (83.5% and 82.6%) of coregistered FDG-PET/CT on a per patient basis (P < 0.05).STIR turbo SE imaging is at least as valid as coregistered FDG-PET/CT for quantitative and qualitative assessment of the N-stage for NSCLC patients.