Purpose To compare predictive capabilities of non-contrast-enhanced (CE)- and dynamic CE-perfusion MRIs, thin-section multidetector computed tomography (CT) (MDCT), and perfusion scan for postoperative lung function in non-small cell lung cancer (NSCLC) patients. Materials and Methods Sixty consecutive pathologically diagnosed NSCLC patients were included and prospectively underwent thin-section MDCT, non-CE-, and dynamic CE-perfusion MRIs and perfusion scan, and had their pre- and postoperative forced expiratory volume in one second (FEV1) measured. Postoperative percent FEV1 (po%FEV1) was then predicted from the fractional lung volume determined on semiquantitatively assessed non-CE- and dynamic CE-perfusion MRIs, from the functional lung volumes determined on quantitative CT, from the number of segments observed on qualitative CT, and from uptakes detected on perfusion scans within total and resected lungs. Predicted po%FEV1s were then correlated with actual po%FEV1s, which were %FEV1s measured postoperatively. The limits of agreement were also determined. Results All predicted po%FEV1s showed significant correlation (0.73-≤-r-≤-0.93, P-<-0.0001) and limits of agreement with actual po%FEV1 (non-CE-perfusion MRI: 0.3-±-10.0%, dynamic CE-perfusion MRI: 1.0-±-10.8%, perfusion scan: 2.2-±-14.1%, quantitative CT: 1.2-±-9.0%, qualitative CT: 1.5-±-10.2%). Conclusion Non-CE-perfusion MRI may be able to predict postoperative lung function more accurately than qualitatively assessed MDCT and perfusion scan.
All Science Journal Classification (ASJC) codes
- Radiology Nuclear Medicine and imaging