Radiological imaging of pulmonary emphysema : Preoperative evaluation of candidates for lung volume reduction surgery

Sammanfattning: Lung volume reduction surgery (LVRS) for pulmonary emphysema, first described by Brantigan in 1957, was re-introduced by Cooper et al in 1995. From a surgical point of view, information on the extent and distribution of emphysema is mandatory and also partly determines the outcome. No consensus exists on the best method for describing emphysema severity and heterogeneity in candidates for LVRS. The aims of this thesis were to define a suitable CT-technique for visual evaluation of severe emphysema, to design an objective model for classification of emphysema heterogeneity and to evaluate the additional information gained by lung perfusion scintigraphy in classification of emphysema. Study I investigated whether radiologists in evaluating severe emphysema subjectively preferred high resolution computed tomography (HRCT) or spiral computed tomography (CT). HRCT was preferred in 56%, spiral CT in 19% and in 25%, the techniques were considered equal. Study II investigated which of three types of images yielded the best results in estimating the degree of emphysema. In the grading of emphysema, the radiologists performed similarly with both HRCT and spiral CT, while "density-masked" images (an image of a quantitative measurement) gave significantly better results. Study III investigated whether spiral CT was superior to HRCT in evaluating the heterogeneity of emphysema and whether the combination of both CT-techniques improved the evaluation. Spiral CT was significantly better compared to HRCT, while the combination of the techniques did not improve the evaluation. Study IV presented a method for classification of emphysema heterogeneity that was calculated objectively, surgically oriented and classifies each lung separately. Study V investigated whether lung perfusion scintigraphy (LPS) contributes to the preoperative classification of emphysema heterogeneity compared to classification based only on CT. The answer was yes, and hence both CT and LPS should be included when assessing emphysema heterogeneity prior to LVRS. To conclude, in contrast to subjective preferences, visual evaluation of emphysema should be based on spiral CT (10 mm slice thickness and reconstruction with a high spatial algorithm) instead of HRCT in potential candidates for LVRS. The presentation of images as "densitymasked" images improves the visual evaluation of degree of emphysema. This type of imaging can easily be performed on all CT scanners and can thus be recommended as part of a complete CT examination. The addition of lung perfusion scintigraphy to spiral CT of the lungs improves the classification of emphysema heterogeneity in candidates for LVRS. A method for objective classification of emphysema heterogeneity has been developed.