Effect of Segmental Bronchoalveolar Lavage on Quantitative Computed Tomography of the Lung

Abstract

With employment of both multidetector computed tomography (MDCT) and endobronchial procedures in multicenter studies, effects of timing of endobronchial procedures on quantitative imaging (Q-MDCT) metrics is a question of increasing importance. Six subjects were studied via MDCT at baseline, immediately following and at 4 hours and 24 hours post-bronchoalveolar lavage (BAL) (right middle lobe and lingula). Through quantitative image analysis, non-air, or "tissue" volume (TV) in each lung and lobe was recorded. Change in TV from baseline was used to infer retention and redistribution of lavage fluid. Bronchoscopist reported unrecovered BAL volume correlated well with Q-MDCT for whole lung measures, but less well with individual lobes indicating redistribution. TV in all lobes except the right lower lobe differed significantly (P < .05) from baseline immediately post lavage. At 24 hours, all lobes except the left lower lobe (small 1% mean difference at 24 hours) returned to baseline. These findings suggest fluid movement affecting Q-MDCT metrics between lobes and between lungs before eventual resolution, and preclude protocols involving the lavage of one lung and imaging of the other to avoid interactions. We demonstrate that Q-MDCT is sensitive to lavage fluid retention and redistribution, and endobronchial procedures should not precede Q-MDCT imaging by less than 24 hours.