Computed Tomography Volumetrics for Size Matching in Lung Transplantation for Restrictive Disease

Prabhu et al. assessed CT-based volumetrics for donor-recipient lung size matching in restrictive lung disease patients undergoing transplantation. Using preoperative CT scans from 218 recipients, they compared traditional predicted total lung capacity (pTLC) ratios with a new CT volumetric (CTVol) ratio, calculated through two segmentation software: Aquarius iNtuition and AVIEW. Results showed AVIEW, using convolutional neural network algorithms, provided more accurate and consistent segmentation, particularly in fibrotic lungs, with lower absolute error rates compared to Aquarius, which uses threshold-based segmentation. Patients were categorized as undersized, reference, or oversized based on each ratio, and regression models assessed associations with primary graft dysfunction (PGD) and survival. Findings indicated that undersizing by CTVol was linked to lower severe PGD risk (OR: 0.42), while oversizing increased mortality risk (HR: 2.27). pTLC ratios, however, did not significantly predict outcomes. This study suggests CT volumetric matching using advanced segmentation may improve transplant outcomes, warranting broader validation.