• C H Park, T H Kim, S Lee, H C Paik, and S J Haam.
• Department of Radiology and the Research Institute of Radiological Science, Yonsei University Health System, Seoul, Republic of Korea.
• Transplant. Proc. 2015 Mar 1; 47 (2): 498-503.

PurposeLung size matching is important in lung transplantation (LT). With advances in computed tomography (CT) technology, multidetector row CT can accurately measure the thoracic cage and lung volumes. The objective of this study was to generate a new regression equation using demographic data based on the measured CT lung volume in a healthy population to predict the CT lung volume of the donor in LT size matching.Materials And MethodsThe medical records of healthy subjects who underwent chest CT scans to screen for lung cancer were retrospectively reviewed. CT lung volume was semi-automatically measured using a threshold-based auto-segmentation technique. New regression equations for CT lung volume were generated by multiple linear regression analysis using demographic data including height (H, cm), weight (W, kg), and age (A, years). The percentage error rate (%) of the equations were calculated as ([Estimated CT lung volume--Measured CT lung volume]/Measured CT lung volume × 100). A percentage error rate within ± 20% was considered acceptable.ResultsA total of 141 men aged 27 to 55 years (mean, 46.7 ± 6.2 years) and 128 women aged 20 to 55 years (mean, 45.4 ± 7.2 years) were enrolled. The final regression equations for CT lung volume were (-5.890 + 0.067 H - 0.030 W + 0.020 A) in men and (-6.698 + 0.072 H - 0.024 W) in women. The mean absolute error rate was 10.9 ± 9.0% and 11.0 ± 8.5% in men and women, respectively. Percentage error rates were within ± 20% in 121 of 141 (85.8%) men and 113 of 128 (88.3%) women.ConclusionThese equations could predict the CT lung volume of healthy subjects using demographic data. Using these equations, the predicted CT lung volume of donors could be matched to the measured CT lung volume of recipients in lung transplantation.Copyright © 2015 Elsevier Inc. All rights reserved.

23 keywords...

hide…