Comparison of Pi10 and Pi-Slope Calculation Methods and Association With Lung Function: Findings from the Canadian Chronic Obstructive Lung Disease (CanCOLD) Study

Abstract Introduction: The average wall thickness of a theoretical airway with a lumen perimeter of 10 mm (Pi10) measured using computed tomography (CT) images is a biomarker for airway remodelling in Chronic Obstructive Pulmonary Disease (COPD). Several methods have been used in the literature to calculate Pi10 leading to significant variability across studies. The objective of this study was to evaluate the consistency between Pi10 calculation methods and their association with lung function. A secondary objective was to extract a parameter called Pi-Slope, investigating whether Pi-Slope provides improved association with lung function.Methods: Participants from CanCOLD were used to calculate Pi10 and Pi-Slope. CT images were acquired at full-inspiration and airway segmentation was performed by VIDA Diagnostics Inc. Luminal perimeter and airway wall thickness were quantified using ten methods from the literature (Table 1). Pi10 was derived by plotting perimeter against the square root of the wall area using linear regression, where the slope represented Pi-Slope. The pairwise Intraclass Correlation Coefficient (ICC) assessed consistency of Pi10 and Pi-Slope between methods yielding 45 comparisons. ICC values were defined as excellent (>0.90), good (0.75-0.90), moderate (0.50-0.75), and poor (<0.50). Multivariate regression models assessed associations for Pi10 and Pi-Slope with Forced Expiratory Volume in 1 second (FEV1) and FEV1/Forced Vital Capacity (FVC), adjusted for age, sex, BMI, smoking status, pack-year, total lung capacity, and CT scanner. Statistical significance was defined using P<0.05.Results: A total of 1,351 participants with and without COPD were evaluated. The pairwise ICC results demonstrated excellent consistency for Pi10 and Pi-Slope across four methods (Patel, Nakano, Jobst, and Bhatt). In contrast, Pi10 from the Park's method showed moderate consistency and Pi10 from Gietema and Telenga's methods displayed poor consistency with those four consistent methods. For Pi-Slope, Park and Telenga's methods exhibited poor consistency and Gietema's method showed moderate consistency with the four consistent methods. Interestingly, these three methods (Gietema, Park, and Telenga) had significantly fewer CT segmented airways than the other methods (p<0.001). In multivariable analyses, five methods demonstrated negative and significant associations between Pi10 and lung function (p<0.05), while the others did not (Table 1). Pi-Slope showed significant associations with lung function across all methods.Conclusion: This study found excellent consistency among Pi10 methods that included a greater number of airways, while those incorporating fewer airways showed greater variability. Additionally, Pi-Slope demonstrated improved association with lung function assessment compared to Pi10 particularly in those methods that included fewer airways.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.