Quantification of Pneumothorax Volume on Chest X-Ray: A More Accurate Index Based On Measurements Made With 3D-printed Models

J Thorac Imaging. 2022 Jul 1;37(4):262-268. doi: 10.1097/RTI.0000000000000636. Epub 2022 Jan 6.

Abstract

Purpose: Owing to the extent of lung collapse estimated on chest radiograph it is still the complementary test most commonly used in the management of patients with pneumothorax. There are several indices to assess the extent of lung collapse. The objective of this study was to develop a more accurate index, using the 3D printing technology.

Materials and methods: We created physical hemithorax models using 3D printing. In this way, we obtained simple radiographs of models for which the lung volume was known accurately. In the first part of the study, we estimated the intraobserver and interobserver agreement as well as the agreement between methods. We created 2 new indices and the results obtained with these; the Light index and the Collins method were compared with data on real lung volume loss using linear regression analysis and by calculating the coefficient of determination (r2). In the second part of the study, we validated the 4 equations, comparing the Light index, the Collins method, and the 2 new indices using regression analysis. For this analysis, we used STATA V14.

Results: Both intraobserver and interobserver agreements were very high (<0.9). The agreement between the Collins method and the Light index was poor, with a mean difference of 18.6%. The equation that best represented real lung collapse was the new equation 2.

Conclusions: This study demonstrates the poor agreement between the Light index and Collins method for measuring the extent of lung collapse in pneumothorax and proposes a more accurate equation for this measurement based on a simple chest radiograph.

MeSH terms

  • Humans
  • Observer Variation
  • Pneumothorax* / diagnostic imaging
  • Printing, Three-Dimensional
  • Pulmonary Atelectasis*
  • Radiography
  • Reproducibility of Results
  • X-Rays