Bronchial wall thickness: appropriate window settings for thin-section CT and radiologic-anatomic correlation

Abstract

Purpose: To analyze the influence of computed tomographic (CT) window settings on bronchial wall thickness and to define appropriate window settings for its evaluation.

Materials and methods: Three inflation-fixed lungs were scanned with a section thickness of 1.5 mm by using a high-spatial-frequency algorithm. Wall thickness in 10 bronchial specimens was measured with planimetry. Window centers were altered in a range of -200 to -900 HU and window widths in a range of 400-1,500 HU. Relative and absolute differences between CT and planimetric values were calculated. CT and planimetric measures were correlated. Inter- and intraobserver variabilities were determined.

Results: Window widths less than 1,000 HU resulted in a substantial overestimation of bronchial wall thickness, whereas widths greater than 1,400 HU resulted in an underestimation of bronchial wall thickness. There was no interaction between "width" and "center" regarding their influence on bronchial walls (F = 0.23; P = .99). Correlation between CT and planimetry was statistically significant (r = .85; P = .0001). Differences between the two observers were not statistically significant; results of the measurements of the two observers correlated well (r = .97; P = .001).

Conclusion: Bronchial wall thickness on thin-section CT scans should be evaluated with window centers between -250 and -700 HU and with window widths greater than 1,000 HU. Other than window settings, notably window widths less than 1,000 HU, can lead to substantial artificial thickening of bronchial walls.