The impact of lung parenchyma attenuation on nodule volumetry in lung cancer screening

Our plain-English summary of The impact of lung parenchyma attenuation on nodule volumetry in lung cancer screening, by D. Penha, E. Pinto, B. Hochhegger, et al (Insights Into Imaging, 2021).

The study

• Authors: D. Penha, E. Pinto, B. Hochhegger, C. Monaghan, E. Marchiori, L. Taborda-Barata, K. Irion
• Journal: Insights Into Imaging
• Published: 2021;12(1):84
• Cited by: 5 articles, as counted by NIH iCite
• Read it: PubMed · Publisher · Free full text

Abstract

Background: Recent recommendations for lung nodule management include volumetric analysis using tools that present intrinsic measurement variability, with possible impacts on clinical decisions and patient safety. This study was conducted to evaluate whether changes in the attenuation of the lung parenchyma adjacent to a nodule affect the performance of nodule segmentation using computed tomography (CT) studies and volumetric tools.

Methods: Two radiologists retrospectively applied two commercially available volumetric tools for the assessment of lung nodules with diameters of 5-8 mm detected by low-dose chest CT during a lung cancer screening program. The radiologists recorded the success and adequacy of nodule segmentation, nodule volume, manually and automatically (or semi-automatically) obtained long- and short-axis measurements, mean attenuation of adjacent lung parenchyma, and presence of interstitial lung abnormalities or disease, emphysema, pleural plaques, and linear atelectasis. Regression analysis was performed to identify predictors of good nodule segmentation using the volumetric tools. Interobserver and intersoftware agreement on good nodule segmentation was assessed using the intraclass correlation coefficient.

Results: In total, data on 1265 nodules (mean patient age, 68.3 ± 5.1 years; 70.2% male) were included in the study. In the regression model, attenuation of the adjacent lung parenchyma was highly significant (odds ratio 0.987, p < 0.001), with a large effect size. Interobserver and intersoftware agreement on good segmentation was good, although one software package performed better and measurements differed consistently between software packages.

Conclusion: For lung nodules with diameters of 5-8 mm, the likelihood of good segmentation declines with increasing attenuation of the adjacent parenchyma.

In plain English

This retrospective study asked whether the density (attenuation) of lung tissue around small nodules affects how well software outlines them on low-dose CT scans. Two radiologists applied two commercial software tools to 1265 solid nodules, 5 to 8 mm across, in 514 patients from a lung cancer screening program in Northeastern England. Readers judged segmentation proper more often with tool 2 (88.1% and 88.4%) than with tool 1 (84.8% and 83.8%). In a regression analysis, the odds of proper segmentation fell by a factor of 0.987 (95% CI 0.985 to 0.988) for each one-Hounsfield-unit increase in the attenuation of the surrounding lung. This factor had the largest effect size. Calcified pleural plaques were linked to lower odds (odds ratio 0.593, 95% CI 0.414 to 0.849) and previous lung surgery to higher odds, but the authors described both effects as negligible. Interstitial lung abnormalities or disease and emphysema were not statistically significant predictors. The study did not examine asbestos exposure. The authors concluded that higher surrounding-lung attenuation was related to poorer segmentation. They noted that their attenuation measure was nonstandard.

We wrote this plain-English summary from the full article with the help of AI, and a second AI model checked each sentence against the article. It is not the authors' wording; the abstract and the article are their own account.

What the study did and found

• Design: Retrospective cross-sectional (observational) study comparing two commercial CT volumetric software tools, each used by two radiologists
• Who was studied: 1265 solid lung nodules (5 to 8 mm) in 514 patients in a lung cancer screening program at a tertiary hospital in Northeastern England, enrolled August 2016 to December 2018. This gave 5060 measurements, 1265 per reader per software tool. Mean age was 68.3 years and 70.2% were male.
• Exposure: No asbestos exposure was studied. The study examined CT image factors: the mean attenuation of lung tissue in a rim about 5 mm thick around each nodule, nodule size and volume, reader, software, and the presence of interstitial lung abnormalities or disease, emphysema, pleural plaques and linear atelectasis.
• Main results:
  • The odds of proper segmentation fell by a factor of 0.987 (95% CI 0.985 to 0.988) for each one-Hounsfield-unit increase in the mean attenuation of the adjacent lung tissue.
  • Of all the predictors, adjacent lung attenuation had the largest effect size (0.195), followed by nodule volume (0.033).
  • Proper segmentation was recorded with tool 2 in 88.1% (reader 1) and 88.4% (reader 2) of nodules, and with tool 1 in 84.8% (reader 1) and 83.8% (reader 2).
  • The odds of proper segmentation rose by a factor of 1.558 (95% CI 1.350 to 1.797) with each 1-mm increase in the average of the manually measured long and short diameters.
  • The odds of proper segmentation fell by a factor of 0.984 (95% CI 0.982 to 0.986) with each unit increase in software-measured nodule volume.
  • The odds of proper segmentation were lower when calcified pleural plaques were present: odds ratio 0.593 (95% CI 0.414 to 0.849), p = 0.004. The authors described this effect size as negligible.
  • The odds of proper segmentation were higher with a previous history of lung surgery: odds ratio 3.414 (95% CI 1.575 to 7.401). The authors described this effect size as negligible.
  • No other variable significantly predicted proper segmentation. This included reader, software, age, sex, interstitial lung abnormalities or disease, emphysema, COPD, tuberculosis and linear atelectasis.
  • Agreement between the two software tools was very high (whole-sample intraclass correlation coefficient 0.905). Agreement between the two readers was reasonable (whole-sample 0.745).
• The authors' conclusion: The authors concluded that, for nodules 5 to 8 mm in long-axis diameter, higher attenuation of the adjacent lung tissue is related to poorer-quality segmentation by volumetric tools, contributing to measurement error. They advised care when interpreting automatic measurements in lungs with abnormal attenuation.
• Limitations the authors state:
  • The measure of lung tissue attenuation was nonstandard, chosen as a compromise because no standard exists.
  • There was no information on where focal changes, such as pleural plaques or post-surgical changes, lay relative to each nodule.
  • Few patients had previous lung surgery, and the effect size was negligible, so the authors cautioned against over-interpreting that finding.

Topics

Medical terms in this study, explained in our Lexicon: Parenchyma, Low-Dose CT Screening, Interstitial, Emphysema, Pleural Plaques, Atelectasis, Odds Ratio.

Related studies

• CT scan screening for lung cancer: risk factors for nodules and malignancy in a high-risk urban cohort — A.K. Greenberg, F. Lu, et al. (2012)
• Probability of cancer in lung nodules using sequential volumetric screening up to 12 months: the UKLS trial — M.W. Marcus, S.W. Duffy, et al. (2019)
• Impact of low-dose CT on lung cancer screening — S. Diederich, D. Wormanns (2004)
• Psychological impact of computed tomography screening for lung cancer and occupational pulmonary disease among asbestos-exposed workers — T. Vierikko, S. Kivistö, et al. (2009)
• Impact of low-dose computed tomography screening on lung cancer mortality among asbestos-exposed workers — F. Barbone, F. Barbiero, et al. (2018)
• Smoking, occupational risk factors, and bronchial tumor location: a possible impact for lung cancer computed tomography scan screening — M. Gonzalez, J.M. Vignaud, et al. (2012)

Browse more research in Insights Into Imaging, or return to medical research.