Chest HRCT signs predict deaths in long-term follow-up among asbestos exposed workers

Our plain-English summary of Chest HRCT signs predict deaths in long-term follow-up among asbestos exposed workers, by T. Vehmas, P. Oksa (European Journal of Radiology, 2014).

The study

• Authors: T. Vehmas, P. Oksa
• Journal: European Journal of Radiology
• Published: 2014;83(10):1983-7
• Cited by: 10 articles, as counted by NIH iCite
• Read it: PubMed · Publisher

Abstract

Objectives: To study associations between chest HRCT signs and subsequent deaths in long-term follow-up.

Methods: Lung and pleural signs of 633 asbestos exposed workers (age 45-86, mean 65) screened with HRCT were recorded by using the International Classification of Occupational and Environmental Respiratory Diseases (ICOERD) system, which contains detailed instructions for use and reference images. Subsequent mortality was checked from the national register. Cox regression adjusted for covariates (age, sex, BMI, asbestos exposure, pack-years) was used to explore the relations between HRCT signs and all-cause deaths, cardiovascular and benign respiratory deaths, and deaths from neoplasms - all according to the ICD-10 diagnostic system.

Results: The follow-up totalled 5271.9 person-years (mean 8.3 y/person, range .04-10.3). 119 deaths were reported. Irregular/linear opacities, honeycombing, emphysema, large opacities, visceral pleural abnormalities and bronchial wall thickening were all significantly related to all-cause deaths. Most of these signs were associated also with deaths from neoplasms and benign respiratory disease. Deaths from cardiovascular disease were predicted by emphysema and visceral pleural abnormalities.

Conclusions: Several HRCT signs predicted deaths. Careful attention should be paid on subjects with radiological signs predictive of deaths and new secondary preventive strategies developed. This calls for further focused studies among different populations.

Copyright © 2014 Elsevier Ireland Ltd. All rights reserved.

In plain English

This Finnish study examined whether signs on high-resolution chest CT (HRCT) scans of 633 asbestos-exposed workers (627 men, 6 women, aged 45 to 86) were linked to later deaths. The scans, taken mostly in 2004, were read using the ICOERD system. Deaths were traced in the national register until 23 August 2013. Analyses were adjusted for age, sex, body mass index, smoking pack-years and an asbestos exposure index. During 5271.9 person-years of follow-up, 119 workers died. Six signs were significantly related to all-cause death: irregular or linear opacities, honeycombing, emphysema, large opacities, visceral pleural abnormalities and bronchial wall thickening. Large opacities had the highest significant hazard ratio (3.165, 95% CI 1.772 to 5.654). Most of these signs were also linked to deaths from neoplasms and benign lung disease. Emphysema and visceral pleural abnormalities were linked to cardiovascular deaths. Suspicion of cancer on CT was not a statistically significant predictor of death in this group, which the authors linked partly to the few confirmed lung cancers. The authors concluded that the images hold death-risk information independent of several other conventional risk factors. They noted the workers were not a random sample.

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: Follow-up (cohort) study linking HRCT findings to later deaths from the national death register, analysed with Cox regression
• Who was studied: 633 asbestos-exposed workers in Finland (627 men, 6 women), aged 45 to 86 (mean 65). They were drawn from the Finnish Institute of Occupational Health asbestos screening program and from occupational wards in Helsinki and Tampere. Of the 633, 11 were imaged during 2003, 619 during 2004 and 3 during 2005.
• Exposure: Occupational asbestos exposure, measured by an asbestos exposure index (available for 632 workers) ranging from 0 to 286, mean 75. No fiber types are specified. The findings studied were HRCT signs classified with the ICOERD system.
• Main results:
  • Over 5271.9 person-years of follow-up (mean 8.3 years per person), 119 deaths occurred among the 633 workers.
  • Causes of the 119 deaths: 38 cardiovascular, 41 neoplasms (19 of them lung cancers or pleural mesotheliomas), 19 benign lung diseases, 18 other causes, and 3 recent deaths with no cause yet available.
  • Irregular/linear opacities, honeycombing, emphysema, large opacities, visceral pleural abnormalities and bronchial wall thickening were significantly related to all-cause deaths, after adjustment for age, sex, BMI, pack-years and asbestos exposure index.
  • For all-cause deaths, large opacities had the highest significant hazard ratio: 3.165 (95% CI 1.772 to 5.654).
  • All-cause death hazard ratios were 2.235 (1.494 to 3.344) for visceral pleural abnormalities and 2.023 (1.351 to 3.030) for bronchial wall thickening.
  • Among the scored (ordinal) signs, honeycombing had the highest all-cause death hazard ratio: 1.493 (1.241 to 1.796).
  • Most of these signs were also related to deaths from neoplasms and benign lung disease. Emphysema and visceral pleural abnormalities also predicted cardiovascular deaths.
  • For deaths from benign respiratory disease, the hazard ratio for large opacities was 10.498 (95% CI 3.610 to 30.529).
  • Suspicion of cancer on CT was not a statistically significant predictor of all-cause death in this group. The hazard ratio was 1.796 (95% CI 0.901 to 3.538), with a p-value of 0.096.
  • The authors attributed the lack of a predictive finding for cancer suspicion partly to the small number of true lung cancers, the low specificity of the finding, and HRCT leaving much of the lungs unstudied.
• The authors' conclusion: The authors concluded that radiological images seem to hold information on death risk that is independent of several other conventional risk factors. They called for more similar follow-up studies in other risk groups.
• Limitations the authors state:
  • The workers were not a random sample of Finnish asbestos-exposed people. They were selected by exposure and smoking habits and were partly a convenience sample.
  • The sample was almost entirely men, so applying the results to women is questionable.
  • Fourteen workers had an asbestos index of zero although all were considered exposed, suggesting the index may not capture all exposure.
  • Compared with the authors' earlier study, the follow-up was shorter, giving fewer deaths and less statistical power.
  • Because of register delays, some recent deaths may be missing or lack a cause. The authors considered this unlikely to bias the results.
  • Other exposures, such as other dusts, fumes or secondary smoking, could not be controlled in the analyses.

Topics

The National Library of Medicine indexes this study under subjects our site covers: Asbestosis.

Medical terms in this study, explained in our Lexicon: High-Resolution CT (HRCT), Asbestos Exposure, Neoplasm, Honeycombing, Emphysema.

Related studies

• Lung and pleural CT signs predict deaths: 10-year follow-up after lung cancer screening of asbestos-exposed workers — T. Vehmas, P. Oksa, et al. (2012)
• Scoring CT/HRCT findings among asbestos-exposed workers: effects of patient's age, body mass index and common laboratory test results — T. Vehmas, L. Kivisaari, et al. (2005)
• Lung function predicts mortality: 10-year follow-up after lung cancer screening among asbestos-exposed workers — T. Vehmas, P. Pallasaho, et al. (2013)
• Visually scored calcifications in thoracic arteries predict death: follow-up study after lung cancer CT screening — T. Vehmas (2012)
• The effects of secondhand smoke exposure on HRCT findings among asbestos-exposed workers — T. Vierikko, R. Järvenpää, et al. (2008)
• Long-term results of chest wall reconstruction with DualMesh — T. Nagayasu, N. Yamasaki, et al. (2010)

Browse more research in European Journal of Radiology, or return to medical research.