Failure to prove asbestos exposure produces obstructive lung disease.

Ohar and colleagues (February 2004)1 analyzed the relationships among asbestos exposure, cigarette smoking, lung function, and chest radiograph abnormalities in 3,383 patients who had been referred primarily by plaintiff attorneys and trade union members. The authors confirmed, as had many others, that cigarette smoking is associated with a greater prevalence of parenchymal opacities on surveillance chest radiographs in asbestos-exposed workers. The important observation was that asbestos exposure causes clinically significant air-flow obstruction as well as restrictive changes in lung function, which were the result of a synergistic effect between cigarette smoking and asbestos exposure. This study is important because the authors were able to obtain lung volume measurements in those patients with abnormal spirometry findings. The majority of the patients with mixed restrictive/obstructive airflow obstruction (73%) had a normal total lung capacity. Only 27% of these patients had coexistent restriction. Later on in the article, the authors stated that 83% of smokers and 71% of nonsmokers with obstruction and a reduced FVC had hyperinflation. Another contradiction later appears in Figure 2 (page 749) concerning the number of patients with obstruction but no restriction. The value n 5,381 or 16.0% referenced in Figure 2 is incorrect (16% of 3,312 is 530, not 5,381). The authors stated that among nonsmokers the FEV1/FVC ratio was 72.8% in those patients with a high International Labor Office (ILO) score (using the ILO scoring system for reading pneumoconiosis chest radiographs), as opposed to an FEV1/FVC ratio of 76.2% in those with low ILO scores (p 0.025 [with Bonferroni correction]). This statistical method has been labeled as imprecise by Rothman and Greenland,2 and it is not clear that there truly is a statistically significant difference between populations. Lung volume values with diffusion measurements are not presented in this group of study results, which makes this information very difficult to understand. What about the diffusing capacity of the lung for carbon monoxide? It is hard to believe that lung volumes were measured without measuring lung diffusion. Certainly, some of these patients might have emphysema that would not have been identified unless diffusion studies had been performed. No information is given about the cause of restrictive disease (ie, diffuse pleural disease vs asbestosis) in this cohort. The total lung capacity should have been used for making clinical correlations instead of the FVC 80% with an FEV1/ FVC ratio of 70%. The use of ILO scoring is fraught with error3 and confounding by cigarette smoking, as is pointed out by the authors. The accuracy of determining the smoking history is critical when evaluating a patient population for obstructive lung disease. Patients with higher ILO perfusion levels found on a chest radiograph are more likely to minimize or deny cigarette smoking when seeking compensation. In a population of patients that has been referred largely by plaintiff attorneys, there is a tendency to exaggerate asbestos exposure and minimize their smoking history. As well, the bibliography is selective, and the discussion fails to reference articles that are in disagreement with the authors conclusions.4,5 I can certainly sympathize with the authors’ struggles to make clinical, physiologic, and radiologic correlations from a very large database. Unfortunately, this article raises more questions than it provides answers, and it fails to prove the authors’ hypothesis that asbestos exposure produces obstructive lung disease.