Pulmonary effects of exposure to fine fibreglass: irregular opacities and small airways obstruction.

Sir,-Kilburn et al (1992;49: 714-20) published a cross sectional prevalence study of workers exposed to fibreglass in an appliance manufacturing plant by means of chest x ray films and spirometry. They concluded that fibreglass used for insulating purposes seems to produce human disease similar to asbestosis. Because of the well known interobserver error in the reading of chest films, it is essential as a matter of good science that the investigation includes "blind" concurrent readings of films from unexposed controls similar to the exposed group in all respects except for the exposure being studied. Kilburn et al neglected to include such a procedure. The use of the standard films for pneumoconiosis is no substitute for proper controls as it tells us nothing about under or over-reading by the investigators. The reading of control films intermixed with films of the exposed group without knowledge by the readers as to which group each film belongs to might have shown no differences if there was matching by at least age, sex, and smoking habits. Smoking habits should include type of tobacco use and degree as well as duration of smoking. Kilburn et al did not include any adjustment for degree of cigarette smoking in their adjustment of their spirographic results and there is no evidence of adjustment for smoking habits with regard to the readings of the chest films in their study. Consequently, their conclusion is not warranted because their method was seriously flawed. In addition, I agree with the other points made by Rossiter. ' The reply by Kilburn et al contains a section labelled "No fibrosis from cigarette smoking alone" which is full of error. They claimed that I used minifilms (70 mm) in a 1972 paper and deny the existence of Rossiter's reference No 5 to my 1991 paper in the British J'ournal of Industial Medicine!2 My original studies of this subject were done with minifilms and were reported in the 1960s but subsequent studies were done with the use of large films, all without my knowledge of smoking habits, and these later reports entailed the use of the 1980 International Labour Office standard films and classification. These reports were confirmatory of the early findings. Kilburn et al are apparently unaware of my papers published in 19843 and 1988.4 These include extensive reviews of relevant publications. Perhaps they have simply ignored this body of work? It is interesting to note that Kilburn5 (Rossiter's reference No 6) in 1981 not only denied the validity of my earlier reports on the relation between cigarette smoking and radiological abnormalities suggestive of mild diffuse pulmonary fibrosis but also castigated my publications since 1971 on the increased frequency of asbestosis in smokers compared with non-smokers. Yet in 1986 he published a study,6 the only reference in his response to Rossiter' in which his own data confirmed this phenomenon. Neither in 1986 nor now has he referred to his 1981 error in denying