In this issue of Thorax Lokke et al report the incidence of COPD in the Copenhagen City Heart Study.¹ They were able to do so because they followed a large cohort of representative citizens of Copenhagen for 25 years with spirometric tests, far longer than any previous such study. Non-asthmatics aged 30–60 years with normal initial spirometric parameters and good smoking histories were studied. About 2500 of them underwent spirometric tests at least twice (25 years apart), and vital status was ascertained in another 5500 who did not undergo comparable spirometric measurements. COPD was assessed in terms of the current standards for spirometry² staged according to GOLD.

More than 24% of the continuous smokers who underwent spirometric tests had COPD (defined as FEV₁<80% of predicted, FEV₁/FVC <70%) after 25 years, in stark contrast to never smokers in whom COPD occurred in <5%. Those who stopped smoking before or during the study did better than those who continued to smoke, and the earlier the cessation occurred, the better the outlook; cessation <10 years before the end of the study did not appear to help. Sex was not a significant risk factor for COPD (most of the subjects who performed spirometric tests were women), but age at entry to the study was, presumably because older smokers had a history of more pack years at entry than younger smokers.

The authors emphasise that the incidence of COPD was unexpectedly high; it must have been over 25%, given that only 48% of those alive at 25 years underwent spirometric tests, and those who did not were almost certainly in worse health than those who did. Furthermore, 109 participants (more than 1% of the total) died of COPD during the observation period and were therefore unavailable for testing. This is distinctly higher than the commonly quoted figure of 15%—that is, 15% of smokers develop symptomatic COPD, a point recently emphasised by another editorial comment.³

The estimate of 15% of smokers developing COPD is usually attributed to Fletcher et al⁴ whose work forms the basis of most of our current understanding of the disease. I am unable to locate an unequivocal statement in their book quoting this number, but I believe that there are some circumstances that might explain a difference between the purported views of Fletcher and the findings of the Copenhagen City Heart Study. Firstly, most of the COPD in the latter study was “moderate”—that is, the participants had FEV₁ values of 50–79% of the predicted normal—so it is likely that many of these were not very symptomatic. Secondly, COPD in the early to mid 1960s was somewhat different from today; in my experience, deaths and hospital admissions caused by COPD were not uncommon in men aged 55–65, which is not the case today. If one imagines the rate of loss of FEV₁ necessary to culminate in values of 0.5–1.0 l at the age of 55, assuming the subject started smoking in late adolescence, the result is spectacular and rare today. This is a somewhat long winded way of saying that the COPD “sensitivity” of the Copenhagen Heart Study was very likely considerably greater than that of Fletcher et al.

Lokke et al¹ emphasise that the high incidence of COPD which they observed was due to the length of their study. The mean age at study entry was about 45 years and, given a 25 year follow up period, many of their participants must have been over 70. The message is that many smokers develop airways obstruction if they live long enough and continue to smoke, and that the number that do so is increasing because of a decline in competing mortality. I heartily agree with this conclusion. If a smoker has an FEV₁ of 4.0 l at the age of 30, his FEV₁ will have to decline by about 120 ml/year to develop respiratory failure (FEV₁ = 1.0 l) at the age of 55, which is uncommon. On the other hand, if the FEV₁ declines at a rate of 60 ml/year—the rate of decline of an “average” smoker^4,5—respiratory failure is likely to occur at the age of 80 in an “average” smoker. It is not uncommon for smokers to reach 80 years of age nowadays.

An argument can therefore be made that many (perhaps most) smokers are “susceptible” to COPD if they live long enough. Indeed, the definition of susceptibility may need to be examined. Furthermore, the GOLD definition of COPD as an “abnormal” response to inhaled toxins may need revision, or at least a rethink. This is not to say that people’s propensity to develop airways obstruction in response to smoking does not vary; approximately 65% of continuous smokers in the Copenhagen City Heart Study had FEV₁/FVC >70% after 25 years. There are such differences, and it is important—though difficult—to work out their mechanisms.