Article Text
Statistics from Altmetric.com
The paper by Schachter et al 1 in the January 2001 issue ofThorax is interesting in that it has a number of unusual and, it is suggested, inexplicable findings that appertain to various indices of ventilatory capacity. With all due deference, we would suggest that there is an explanation for these unusual findings.
Firstly, mild, moderate and severe obesity are all associated with an incremental reduction in both the forced expiratory volume in 1 second (FEV1) and the forced vital capacity (FVC).2-4 Secondly, in normal subjects and in those who have pure restrictive impairment, the FVC and FEV1 are within 2–3% of each other when expressed as a percentage of predicted. The FVC cannot be significantly smaller than the FEV1 when expressed as a percentage of predicted except in certain neurological diseases. It is noted that the criterion for acceptance of the spirometric volumes was “two measurements of the FEV1 within 100 ml of each other”, suggesting the FVC was ignored. Table 3 in the paper by Schachter et al shows that, when expressed as a percentage of predicted, the FVC in every instance is less than the FEV1. In most groups there is a relatively small difference except for those who are moderately or severely obese.
The reason for the disparity in the FEV1 and the FVC is that the FVC manoeuvre was likely to be incomplete, especially in those who are overweight. Some normal large men over 74 inches in height take 12–16 seconds to complete their FVC manoeuvre. Unfortunately, these days few physicians spend any time doing routine spirometric testing themselves as they rely on their technicians. “Shoe leather” epidemiologists such as Archie Cochrane and Ian Higgins have been replaced by computer addicted statisticians who are thrown into ecstasy by what they can do with a computer, but who fail to realise that their original data may be flawed.5 Were Dr Schachter and her colleagues to review their tracings, we suspect that they would find that at least some of the FVC manoeuvres had been aborted prematurely. If only flow-volume loops are relied on, it needs to be borne in mind that it is difficult—and, indeed, usually impossible—to know whether the FVC manoeuvre has been completed.
The other surprise in the study is that the smaller the FVC when expressed as a percentage of predicted, the higher the FEF25–75. What is abundantly clear, however, is that, when the FVC manoeuvre is incomplete, then the FEF25–75 is “pushed” further up the steeper portion of the FVC curve so that the FEF25–75 is artefactually increased—that is, the more premature the termination of the FVC, the higher the FEF25–75.6 ,7
The findings of wheeze in those who are obese is not surprising, especially in cigarette smokers. When a markedly obese subject exercises on the treadmill wheezes are frequently heard, providing he can continue for a sufficient length of time. In a subject who has asthma the FEV1 is generally reduced appreciably more than is the FVC. The exact opposite applies in overweight patients included in the study by Schachter et al. It can be seen from table 3 that the FVC in both the moderately obese and the severely obese is significantly lower than the FEV1measurement. Finally, one would expect the most obese subjects to become short of breath much more quickly, especially if some of them had exercise induced asthma. This would explain the much more frequent medication usage.
We suggest the disparate reduction in the FVC and FEV1 seen in obese subjects has little to do with asthma, but is a direct effect of their obesity and the fact that some of the FVC measurements have been significantly underestimated.
References
author's reply I am pleased to note that Drs Ahmad and Morgan were able to pick up on one of the main points of our paper—that reduced lung function and respiratory symptoms in obese people may well be an effect of their obesity and are not necessarily an indication of asthma. However, they imply that the differences in spirometric function that we observed were due to technical error rather than to an effect of obesity. As a physician who has spent a substantial amount of time measuring routine spirometric parameters on over 300 severely obese patients, I disagree with their suggestion.
As with a number of previous studies,1-1-1-3 we found that moderate and severe obesity were associated with an incremental reduction in both FEV1 and FVC. In our normal subjects the FEV1 and FVC, expressed as percentage predicted, were within 2.6% of each other. The mean absolute values for FEV1 and FVC in this group were 3.5 l and 4.0 l, respectively. The mean FEV1/FVC% in all groups was 85.8–87%, which is well within the normal range for this age group.
Our results show that most patients with severe obesity have FVC within the normal range, although it is reduced when compared with patients with normal body mass index. We do not have other measurements of lung volumes to confirm further the presence of restriction, but these findings are consistent with those of other studies.
It is unlikely that our results are due to a systematic underestimation of FVC in the obese groups. In my experience, obese patients who are otherwise healthy do not usually have evidence of airway obstruction or a need for prolonged expiration times to complete their FVC manoeuvres. Their spirometric tracings show that the expiration reaches a clear plateau within 2–3 seconds in the same way as is seen in non-obese subjects.
The technical staff involved in the collection of the data are extremely well trained and the measurement methods are well standardised. The same two senior researchers were present at all studies and trained and supervised all other staff involved. Our senior researchers and technicians are very experienced, having performed many large epidemiology studies involving thousands of subjects. The FVC manoeuvre was performed to a minimum of 3 seconds. The criterion for acceptance of the spirometric volumes included both FEV1and FVC and required both parameters to be repeatable to within 100 ml. These procedures are stricter than the ATS guidelines which allow for 5% variability between blows. If it appeared that the patient was obstructed, then FVC was performed until expiration was complete.
In reporting our results we did not attempt to draw any conclusions from the very small differences between the percentage predicted FEV1 and FVC values. Instead, we limited our discussion to the more substantial differences between groups based on body mass index—the hypothesis that we set out to test.