Statistics from Altmetric.com
Chronic obstructive pulmonary disease (COPD) is a common disease usually treated in general practice, especially in the early stages.1 The recently published British Thoracic Society guidelines encourage a systematic approach to the management of COPD as is widely used in asthma.2 Lung function measurements are regarded as central to the correct implementation of the guidelines.The guidelines are unequivocal in advising the use of forced expiratory volume in one second (FEV1) rather than peak expiratory flow (PEF) in the management of COPD: “. . . in COPD the relationship between PEF and FEV1 is poor and it is not possible to predict FEV1 from the PEF or vice versa.” This is a key issue for GPs who have to decide now whether or not to purchase a spirometer, and whether they have the organisational capacity to cope with the maintenance, calibration, and interpretation demands of modern spirometers.
We have investigated the literature examining the relationship between FEV1 and PEF and exploring their use in COPD. We have been unable to find substantive evidence to support the statement in the BTS guidelines regarding the superiority of FEV1 over PEF. The only citation among the 171 references offered in the guidelines to support their position is a paper by Kelly and Gibson.3 In fact, Kelly and Gibson state the opposite view and report a very strong correlation between FEV1 and PEF with anr value of 0.95 (p<0.001). A similarly strong relationship between the two parameters has been reported by others.4 5
The close relationship between FEV1 and PEF is reassuring to us because the arguments put forward by the COPD guidelines seem counter-intuitive to GPs working daily with PEF in asthma. We recognise the role of spirometry as a whole in the diagnosis of COPD, especially in distinguishing primarily restrictive from obstructive disease. In the continuing management of COPD, however,we suspect that spirometry has little additional value to offer over PEF, but considerable practical disadvantages.
authors’ reply Drs Nolan and White are concerned that GPs might be persuaded to buy spirometers when peak flow meters might perform just as well. The guidelines list a number of reasons why FEV1 is preferable to PEF in managing COPD. They attack one specific aspect but unfortunately misquote their references. The figure they quote from Kelly and Gibson1-1applies to a previous 61 patients undergoing routine testing (not all with COPD) and not to the 10 subjects with COPD and a positive steroid trial in whom the relative changes in PEF and FEV1 do not exhibit the same slopes. Liebowitz1-2 studied 10 healthy individuals and none with COPD. The 1962Lancet article by Ritchie1-3dates from the infancy of FEV1 and long before COPD was defined as a discrete entity.
The most important and fundamental point is that PEF cannot differentiate between obstructive and restrictive patterns of abnormal function. If the diagnosis is not made correctly then the GP cannot hope to manage the patient correctly. The implications of a restrictive defect will often necessitate referral to secondary care to assess the cause, whereas most cases of mild to moderate COPD are manageable within primary care.
The FEV1 is a more reproducible measurement so that measurements on a single occasion can be of value whereas, to obtain similar accuracy with PEF, serial measurements are required. By inspecting the FEV1 traces it is possible to know whether a patient has performed the manoeuvre correctly, whereas no such confirmation exists for PEF. Even in asthma, studies of repeated measurements of serial PEF using computerised measurements confirm that up to 50% of readings may be non-valid.1-4 1-5
To understand the relationship between the level of PEF and the level of FEV1 it is necessary to go—not to epidemiology—but to the physiology underlying the shape of the flow-volume loop in COPD. In the first draft of the guidelines we included a figure illustrating how the FEV1 could be reduced to 33% of predicted at a time when the PEF remains relatively preserved at 60% of predicted. The discrepancy arises because of the airway collapsibility present in COPD secondary to the loss of elastic tissue. The PEF is generated by the instantaneous flow of air leaving the trachea in the first 0.1 seconds of expiration, while the FEV1 includes air leaving the airways through airways that have collapsed after about 0.2 seconds of expiration (fig 1-1). In the example shown a patient with severe COPD (lower line) is compared with the predicted normal pattern (upper line). The patient’s FEV1 is markedly reduced to 0.8 l (33%) while the PEF is relatively preserved at 5.7 l/s (or 340 l/min) which is 80% predicted. As expiration begins (point “a”) there is a rapid increase in expiratory flow until the flow becomes limited by the airway dimensions and peak flow is reached (point “b”). As expiration continues in the healthy subject so flow decreases slowly and progressively until the residual volume is reached when flow ceases (point “c”). In the patient with COPD the initial rapid rise in expiratory flow is similar but, as the intrathoracic pressure increases, that pressure is transmitted to the segmental and other largeairways which have lost the elastic attachments. The airways therefore “collapse” and obstruct the passage of air through those airways. This results in the rapid reduction in flow after the peak has been attained (point “d”). Flow in the remainder of the expiration remains limited. The effect of the expiratory airway collapsibility is shown by the time points marked. The subject with COPD reaches peak flow at about 100 ms and has reached the point of expiratory collapse (point “d”) within 0.25 s. Thus, for the remaining 0.75 s contribution to the FEV1 measurement, flow is at the very low level shown. As the airway collapsibility varies between COPD patients, the relationship between PEF and FEV1 will also vary. Because PEF can be misleadingly optimistic it is severely limited as a diagnostic tool. This figure was edited out of the guidelines, possibly on the mistaken grounds that it was too obvious a point.
Last FEV1 is a measure both of current severity of disease (which dictates likely treatments to be considered) and also of prognosis. Indeed, FEV1 has a prognostic value even beyond COPD as can been seen from the Renfrewshire 21 year prospective study1-6 where FEV1 had greater prognostic value than many other frequently measured variables including serum cholesterol.
The FEV1 is here to stay.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.