The FEV1/FVC ratio is almost universally used to determine the presence or absence of airway obstruction. There are several different threshold values for this, however.
Lower Limit of Normal (LLN):
The ATS/ERS statement on interpretation recommends the use of the Lower Limit of Normal (LLN) for the FEV1/FVC ratio as well as the FVC and FEV1. The LLN is derived from the statistical analysis of a study population and specifically demarks the bottom 5th percentile. It is calculated as 1.645 times the standard error of the estimate (SEE).
A fixed percent of predicted:
A percent of predicted, which for the FEV1/FVC ratio is most commonly either 95% or 90% of predicted has frequently been used as a threshold. This approach is not supported by any official group and has its roots in the ITS Snowbird workshops in the early 1970’s and NIH recommendations around that time.
Fixed ratio (0.70):
The Global Initiative for Chronic Obstructive Lung Disease (GOLD) has stated that an absolute (not percent predicted) post-bronchodilator FEV1/FVC ratio less than 0.70 should be used to indicate the presence of airway obstruction and this is applied to individuals of all ages, genders, heights and ethnicities. Although this is intended for the diagnosis of COPD it has been used in routine clinical spirometry as well.
Which approach is correct?
This subject has been debated for decades and continues to be debated. There are proponents and opponents for each approach and the arguments for and against each approach are as follows:
LLN:
For:
The primary decision behind the LLN is that the bottom 5th percentile of any study population has a high probability of being abnormal. This concept is frequently used in biological research and appears to have a significant level of statistical relevance.
Against:
Because it is dependent upon a specific study it will be different depending on which reference equations are in use. The decision that the bottom 5% of the population is abnormal is somewhat arbitrary and may underestimate the presence of airway obstruction.
Fixed %Predicted:
For:
Easy to remember and use. Appears to have a reasonable level of clinical relevance.
Against:
Also dependent on specific reference equation and will differ depending on which reference equations are in use. In addition the population distribution for pulmonary function values is usually considered to be homoscedastic (i.e. equal distribution away from the mean value throughout the range), this means that a fixed percent predicted will tend to overestimate the normal range for younger and taller individuals and underestimate it for older and shorter individuals.
Fixed ratio:
For:
Easy to remember and use. Not dependent on any specific reference equations. A reasonably level of clinical relevance when applied correctly (post-bronchodilator spirometry only and for a diagnosis of COPD).
Against:
Underestimates airway obstruction in the young and overestimates it in the elderly.
In the final analysis each approach is relatively arbitrary to one degree or another and none are supported by clear clinical correlations. The concept of normalcy however, is not easily pinned down and all approaches should be considered a conditional starting point for the interpretation of pulmonary function results.
In addition, the accuracy of the FEV1/FVC ratio is dependent on the individual accuracy of the FVC and FEV1 measurements and this places a limitation on the accuracy of any FEV1/FVC ratio threshold.
Recommendation:
To some extent any threshold value for the FEV1/FVC ratio is arbitrary and there are limitations to any approach. At this time however, the preponderance of evidence and opinion is in favor of the LLN so the recommendation has to be for those interpreting pulmonary function tests to use the LLN for all reference values, including the FEV1/FVC ratio, unless there are clear and overwhelming reasons not to. At the same time it is also important to remember there is a certain amount of overlap between normal and abnormal in pulmonary function results and for this reason individuals on the borderline are not clearly one or the other without accompanying symptoms.
PFT Interpretation by Richard Johnston is licensed under a Creative Commons Attribution-NonCommercial 4.0 International License