The aim of workplace spirometry is to identify workers who may have lung disease and require further evaluation. Determining what constitutes an abnormal, versus a normal, spirometry result is particularly important when spirometry is performed in relation to the workplace. An ‘abnormal’ spirometry result can impact a worker’s job (e.g., determining job placement)’1, as well as prompting further evaluation of the worker’s lung health and possible workplace exposures. Therefore, to determine what is normal and abnormal, the approach of interpretation needs to be decided.
The occupational health practitioner performing spirometry should ensure that technically acceptable and repeatable spirometry results are obtained according to the current 2019 ATS/ERS spirometry guidelines. It is important to highlight interpretation of spirometry doesn’t provide full diagnosis of lung diseases, but it can begin to paint the overall diagnostic picture. Spirometry interpretation should specify whether the workers lung function is within the normal range or whether it shows an obstructive, restrictive, or mixed impairment pattern using a system which compares the measured result to the predicted result.
Selecting reference values
The interpretation involved in spirometry depends on two factors. First, is selection of an appropriate predictive value reference equation to determine the normal range. The reference equations are summarized data of previous spirometry tests collated from healthy, non-smoking individuals of varied age, gender, and ethnic backgrounds. The reference equations create the predicted value, lower and upper boundaries of the normal range based on the worker’s gender, age, height, and ethnicity. Overall, looking at whether the lung function is within normal range. The second factor is use of an appropriate algorithm to categorize the worker’s spirometry results as normal or abnormal, with the commonly used algorithm being the ‘ATS/ERS interpretive strategies for lung function’ 2. The most widely used predictive reference equation for spirometry is the 2012 Global lung initiative equation (GLI,2012). This equation can be applied more accurately to our current population, as it is multiethnic and has a wide age range from 3 – 95 years.
Use of percent predicted
After choosing the predictive reference equation, the next step is choosing a way to define what is normal and abnormal. Percent predicted was widely used in the interpretation of spirometry results, by using a fixed value of 0.70 for the FEV1/FVC ratio, and 80% of predicted for FEV1, FVC, as a cut off for the lower limit of normal (LLN). It has been highlighted that the use of the fixed value of 0.70 as the LLN for the FEV1/FVC ratio is not accurate as the ratio declines with age. Using the fixed value will result in false-negative results in younger workers (25-45 years age) and false-positive results in older workers (men older than 45 year and women younger than 55years)3. This is also evident for FEV1, A scatter plot showing the relationship between FEV1 and age in healthy non-smokers shows a change in the scatter as the FEV1 declines with age. Therefore, the method for defining the LLN should follow the same pattern. The raw data scatter (Graph 1) shows the use of 80% predicted LLN, as well as a progressively higher increase in LLN as age increases. Using this as a marker of abnormality would create a large percentage of false-positive results in both young and elderly people. With an ever-increasing ageing population, it could be regarded as problem when diagnosing and monitoring respiratory disease. Whereas the ‘1.64SD LLN (5th Percentile)’ shows a lesser increase in FEV1 as age increases.
Figure 1. Showing data scatter of FEV1 and age. The comparison of the 80% Predicted and 1.64SD LLN (5th Percentile).Use of LLN and Z-score
Z-scores provide a comparison between measured and predicted values by determining whether the difference between the measured and predicted is compatible with the “normal distribution” of the data from predictive value reference equations. Z-scores provide a way of viewing the 5th percentile LLN of -1.645, 5% of abnormal results are expected to fall within the area and 90% represents the majority of the healthy population (Figure 2). The method of using z-scores determines how normal a result is, how far the result is from the mean (predicted) value and expresses it in the terms of a dimensionless number. The z-score represents how many standard deviations the result differs from the predictive mean value.
The ATS/ERS spirometry guidelines have recommended the LLN as 1.645 SD below the mean value applying 5th percentile LLN. The ‘Z-score’ or Standard Residual (SR) shows the number of standard deviations a result is from the mean value. By using LLN of 1.64 SD abnormal results lower than -1.64 would occur in 5% of the reflected population.
Figure 2. A bell-curve of normal distribution showing the commonly used 5th percentile LLN of 1.64 SD accepting a false positive of 5%. Standard deviation “The STANDARD DEVIATION (SD) is a measure of how spread-out numbers are from their expected mean value”Benefits of LLN and Z-score
Using 1.64 SD as a statistical marker for abnormality of a result could have a beneficial effect in assistance in respiratory diagnosis and monitoring, potentially having an effect on the worker respiratory evaluation. Whereas percent predicted can have the potential to over diagnose those older and underdiagnose those younger. An example of this is shown in table 1.
Table 1 shows that subject (a) had a measured FEV1 which would be considered ‘normal’ when implementing an 80% LLN but shows that this result would be considered ‘abnormal’ as it is in the 5th percentile of their population normality. Conversely, subject (b) shows an FEV1 of 75% predicted but is within a statistical LLN (90th percentile) based on their given population normality.It is recommended to implement the use of Z-score when considering a workers spirometry result. A non-proven LLN of 80% predicted for FEV1 and fixed value of 0.70 for FEV1/FVC as a ‘cut off’ for normality may prove to be costly for both the worker and job role. Visualizations of Z-score on reports can aid in increased use of the z-score, as it provides a visual reference as shown in figure 3.
Figure 3. Showing a Z-score pictogram.References
1Townsend, M., 2020. Spirometry in Occupational Health—2020. Journal of Occupational & Environmental Medicine, 62(5), pp.e208-e230.
2Quanjer, P., Stanojevic, S., Cole, T., Baur, X., Hall, G., Culver, B., Enright, P., Hankinson, J., Ip, M., Zheng, J. and Stocks, J., 2012. Multi-ethnic reference values for spirometry for the 3–95-yr age range: the global lung function 2012 equations. European Respiratory Journal, 40(6), pp.1324-1343.
Pellegrino, R., 2005. Interpretative strategies for lung function tests. European Respiratory Journal, 26(5), pp.948-968.
3Redlich, C., Tarlo, S., Hankinson, J., Townsend, M., Eschenbacher, W., Von Essen, S., Sigsgaard, T. and Weissman, D., 2014. Official American Thoracic Society Technical Standards: Spirometry in the Occupational Setting. American Journal of Respiratory and Critical Care Medicine, 189(8), pp.983-993.
Quanjer, P., Pretto, J., Brazzale, D. and Boros, P., 2013. Grading the severity of airways obstruction: new wine in new bottles. European Respiratory Journal, 43(2), pp.505-512.
Share
