Height seems to be the only variable that determines lung oscillometry indices

To the Editor:

We read the article by Gochicoa-Rangel et al. [1] with great interest, wherein the authors reported the reference equations for the impulse oscillometry system (IOS) indices. Their study, encompassing a large healthy population of 830 male and female subjects between the ages of 2.7 to 90 years, fills a crucial gap in the current literature on oscillometry. While numerous reference equations exist globally, none has covered such a broad age range. Interestingly, the authors also provide cut-off values for IOS indices, which is in line with the proposition that in comparison to predicted values or z-scores, fixed cut-off values will be more accurate to discriminate between normal and abnormal oscillometric outcomes in adults, with a reasonably good sensitivity, specificity and overall accuracy [2]. Aligning with current expectations, the authors incorporated sex, height, age and body mass index as key physiological variables into their reference equation. This approach reflects the recognised dependence of spirometry indices on these variables. It seems almost imperative to use these variables when interpreting lung function.

We are grateful to the authors for sharing their rich source of data (both in the main manuscript as well as in the supplement). We took the opportunity to dig deeper into their data to gain additional insights. Notably, we observed that resistance at 5 Hz (R₅) were highest at the youngest ages and then exhibited a sharp decline until the age of 18 years. There were no difference at all between boys and girls from 2.7 to 18 years (figure 1). After 19 years of age, the values remained relatively constant throughout adulthood, with females showing a consistently higher value than males from 19 years to 90 years. There was a small increase in R₅ in both males and females towards the latter ages (figure 1).

• Download figure
• Open in new tab
• Download powerpoint

FIGURE 1

Resistance measured at 5 Hz (R₅) across different ages for males and females.

Most interestingly, the difference in R₅ between males and females (0.7 cmH₂O·L⁻¹·s⁻¹) from 19 to 90 years was exactly equivalent to the expected difference in R₅ between a height of 158 and 168 cm, which happens to be the exact average heights of females and males in Mexico, respectively [1]. The slight increase in R₅ (0.8 cmH₂O·L⁻¹·s⁻¹) during the latter years in both males and females can at least be partly explained by the reduction in average height in this age range (see data from table 1 of [1]). This interesting observation led us to infer that the difference in R₅ between adult males and females is only because of height differences between them, rather than gender-related changes.

The same phenomenon seems to occur in the reactance at 5 Hz (X₅). X₅ is lowest at the youngest ages and becomes rapidly less negative up the age of 18 years, with no difference between boys and girls, and thereafter remains almost a straight line from 19 to 90 years, with females showing consistently lower X₅ than males. The differences in X₅ between females and males in this age range seems to match exactly with the X₅ values for the differences caused by their heights.

What this means is that gender and age have no impact on R₅ and X₅ between the ages of 2.7 to 18 years, only height does. This inference is very similar to the results of the study of 692 children from Africa reported by Chaya et al. [3], who showed that height was the only dependent variable in children between 3 and 17 years. Gender and age have no impact on R₅ and X₅ between the ages of 19 and 90 years, only height does. Studies in adults who have derived reference equations for oscillometry from different parts of the world have consistently reported that height is the most important, if not the only, variable affecting R₅ and X₅ [4]. Almost all adult studies have reported significant differences between males and females for R₅ and X₅, but no one attributed these differences to their heights. Our inferences based on the data of Gochicoa-Rangel et al. [1] and the observations made by others underscore the crucial role of height as the most important, if not the sole, determinant of lung oscillometry indices.

The implications of these observations in clinical practice are monumental. The only variable that needs to be measured to interpret oscillometry indices seems to be height. Our previous assertion that interpreting oscillometry necessitates only the use of cut-off values rather than reference values aligns seamlessly with this notion [2].

In light of these newer insights, oscillometry presents four distinct advantages over spirometry: 1) it is a simpler test to perform; 2) height is the only variable that determines R₅ and X₅ (it seems most likely that all other variables will follow the same pattern); 3) single cut-off values using height as the only variable will help differentiate between healthy and lung disease states; and 4) oscillometry is not only more sensitive to diagnose small airways obstruction as reported by several authors, but also provides comprehensive information about the mechanical properties of the lung. These developments herald exciting times for lung oscillometry, showcasing its potential to revolutionise respiratory assessments in clinical practice.