Through the comparison between MFEV curves in patients with ILAO and COPD, it was found that flow parameters such as PEF, FEF50%, FEF75% and FEF25 − 75% were conducive to better classifying the characteristics of MEFV curve in ILAO. This study is the first to analyze FEF ratios, specifically FEF50%/FEF25%, FEF75%/FEF50% and FEF75%/FEF25%. These ratios play a distinguishing role in airflow limitation resulting from large or small airway lesions. In particular, FEF50%/FEF25% and Empey’s index have satisfactory applicability for diagnosing the early ILAO.
When a large airway lesion is obstructed, factors such as the size and location of the airway obstruction, the property of the lesion and the phase of forced expiration can all influence the MEFV curve(8). As early as 1973, Miller and Hyatt described three different abnormal patterns of MEFV curves in their research: the variable extrathoracic, variable intrathoracic and fixed ILAO(9). Therefore, clinicians make initial assessments of the location of airway lesions(7, 10) by evaluating the characteristics of different shapes of MEFV curves. In the case of central fixed airway obstruction and central variable airway obstruction, the MEFV curve shows a characteristic plateau-like change in the expiratory phase(7, 10, 11). This change featuring a near-zero slope often occurs rapidly after the initial peak. The slope reproduces in the degree of unrelated forced expiration(9). COPD is characterized by lung hyperinflation caused by expiratory flow limitation, which is primarily due to increased resistance in peripheral small airways and decreased lung elastic recoil(12). MEFV curves in COPD typically exhibit reduced PEF at the beginning of the expiratory phase and a decrease in the descending limb flow. This flow reduction is often non-linear and correlates with the severity of obstruction. The damage to small airway structures will result in the intensification of airway closure and the decrease of both residual and functional residual volumes. The expiratory limb of the MEFV curve also demonstrates a distinct inflection point, followed by a sustained low-flow state(13). This presentation can sometimes resemble the severe ILAO, as shown in Fig. 3.
To distinguish ILAO, the use of classic indicators such as FEV1, FEV1/FVC and FVC alone cannot comprehensively and objectively assess lung ventilation function, which potentially leads to the missing of valuable clinical information. It was observed that seven of the 63 patients with ILAO were misclassified as having restrictive ventilatory impairment due to the early termination of expiration leading to reduced FVC. Additionally, three patients were mistakenly categorized as having essentially normal lung ventilation because they only exhibited reduced PEF and FEF25%. Moreover, 53 patients accounted for 84%. FEV1/FVC reduction is normally the primary focus in lung function reports, which gives rise to a classification of obstructive ventilatory impairment. That may cause ILAO to be easily overlooked by clinicians. This finding is consistent with previous research(14, 15) that FEV1 is not particularly sensitive in diagnosing ILAO compared with PEF. As early as 1969, Miller and Hyatt conducted simulated experiments on healthy individuals and discovered that FEV1 only showed a significant decrease when the tracheal diameter was less than 6 mm. The sensitivity of MEFV curves for detecting ILAO is not very high because the curve shape only becomes abnormal when the tracheal diameter is smaller than 8 mm(14). Stoller also noticed that characteristic plateau-like changes are more likely to occur in the MEFV curve when the tracheal diameter is less than 8 mm (indicating the airway narrowing of more than 50%)(16).
In the case of ILAO, PEF is affected at the start of expiration at a high lung volume during a forceful exhalation, and FEV1% may still fall within the normal range at the moment(14, 15). According to a study from 1972, ILAO is likely present when Empey’s index exceeds 10, and a higher value suggests a more severe degree of ILAO(6, 17). Similar results were confirmed by Brookes and Fairfax(18). In addition, some research suggests that Empey’s index ≥ 8 is related to the location, type and severity of ILAO(19). This study determined the critical value of Empey’s index for diagnosing ILAO as 7.415 by ROC curves, with a specificity of 100%. This finding is in line with the discovery of Miller, who found that specificity reaches 94% when Empey’s index exceeds 8(20).
In this study, it was found that patients with ILAO have significantly lower PEF but higher FEF50%, FEF75% and FEF25 − 75% than those with COPD, as shown in Fig. 3. When ILAO is present, the expiratory flow at high lung volumes is dependent on the strength and influenced by the whole chest cage resistance, and the increased resistance at the narrowed site significantly reduces PEF at high lung volumes. Expiratory flow depends more on the resistance of peripheral airways at low lung volumes. During this phase, the degree of strength and resistance in central airways has no impact on expiratory flow(6).
Specific quantitative values were also used to show the shape characteristics of the MEFV curve, and a predictive model for ILAO was established for the first time. A comparison was made between the application of FEF ratios and Empey’s index for ILAO by ROC curves. It was noticed that the AUC for FEF50%/FEF25% is similar to that for Empey’s index and even larger than Empey’s index, and both of them exceeded 0.9. This suggests that FEF50%/FEF25% may play a similar role to Empey’s index in airway obstruction(6, 17–19).
The specificity of the three predictive models for ILAO is higher than 78% found in previous studies based on the characteristics of MEFV curves(20). Using the combination of qualitative and quantitative indicators for assessment can improve specificity and sensitivity(19).
In the overall sample, the correlation study also found that FEF50%/FEF25% in patients with COPD was negatively correlated with FEV1. In the case of severe COPD obstruction, a significant increase occurs in small airway resistance(21), which ultimately leads to airway closure(22). Additionally, the MEFV curve exhibits a noticeable inflection point in PEF, followed by a low-flow plateau(23–25). As a result, FEF50%/FEF25% approaches 1. In contrast, FEF50%/FEF25% is unrelated to FEV1 for patients with ILAO. When FEF50%/FEF25% is close to 1, it signifies that the decrease in FEF50% and FEF25% occurs in proportion. This concurs with the characteristics of a plateau-like change in the MEFV curve when expiratory airflow encounters ILAO in the first half of the forced expiration(7, 10). When FEF50%/FEF25% is greater than 1, it further suggests the possibility of ILAO.
The predictive performance of the three models for ILAO was separately validated from the validation group. It showed good consistency with the model group, which indicated the strong discriminative ability and certain generalizability and clinical application of these three ILAO prediction models.
Certain limitations exist in this study. The value of FEF50%/FEF25% and Empey’s index in the context of ILAO still require further validation in larger clinical cases. Furthermore, the relationships between the degree and length of tracheal stenosis caused by central airway disease and the parameters of the FEF ratio (FEF50%/FEF25%), Empey’s index and characteristic changes in the MEFV curve warrant subgroup investigations.
In summary, the MEFV curve is a simple, easily accessible, non-invasive and cost-effective pulmonary function test. Characteristic changes in the shape of MEFV curve help distinguish ILAO from peripheral small airway obstruction. The prediction models for ILAO established using FEF50%/FEF25% and Empey’s index exhibit good clinical value. It is expected that FEF50%/FEF25% and Empey’s index become a potential indicator to diagnose ILAO.