Improved spirometric detection of small airway narrowing: concavity in the expiratory flow-volume curve in people aged over 40 years

Curve-Modelling and Machine Learning for a Better COPD Diagnosis

Background

Development of new tools in artificial intelligence has an outstanding performance in the recognition of multidimensional patterns, which is why they have proven to be useful in the diagnosis of Chronic Obstructive Pulmonary Disease (COPD).

Methods

This was an observational analytical single-centre study in patients with spirometry performed in outpatient medical care. The segment that goes from the peak expiratory flow to the forced vital capacity was modelled with quadratic polynomials, the coefficients obtained were used to train and test neural networks in the task of classifying patients with COPD.

Results

A total of 695 patient records were included in the analysis. The COPD group was significantly older than the No COPD group. The pre-bronchodilator (Pre BD) and post-bronchodilator (Post BD) spirometric curves were modelled with a quadratic polynomial, and the coefficients obtained were used to feed three neural networks (Pre BD, Post BD and all coefficients). The best neural network was the one that used the post-bronchodilator coefficients, which has an input layer of 3 neurons and three hidden layers with sigmoid activation function and two neurons in the output layer with softmax activation function. This system had an accuracy of 92.9% accuracy, a sensitivity of 88.2% and a specificity of 94.3% when assessed using expert judgment as the reference test. It also showed better performance than the current gold standard, especially in specificity and negative predictive value.

Conclusion

Artificial Neural Networks fed with coefficients obtained from quadratic and cubic polynomials have interesting potential of emulating the clinical diagnostic process and can become an important aid in primary care to help diagnose COPD in an early stage.

The Challenges of Spirometric Diagnosis of COPD

Chronic obstructive pulmonary disease (COPD) is one of the top causes of morbidity and mortality worldwide. Although for many years its accurate diagnosis has been a focus of intense research, it is still challenging. Due to its simplicity, portability, and low cost, spirometry has been established as the main tool to detect this condition, but its flawed performance makes it an imperfect COPD diagnosis gold standard. This review aims to provide an up-to-date literature overview of recent studies regarding COPD diagnosis; we seek to identify their limitations and establish perspectives for spirometric diagnosis of COPD in the XXI century by combining deep clinical knowledge of the disease with advanced computer analysis techniques.

Effect of exercise-induced bronchoconstriction on the configuration of the maximal expiratory flow-volume curve in adults with asthma

We determined the effect of exercise-induced bronchoconstriction (EIB) on the shape of the maximal expiratory flow-volume (MEFV) curve in asthmatic adults. The slope-ratio index (SR) was used to quantitate the shape of the MEFV curve. We hypothesized that EIB would be accompanied by increases in SR and thus increased curvilinearity of the MEFV curve. Adult asthmatic ( n  = 10) and non-asthmatic control subjects ( n  = 9) cycled for 6-8 min at 85% of peak power. Following exercise, subjects remained on the ergometer and performed a maximal forced exhalation every 2 min for a total 20 min. In each MEFV curve, the slope-ratio index (SR) was calculated in 1% volume increments beginning at peak expiratory flow (PEF) and ending at 20% of forced vital capacity (FVC). Baseline spirometry was lower in asthmatics compared to control subjects (FEV₁ % predicted, 89.1 ± 14.3 vs. 96.5 ± 12.2% [SD] in asthma vs. control; p  < 0.05). In asthmatic subjects, post-exercise FEV₁ decreased by 29.9 ± 13.2% from baseline (3.48 ± 0.74 and 2.24 ± 0.59 [SD] L for baseline and post-exercise nadir; p  < 0.001). At baseline and at all timepoints after exercise, average SR between 80 and 20% of FVC was larger in asthmatic than control subjects (1.48 ± 0.02 vs. 1.23 ± 0.02 [SD] for asthma vs. control; p < 0.005). This averaged SR did not change after exercise in either subject group. In contrast, post-exercise SR between PEF and 75% of FVC was increased from baseline in subjects with asthma, suggesting that airway caliber heterogeneity increases with EIB. These findings suggest that the SR-index might provide useful information on the physiology of acute airway narrowing that complements traditional spirometric measures.

Deep learning using multilayer perception improves the diagnostic acumen of spirometry: a single-centre Canadian study

RATIONALE

Spirometry and plethysmography are the gold standard pulmonary function tests (PFT) for diagnosis and management of lung disease. Due to the inaccessibility of plethysmography, spirometry is often used alone but this leads to missed or misdiagnoses as spirometry cannot identify restrictive disease without plethysmography. We aimed to develop a deep learning model to improve interpretation of spirometry alone.

METHODS

We built a multilayer perceptron model using full PFTs from 748 patients, interpreted according to international guidelines. Inputs included spirometry (forced vital capacity, forced expiratory volume in 1 s, forced mid-expiratory flow_25-75), plethysmography (total lung capacity, residual volume) and biometrics (sex, age, height). The model was developed with 2582 PFTs from 477 patients, randomly divided into training (80%), validation (10%) and test (10%) sets, and refined using 1245 previously unseen PFTs from 271 patients, split 50/50 as validation (136 patients) and test (135 patients) sets. Only one test per patient was used for each of 10 experiments conducted for each input combination. The final model was compared with interpretation of 82 spirometry tests by 6 trained pulmonologists and a decision tree.

RESULTS

Accuracies from the first 477 patients were similar when inputs included biometrics+spirometry+plethysmography (95%±3%) vs biometrics+spirometry (90%±2%). Model refinement with the next 271 patients improved accuracies with biometrics+pirometry (95%±2%) but no change for biometrics+spirometry+plethysmography (95%±2%). The final model significantly outperformed (94.67%±2.63%, p<0.01 for both) interpretation of 82 spirometry tests by the decision tree (75.61%±0.00%) and pulmonologists (66.67%±14.63%).

CONCLUSIONS

Deep learning improves the diagnostic acumen of spirometry and classifies lung physiology better than pulmonologists with accuracies comparable to full PFTs.

Relationship between concavity of the flow-volume loop and small airway measures in smokers with normal spirometry

Background

There is increasing evidence of small airway abnormalities in smokers despite normal spirometry. The concavity in the descending limb of the maximum expiratory flow curve (MEFV) is a recognised feature of obstruction and can provide information beyond FEV₁, and potentially early smoking-related damage. We aimed to evaluate concavity measures compared to known small airway measurements.

Methods

Eighty smokers with normal spirometry had small airway function assessed: multiple breath nitrogen washout (MBNW) from which ventilation heterogeneity in the diffusion-dependent acinar (Sacin) and convection-dependent conductive (Scond) airways were assessed, and impulse oscillometry system (IOS) from which respiratory resistance and reactance at 5 Hz (R5 and X5) were measured. Concavity measures were calculated from the MEFV, partitioned into global and peripheral concavity.

Results

We found abnormal peripheral and global concavity as well as acinar ventilation heterogeneity are common in “normal” smokers. Concavity measures were not related to either MBNW or IOS measurements.

Conclusion

Abnormalities in concavity indices and MBNW or oscillometry parameters are common in smokers despite normal spirometry. However, these measures likely reflect different mechanisms of peripheral airway dysfunction.

Spirometry parameters used to define small airways obstruction in population-based studies: systematic review

Background

The assessment of small airways obstruction (SAO) using spirometry is practiced in population-based studies. However, it is not clear what are the most used parameters and cut-offs to define abnormal results.

Methods

We searched three databases (Medline, Web of Science, Google Scholar) for population-based studies, published by 1 May 2021, that used spirometry parameters to identify SAO and/or provided criteria for defining SAO. We systematically reviewed these studies and summarised evidence to determine the most widely used spirometry parameter and criteria for defining SAO. In addition, we extracted prevalence estimates and identified associated risk factors. To estimate a pooled prevalence of SAO, we conducted a meta-analysis and explored heterogeneity across studies using meta regression.

Results

Twenty-five studies used spirometry to identify SAO. The most widely utilised parameter (15 studies) was FEF_25–75, either alone or in combination with other measurements. Ten studies provided criteria for the definition of SAO, of which percent predicted cut-offs were the most common (5 studies). However, there was no agreement on which cut-off value to use. Prevalence of SAO ranged from 7.5% to 45.9%. As a result of high heterogeneity across studies (I² = 99.3%), explained by choice of spirometry parameter and WHO region, we do not present a pooled prevalence estimate.

Conclusion

There is a lack of consensus regarding the best spirometry parameter or defining criteria for identification of SAO. The value of continuing to measure SAO using spirometry is unclear without further research using large longitudinal data.

PROSPERO registration number CRD42021250206

Supplementary Information

The online version contains supplementary material available at 10.1186/s12931-022-01990-2.

The Underappreciated Role of Epithelial Mesenchymal Transition in Chronic Obstructive Pulmonary Disease and Its Strong Link to Lung Cancer

The World Health Organisation reported COPD to be the third leading cause of death globally in 2019, and in 2020, the most common cause of cancer death was lung cancer; when these linked conditions are added together they come near the top of the leading causes of mortality. The cell-biological program termed epithelial-to-mesenchymal transition (EMT) plays an important role in organ development, fibrosis and cancer progression. Over the past decade there has emerged a substantial literature that also links EMT specifically to the pathophysiology of chronic obstructive pulmonary disease (COPD) as primarily an airway fibrosis disease; COPD is a recognised strong independent risk factor for the development of lung cancer, over and above the risks associated with smoking. In this review, our primary focus is to highlight these linkages and alert both the COPD and lung cancer fields to these complex interactions. We emphasise the need for inter-disciplinary attention and research focused on the likely crucial roles of EMT (and potential for its inhibition) with recognition of its strategic place mechanistically in both COPD and lung cancer. As part of this we discuss the future potential directions for novel therapeutic opportunities, including evidence-based strategic repurposing of currently used familiar/approved medications.

Pulmonary function testing in COPD: looking beyond the curtain of FEV1

Chronic obstructive pulmonary disease (COPD) management remains challenging due to the high heterogeneity of clinical symptoms and the complex pathophysiological basis of the disease. Airflow limitation, diagnosed by spirometry, remains the cornerstone of the diagnosis. However, the calculation of the forced expiratory volume in the first second (FEV1) alone, has limitations in uncovering the underlying complexity of the disease. Incorporating additional pulmonary function tests (PFTs) in the everyday clinical evaluation of COPD patients, like resting volume, capacity and airway resistance measurements, diffusion capacity measurements, forced oscillation technique, field and cardiopulmonary exercise testing and muscle strength evaluation, may prove essential in tailoring medical management to meet the needs of such a heterogeneous patient population. We aimed to provide a comprehensive overview of the available PFTs, which can be incorporated into the primary care physician's practice to enhance the efficiency of COPD management.

Short-term exposure to air pollution: Associations with lung function and inflammatory markers in non-smoking, healthy adults

INTRODUCTION

Air pollution impacts health by increasing mortality and the incidence of acute events in unhealthy individuals. In contrast, the acute effects of pollution in healthy individuals are less obvious. The present study was designed to evaluate the associations between short-term exposure to air pollution on one hand and lung function, and inflammatory markers on the other in middle-aged, non-smoking adults with no respiratory disease, in two urban areas in northern France.

METHODS

A sample of 1506 non-smoking adults (aged from 40 to 65) with no respiratory disease was selected from the participants in the 2011-2013 cross-sectional Enquête Littoral Souffle Air Biologie Environnement (ELISABET) survey in two urban areas in the northern France. We evaluated the associations between (i) mean levels of particulate matter with aerodynamic diameter < 10 μm (PM₁₀), nitrogen dioxide (NO₂) and ozone (O₃) exposure on the day and the day before the study examination for each participant, and (ii) spirometry data and levels of inflammatory markers. Coefficients of multiple linear regression models were expressed (except for the forced expiratory volume in 1 s (FEV1)/forced vital capacity (FVC) ratio) as the percentage change [95% confidence interval] per 10 μg increment in each pollutant.

RESULTS

Levels of PM₁₀, NO₂ and O₃ exposure were below or only close to the World Health Organization's recommended limits in our two study areas. An increment in NO₂ levels was significantly associated with a lower FEV1/FVC ratio (-0.38 [-0.64; -0.12]), a lower forced expiratory flow between 25% and 75% of FVC (FEF25-75%) (-1.70 [-3.15; -0.23]), and a lower forced expiratory flow measured at 75% of FVC (FEF75%) (-3.07 [-4.92; -1.18]). An increment in PM₁₀ levels was associated with lower FEF75% (-1.41 [-2.79; -0.01]) and a non-significant elevation in serum levels of high-sensitivity C-reactive protein (+3.48 [-0.25; 7.36], p = 0.07). Lastly, an increment in O₃ levels was associated with a significantly higher blood eosinophil count (+2.41 [0.10; 4.77]) and a non-significant elevation in fractional exhaled nitric oxide (+2.93 [-0.16; 6.13], p = 0.06).

CONCLUSION

A short-term exposure to air pollution was associated with a subclinical decrement in distal lung function and increment in inflammatory markers in healthy inhabitants of two urban areas in France. If these exploratory results are confirmed, this could suggest that even moderate levels of air pollution could have an impact on respiratory health on the general population, and not solely on susceptible individuals.