No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
COPD phenotypes and machine learning cluster analysis: A systematic review and future research agenda
Abstract
Chronic Obstructive Pulmonary Disease (COPD) is a highly heterogeneous condition projected to become the third leading cause of death worldwide by 2030. To better characterize this condition, clinicians have classified patients sharing certain symptomatic characteristics, such as symptom intensity and history of exacerbations, into distinct phenotypes. In recent years, the growing use of machine learning algorithms, and cluster analysis in particular, has promised to advance this classification through the integration of additional patient characteristics, including comorbidities, biomarkers, and genomic information. This combination would allow researchers to more reliably identify new COPD phenotypes, as well as better characterize existing ones, with the aim of improving diagnosis and developing novel treatments. Here, we systematically review the last decade of research progress, which uses cluster analysis to identify COPD phenotypes. Collectively, we provide a systematized account of the extant evidence, describe the strengths and weaknesses of the main methods used, identify gaps in the literature, and suggest recommendations for future research.
... 20 Studies that utilize cluster analysis combining transcriptomic datasets with COPD-related clinical characteristics, comorbidities and biomarker are important in helping better understand mechanisms underlying the disease as well as strengthening the robustness of any identified COPD phenotypes. 28 This study used unsupervised hierarchical clustering of induced sputum gene expression profiles of 72 stable COPD patients from Newcastle area (Australia) to identify distinct and clinically relevant transcriptional COPD phenotypes, and the driving factors behind these cluster phenotypes. ...
... To this end, the use of techniques such as cluster analysis to identify groups of COPD patients with similar clinical or physiological characteristics has been the focus of studies. 6,28 For instance, in a study that utilized hierarchical cluster analysis of clinical, functional and imaging data of stable COPD patients, Burgel et al 42 identified 3 distinct phenotypes of COPD with varying COPD severity and risk of mortality. Garcia-Aymerich et al 43 also identified three clusters of COPD, namely moderate, severe and systemic COPD. ...
... Studies that have utilized cluster analysis integrating genetic 44 or transcriptomic data 15,[45][46][47] in COPD cohorts, like ours, have great benefit in that they provide the opportunity to analytically and jointly assess COPD-related clinical characteristics, comorbidities, and biomarker data to strengthen the robustness of the COPD phenotypes as well as to better understand the underlying biological mechanisms of the condition. 28 Our study has linked transcriptomic profiles underlying 2 main COPD clusters, and with thorough investigations we identified the main driving factor to these differences in airway gene expression patterns being the degree of lung function impairment. A better understanding of phenotypes of COPD on a deep cellular and molecular level will lead to the development of more targeted strategies for personalized COPD treatment and management. ...
Purpose
This study sought to characterize transcriptional phenotypes of COPD through unsupervised clustering of sputum gene expression profiles, and further investigate mechanisms underlying the characteristics of these clusters.
Patients and methods
Induced sputum samples were collected from patients with stable COPD (n = 72) and healthy controls (n = 15). Induced sputum was collected for inflammatory cell counts, and RNA extracted. Transcriptional profiles were generated (Illumina Humanref-8 V2) and analyzed by GeneSpring GX14.9.1. Unsupervised hierarchical clustering and differential gene expression analysis were performed, and gene alterations validated in the ECLIPSE dataset (GSE22148).
Results
We identified 2 main clusters (Cluster 1 [n = 35] and Cluster 2 [n = 37]), which further divided into 4 sub-clusters (Sub-clusters 1.1 [n = 14], 1.2 [n = 21], 2.1 [n = 20] and 2.2 [n = 17]). Compared with Cluster 1, Cluster 2 was associated with significantly lower lung function (p = 0.014), more severe disease (p = 0.009) and breathlessness (p = 0.035), and increased sputum neutrophils (p = 0.031). Sub-cluster 1.1 had significantly higher proportion of people with comorbid cardiovascular disease compared to the other 3 sub-clusters (92.5% vs 57.1%, 50% and 52.9%, p < 0.013). Through supervised analysis we determined that degree of airflow limitation (GOLD stage) was the predominant factor driving gene expression differences in our transcriptional clusters. There were 452 genes (adjusted p < 0.05 and ≥2 fold) altered in GOLD stage 3 and 4 versus 1 and 2, of which 281 (62%) were also found to be significantly expressed between these GOLD stages in the ECLIPSE data set (GSE22148). Differentially expressed genes were largely downregulated in GOLD stages 3 and 4 and connected in 5 networks relating to lipoprotein and cholesterol metabolism; metabolic processes in oxidation/reduction and mitochondrial function; antigen processing and presentation; regulation of complement activation and innate immune responses; and immune and metabolic processes.
Conclusion
Severity of lung function drives 2 distinct transcriptional phenotypes of COPD and relates to immune and metabolic processes.
... Research efforts have been made to advance knowledge in this field, namely through the identification of homogeneous subgroups of patients with COPD [3,4], the so-called clinical phenotypes or profiles, grouped by different type of personal characteristics (e.g., genetic, clinical, biochemical, radiological) for prognostic and therapeutic purposes [3][4][5][6][7][8][9][10][11][12]. More recently, a new approach, "treatable traits", i.e., pulmonary, extra-pulmonary and behaviour/lifestyle characteristics of each person that are clinically relevant, identifiable and treatable, emerged [13][14][15]. ...
... Research efforts have been made to advance knowledge in this field, namely through the identification of homogeneous subgroups of patients with COPD [3,4], the so-called clinical phenotypes or profiles, grouped by different type of personal characteristics (e.g., genetic, clinical, biochemical, radiological) for prognostic and therapeutic purposes [3][4][5][6][7][8][9][10][11][12]. More recently, a new approach, "treatable traits", i.e., pulmonary, extra-pulmonary and behaviour/lifestyle characteristics of each person that are clinically relevant, identifiable and treatable, emerged [13][14][15]. ...
... More recently, a new approach, "treatable traits", i.e., pulmonary, extra-pulmonary and behaviour/lifestyle characteristics of each person that are clinically relevant, identifiable and treatable, emerged [13][14][15]. Although studies on clinical profiles and treatable traits have been conducted, their crosssectional nature, narrow eligibility criteria, main focus on physiological/pulmonary measures often not available across settings [5,6], absence of decision trees and lack of validation with independent samples [3], limits our understanding of the heterogeneous manifestations of COPD and hinders their applicability in daily clinical practice. ...
Abstract Background and objective Profiles of people with chronic obstructive pulmonary disease (COPD) often do not describe treatable traits, lack validation and/or their stability over time is unknown. We aimed to identify COPD profiles and their treatable traits based on simple and meaningful measures; to develop and validate a decision tree and to explore profile stability over time. Methods An observational, prospective study was conducted. Clinical characteristics, lung function, symptoms, impact of the disease (COPD Assessment Test—CAT), health-related quality of life, physical activity, lower-limb muscle strength and functional status were collected cross-sectionally and a subsample was followed-up monthly over six months. A principal component analysis and a clustering procedure with k-medoids were applied to identify profiles. A decision tree was developed and validated cross-sectionally. Stability was explored over time with the ratio between the number of timepoints that a participant was classified in the same profile and the total number of timepoints (i.e., 6). Results 352 people with COPD (67.4 ± 9.9 years; 78.1% male; FEV1 = 56.2 ± 20.6% predicted) participated and 90 (67.6 ± 8.9 years; 85.6% male; FEV1 = 52.1 ± 19.9% predicted) were followed-up. Four profiles were identified with distinct treatable traits. The decision tree included CAT (
... Some articles have reviewed the progress of AI techniques in COPD [56][57][58]. Exarchos et al. reviewed the general adoption of AI in COPD research, categorizing the studies into 'COPD diagnosis' , 'COPD prognosis' , 'Patient classification' , and 'COPD management' . It identified an acceleration of AI use in COPD research and calls for broader adoption due to the large and complex data involved [57]. ...
... The article published by Nikolaou et al. focused on the use of machine learning algorithms, specifically cluster analysis, to better characterize COPD through integration of patient characteristics like symptoms, comorbidities, biomarkers, and genomic information. It reviewed the progress of research in the past decade using cluster analysis for COPD phenotypes [56]. Estépar's article provided an introduction to AI and deep learning, discussing their role in understanding the evolution and divergent trajectories of COPD. ...
Chronic obstructive pulmonary disease (COPD) stands as a significant global health challenge, with its intricate pathophysiological manifestations often demanding advanced diagnostic strategies. The recent applications of artificial intelligence (AI) within the realm of medical imaging, especially in computed tomography, present a promising avenue for transformative changes in COPD diagnosis and management. This review delves deep into the capabilities and advancements of AI, particularly focusing on machine learning and deep learning, and their applications in COPD identification, staging, and imaging phenotypes. Emphasis is laid on the AI-powered insights into emphysema, airway dynamics, and vascular structures. The challenges linked with data intricacies and the integration of AI in the clinical landscape are discussed. Lastly, the review casts a forward-looking perspective, highlighting emerging innovations in AI for COPD imaging and the potential of interdisciplinary collaborations, hinting at a future where AI doesn’t just support but pioneers breakthroughs in COPD care. Through this review, we aim to provide a comprehensive understanding of the current state and future potential of AI in shaping the landscape of COPD diagnosis and management.
... Although patients with acute exacerbations often reported a poor HRQoL (72), minor statistically significant changes in participant's QoL were observed in both our study clusters. In terms of the clinical features used to define the COPD phenotypes, no standardization exists; however, Nikolaou et al. (71) suggested to complement risk-stratification models based on clinical severity, as was performed in this study to define both clusters of COPD patients, with other determinants, such as physiological characteristics (e.g., age, body mass index, waist circumference), comorbidities, pulmonary function tests, biomarkers, and genetic variants. In addition, Parikh et al. (73) advised including social, economic, behavioral, and environmental determinants of health when defining phenotypes. ...
... Regardless of the clustering method used, Nikolaou et al. (71) recommended the use of prospective longitudinal data with large samples to develop clinically meaningful COPD-derived phenotypes as clustering methods are data driven techniques. Although there is a known bias of estimates with small samples, there is no theorem supporting the rule of thumb for the size of the dataset for cluster analysis; yet hierarchical clustering works (74). ...
Introduction
Inconclusive results exist around the effectiveness of telemonitoring for patients with COPD, and studies recommended conducting subgroup analyses to identify patient phenotypes that could benefit from these services. This exploratory study investigated what type of COPD patients were receiving which type of telenursing interventions more frequently using the telemonitoring platform.
Methods
A sample of 36 older adults with COPD were receiving telenursing services for 12 months and were asked to answer five COPD-symptom related questions and submit their vital signs daily.
Results
Findings revealed two phenotypes of older adults for whom the frequency of telenursing calls and related interventions differed. Although no statistically significant differences were observed in participants' GOLD grades and hospitalizations, cluster one participants used their COPD action plan significantly more frequently, and were in frequent contact with the telenurse.
Discussion
It is paramount that further research is needed on the development of patient phenotypes who may benefit from telemonitoring.
... In order to gain an overview of the common practices with regards to mixed data handling in subtyping for COPD, we examined three systematic review [10][11][12] and expanded the search (non-systematic) to further studies from the most recent literature. We also discuss the findings of a relevant systematic review by Horne et al. [13], which focused on the methodological challenges of cluster analysis using mixed-type data in asthma. ...
... An exhaustive hyperparameter search was run in order to select the best autoencoder with regards to number of layers (ranging in [2][3][4][5]), number of hidden units (ranging in [10,12,16] ...
Background
The ever-growing size, breadth, and availability of patient data allows for a wide variety of clinical features to serve as inputs for phenotype discovery using cluster analysis. Data of mixed types in particular are not straightforward to combine into a single feature vector, and techniques used to address this can be biased towards certain data types in ways that are not immediately obvious or intended. In this context, the process of constructing clinically meaningful patient representations from complex datasets has not been systematically evaluated.
Aims
Our aim was to a) outline and b) implement an analytical framework to evaluate distinct methods of constructing patient representations from routine electronic health record data for the purpose of measuring patient similarity. We applied the analysis on a patient cohort diagnosed with chronic obstructive pulmonary disease.
Methods
Using data from the CALIBER data resource, we extracted clinically relevant features for a cohort of patients diagnosed with chronic obstructive pulmonary disease. We used four different data processing pipelines to construct lower dimensional patient representations from which we calculated patient similarity scores. We described the resulting representations, ranked the influence of each individual feature on patient similarity and evaluated the effect of different pipelines on clustering outcomes. Experts evaluated the resulting representations by rating the clinical relevance of similar patient suggestions with regard to a reference patient.
Results
Each of the four pipelines resulted in similarity scores primarily driven by a unique set of features. It was demonstrated that data transformations according to each pipeline prior to clustering can result in a variation of clustering results of over 40%. The most appropriate pipeline was selected on the basis of feature ranking and clinical expertise. There was moderate agreement between clinicians as measured by Cohen’s kappa coefficient.
Conclusions
Data transformation has downstream and unforeseen consequences in cluster analysis. Rather than viewing this process as a black box, we have shown ways to quantitatively and qualitatively evaluate and select the appropriate preprocessing pipeline.
... Pentru a caracteriza mai bine această condiție, clinicienii au clasificat pacienții care împărtășesc anumite caracteristici simptom-atice, cum ar fi intensitatea simptomelor și istoricul exacerbărilor, în fenotipuri distincte [8]. Utilizarea tot mai mare a algoritmilor de învățare automată și, în special, a analizei clusterelor, a promis să avanseze această clasificare prin integrarea caracterului suplimentar al pacientului, inclusiv comorbidități, biomarkeri și informații genomice [9]. ...
... În ultimii ani, s-au făcut progrese în ceea ce privește clasificarea PH, evaluarea diagnostică și algoritmul de management la pacienții cu suspiciune de PH, cu multe provocări clinice rămase, în pofida publicării recente a două ghiduri internaționale [6,8]. În prezent, diagnosticul este lăsat fie la latitudinea competenței specialiștilor pneumologi, fie stabilit în cadrul discuției unei echipe multidisciplinare (MDT), în baza unui istoric detaliat de expuneri, a datelor clinice și imagistice, a lavajului bronhoalveolar (LBA) și a testelor serologice precum și histologice [9]. ...
... Phenotyping can be aided by using descriptive statistics, such as cluster analysis to identify separate patient groups according to preselected variables [4]. With regards to these variables, patients within a certain cluster are more similar to each other than to patients in different clusters [5]. ...
... Several attempts have been made to develop a useful classification of phenotypes of COPD patients [4]. To be potentially useful in clinical practice, the identity of the defined clusters needs to be confirmed in different, independent cohorts of COPD patients, but to the best of our knowledge such replication studies have not been performed yet. ...
Background
The global initiative for chronic obstructive lung disease (GOLD) 2020 emphasizes that there is only a weak correlation between FEV1, symptoms and impairment of the health status of patients with chronic obstructive pulmonary disease (COPD). Various studies aimed to identify COPD phenotypes by cluster analyses, but behavioral aspects besides smoking were rarely included.
Methods
The aims of the study were to investigate whether (i) clustering analyses are in line with the classification into GOLD ABCD groups; (ii) clustering according to Burgel et al. (Eur Respir J. 36(3):531–9, 2010) can be reproduced in a real-world COPD cohort; and (iii) addition of new behavioral variables alters the clustering outcome. Principal component and hierarchical cluster analyses were applied to real-world clinical data of COPD patients newly referred to secondary care (n = 155). We investigated if the obtained clusters paralleled GOLD ABCD subgroups and determined the impact of adding several variables, including quality of life (QOL), fatigue, satisfaction relationship, air trapping, steps per day and activities of daily living, on clustering.
Results
Using the appropriate corresponding variables, we identified clusters that largely reflected the GOLD ABCD groups, but we could not reproduce Burgel’s clinical phenotypes. Adding six new variables resulted in the formation of four new clusters that mainly differed from each other in the following parameters: number of steps per day, activities of daily living and QOL.
Conclusions
We could not reproduce previously identified clinical COPD phenotypes in an independent population of COPD patients. Our findings therefore indicate that COPD phenotypes based on cluster analysis may not be a suitable basis for treatment strategies for individual patients.
... According to the World Health Organisation (WHO), COPD is projected to become the third leading cause of death by 2030 [2] because our ability to diagnose early and treat effectively has been relatively static. To better understand the heterogeneity of COPD, recent and ongoing research [3] is applying a wide range of machine learning methods, which can integrate patients' demographic and clinical characteristics to derive underlying disease traits that often occur together (i.e., COPD phenotypes). Among these, the cardiovascular phenotype remains one of the most relevant phenotypes to analyse, given that cardiovascular disease is the major contributor to morbidity and mortality in patients with COPD [4]. ...
... To the best of our knowledge, this study is the first to implement machine learning to identify clinically meaningful phenotypes of cardiovascular comorbidities that develop after a COPD diagnosis, though we are not the first to apply machine learning to COPD in general [3]. ...
Background
Chronic Obstructive Pulmonary Disease (COPD) is a heterogeneous group of lung conditions that are challenging to diagnose and treat. As the presence of comorbidities often exacerbates this scenario, the characterization of patients with COPD and cardiovascular comorbidities may allow early intervention and improve disease management and care.
Methods
We analysed a 4-year observational cohort of 6,883 UK patients who were ultimately diagnosed with COPD and at least one cardiovascular comorbidity. The cohort was extracted from the UK Royal College of General Practitioners and Surveillance Centre database. The COPD phenotypes were identified prior to diagnosis and their reproducibility was assessed following COPD diagnosis. We then developed four classifiers for predicting cardiovascular comorbidities.
Results
Three subtypes of the COPD cardiovascular phenotype were identified prior to diagnosis. Phenotype A was characterised by a higher prevalence of severe COPD, emphysema, hypertension. Phenotype B was characterised by a larger male majority, a lower prevalence of hypertension, the highest prevalence of the other cardiovascular comorbidities, and diabetes. Finally, phenotype C was characterised by universal hypertension, a higher prevalence of mild COPD and the low prevalence of COPD exacerbations. These phenotypes were reproduced after diagnosis with 92% accuracy. The random forest model was highly accurate for predicting hypertension while ruling out less prevalent comorbidities.
Conclusions
This study identified three subtypes of the COPD cardiovascular phenotype that may generalize to other populations. Among the four models tested, the random forest classifier was the most accurate at predicting cardiovascular comorbidities in COPD patients with the cardiovascular phenotype.
... This heterogeneity has prompted a shift towards a phenotypical approach in managing COPD, which aims to tailor treatment strategies based on individual patient profiles. 1,2 Recent studies have focused on identifying clinically relevant subtypes of COPD, revealing that patients can be grouped based on their symptoms, lung function, and other clinical parameters. 3 The quantitative imaging methods technical development has led to their increased use in the diagnosis and management of patients with COPD. ...
Purpose
To explore the quantitative imaging phenotype differences for CT-defined subtypes classified by the Fleischner Society in patients with chronic obstructive pulmonary disease (COPD).
Patients and Methods
A total of 228 COPD patients who underwent non-enhanced chest CT screening from 2018 to 2024 were included. All patients were divided into type-A (Absent emphysema that no or mild emphysema, Goddard score ≤8, regardless of bronchial wall thickening), type-E (Emphysema that significant emphysema, Goddard score >8, without bronchial wall thickening), and type-M (Mixed emphysema and bronchial wall thickening that both significant emphysema, Goddard score >8, and bronchial wall thickening ≥ grade 1 in ≥1 lung lobe). Imaging phenotype parameters included lung airspace analysis (LAA) and LAA size analysis (LAASA) in emphysema, airway wall, lung vessels and interstitial lung disease (ILD) extracted by a COPD-specific analysis software were analysis among three groups.
Results
Quantitative assessment of emphysema among three image phenotypes showed significant differences in full emphysema and full emphysema ratio based on LAA among three groups (P < 0.05). The areas of consolidation, ground-glass opacity, and reticular patterns were significantly larger in type-M than the other two types (P < 0.05). Quantitative assessment of small airways disease and small vessel parameters found smaller lumen-volume and larger wall-volume in whole lung level in the emphysema phenotype of type-M (P < 0.05) were found in the small vessel count in distance of 6 mm and 9mm from the pleura were significant differences among three groups (P < 0.05). The multivariate logistic regression analysis showed that the higher proportion of full emphysema ratio and wall-volume, a proportion of smaller lumen-volume, and a more noticeable interstitial lung alterations were associated with type-M.
Conclusion
A quantitative CT evaluation can further delineate the imaging phenotypes characteristics thereby in guiding to early diagnosis, severity assessment, and therapeutic recommendations in COPD patients.
... Machine learning (ML)-based phenotyping is increasingly utilized in research, particularly with so-called unsupervised cluster analysis, as it enables discovery of latent patterns in high-dimensional data, with theoretically less bias than the historically common practice of deriving disease subtypes through clinical experience [8,9]. Previous ML-based work has identified meaningful and distinct phenotypes of e.g., asthma [10] and chronic obstructive pulmonary disease (COPD) [11]. However, such analyses have been limited to subjects with pre-defined and diagnosed disease. ...
Background
Respiratory symptoms are common in the general adult population. Increased burden of respiratory symptoms may increase the risk of mortality. We assessed the association between respiratory symptom clusters and mortality.
Methods
Participants were derived from two population-based Swedish adult cohorts (N = 63,060). Cluster analysis was performed with Locality Sensitive Hashing (LSH)-k-prototypes in subjects with ≥ 1 self-reported respiratory symptom. Linked mortality register data (up to 21 years of follow-up, > 600,000 person-years) were used. Associations between clusters and all-cause/cause-specific mortality were assessed using asymptomatic subjects as reference.
Results
Over 60% reported ≥ 1 respiratory symptom and ~ 30% reported ≥ 5 respiratory symptoms. Five clusters were identified, partly overlapping with established respiratory disease phenotypes but many individuals were undiagnosed: (1) "low-symptomatic" (30.3%); (2) "allergic nasal symptoms" (10.7%); (3) "allergic nasal symptoms, wheezing, and dyspnea attacks" (4.7%); (4) "wheezing and dyspnea attacks" (6.6%); (5) "recurrent productive cough and wheezing" (4.1%). All but Cluster 2 were associated with all-cause mortality, highest risk for Cluster 3 (hazard ratio 1.4, 95% confidence interval 1.13–1.73) and Cluster 5 (1.4, 1.22–1.61). Comparable associations were seen for cardiovascular mortality. For respiratory mortality, Cluster 4 (2.02, 1.18–3.46) and Cluster 5 (1.89, 1.1–3.25) were most strongly associated.
Conclusions
Respiratory symptoms are common in the general adult population, with identifiable clusters. These clusters have clinical relevancy as they are differentially associated with mortality and relatively weakly correlated with diagnosed respiratory disease.
... Some studies also introduced AI and deep learning as tools to recognize COPD's progression and diverse paths. The study examined AI's effectiveness in making medical decisions, radiographic comprehension, and prognosis, as well as its potential benefits, problems, and drawbacks in COPD [26]. ...
Artificial intelligence refers to the capacity of computers to execute functions that typically demand human cognition, such as acquiring knowledge, logical thinking, and resolving challenges. Artificial intelligence is advancing across various domains, with health care being one of the most significant. Artificial intelligence has revolutionised respiratory care through early recognition of the diseases, patient tailored medical interventions as well as patient tracking using machine learning and deep learning technologies, not only this but artificial intelligence can also be used in early diagnosis and a better prognosis of respiratory conditions. Algorithms that are developed using machine learning can analyse enormous number of hospital information as well as healthcare information to identify patterns that normal diagnosis procedures frequently overlook. Various respiratory diseases including Idiopathic pulmonary fibrosis, pneumoconiosis, lung cancer, chronic obstructive pulmonary disease as well as asthma have all benefited from the use of artificial intelligence as it not only leads to early detection, but nowadays, various rehabilitative programmes are also available that incorporate the use of artificial intelligence to provide better healthcare facilities for individuals suffering from these condition.
... 2,3 The incidence rate of COPD is high and is expected to rise to become the world's third leading cause of death by 2030. 4 In China alone, the prevalence of COPD in adults is 8.6%, especially in people aged ≥ 40 years, with a prevalence rate as high as 13.7% and affecting nearly 100 million people. 5 Presently, the clinical use of bronchodilators and inhaled glucocorticoids does not effectively interrupt the progress of chronic inflammation in COPD, and it is associated with adverse reactions and drug resistance, which limits the overall clinical application of these drugs. ...
... Although the use of cluster analysis is not new in the field of COPD, in the vast majority of the already published studies, it has only been applied to clinical variables, including some co-authored by members of our group [38][39][40]. This strategy is fully justified to identify new clinical phenotypes, but it does not consider the contribution of the corresponding biological substrates (endotypes) and, therefore, that of their corresponding pathophysiological mechanisms. ...
Although Chronic Obstructive Pulmonary Disease (COPD) is highly prevalent, it is often underdiagnosed. One of the main characteristics of this heterogeneous disease is the presence of periods of acute clinical impairment (exacerbations). Obtaining blood biomarkers for either COPD as a chronic entity or its exacerbations (AECOPD) will be particularly useful for the clinical management of patients. However, most of the earlier studies have been characterized by potential biases derived from pre-existing hypotheses in one or more of their analysis steps: some studies have only targeted molecules already suggested by pre-existing knowledge, and others had initially carried out a blind search but later compared the detected biomarkers among well-predefined clinical groups. We hypothesized that a clinically blind cluster analysis on the results of a non-hypothesis-driven wide proteomic search would determine an unbiased grouping of patients, potentially reflecting their endotypes and/or clinical characteristics. To check this hypothesis, we included the plasma samples from 24 clinically stable COPD patients, 10 additional patients with AECOPD, and 10 healthy controls. The samples were analyzed through label-free liquid chromatography/tandem mass spectrometry. Subsequently, the Scikit-learn machine learning module and K-means were used for clustering the individuals based solely on their proteomic profiles. The obtained clusters were confronted with clinical groups only at the end of the entire procedure. Although our clusters were unable to differentiate stable COPD patients from healthy individuals, they segregated those patients with AECOPD from the patients in stable conditions (sensitivity 80%, specificity 79%, and global accuracy, 79.4%). Moreover, the proteins involved in the blind grouping process to identify AECOPD were associated with five biological processes: inflammation, humoral immune response, blood coagulation, modulation of lipid metabolism, and complement system pathways. Even though the present results merit an external validation, our results suggest that the present blinded approach may be useful to segregate AECOPD from stability in both the clinical setting and trials, favoring more personalized medicine and clinical research.
... Chronic obstructive pulmonary disease According to WHO estimates, COPD will become the third leading cause of death worldwide by 2030, probably due to difficulties in early and accurate diagnosis . (6) The classification of causes of death published by the WHO lists individual chronic diseases as the sole causes responsible for the event. The reality is different, as multimorbidity, defined as the presence of two or more chronic diseases, is seen in almost 50% of adults . ...
Patients with chronic obstructive pulmonary disease (COPD) frequently suffer from multimorbidity, defined as the presence of 2 or more comorbidities in a patient, the most common of which are cardiovascular and cerebrovascular disease, lung cancer, diabetes, muscle weakness, osteoporosis, anxiety and depression. In patients with COPD, comorbidities can develop syndemically, that is, they can evolve simultaneously in response to common risk factors and through common pathogenetic mechanisms. The authors present the case of a patient with multimorbidity, with a history of a common risk factor (smoking) for some of the associated comorbidities and challenges occurred in his therapeutic approach.
... COPD is the leading cause of lung diseaseassociated morbidity and mortality, and its incidence has been increasing globally (Singh et al., 2019). It has been predicted that by 2030, COPD will be the third leading cause of death worldwide, imposing a heavy socioeconomic burden (Nikolaou et al., 2020). COPD onset is closely correlated with airway and lung inflammation caused by harmful particles and smoke (Lareau et al., 2019;Global Initiative for Chronic Obstruc, 2021). ...
Background: Chronic obstructive pulmonary disease (COPD) affects approximately 400 million people worldwide and is associated with high mortality and morbidity. The effect of EPHX1 and GSTP1 gene polymorphisms on COPD risk has not been fully characterized.
Objective: To investigate the association of EPHX1 and GSTP1 gene polymorphisms with COPD risk.
Methods: A systematic search was conducted on 9 databases to identify studies published in English and Chinese. The analysis was conducted following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses reporting guidelines (PRISMA). The pooled OR and 95% CI were calculated to evaluate the association of EPHX1 and GSTP1 gene polymorphisms with COPD risk. The I² test, Q test, Egger’s test, and Begg’s test were conducted to determine the level of heterogeneity and publication bias of the included studies.
Results: In total, 857 articles were retrieved, among which 59 met the inclusion criteria. The EPHX1 rs1051740 polymorphism (homozygote, heterozygote, dominant, recessives, and allele model) was significantly associated with high risk of COPD risk. Subgroup analysis revealed that the EPHX1 rs1051740 polymorphism was significantly associated with COPD risk among Asians (homozygote, heterozygote, dominant, and allele model) and Caucasians (homozygote, dominant, recessives, and allele model). The EPHX1 rs2234922 polymorphism (heterozygote, dominant, and allele model) was significantly associated with a low risk of COPD. Subgroup analysis showed that the EPHX1 rs2234922 polymorphism (heterozygote, dominant, and allele model) was significantly associated with COPD risk among Asians. The GSTP1 rs1695 polymorphism (homozygote and recessives model) was significantly associated with COPD risk. Subgroup analysis showed that the GSTP1 rs1695 polymorphism (homozygote and recessives model) was significantly associated with COPD risk among Caucasians. The GSTP1 rs1138272 polymorphism (heterozygote and dominant model) was significantly associated with COPD risk. Subgroup analysis suggested that the GSTP1 rs1138272 polymorphism (heterozygote, dominant, and allele model) was significantly associated with COPD risk among Caucasians.
Conclusion: The C allele in EPHX1 rs1051740 among Asians and the CC genotype among Caucasians may be risk factors for COPD. However, the GA genotype in EPHX1 rs2234922 may be a protective factor against COPD in Asians. The GG genotype in GSTP1 rs1695 and the TC genotype in GSTP1 rs1138272 may be risk factors for COPD, especially among Caucasians.
... A 2022 review of artificial intelligence (AI) techniques in COPD yielded 156 articles relevant to the application of AI in COPD research, including 56 concerning diagnosis, 65 on its prognosis, 54 on COPD severity classification, and 17 on the management of the disease [27]. Most studies have used a variety of features, including patient physiological characteristics, comorbidities, symptoms, vital signs, biomarkers, genomic information, pulmonary function tests, CT images, hospitalization information, and/or breath sounds [28,29]. Regardless of the method(s) chosen, COPD remains an incurable and progressive disease and diagnosis at the early risk stage is important. ...
Background
Interstitial lung diseases (ILD), such as idiopathic pulmonary fibrosis (IPF) and non-specific interstitial pneumonia (NSIP), and chronic obstructive pulmonary disease (COPD) are severe, progressive pulmonary disorders with a poor prognosis. Prompt and accurate diagnosis is important to enable patients to receive appropriate care at the earliest possible stage to delay disease progression and prolong survival. Artificial intelligence-assisted lung auscultation and ultrasound (LUS) could constitute an alternative to conventional, subjective, operator-related methods for the accurate and earlier diagnosis of these diseases. This protocol describes the standardised collection of digitally-acquired lung sounds and LUS images of adult outpatients with IPF, NSIP or COPD and a deep learning diagnostic and severity-stratification approach.
Methods
A total of 120 consecutive patients (≥ 18 years) meeting international criteria for IPF, NSIP or COPD and 40 age-matched controls will be recruited in a Swiss pulmonology outpatient clinic, starting from August 2022. At inclusion, demographic and clinical data will be collected. Lung auscultation will be recorded with a digital stethoscope at 10 thoracic sites in each patient and LUS images using a standard point-of-care device will be acquired at the same sites. A deep learning algorithm (DeepBreath) using convolutional neural networks, long short-term memory models, and transformer architectures will be trained on these audio recordings and LUS images to derive an automated diagnostic tool. The primary outcome is the diagnosis of ILD versus control subjects or COPD. Secondary outcomes are the clinical, functional and radiological characteristics of IPF, NSIP and COPD diagnosis. Quality of life will be measured with dedicated questionnaires. Based on previous work to distinguish normal and pathological lung sounds, we estimate to achieve convergence with an area under the receiver operating characteristic curve of > 80% using 40 patients in each category, yielding a sample size calculation of 80 ILD (40 IPF, 40 NSIP), 40 COPD, and 40 controls.
Discussion
This approach has a broad potential to better guide care management by exploring the synergistic value of several point-of-care-tests for the automated detection and differential diagnosis of ILD and COPD and to estimate severity.
Trial registration Registration: August 8, 2022. ClinicalTrials.gov Identifier: NCT05318599.
... It is also one of the important causes of disease and death globally. 1,2 According to available statistics, some 600 million people worldwide have been diagnosed with COPD, while about 100 million COPD patients in China. ...
Objective:
To integrate Meta-analysis and bioinformatics strategies in the preliminary exploration of the potential mechanism of Yinyanghuo () and its extract in treating chronic obstructive pulmonary disease (COPD).
Methods:
First, Meta-analysis was carried out. The Chinese and English literature of Yinyanghuo () in treating COPD was searched using the systematic strategy of combining subject words with free words. The included studies were evaluated by the SYRCLE risk bias assessment tool, after which the review manager software was used to combine the effect quantities for statistical analysis. Then, based on bioinformatics technology, the active ingredients and their targets of Yinyanghuo () were screened, and the intersection genes were obtained by mapping and comparing with the targets of COPD. The "medicinal materials-compounds-targets model" was constructed, and the key pathways were annotated. Finally, the core target was docked with important compounds.
Results:
A total of 8 studies were included in the Meta-analysis. The results showed that the Yinyanghuo (Herba Epimedii Brevicornus) group could significantly down-regulate pro-inflammatory factors such as tumor necrosis factor-α (TNF-α) and interleukin (IL)-8 and increase the expression of anti-inflammatory factors and antioxidant factors such as IL-10 and phospho-protein kinase B (p-AKT) in the COPD model (all P < 0.05). A total of 23 active components and 102 corresponding target genes of Yinyanghuo (Herba Epimedii Brevicornus) were obtained by bioinformatics technology, among which 17 compounds and 63 targets were closely related to COPD. The results of enrichment analysis mainly included TNF signaling pathway, phosphoinositide 3-kinase (PI3K)/Akt signaling pathway, cancer signaling pathway, and other inflammatory reactions, oxidative stress, and tumor-related pathways. The molecular docking results showed that the binding energy fractions of the top five components of 24-epicampesterol with 10 core targets such as IL-6 were all less than ﹣5.0 kcal/mol, suggesting good binding ability.
Conclusions:
Meta-analysis and bioinformatics results indicated that the therapeutic effect of Yinyanghuo () and its components on COPD might be related to antagonizing inflammation and oxidative stress. The above findings provide a preliminary basis for the development of Yinyanghuo () as a natural drug for preventing and treating COPD.
... En los últimos años se ha observado un aumento del uso de algoritmos de AA para clasificar con mayor precisión los fenotipos de la enfermedad a través de clústeres que permiten integrar información de los pacientes y encontrar estructuras en los datos que sean similares entre sí (10). En este ámbito de investigación también se han establecido puntos de buenas prácticas como son el uso de datos longitudinales prospectivos, cohortes que incluyan diferentes poblaciones y escenarios, inclusión de información genética como la disponible en las bases de datos de COPDgene y la garantía de claridad en la selección de las variables que se usen para identificar los fenotipos (10). ...
Cada día encontramos con mayor frecuencia el término inteligencia artificial en todos los escenarios de la medicina incluyendo la neumología, siendo una necesidad el conocimiento en el tema y el acercamiento del profesional de salud al uso de estas herramientas. La aplicabilidad de la inteligencia artificial en neumología tuvo sus inicios en la interpretación de pruebas de función pulmonar y en ámbitos de diagnóstico. Sin embargo, rápidamente vemos la importancia del uso en soporte de decisiones, monitoreo y vigilancia de desenlaces en la práctica diaria y en el campo de investigación, modelos de predicción clínicos, solo para mencionar algunos. Este artículo es una revisión narrativa del papel de la inteligencia artificial en algunas de las patologías más frecuentemente vistas en nuestra práctica diaria. Describimos el uso de inteligencia artificial en el diagnóstico clínico, funcional e imagenológico de la EPOC y el asma así como en el monitoreo remoto de los pacientes, los avances en la interpretación de imágenes y de patología de las enfermedades intersticiales con el uso de Aprendizaje Automatizado (AA) y Aprendizaje Profundo (AP), la aplicación en tamización de hipertensión pulmonar a partir de pruebas diferentes al cateterismo cardiaco derecho, y finalmente la amplia gama de aplicaciones en medicina del sueño, en la que el avance es asombroso.
... There is an intermediate group including diseases of the circulatory system (N = 9) [65][66][67][68][69][70][71][72][73], diseases of the musculoskeletal system (N = 8) [74][75][76][77][78][79][80][81], diseases of the digestive system (N = 7) [82][83][84][85][86][87][88] and the nervous system (N = 7) [89][90][91][92][93][94][95]. There are a few reviews focused on other categories such as diseases of respiratory system (N = 4) [96][97][98][99], visual system (N = 3) [100][101][102] and the other chapters include only one or two systematic reviews [103][104][105][106][107][108][109][110][111][112][113][114][115][116]. The number of studies included in the reviews is 41.29 ± 41.9 (mean ± standard deviation) with an IQR = . ...
Background
Artificial intelligence is fueling a new revolution in medicine and in the healthcare sector. Despite the growing evidence on the benefits of artificial intelligence there are several aspects that limit the measure of its impact in people’s health. It is necessary to assess the current status on the application of AI towards the improvement of people’s health in the domains defined by WHO’s Thirteenth General Programme of Work (GPW13) and the European Programme of Work (EPW), to inform about trends, gaps, opportunities, and challenges
Objective
To perform a systematic overview of systematic reviews on the application of artificial intelligence in the people’s health domains as defined in the GPW13 and provide a comprehensive and updated map on the application specialties of artificial intelligence in terms of methodologies, algorithms, data sources, outcomes, predictors, performance, and methodological quality.
Methods
A systematic search in MEDLINE, EMBASE, Cochrane and IEEEXplore was conducted between January 2015 and June 2021 to collect systematic reviews using a combination of keywords related to the domains of universal health coverage, health emergencies protection, and better health and wellbeing as defined by the WHO’s PGW13 and EPW. Eligibility criteria was based on methodological quality and and the inclusion of practical implementation of artificial intelligence. Records were classified and labeled using ICD-11 categories into the domains of the GPW13. Descriptors related to the area of implementation, type of modeling, data entities, outcomes and implementation on care delivery were extracted using a structured form and methodological aspects of the included reviews studies was assessed using the AMSTAR checklist.
Results
The search strategy resulted in the screening of 815 systematic reviews from which 203 were assessed for eligibility and 129 were included in the review. The most predominant domain for artificial intelligence applications was Universal Health Coverage (N=98) followed by Health Emergencies (N=16) and Better Health and Wellbeing (N=15). Neoplasms area on Universal Health Coverage was the disease area featuring most of the applications (21.7%, N=28). The reviews featured analytics primarily over both public and private data sources (67.44%, N=87). The most used type of data was medical imaging (31.8%, N=41) and predictors based on regions of interest and clinical data. The most prominent subdomain of Artificial Intelligence was Machine Learning (43.4%, N=56), in which Support Vector Machine method was predominant (20.9%, N=27). Regarding the purpose, the application of Artificial Intelligence I is focused on the prediction of the diseases (36.4%, N=47). With respect to the validation, more than a half of the reviews (54.3%, N=70) did not report a validation procedure and, whenever available, the main performance indicator was the accuracy (28.7%, N=37). According to the methodological quality assessment, a third of the reviews (34.9%, N=45) implemented methods for analysis the risk of bias and the overall AMSTAR score below was 5 (4.01±1.93) on all the included systematic reviews.
Conclusion
Artificial intelligence is being used for disease modelling, diagnose, classification and prediction in the three domains of GPW13. However, the evidence is often limited to laboratory. and the level of adoption is largely unbalanced between ICD-11 categoriesand diseases. Data availability is a determinant factor on the developmental stage of artificial intelligence applications. Most of the reviewed studies show a poor methodological quality and are at high risk of bias, which limits the reproducibility of the results and the reliability of translating these applications to real clinical scenarios. The analyzed papers show results only in laboratory and testing scenarios and not in clinical trials nor case studies, limiting the supporting evidence to transfer artificial intelligence to actual care delivery.
... En los últimos años se ha observado un aumento del uso de algoritmos de AA para clasificar con mayor precisión los fenotipos de la enfermedad a través de clústeres que permiten integrar información de los pacientes y encontrar estructuras en los datos que sean similares entre sí (10). En este ámbito de investigación también se han establecido puntos de buenas prácticas como son el uso de datos longitudinales prospectivos, cohortes que incluyan diferentes poblaciones y escenarios, inclusión de información genética como la disponible en las bases de datos de COPDgene y la garantía de claridad en la selección de las variables que se usen para identificar los fenotipos (10). ...
Cada día encontramos con mayor frecuencia el término inteligencia artificial en todos los escenarios de la medicina incluyendo la neumología, siendo una necesidad el conocimiento en el tema y el acercamiento del profesional de salud al uso de estas herramientas. La aplicabilidad de la inteligencia artificial en neumología tuvo sus inicios en la interpretación de pruebas de función pulmonar y en ámbitos de diagnóstico. Sin embargo, rápidamente vemos la importancia del uso en soporte de decisiones, monitoreo y vigilancia de desenlaces en la práctica diaria y en el campo de investigación, modelos de predicción clínicos, solo para mencionar algunos. Este artículo es una revisión narrativa del papel de la inteligencia artificial en algunas de las patologías más frecuentemente vistas en nuestra práctica diaria. Describimos el uso de inteligencia artificial en el diagnóstico clínico, funcional e imagenológico de la EPOC y el asma así como en el monitoreo remoto de los pacientes, los avances en la interpretación de imágenes y de patología de las enfermedades intersticiales con el uso de Aprendizaje Automatizado (AA) y Aprendizaje Profundo (AP), la aplicación en tamización de hipertensión pulmonar a partir de pruebas diferentes al cateterismo cardiaco derecho, y finalmente la amplia gama de aplicaciones en medicina del sueño, en la que el avance es asombroso
... A precise survey of articles that utilize AI techniques to distinguish clinically significant COPD phenotypes was performed in [40]. Lately, the developing utilization of AI calculations, bunch investigations specifically, has the potential to establish this grouping via joining other explanatory attributes, comorbidities, genomic data, and biomarkers. ...
Chest diseases can be dangerous and deadly. They include many chest infections such as pneumonia, asthma, edema, and, lately, COVID-19. COVID-19 has many similar symptoms compared to pneumonia, such as breathing hardness and chest burden. However, it is a challenging task to differentiate COVID-19 from other chest diseases. Several related studies proposed a computer-aided COVID-19 detection system for the single-class COVID-19 detection, which may be misleading due to similar symptoms of other chest diseases. This paper proposes a framework for the detection of 15 types of chest diseases, including the COVID-19 disease, via a chest X-ray modality. Two-way classification is performed in proposed Framework. First, a deep learning-based convolutional neural network (CNN) architecture with a soft-max classifier is proposed. Second, transfer learning is applied using fully-connected layer of proposed CNN that extracted deep features. The deep features are fed to the classical Machine Learning (ML) classification methods. However, the proposed framework improves the accuracy for COVID-19 detection and increases the predictability rates for other chest diseases. The experimental results show that the proposed framework, when compared to other state-of-the-art models for diagnosing COVID-19 and other chest diseases, is more robust, and the results are promising.
... In recent years, attempts to subtyping of obstructive pulmonary diseases have been shifted to more data-driven methods. 13,14 A system of classification of obstructive lung diseases that integrate the multidimensionality of asthma and COPD on clinical, cellular and molecular levels may be a tool for identifying numerous distinct phenotypes, with specific pathobiological components that respond to particular therapy. Phenotyping of obstructive pulmonary diseases has usually been studied in severe stage of the diseases. ...
Purpose
Asthma and chronic obstructive pulmonary disease (COPD) are complex and heterogeneous inflammatory diseases. We sought to investigate distinct disease profiles based on clinical, cellular and molecular data from patients with mild-to-moderate obstructive pulmonary diseases.
Patients and Methods
Patients with mild-to-moderate allergic asthma (n=30) and COPD (n=30) were prospectively recruited. Clinical characteristics and induced sputum were collected. In total, 35 mediators were assessed in induced sputum. Logistic regression analysis was conducted to identify the optimal factors that were able to discriminate between asthma and COPD. Further, the data were explored using hierarchical clustering in order to discover and compare clusters of combined samples of asthma and COPD patients. Clinical parameters, cellular composition, and sputum mediators of asthma and COPD were assessed between and within obtained clusters.
Results
We found five clinical and biochemical variables, namely IL-6, IL-8, CCL4, FEV1/VC ratio pre-bronchodilator (%), and sputum neutrophils (%) that differentiated asthma and COPD and were suitable for discrimination purposes. A combination of those variables yielded high sensitivity and specificity in the differentiation between asthma and COPD, although only FEV1/VC ratio pre-bronchodilator (%) proven significant in the combined model. In cluster analysis, two main clusters were identified: cluster 1, asthma predominant with evidence of eosinophilic airway inflammation and low level of Th1 and Th2 cytokines; and cluster 2, COPD predominant with elevated levels of Th1 and Th2 mediators.
Conclusion
The inflammatory profile of sputum samples from patients with stable mild-to-moderate asthma and COPD is not disease specific, varies within the disease and might be similar between these diseases. This study highlights the need for phenotyping the mild-to-moderate stages according to their clinical and molecular features.
... According to the Global Obstructive Lung Disease Initiative, COPD patients are classified into four phenotypes based on their symptomatic assessment, exacerbation and hospitalization history [83]. However, the discriminatory ability of this method is insufficient, leading to the AI/ML-based integration of additional information, including physiological features, lung function test results, comorbidities, genome, and biomarkers, for precise phenotype classification, severity assessment, and therapeutic guidance [84][85][86][87][88][89]. ...
Chronic airway diseases are characterized by airway inflammation, obstruction, and remodeling and show high prevalence, especially in developing countries. Among them, asthma and chronic obstructive pulmonary disease (COPD) show the highest morbidity and socioeconomic burden worldwide. Although there are extensive guidelines for the prevention, early diagnosis, and rational treatment of these lifelong diseases, their value in precision medicine is very limited. Artificial intelligence (AI) and machine learning (ML) techniques have emerged as effective methods for mining and integrating large-scale, heterogeneous medical data for clinical practice, and several AI and ML methods have recently been applied to asthma and COPD. However, very few methods have significantly contributed to clinical practice. Here, we review four aspects of AI and ML implementation in asthma and COPD to summarize existing knowledge and indicate future steps required for the safe and effective application of AI and ML tools by clinicians.
... 1 The high incidence of COPD, which is expected to rise to the third leading cause of death in the world by 2030, will impose a heavy social and economic burden. 2 It is important to note that most of the world's COPD morbidity and mortality occurs in low -and middle-income countries, and there are significant urban-rural differences-the rural situation is more severe than the urban one. There is even a growing body of research confirming that rural residence is an independent and important risk factor for COPD. ...
Objective
This study aimed to construct and evaluate a clinical predictive model for the development of COPD in northwest China’s rural areas.
Methods
A cross-sectional study of a natural population was performed in rural northwest China. After assessing demographic and disease characteristics, a clinical prediction model was developed. First, we used the least absolute shrinkage and selection operator regression model to screen possible factors influencing COPD. Then construct a logistic regression model and draw a nomogram. The discriminability of the model was further evaluated by the calibration diagram, C-index and ROC curve system. Clinical benefit was analyzed using the decision curve. Finally, the 1000 bootstrap resamples and Harrell’s C-index was used for internal verification of the nomogram.
Results
Among 3249 patients in the local rural natural population, 394 (12.13%) were diagnosed with COPD. The LASSO regression model was used to find the optimal combination of parameters, and the screened influencing factors included age, gender, barbeque, smoking, passive smoking, energy type, ventilation system and Post-Bronchodilator FEV1. These predictors are used to construct a nomogram. C index is 0.81 (95% confidence interval:0.79–0.83). The combination of the calibration curve and ROC curve indicates that the model has high discriminability. The decision curve shows benefits in clinical practice when the threshold probability is >6% and <58%, respectively. The internal verification results using Harrell’s C-Index were 0.80 (95% confidence interval: 0.78–0.83).
Conclusion
Combining information such as age, sex, barbeque, smoking, passive smoking, type of energy, ventilation systems, and Post-Bronchodilator FEV1 can be easily used to predict the risk of COPD in local rural areas.
... Chronic obstructive pulmonary disease (COPD) is a serious chronic respiratory disease characterized by incomplete reversibility and progressive exacerbation of air ow restriction. As the most common chronic respiratory disease, it is expected to become the third leading cause of death worldwide in 2030 [1] . The prevalence of COPD increased by 44.2% from 1990 to 2015, and 3.2 million people died from COPD worldwide in 2015 with an increase of 11.6% compared to 1990 [2] . ...
Background: Chronic obstructive pulmonary disease (COPD) is the most common chronic respiratory disease which is expected to become the third leading cause of death worldwide in 2030. Series of susceptibility genes and single nucleotide polymorphism (SNPs) play an important role in the occurrence and development of COPD.
Methods: In our study, 98 COPD patients and 90 healthy volunteers were enrolled. The +869 SNP (SNP, Single Nucleotide Polymorphisms) of TGF-β1 was detected in 98 COPD patients and 90 healthy volunteers by PCR-DNA sequencing. The effects of different genotypes of +869 locus on the susceptibility of COPD, pulmonary function and airflow limitation of COPD patients were analyzed.
Results: Allele C of +869 locus was associated with the susceptibility of COPD (OR:1.913, 95% CI: 1.251-2.926). The predicted value of FEV1% (FEV1, Forced Expiratory Volume in One Second) in patients with CC of +869 locus was significantly lower than that in patients with TT (P < 0.05). The genotype frequencies of CC, CT and TT were 6.5%, 58.7% and 34.8% in Mild-to-Moderate airflow restriction patients. In severe airflow restriction patients, the genotype frequencies were CC 23.1%, CT 57.7% and TT 19.2%. The distribution of CC genotype in severe airflow restriction COPD patients was significantly higher than that in Mild-to-Moderate airflow restriction COPD patients (P = 0.037). Moreover, the frequency of C allele was significantly higher in patients with severe airflow restriction than that patients with Mild-to-Moderate airflow restriction (P = 0.024).
Conclusions: The SNP of +869 T/C in TGF-β1 is closely related to the susceptibility of COPD and the airflow restriction of COPD patients.
... Furthermore, a future goal could be that the models in this study provide a framework for the integration of this information into electronic healthcare records to ultimately inform decision making in the management of patients with COPD. Further research into machine learning algorithms and artificial intelligence applications is ongoing [38]. ...
Introduction:
Outcomes in chronic obstructive pulmonary disease (COPD) such as symptoms, hospitalisations and mortality rise with increasing disease severity. However, the heterogeneity of electronic medical records presents a significant challenge in measuring severity across geographies. We aimed to develop and validate a method to approximate COPD severity using the Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2011 classification scheme, which categorises patients based on forced expiratory volume in 1 s, hospitalisations and the modified Medical Research Council dyspnoea scale or COPD Assessment Test.
Methods:
This analysis was part of a comprehensive retrospective study, including patients sourced from the IQVIA Medical Research Data [IMRD; incorporating data from The Health Improvement Network (THIN), a Cegedim database] and the Clinical Practice Research Datalink (CPRD) in the UK, the Disease Analyzer in Germany and the Longitudinal Patient Data in Italy, France and Australia. Patients in the CPRD with the complete set of information required to calculate GOLD 2011 groups were used to develop the method. Ordinal logistic models at COPD diagnosis and at index (first episode of triple therapy) were then used to validate the method to estimate COPD severity, and this was applied to the full study population to estimate GOLD 2011 categories.
Results:
Overall, 4579 and 12,539 patients were included in the model at COPD diagnosis and at index, respectively. Models correctly classified 74.4% and 75.9% of patients into severe and non-severe categories at COPD diagnosis and at index, respectively. Age, gender, time between diagnosis and start of triple therapy, healthcare resource use, comorbid conditions and prescriptions were included as covariates.
Conclusion:
This study developed and validated a method to approximate disease severity based on GOLD 2011 categories that can potentially be used in patients without all the key parameters needed for this calculation.
Chronic obstructive pulmonary disease (COPD) is a heterogeneous and multisystem disease with multiple phenotypes and a progressive increase in morbidity and mortality. This article provides a review of the current data on the identification, characterization, and features of therapy for the most common phenotypes of the disease. A literature review was conducted using medical resources such as PubMed, Google Scholar, and UpToDate, addressing issues related to phenotyping in COPD.
Introduction:
Asthma and chronic obstructive pulmonary disease (COPD) are leading causes of morbidity and mortality worldwide. Despite all available diagnostics and treatments, these conditions pose a significant individual, economic and social burden. Artificial intelligence (AI) promises to support clinical decision-making processes by optimizing diagnosis and treatment strategies of these heterogeneous and complex chronic respiratory diseases. Its capabilities extend to predicting exacerbation risk, disease progression and mortality, providing healthcare professionals with valuable insights for more effective care. Nevertheless, the knowledge gap between respiratory clinicians and data scientists remains a major constraint for wide application of AI and may hinder future progress. This narrative review aims to bridge this gap and encourage AI deployment by explaining its methodology and added value in asthma and COPD diagnosis and treatment.
Areas covered:
This review offers an overview of the fundamental concepts of AI and machine learning, outlines the key steps in building a model, provides examples of their applicability in asthma and COPD care, and discusses barriers to their implementation.
Expert opinion:
Machine learning can advance our understanding of asthma and COPD, enabling personalized therapy and better outcomes. Further research and validation are needed to ensure the development of clinically meaningful and generalizable models.
The significance of medical diagnosis in directing treatment options and improving patient outcomes is critical. With the rapid growth of machine learning, there has been an increase in interest in leveraging its potential to improve diagnostic capabilities. The aim of this research was to conduct a survey on current improvements in medical diagnosis and telemedicine using machine learning techniques, Machine learning and deep learning has shown exceptional success in analyzing medical images in a variety of modalities, including radiology, pathology, dermatology, ophthalmology, Neuro-science, Neuro-computing and Neuro-imaging. Machine learning algorithms have outperformed human specialists in tasks such as tumor identification, segmentation, and disease categorization in some circumstances. The incorporation of machine learning in telemedicine and remote monitoring has allowed for remote access to healthcare services as well as continuous patient monitoring. These advances have resulted in greater accuracy and fewer diagnostic errors in medical diagnosis. Machine learning algorithms have shown excellent sensitivity in diagnosing diseases such diabetic retinopathy, skin cancer, breast cancer metastases, and lung nodules. The successful creation, validation, and implementation of machine learning models in medical diagnostics requires collaboration between machine learning experts and medical professionals. This partnership brings together subject expertise, clinical competence and technical capabilities, resulting in more accurate, reliable, and clinically useful diagnostic tools. We can continue to uncover the full potential of machine learning in medical diagnostics and achieve transformative advances in healthcare by tackling the difficulties and fostering collaboration.
span lang="EN-US">This paper presents a study on an embedded spirometer using the low-cost MPX5100DP pressure sensor and an Arduino Uno board to measure the air exhaled flow rate and calculate force vital capacity (FVC), forced expiratory volume in 1 s (FEV1), and the FEV1/FVC ratio of human lungs volume. The exhaled air flow rate was measured from differential pressure in the sections of a mouthpiece tube using the venturi effect equation. This constructed mouthpiece and the embedded spirometer resulted in a 96.27% FVC reading accuracy with a deviation of 0.09 L and 98.05% FEV1 accuracy with a deviation of 0.05 L compared to spirometry. This spirometer integrates an HC-05 Bluetooth module for spirometry data transceiving to a smartphone for display and recording in an Android application for further chronic obstructive pulmonary disease (COPD) diagnosis.</span
Background:
Asthma and chronic obstructive pulmonary disease (COPD) are complex diseases whose definitions overlap.
Objective:
To investigate clustering of clinical/physiological features and readily available biomarkers in patients with physician-assigned diagnoses of asthma and/or COPD in NOVELTY (NCT02760329).
Methods:
Two approaches were taken to variable selection, using baseline data: approach A was data-driven, hypothesis-free, using Pearson's dissimilarity matrix; approach B used an unsupervised Random Forest guided by clinical input. Cluster analyses were conducted across 100 random resamples using partitioning around medoids, followed by consensus clustering.
Results:
Approach A included 3,796 individuals (mean age 59.5 years, 54% female); approach B included 2,934 patients (mean age 60.7 years, 53% female). Each identified six mathematically stable clusters, which had overlapping characteristics. Overall, 67-75% of asthma patients were in three clusters, and ∼90% of COPD patients in three clusters. Although traditional features like allergies and current/ex-smoking (respectively) were higher in these clusters, there were differences between clusters and approaches in features such as sex, ethnicity, breathlessness, frequent productive cough and blood cell counts. The strongest predictors of approach A cluster membership were age, weight, childhood onset, pre-bronchodilator FEV1, duration of dust/fume exposure and number of daily medications.
Conclusion:
Cluster analyses in NOVELTY patients with asthma and/or COPD yielded identifiable clusters, with several discriminatory features that differed from conventional diagnostic characteristics. The overlap between clusters suggests that they do not reflect discrete underlying mechanisms, and points to the need for identification of molecular endotypes and potential treatment targets across asthma and/or COPD.
Background
Lymphangioleiomyomatosis (LAM) is a rare multisystem disease with variable manifestations and differing rates of progression among individuals. Classification of its phenotypes is an issue for consideration. We hypothesized that clinical manifestations associated with LAM cluster together and identifying these associations would be useful for identifying phenotypes.
Methods
Using cross-sectional data from the National Database of Designated Intractable Diseases of Japan, we performed a hierarchical cluster analysis based on disease-associated manifestations.
Results
Four clusters were identified from 404 patients (50.4% of 801 LAM patients registered in 2016). Patients in cluster 1 had only dyspnea on exertion, relatively low lung function, the earliest onset age, and the lowest prevalence of tuberous sclerosis complex (TSC). Those in cluster 2 had various manifestations with the highest prevalence of TSC. Patients in cluster 3 had major respiratory symptoms (cough, sputum, or dyspnea on exertion) or fatigue and the lowest lung function. Those in cluster 4 were asymptomatic and had the latest onset age, shortest disease duration, and relatively high prevalence of TSC. Patients in cluster 1 had the highest rate of receiving mechanistic target of rapamycin (mTOR) inhibitor treatment, suggesting that cluster 1 included those with declining lung function for which mTOR inhibitor treatment was required.
Conclusions
Hierarchical cluster analysis based on manifestations data identified four clusters. The characteristics of cluster 1 are noteworthy in relation to the indication for mTOR inhibitor treatment. A cluster analysis of accumulated and longitudinal data that allows valid clustering and outcome comparisons is required in the future.
Chronic obstructive pulmonary disease (COPD) remains a leading cause of morbidity and mortality despite current treatment strategies which focus on smoking cessation, pulmonary rehabilitation, and symptomatic relief. A focus of COPD care is to encourage self-management, particularly during COVID-19, where much face-to-face care has been reduced or ceased. Digital health solutions may offer affordable and scalable solutions to support COPD patient education and self-management, such solutions could improve clinical outcomes and expand service reach for limited additional cost. However, optimal ways to deliver digital medicine are still in development, and there are a number of important considerations for clinicians, commissioners, and patients to ensure successful implementation of digitally augmented care. In this narrative review, we discuss advantages, pitfalls, and future prospects of digital healthcare, which offer a variety of tools including self-management plans, education videos, inhaler training videos, feedback to patients and healthcare professionals (HCPs), exacerbation monitoring, and pulmonary rehabilitation. We discuss the key issues with sustaining patient and HCP engagement and limiting attrition of use, interoperability with devices, integration into healthcare systems, and ensuring inclusivity and accessibility. We explore the essential areas of research beyond determining safety and efficacy to understand the acceptability of digital healthcare solutions to patients, clinicians, and healthcare systems, and hence ways to improve this and sustain engagement. Finally, we explore the regulatory challenges to ensure quality and engagement and effective integration into current healthcare systems and care pathways, while maintaining patients’ autonomy and privacy. Understanding and addressing these issues and successful incorporation of an acceptable, simple, scalable, affordable, and future-proof digital solution into healthcare systems could help remodel global chronic disease management and fractured healthcare systems to provide best patient care and optimisation of healthcare resources to meet the global burden and unmet clinical need of COPD.
Background
Chronic obstructive pulmonary disease (COPD) is a heterogeneous group of lung conditions challenging to diagnose and treat. Identification of phenotypes of patients with lung function loss may allow early intervention and improve disease management. We characterised patients with the ‘fast decliner’ phenotype, determined its reproducibility and predicted lung function decline after COPD diagnosis.
Methods
A prospective 4 years observational study that applies machine learning tools to identify COPD phenotypes among 13 260 patients from the UK Royal College of General Practitioners and Surveillance Centre database. The phenotypes were identified prior to diagnosis (training data set), and their reproducibility was assessed after COPD diagnosis (validation data set).
Results
Three COPD phenotypes were identified, the most common of which was the ‘fast decliner’—characterised by patients of younger age with the lowest number of COPD exacerbations and better lung function—yet a fast decline in lung function with increasing number of exacerbations. The other two phenotypes were characterised by (a) patients with the highest prevalence of COPD severity and (b) patients of older age, mostly men and the highest prevalence of diabetes, cardiovascular comorbidities and hypertension. These phenotypes were reproduced in the validation data set with 80% accuracy. Gender, COPD severity and exacerbations were the most important risk factors for lung function decline in the most common phenotype.
Conclusions
In this study, three COPD phenotypes were identified prior to patients being diagnosed with COPD. The reproducibility of those phenotypes in a blind data set following COPD diagnosis suggests their generalisability among different populations.
Background
This paper is aimed to (i) develop an innovative classification of COPD, multi-dimensional phenotype, based on a multidimensional assessment; (ii) describe the identified multi-dimensional phenotypes.
Methods
An exploratory factor analysis to identify the main classificatory variables and, then, a cluster analysis based on these variables were run to classify the COPD-diagnosed 514 patients enrolled in the STORICO (trial registration number: NCT03105999) study into multi-dimensional phenotypes.
Results
The circadian rhythm of symptoms and health-related quality of life, but neither comorbidity nor respiratory function, qualified as primary classificatory variables. Five multidimensional phenotypes were identified: the MILD COPD characterized by no night-time symptoms and the best health status in terms of quality of life, quality of sleep, level of depression and anxiety, the MILD EMPHYSEMATOUS with prevalent dyspnea in the early-morning and day-time, the SEVERE BRONCHITIC with nocturnal and diurnal cough and phlegm, the SEVERE EMPHYSEMATOUS with nocturnal and diurnal dyspnea and the SEVERE MIXED COPD distinguished by higher frequency of symptoms during 24h and worst quality of life, of sleep and highest levels of depression and anxiety.
Conclusions
Our results showed that properly collected respiratory symptoms play a primary classificatory role of COPD patients. The longitudinal observation will disclose the discriminative and prognostic potential of the proposed multidimensional phenotype.
Trial registration
Trial registration number: NCT03105999, date of registration: 10th April 2017.
Background:
Because ACO (Asthma-COPD-Overlap) does not fill out asthma or COPD (Chronic Obstructive Pulmonary Disease) criteria, such patients are poorly evaluated. The aim of this study was to screen asthma and COPD for an alternative diagnosis of ACO, then to determine subgroups of patients, using cluster analysis.
Material and methods:
Using GINA-GOLD stepwise approach, asthmatics and COPD were screened for ACO. Clusterization was then performed employing Multiple Correspondent Analysis (MCA) model, encompassing 9 variables (age, symptoms onset, sex, BMI (Body Mass Index), smoking, FEV-1, dyspnea, exacerbation, comorbidity). Finally, clusters were compared to determine phenotypes.
Results:
MCA analysis was performed on 172 ACO subjects. To better distinguish clusters, the analysis was then focused on 55 subjects, having at least one cosine squared >0.3. Six clusters were identified, allowing the description of 4 phenotypes. Phenotype A represented overweighed heavy smokers, with an early onset and a severe disease (27% of ACO patients). Phenotype B gathered similar patients, with a late onset (29%). Patients from Phenotypes C-D were slighter smokers, presenting a moderate disease, with early and late onset respectively (respectively 13% and 31%).
Conclusions:
By providing evidences for clusters within ACO, our study confirms its heterogeneity, allowing the identification of 4 phenotypes. Further prospective studies are mandatory to confirm these data, to determine both specific management requirements and prognostic value.
Background:
The 2019 Global Initiative for Chronic Obstructive Lung Disease (GOLD) report made recommendations for the assessment, initial and subsequent treatment chronic obstructive pulmonary disease (COPD) based on biomarkers, including blood eosinophil counts.
Methods:
We evaluated the distribution of UK COPD patients initiating maintenance therapy and established patients by GOLD group, the prevalence of comorbidities and appropriateness of therapy using electronic patient records from the Optimum Patient Care Research Database (OPCRD). Changes in effective GOLD group, therapy and exacerbation rates over the next 2 years were analysed.
Findings:
11,409 established COPD patients and 699 starting therapy were studied. 44·3%, 25·7%, 13·8% & 16·2% of established COPD patients and 45·2%, 28·5%, 15·7% & 10·6% initiating therapy were in GOLD groups A, B, C & D respectively.The overall proportion in each GOLD group was similar after 2 years but there was substantial movement of patients between groups. Diabetes and cardiovascular disease were the most common comorbidities in all groups in both cohorts.LAMA monotherapy was the commonest initial therapy in all GOLD groups. In both cohorts there was over-treatment with escalation, de-escalation or switching in nearly 50% during follow-up.In both cohorts, exacerbation rates were highest in group D and appeared higher in over-treated patients.
Interpretation:
Most patients are not at risk of exacerbations and co-morbidities are common. Many patients change effective GOLD group and therapy over time. Prescribing is not in accordance with guideline recommendations and many patients still appear over treated.
Background
Chronic obstructive pulmonary disease (COPD) is an inflammatory lung disorder associated with lung microbiome dysbiosis. Although the upper airway microbiome is the source of the lung microbiome, the relationships between the oral, nasal, and sputum microbiota are incompletely understood. Our objective was to determine features that differentiate the oral, nasal, and sputum microbiome among subjects with stable COPD.
Methods
We recruited 15 current or former smokers to provide oral and sputum samples on day 1. On day 2, another oral sample and a nasal sample were obtained. Each sample and control underwent DNA extraction, 16S V4 rRNA amplification, 16S V4 sequencing, and qPCR of 16S rRNA. Data were analyzed using dada2 and R.
Results
Most (14 of 15) subjects were male with a mean age of 65.2. One subject had no pulmonary obstruction, while 5 had mild COPD, 7 had moderate COPD, and 2 had severe COPD. Three subjects (20%) were current tobacco users and 2 subjects (13%) used inhaled corticosteroids (ICS). Subjects had a mean of 49.1 pack-years of tobacco exposure. Bacterial biomass was associated with anatomic site, but no differences in biomass were observed with age, FEV1 percent predicted (FEV1pp), ICS use, smoking status, or edentulous state. Shannon index was associated with site (lower nasal diversity than oral and sputum diversity, p<0.001), but not age, ICS use, FEV1pp, tobacco use, or edentulous state. β-diversity was illustrated by principal coordinate analysis using Bray-Curtis dissimilarity and PERMANOVA analyses, showing sample clustering by anatomic site (p = 0.001) with nasal samples forming a cluster separate from the combined oral wash samples and sputum samples. Clustering was also observed with ICS use (p = 0.029) and edentulous state (p = 0.019), while FEV1pp and current tobacco use were not significant. In an amplicon sequencing variant (ASV)-level analysis of oral samples using a linear regression model with Benjamini-Hochberg correction at an FDR<0.10, 10 ASVs were associated with age while no ASVs were associated with FEV1pp or smoking status. Sputum sample analysis demonstrated that 51 ASVs (25 unique genera) were associated with age, 61 ASVs (32 genera) were associated with FEV1pp, and no ASVs were associated with smoking status. In a combined dataset, the frequent exacerbator phenotype, rather than ICS use, was associated with decreased sputum Shannon diversity.
Conclusions
Among the upper airway microbiota of COPD subjects, anatomic site was associated with bacterial biomass, Shannon diversity, and β-diversity. ICS use and edentulous state were both associated with β-diversity. Age was associated with taxa relative abundance in oral and sputum samples, while FEV1pp was associated with taxa relative abundance in sputum samples only.
Background:
Quantitative computed tomographic (QCT) imaging-based metrics enable to quantify smoking induced disease alterations and to identify imaging-based clusters for current smokers. We aimed to derive clinically meaningful sub-groups of former smokers using dimensional reduction and clustering methods to develop a new way of COPD phenotyping.
Methods:
An imaging-based cluster analysis was performed for 406 former smokers with a comprehensive set of imaging metrics including 75 imaging-based metrics. They consisted of structural and functional variables at 10 segmental and 5 lobar locations. The structural variables included lung shape, branching angle, airway-circularity, airway-wall-thickness, airway diameter; the functional variables included regional ventilation, emphysema percentage, functional small airway disease percentage, Jacobian (volume change), anisotropic deformation index (directional preference in volume change), and tissue fractions at inspiration and expiration.
Results:
We derived four distinct imaging-based clusters as possible phenotypes with the sizes of 100, 80, 141, and 85, respectively. Cluster 1 subjects were asymptomatic and showed relatively normal airway structure and lung function except airway wall thickening and moderate emphysema. Cluster 2 subjects populated with obese females showed an increase of tissue fraction at inspiration, minimal emphysema, and the lowest progression rate of emphysema. Cluster 3 subjects populated with older males showed small airway narrowing and a decreased tissue fraction at expiration, both indicating air-trapping. Cluster 4 subjects populated with lean males were likely to be severe COPD subjects showing the highest progression rate of emphysema.
Conclusions:
QCT imaging-based metrics for former smokers allow for the derivation of statistically stable clusters associated with unique clinical characteristics. This approach helps better categorization of COPD sub-populations; suggesting possible quantitative structural and functional phenotypes.
Purpose
In patients with COPD, acute exacerbation (AE) is not only an important determinant of prognosis, but also an important factor in choosing therapeutic agents. In this study, we evaluated the usefulness of COPD subtypes identified through cluster analysis to predict the first AE.
Patients and methods
Among COPD patients in the Korea COPD Subgroup Study (KOCOSS) cohort, 1,195 who had follow-up data for AE were included in our study. We selected seven variables for cluster analysis – age, body mass index, smoking status, history of asthma, COPD assessment test (CAT) score, post-bronchodilator (BD) FEV1 % predicted, and diffusing capacity of carbon monoxide % predicted.
Results
K-means clustering identified four clusters for COPD that we named putative asthma-COPD overlap (ACO), mild COPD, moderate COPD, and severe COPD subtypes. The ACO group (n=196) showed the second-best post-BD FEV1 (75.5% vs 80.9% [mild COPD, n=313] vs 52.4% [moderate COPD, n=345] vs 46.7% [severe COPD, n=341] predicted), the longest 6-min walking distance (424 m vs 405 m vs 389 m vs 365 m), and the lowest CAT score (12.2 vs 13.7 vs 15.6 vs 17.5) among the four groups. ACO group had greater risk for first AE compared to the mild COPD group (HR, 1.683; 95% CI, 1.175–2.410). The moderate COPD and severe COPD group HR values were 1.587 (95% CI, 1.145–2.200) and 1.664 (95% CI, 1.203–2.302), respectively. In addition, St. George’s Respiratory Questionnaire score (HR: 1.019; 95% CI, 1.014–1.024) and gastroesophageal reflux disease were independent factors associated with the first AE (HR: 1.535; 95% CI, 1.116–2.112).
Conclusion
Our cluster analysis revealed an exacerbator subtype of COPD independent of FEV1. Since these patients are susceptible to AE, a more aggressive treatment strategy is needed in these patients.
Purpose
The diagnosis and severity of chronic obstructive pulmonary disease (COPD) are defined by airflow limitation using spirometry. However, COPD has diverse clinical features, and several phenotypes based on non-spirometric data have been investigated. To identify novel phenotypes of COPD using radiologic data obtained by three-dimensional computed tomography (3D-CT).
Patients and methods
The inner luminal area and wall thickness of third- to sixth-generation bronchi and the percentage of the low-attenuation area (less than −950 HU) of the lungs were measured using 3D-CT in patients with COPD. Using the radiologic data, hierarchical clustering was performed. Respiratory reactance and resistance were measured to evaluate functional differences among the clusters.
Results
Four clusters were identified among 167 patients with COPD: Cluster I, mild emphysema with severe airway changes, severe airflow limitation, and high exacerbation risk; Cluster II, mild emphysema with moderate airway changes, mild airflow limitation, and mild dyspnea; Cluster III, severe emphysema with moderate airway changes, severe airflow limitation, and increased dyspnea; and Cluster IV, moderate emphysema with mild airway changes, mild airflow limitation, low exacerbation risk, and mild dyspnea. Cluster I had the highest respiratory resistance among the four clusters. Clusters I and III had higher respiratory reactance than Clusters II and IV.
Conclusions
The 3D-CT-based radiologic phenotypes were associated with the clinical features of COPD. Measurement of respiratory resistance and reactance may help to identify phenotypic differences.
Background:
Chronic obstructive pulmonary disease (COPD) frequent exacerbators (FE) suffer increased morbidity and mortality compared to infrequent exacerbators (IE). The association between the oral and sputum microbiota and exacerbation phenotype is not well defined. The objective of this study was to determine key features that differentiate the oral and sputum microbiota of FEs from the microbiota of IEs during periods of clinical stability.
Methods:
We recruited 11 FE and 11 IE who had not used antibiotics or systemic corticosteroids in the last 1 month. Subjects provided oral wash and sputum samples, which underwent 16S V4 MiSeq sequencing and qPCR of 16S rRNA. Data were analyzed using Dada2 and R.
Results:
FE and IE were similar in terms of age, FEV1 percent predicted (FEV1pp), pack-years of tobacco exposure, and St. George's Respiratory Questionnaire score. 16S copy numbers were significantly greater in sputum vs. oral wash (p = 0.01), but phenotype was not associated with copy number. Shannon diversity was significantly greater in oral samples compared to sputum (p = 0.001), and IE samples were more diverse than FE samples (p < 0.001). Sputum samples from FE had more Haemophilus and Moraxella compared to IE sputum samples, due to dominance of these COPD-associated taxa in three FE sputum samples. Amplicon sequencing variant (ASV)-level analysis of sputum samples revealed one ASV (Actinomyces) was significantly more abundant in IE vs. FE sputum (padj = 0.048, Wilcoxon rank-sum test), and this persisted after controlling for FEV1pp. Principal coordinate analysis using Bray-Curtis distance with PERMANOVA analyses demonstrated clustering by anatomic site, phenotype, inhaled corticosteroid use, current tobacco use, COPD severity, and last professional dental cleaning.
Conclusions:
FE have less diverse oral and sputum microbiota than IE. Actinomyces was significantly more abundant in IE sputum than FE sputum. The oral and sputum microbiota of COPD subjects cluster based on multiple clinical factors, including exacerbation phenotype. Even during periods of clinical stability, the frequent exacerbator phenotype is associated with decreased alpha diversity, beta-diversity clustering, and changes in taxonomic abundance.
Background:
COPD is a highly heterogeneous disease composed of different phenotypes with different aetiological and prognostic profiles and current classification systems do not fully capture this heterogeneity. In this study we sought to discover, describe and validate COPD subtypes using cluster analysis on data derived from electronic health records.
Methods:
We applied two unsupervised learning algorithms (k-means and hierarchical clustering) in 30,961 current and former smokers diagnosed with COPD, using linked national structured electronic health records in England available through the CALIBER resource. We used 15 clinical features, including risk factors and comorbidities and performed dimensionality reduction using multiple correspondence analysis. We compared the association between cluster membership and COPD exacerbations and respiratory and cardiovascular death with 10,736 deaths recorded over 146,466 person-years of follow-up. We also implemented and tested a process to assign unseen patients into clusters using a decision tree classifier.
Results:
We identified and characterized five COPD patient clusters with distinct patient characteristics with respect to demographics, comorbidities, risk of death and exacerbations. The four subgroups were associated with 1) anxiety/depression; 2) severe airflow obstruction and frailty; 3) cardiovascular disease and diabetes and 4) obesity/atopy. A fifth cluster was associated with low prevalence of most comorbid conditions.
Conclusions:
COPD patients can be sub-classified into groups with differing risk factors, comorbidities, and prognosis, based on data included in their primary care records. The identified clusters confirm findings of previous clustering studies and draw attention to anxiety and depression as important drivers of the disease in young, female patients.
Background
The clinical course of IPF varies. This study sought to identify phenotyping with quantitative computed tomography (CT) fibrosis and emphysema features using a cluster analysis and to assess prognostic impact among identified clusters in patient with idiopathic pulmonary fibrosis (IPF). Furthermore, we evaluated the impact of fibrosis and emphysema on lung function with development of a descriptive formula.
Methods
This retrospective study included 205 patients with IPF. A texture-based automated system was used to quantify areas of normal, emphysema, ground-glass opacity, reticulation, consolidation, and honeycombing. Emphysema index was obtained by calculating the percentage of low attenuation area lower than -950HU. We used quantitative CT features and clinical features for clusters and assessed the association with prognosis. A formula was derived using fibrotic score and emphysema index on quantitative CT.
Results
Three clusters were identified in IPF patients using a quantitative CT score and clinical values. Prognosis was better in cluster1, with a low extent of fibrosis and emphysema with high forced vital capacity (FVC) than cluster2 and cluster3 with higher fibrotic score and emphysema (p = 0.046, and p = 0.026). In the developed formula [1.5670—fibrotic score(%)*0.04737—emphysema index*0.00304], a score greater ≥ 0 indicates coexisting of pulmonary fibrosis and emphysema at a significant extent despite of normal spirometric result.
Conclusions
Cluster analysis identified distinct phenotypes, which predicted prognosis of clinical outcome. Formula using quantitative CT values is useful to assess extent of pulmonary fibrosis and emphysema with normal lung function in patients with IPF.
Asthma and chronic obstructive pulmonary disease (COPD) have overlapping clinical features and share pathobiological mechanisms but are often considered distinct disorders. Prospective, observational studies across asthma, COPD and asthma–COPD overlap are limited. NOVELTY is a global, prospective observational 3-year study enrolling ∼12 000 patients ≥12 years of age from primary and specialist clinical practices in 19 countries (ClinicalTrials.gov identifier: NCT02760329 ).
NOVELTY's primary objectives are to describe patient characteristics, treatment patterns and disease burden over time, and to identify phenotypes and molecular endotypes associated with differential outcomes over time in patients with a diagnosis/suspected diagnosis of asthma and/or COPD. NOVELTY aims to recruit real-world patients, unlike clinical studies with restrictive inclusion/exclusion criteria.
Data collected at yearly intervals include clinical assessments, spirometry, biospecimens, patient-reported outcomes (PROs) and healthcare utilisation (HCU). PROs and HCU will also be collected 3-monthly via internet/telephone. Data will be used to identify phenotypes and endotypes associated with different trajectories for symptom burden, clinical progression or remission and HCU. Results may allow patient classification across obstructive lung disease by clinical outcomes and biomarker profile, rather than by conventional diagnostic labels and severity categories.
NOVELTY will provide a rich data source on obstructive lung disease, to help improve patient outcomes and aid novel drug development.
Background
Pulmonary rehabilitation (PR) is a cornerstone in the management of chronic obstructive pulmonary disease (COPD), targeting skeletal muscle to improve functional performance. However, there is substantial inter‐individual variability in the effect of PR on functional performance, which cannot be fully accounted for by generic phenotypic factors. We performed an unbiased integrative analysis of the skeletal muscle molecular responses to PR in COPD patients and comprehensively characterized their baseline pulmonary and physical function, body composition, blood profile, comorbidities, and medication use.
Methods
Musculus vastus lateralis biopsies were obtained from 51 COPD patients (age 64 ± 1 years, sex 73% men, FEV1, 34 (26–41) %pred.) before and after 4 weeks high‐intensity supervised in‐patient PR. Muscle molecular markers were grouped by network‐constrained clustering, and their relative changes in expression values—assessed by qPCR and western blot—were reduced to process scores by principal component analysis. Patients were subsequently clustered based on these process scores. Pre‐PR and post‐PR functional performance was assessed by incremental cycle ergometry and 6 min walking test (6MWT).
Results
Eight molecular processes were discerned by network‐constrained hierarchical clustering of the skeletal muscle molecular rehabilitation responses. Based on the resulting process scores, four clusters of patients were identified by hierarchical cluster analysis. Two major patient clusters differed in PR‐induced autophagy (P < 0.001), myogenesis (P = 0.014), glucocorticoid signalling (P < 0.001), and oxidative metabolism regulation (P < 0.001), with Cluster 1 (C1; n = 29) overall displaying a more pronounced change in marker expression than Cluster 2 (C2; n = 16). General baseline characteristics did not differ between clusters. Following PR, both 6 min walking distance (+26.5 ± 8.3 m, P = 0.003) and peak load on the cycle ergometer test (+9.7 ± 1.9 W, P < 0.001) were improved. However, the functional improvement was more pronounced in C1, as a higher percentage of patients exceeded the minimal clinically important difference in peak workload (61 vs. 21%, P = 0.022) and both peak workload and 6 min walking test (52 vs. 8%, P = 0.008) upon PR.
Conclusions
We identified patient groups with distinct skeletal muscle molecular responses to rehabilitation, associated with differences in functional improvements upon PR.
Abstract Digital health interventions have been associated with reduced rescue inhaler use and improved controller medication adherence. This quality improvement project assessed the benefit of these interventions on asthma-related healthcare utilizations, including hospitalizations, emergency department (ED) utilization and outpatient visits. The intervention consisted of electronic medication monitors (EMMs) that tracked rescue and controller inhaler medication use, and a digital health platform that presented medication use information and asthma control status to patients and providers. In 224 study patients, the number of asthma-related ED visits and combined ED and hospitalization events 365 days pre- to 365 days post-enrollment to the intervention significantly decreased from 11.6 to 5.4 visits (p
Introduction
People with chronic obstructive pulmonary disease (COPD) present high prevalence of physical inactivity that leads to a negative effect on health-related quality of life (HRQoL). The present study investigated COPD phenotypes according to their levels of physical activity and sedentary behaviour, as well as body composition and skeletal muscle strength.
Methods
This is an observational and cross-sectional study. Anthropometric data and COPD clinical control were collected and all participants underwent assessments of lung function, HRQoL, dyspnoea, levels of physical activity and sedentary behaviour, body composition and skeletal muscle strength. Participants were classified using hierarchical cluster analysis. Age, dyspnoea and obstruction (ADO) index was used to determine prognosis and calculated for each cluster.
Results
One hundred and fifty-two participants were included. Three distinct phenotypes were identified. Participants in phenotype 1 were more physically active, less sedentary and had better body composition and lower ADO index (p < 0.0001 for all variables). Overall, participants in phenotypes 2 and 3 were less physically active, more sedentary having a higher ADO index. However, participants in phenotype 2 were older, whereas participants in phenotype 3 had worse HRQoL, clinical control and body composition. Lung function did not differ across the three phenotypes.
Conclusions
Our results show that physical activity, sedentary behaviour and body composition should be considered to determine phenotypes in people with COPD and are involved in the prognosis of the disease. Less sedentary patients have better prognosis while age, body composition and clinical control seems to differentiate physically inactive patients.
Background:
Chronic obstructive pulmonary disease (COPD) is an important cause of morbidity and mortality around the world. The aim of our study was to determine the association between specific comorbidities and COPD severity.
Methods:
Pulmonologists included patients with COPD using a web-site questionnaire. Diagnosis of COPD was made using spirometry post-bronchodilator FEV1/FVC < 70%. The questionnaire included the following domains: demographic criteria, clinical symptoms, functional tests, comorbidities and therapeutic management. COPD severity was classified according to GOLD 2011. First we performed a principal component analysis and a non-hierarchical cluster analysis to describe the cluster of comorbidities.
Results:
One thousand, five hundred and eighty-four patients were included in the cohort during the first 2 years. The distribution of COPD severity was: 27.4% in group A, 24.7% in group B, 11.2% in group C, and 36.6% in group D. The mean age was 66.5 (sd: 11), with 35% of women. Management of COPD differed according to the comorbidities, with the same level of severity. Only 28.4% of patients had no comorbidities associated with COPD. The proportion of patients with two comorbidities was significantly higher (p < 0.001) in GOLD B (50.4%) and D patients (53.1%) than in GOLD A (35.4%) and GOLD C ones (34.3%). The cluster analysis showed five phenotypes of comorbidities: cluster 1 included cardiac profile; cluster 2 included less comorbidities; cluster 3 included metabolic syndrome, apnea and anxiety-depression; cluster 4 included denutrition and osteoporosis and cluster 5 included bronchiectasis. The clusters were mostly significantly associated with symptomatic patients i.e. GOLD B and GOLD D.
Conclusions:
This study in a large real-life cohort shows that multimorbidity is common in patients with COPD.
Background:
Chronic obstructive pulmonary disease (COPD) is a heterogeneous condition that can differ in its clinical manifestation, structural changes and response to treatment.
Objective:
To identify subgroups of COPD with distinct phenotypes, evaluate the distribution of phenotypes in four related regions and calculate the 1-year change in lung function and quality of life according to subgroup.
Methods:
Using clinical characteristics, we performed factor analysis and hierarchical cluster analysis in a cohort of 1676 COPD patients from 13 Asian cities. We compared the 1-year change in forced expiratory volume in one second (FEV1), modified Medical Research Council dyspnoea scale score, St George's Respiratory Questionnaire (SGRQ) score and exacerbations according to subgroup derived from cluster analysis.
Results:
Factor analysis revealed that body mass index, Charlson comorbidity index, SGRQ total score and FEV1 were principal factors. Using these four factors, cluster analysis identified three distinct subgroups with differing disease severity and symptoms. Among the three subgroups, patients in subgroup 2 (severe disease and more symptoms) had the most frequent exacerbations, most rapid FEV1 decline and greatest decline in SGRQ total score.
Conclusion:
Three subgroups with differing severities and symptoms were identified in Asian COPD subjects.
Genome‐wide association studies (GWAS) have become a very effective research tool to identify genetic variants of underlying various complex diseases. In spite of the success of GWAS in identifying thousands of reproducible associations between genetic variants and complex disease, in general, the association between genetic variants and a single phenotype is usually weak. It is increasingly recognized that joint analysis of multiple phenotypes can be potentially more powerful than the univariate analysis, and can shed new light on underlying biological mechanisms of complex diseases. In this paper, we develop a novel variable reduction method using hierarchical clustering method (HCM) for joint analysis of multiple phenotypes in association studies. The proposed method involves two steps. The first step applies a dimension reduction technique by using a representative phenotype for each cluster of phenotypes. Then, existing methods are used in the second step to test the association between genetic variants and the representative phenotypes rather than the individual phenotypes. We perform extensive simulation studies to compare the powers of multivariate analysis of variance (MANOVA), joint model of multiple phenotypes (MultiPhen), and trait‐based association test that uses extended simes procedure (TATES) using HCM with those of without using HCM. Our simulation studies show that using HCM is more powerful than without using HCM in most scenarios. We also illustrate the usefulness of using HCM by analyzing a whole‐genome genotyping data from a lung function study.
Background
There is limited knowledge of the scale and impact of multimorbidity for patients who have had an acute myocardial infarction (AMI). Therefore, this study aimed to determine the extent to which multimorbidity is associated with long-term survival following AMI.
Methods and findings
This national observational study included 693,388 patients (median age 70.7 years, 452,896 [65.5%] male) from the Myocardial Ischaemia National Audit Project (England and Wales) who were admitted with AMI between 1 January 2003 and 30 June 2013. There were 412,809 (59.5%) patients with multimorbidity at the time of admission with AMI, i.e., having at least 1 of the following long-term health conditions: diabetes, chronic obstructive pulmonary disease or asthma, heart failure, renal failure, cerebrovascular disease, peripheral vascular disease, or hypertension. Those with heart failure, renal failure, or cerebrovascular disease had the worst outcomes (39.5 [95% CI 39.0–40.0], 38.2 [27.7–26.8], and 26.6 [25.2–26.4] deaths per 100 person-years, respectively). Latent class analysis revealed 3 multimorbidity phenotype clusters: (1) a high multimorbidity class, with concomitant heart failure, peripheral vascular disease, and hypertension, (2) a medium multimorbidity class, with peripheral vascular disease and hypertension, and (3) a low multimorbidity class. Patients in class 1 were less likely to receive pharmacological therapies compared with class 2 and 3 patients (including aspirin, 83.8% versus 87.3% and 87.2%, respectively; β-blockers, 74.0% versus 80.9% and 81.4%; and statins, 80.6% versus 85.9% and 85.2%). Flexible parametric survival modelling indicated that patients in class 1 and class 2 had a 2.4-fold (95% CI 2.3–2.5) and 1.5-fold (95% CI 1.4–1.5) increased risk of death and a loss in life expectancy of 2.89 and 1.52 years, respectively, compared with those in class 3 over the 8.4-year follow-up period. The study was limited to all-cause mortality due to the lack of available cause-specific mortality data. However, we isolated the disease-specific association with mortality by providing the loss in life expectancy following AMI according to multimorbidity phenotype cluster compared with the general age-, sex-, and year-matched population.
Conclusions
Multimorbidity among patients with AMI was common, and conferred an accumulative increased risk of death. Three multimorbidity phenotype clusters that were significantly associated with loss in life expectancy were identified and should be a concomitant treatment target to improve cardiovascular outcomes.
Trial registration
ClinicalTrials.gov NCT03037255.
Apart from the refined management-oriented clinical stratification of chronic obstructive pulmonary disease (COPD), the molecular pathologies behind this highly prevalent disease have remained obscure. The aim of this study was the characterization of patients with COPD, based on the metabolomic profiling of peripheral blood and exhaled breath condensate (EBC) within the context of defined clinical and demographic variables. Mass-spectrometry-based targeted analysis of serum metabolites (mainly amino acids and lipid species), untargeted profiles of serum and EBC of patients with COPD of different clinical characteristics (n = 25) and control individuals (n = 21) were performed. From the combined clinical/demographic and metabolomics data, associations between clinical/demographic and metabolic parameters were searched and a de novo phenotyping for COPD was attempted. Adjoining the clinical parameters, sphingomyelins were the best to differentiate COPD patients from controls. Unsaturated fatty acid-containing lipids, ornithine metabolism and plasma protein composition-associated signals from the untargeted analysis differentiated the Global Initiative for COPD (GOLD) categories. Hierarchical clustering did not reveal a clinical-metabolomic stratification superior to the strata set by the GOLD consensus. We conclude that while metabolomics approaches are good for finding biomarkers and clarifying the mechanism of the disease, there are no distinct co-variate independent clinical-metabolic phenotypes.
Background:
Chronic obstructive pulmonary disease (COPD) is a heterogeneous disease with variable clinical manifestations, structural changes, and treatment responses. In a cohort study, we performed a baseline cluster analysis to identify the subgroups of COPD and to assess the clinical outcomes of each subgroup during a 1-year follow-up.
Methods:
We analyzed dusty areas cohort comprising 272 patients with COPD. The main factors with the highest loading in 15 variables were selected using principal component analysis (PCA) at baseline. The COPD patients were classified by hierarchical cluster analysis using clinical, physiological, and imaging data based on PCA-transformed data. The clinical parameters and outcomes during the 1-year follow-up were evaluated among the subgroups.
Results:
PCA revealed that six independent components accounted for 77.3% of variance. Three distinct subgroups were identified through the cluster analysis. Subgroup 1 included younger subjects with fewer symptoms and mild airflow obstruction, and they had fewer exacerbations during the 1-year follow-up. Subgroup 2 comprised subjects with additional symptoms and moderate airflow obstruction, and they most frequently experienced exacerbations requiring hospitalization during the 1-year follow-up. Subgroup 3 included subjects with additional symptoms and mild airflow obstruction; this group had more female patients and a modest frequency of exacerbations requiring hospitalization.
Conclusions:
Cluster analysis using the baseline data of a COPD cohort identified three distinct subgroups with different clinical parameters and outcomes. These findings suggest that the identified subgroups represent clinically meaningful subtypes of COPD.
Background
While a subgroup of patients with exacerbations of chronic obstructive pulmonary disease (COPD) clearly benefit from antibiotics, their identification remains challenging. We hypothesised that selective assessment of the balance between the two dominant bacterial groups (Gammaproteobacteria (G) and Firmicutes (F)) in COPD sputum samples might reveal a subgroup with a bacterial community structure change at exacerbation that was restored to baseline on recovery and potentially reflects effective antibiotic treatment.
Methods
Phylogenetically specific 16S rRNA genes were determined by quantitative real time PCR to derive a G:F ratio in serial sputum samples from 66 extensively-phenotyped COPD exacerbation episodes.
Results
Cluster analysis based on Euclidean distance measures, generated across the 4 visit times (stable and exacerbation day: 0,14 and 42) for the 66 exacerbation episodes, revealed three subgroups designated HG, HF, and GF reflecting predominance or equivalence of the two target bacterial groups. While the other subgroups showed no change at exacerbation, the HG cluster (n = 20) was characterized by G:F ratios that increased significantly at exacerbation and returned to baseline on recovery (p<0.00001); ratios in the HG group also correlated positively with inflammatory markers and negatively with FEV1. At exacerbation G:F showed a significant receiver-operator-characteristic curve to identify the HG subgroup (AUC 0.90, p<0.0001).
Conclusions
The G:F ratio at exacerbation can be determined on a timescale compatible with decisions regarding clinical management. We propose that the G:F ratio has potential for use as a biomarker enabling selective use of antibiotics in COPD exacerbations and hence warrants further clinical evaluation.
Background
Asthma and chronic obstructive pulmonary disease (COPD) are heterogeneous diseases. The phenotypes that have clinical features of both asthma and COPD are still incompletely understood.
Objective
To clarify the best discriminators of the asthma-COPD overlap phenotype from asthma and COPD subgroups using a clustering approach.
Methods
This study assessed pathophysiological parameters, including mRNA expression levels of T helper cell-related transcription factors, namely, TBX21 (Th1), GATA3 (Th2), RORC (Th17), and FOXP3 (Treg), in peripheral blood mononuclear cells in asthma patients (n = 152) and in COPD patients (n = 50). Clusters were determined using k-means clustering. Exacerbations of asthma and COPD were recorded during the 1-year follow-up period.
Results
The cluster analysis revealed four biological clusters: cluster 1, predominantly patients with COPD; cluster 2, patients with an asthma-COPD overlap phenotype; cluster 3, patients with non-atopic and late-onset asthma; and cluster 4, patients with early-onset atopic asthma. Hazard ratios for exacerbation were 2.5 (95% confidence interval [CI], 1.1–5.6) in cluster 1 and 2.3 (95% CI, 1.0–5.0) in cluster 2 compared with patients in other clusters. Cluster 2 was discriminated from other clusters by total serum IgE level ≥ 310 IU/mL, blood eosinophil counts ≥ 280 cells/μL, a higher ratio of TBX21/GATA3, FEV1/FVC ratio < 0.67, and smoking ≥ 10 pack-years with an area under the curve of 0.94 (95% CI, 0.90–0.98) in the receiver operating characteristic analysis.
Conclusions & Clinical Relevance
The asthma-COPD overlap phenotype was characterized by peripheral blood eosinophilia and higher levels of IgE despite the Th2-low endotype.
T-helper cell 17 (Th17) mediated inflammation is associated with various diseases including autoimmune encephalitis, inflammatory bowel disease and lung diseases such as chronic obstructive pulmonary disease and asthma. Differentiation into distinct Th subtypes needs to be tightly regulated to ensure an immunological balance. As microRNAs (miRNAs) are critical regulators of signaling pathways, we aimed to identify specific miRNAs implicated in controlling Th17 differentiation. We were able to create a regulatory network model of murine Th cell differentiation by combining Affymetrix mRNA and miRNA arrays and in-silico analysis. In this model, the miR-212~132 and miR-182~183 clusters were significantly up-regulated upon Th17 differentiation, while the entire miR-106~363 cluster was down regulated and predicted to target well-known Th17 cell differentiation pathways. In-vitro transfection of miR-18b, miR-106 and miR-363 into primary murine CD4(+) lymphocytes decreased expression of retinoid-related orphan receptor c (Rorc), Rora, Il17a and Il17f, and abolished secretion of Th17 mediated Il17-a. Moreover, we demonstrated target site-specific regulation of the Th17 transcription factors Rora and nuclear factor of activated T-cells (Nfat) 5 by miR-18b, miR-106a and miR-363-3p through luciferase reporter assays. Here, we provide evidence that miRNAs are involved in controlling the differentiation and function of T-helper cells, offering useful tools to study and modify Th17 mediated inflammation. This article is protected by copyright. All rights reserved.
Purpose
The purpose of this study was to identify subtypes in patients with mild-to-moderate airflow limitation and to appreciate their clinical and socioeconomic implications.
Methods
Subjects who were aged ≥20 years and had forced expiratory volume in 1 second (FEV1) ≥60% predicted and FEV1/forced vital capacity <0.7 were selected from the fourth Korea National Health and Nutrition Examination Survey (KNHANES) in 2007–2012. The data were merged to the National Health Insurance reimbursement database during the same period. k-Means clustering was performed to explore subtypes. For clustering analysis, six key input variables – age, body mass index (BMI), FEV1% predicted, the presence or absence of self-reported wheezing, smoking status, and pack-years of smoking – were selected.
Results
Among a total of 2,140 subjects, five groups were identified through k-means clustering, namely putative “near-normal (n=232),” “asthmatic (n=392),” “chronic obstructive pulmonary disease (COPD) (n=37),” “asthmatic-overlap (n=893),” and “COPD-overlap (n=586)” subtypes. Near-normal group showed the oldest mean age (72±7 years) and highest FEV1 (102%±8% predicted), and asthmatic group was the youngest (46±9 years). COPD and COPD-overlap groups were male predominant and all current or ex-smokers. While asthmatic group had the lowest prescription rate despite the highest proportion of self-reported wheezing, COPD, asthmatic-overlap, and COPD-overlap groups showed high prescription rate of respiratory medicine. Although COPD group formed only 1.7% of total subjects, they showed the highest mean medical cost and health care utilization, comprising 5.3% of the total medical cost. When calculating a ratio of total medical expense to household income, the mean ratio was highest in the COPD group.
Conclusion
Clinical and epidemiological heterogeneities of subjects with mild-to-moderate airflow limitation and a different level of health care utilization by each subtype are shown. Identification of a subtype with high health care demand could be a priority for effective utilization of limited resources.
Background: Osteoporosis, the most common extra-pulmonary complication of chronic obstructive pulmonary disease (COPD), may be related to general causes or COPD-specific causes such as low forced expiratory volume in 1 s (FEV1) and hypoxia. A few studies reported that emphysema is an independent risk factor for osteoporosis. However, other workers considered the association to be confounded by low FEV1 and low body mass index (BMI) which cluster with emphysema.
Aims: To study the association between osteoporosis and emphysema in a model that includes these potentially confounding factors.
Methods: We studied prospectively 52 COPD patients with both high resolution computed tomography and carbon monoxide diffusion coefficient as diagnostic markers of emphysema. Dual-energy X-ray absorptiometry was used to measure the bone mass density (BMD) of lumbar vertebrae and neck of the femur. Vertebral fractures were evaluated using the Genant semiquantitative score. Multiple linear regression analysis was used to identify the following independent variables: age, BMI, FEV1% predicted, PaO2, emphysema score, C-reactive protein (CRP), and dyspnea score as related to BMD. P ≤ 0.05 was considered statistically significant.
Results: There was no significant difference in the serum Vitamin D levels, vertebral fracture score, or BMD between the emphysematous and nonemphysematous patients. Multivariate analysis showed that (in a model including age, BMI, FEV1, PaO2, emphysema score, CRP, and dyspnea score) only reduced BMI, FEV1, and PaO2were independent risk factors for low BMD.
Conclusions: The emphysematous phenotype is not a risk factor for osteoporosis independently of BMI, FEV1, and PaO2.
Bronchiectasis is a long-neglected disease currently experiencing a surge in interest. It is a highly complex condition with numerous aetiologies, co-morbidities and a heterogeneous disease presentation and clinical course. The past few years have seen major advances in our understanding of the disease, primarily through large real-life cohort studies. The main outcomes of interest in bronchiectasis are symptoms, exacerbations, treatment response, disease progression and death. We are now more able to identify clearly the radiological, clinical, microbiological and inflammatory contributors to these outcomes. Over the past couple of years, multidimensional scoring systems such as the Bronchiectasis Severity Index have been introduced to predict disease severity and mortality. Although there are currently no licensed therapies for bronchiectasis, an increasing number of clinical trials are planned or ongoing. While this emerging evidence is awaited, bronchiectasis guidelines will continue to be informed largely by real-life evidence from observational studies and patient registries. Key developments in the bronchiectasis field include the establishment of international disease registries and characterisation of disease phenotypes using cluster analysis and biological data.
Purpose:
The purposes of the study are to identify clinical phenotypes that reflect the level of adaptation to the disease and to examine whether these clinical phenotypes respond differently to treatment as usual (TAU) and pulmonary rehabilitation (PR), the latter with its strong emphasis on improving adaptation.
Methods:
Clusters were identified by a cluster analysis using data on many subdomains of the four domains of health status (HS) (physiological functioning, functional impairment, symptoms and quality of life) in 160 outpatients with chronic obstructive pulmonary disease (COPD) receiving TAU. By discriminant analysis in the TAU sample, all 459 PR patients could be assigned to one of the identified clusters. The effect of TAU and PR on HS was examined with paired t tests.
Results:
Three distinct phenotypes were identified in the TAU sample. Two types were labelled adapted: phenotype 1 (moderate COPD-low impact on HS, n = 53) and phenotype 3 (severe COPD-moderate impact on HS, n = 73). One type was labelled non-adapted: phenotype 2 (moderate COPD-high impact on HS, n = 34). After 1-year TAU, the integral health status of all patients did not improve in any subdomain. In contrast, at the end of PR, significant improvements in HS were found in all three phenotypes especially the non-adapted.
Conclusions:
Different phenotypes exist in COPD that are based on behavioural aspects (i.e. the level of adaptation to the disease). Non-adapted patient responds better to treatments with a strong emphasis on improving adaptation by learning the patient better self-management skills. Knowing to which clinical phenotype a patient belongs helps to optimize patient-tailored treatment.
Background: Patients with heart failure (HF) are heterogeneous with multiple complex phenotypes across the ejection fraction (EF) spectrum. Phenotype-specific response to various treatments have not been well-described.
Hypothesis: Clinical response to specific interventions will vary according to HF phenotype.
Methods: Using latent class analysis, six cluster-based HF phenotypes across the EF spectrum were previously identified using patient data from 2130 patients enrolled in HF-ACTION (LVEF ≤ 35%) and 1767 patients enrolled from the Americas in TOPCAT (LVEF ≥ 45%) based on age, sex, race, CAD, BMI, hyperlipidemia, hypertension, diabetes mellitus, atrial fibrillation, COPD, anemia, and renal function, but not LVEF. Response to aerobic exercise training vs. usual care (HF-ACTION) and spironolactone vs. placebo (TOPCAT) were quantified by phenotype. The primary outcome was a composite of cardiovascular mortality (CVM) or HF hospitalization (HFH). Secondary outcomes included CVM, HFH, and all-cause mortality (ACM). Change in peak VO2 at 3 and 12 months were also analyzed in HF-ACTION.
Results: Of the established phenotypes, the phenotype composed of elderly non-ischemic patients as well as the non-white/non-ischemic/hypertensive phenotype experienced improvement in combined CVM and HFH, ACM and exercise capacity (28% vs 38%, HR: 0.66 [0.46-0.94], 8% vs 16% HR: 0.49 [0.27-0.91], change in VO2: 1.06±3.01 vs 0.04±3.14, p<0.05). Elderly patients with non-ischemic HFrEF enrolled in HF-ACTION randomized to therapeutic exercise program demonstrated significantly improved exercise capacity compared to usual care (change in VO2: 1.45±2.82 vs -0.09±2.49 and 1.25±3.18 vs 0.66±3.64 respectively, p<0.05). Elderly non-ischemic patients treated with spironolactone in TOPCAT had a lower risk of the primary outcome CVM and HFH (20% vs 27%, HR: 0.67 [0.48-0.95]), driven mostly by reduced CVM (9% vs 17%, HR: 0.52 [0.32-0.84]).
Conclusions: Response to varied treatments such as exercise training and spironolactone varies among complex HF phenotypes in both HFpEF and HFrEF. Additional investigation which further characterizes phenotype-specific treatments may help select specific interventions most likely to benefit specific phenotypes.
A basic model of factor analysis is employed in the estimation of multiple correlation coefficients and partial regression weights. Estimators are derived for situations in which some or all of the independent variates are subject to errors in measurement. The effect of the errors is indicated and the problem of bias in the estimators is considered. In one special case it is shown how a best subset of the independent variates of any size can readily be found for data under analysis.
This study looks at the assessment parameters and their informing value for the phenotypic stratification of elderly patients with chronic obstructive pulmonary disease (COPD). Using cluster analysis all patients were divided into subgroups (phenotypes). For women, phenotypic cluster K1: patients with a normal body weight, with disease duration more than 5 years; with a frequency of exacerbations less than 2 times a year. Phenotypic cluster K2: elderly patients, but younger than in K1, overweight, with a disease duration less than 5 years, with a frequency of exacerbations less than 2 times a year, but lesser than in K1, with a history of asthma in 66% of cases. 3 phenotypes were identified for men: K1 - overweight, disease duration of about 6 years, with a frequency of exacerbations more than 2 times a year; K2 - patients with body weight deficiency, disease duration more than 7 years, frequency of exacerbations less than 2 times a year; K3 - overweight, disease duration more than 8 years, with a frequency of exacerbations less than 2 times a year. COPD main signs of phenotyping in elderly patients were determined: gender, disease duration, body mass index, frequency of exacerbations. This allowed to identify 2 phenotypes in women: COPD with in frequent exacerbations and phenotype with presence of syndrome BA-COPD; in men - 3 phenotypes: bronchitis's, emphysematous and COPD with a slowly progressive course.
Background:
Previous studies have highlighted a relationship between reduction in rate of exacerbations with therapies containing inhaled corticosteroids (ICS) and baseline blood eosinophil count in patients with chronic obstructive pulmonary disease (COPD). The IMPACT trial showed that once-daily single-inhaler triple therapy significantly reduced exacerbations versus dual therapies. Blood eosinophil counts and smoking status could be important modifiers of treatment response to ICS. We aimed to model these relationships and their interactions, including outcomes other than exacerbations.
Methods:
IMPACT was a phase 3, randomised, double-blind, parallel-group, 52-week global study comparing once-daily single-inhaler triple therapy (fluticasone furoate-umeclidinium-vilanterol) with dual inhaled therapy (fluticasone furoate-vilanterol or umeclidinium-vilanterol). Eligible patients had moderate-to-very-severe COPD and at least one moderate or severe exacerbation in the previous year. We used fractional polynomials to model continuous blood eosinophil counts. We used negative binomial regression for numbers of moderate and severe exacerbations, severe exacerbations, and pneumonia. We modelled differences at week 52 in trough FEV1, St George's Respiratory Questionnaire (SGRQ) total score, and Transition Dyspnoea Index using repeated measurements mixed effect models. IMPACT was registered with ClinicalTrials.gov, number NCT02164513.
Findings:
The magnitude of benefit of regimens containing ICS (fluticasone furoate-umeclidinium-vilanterol n=4151 and fluticasone furoate-vilanterol n=4134) in reducing rates of moderate and severe exacerbations increased in proportion with blood eosinophil count, compared with a non-ICS dual long-acting bronchodilator (umeclidinium-vilanterol n=2070). The moderate and severe exacerbation rate ratio for triple therapy versus umeclidinium-vilanterol was 0·88 (95% CI 0·74 to 1·04) at blood eosinophil count less than 90 cells per μL and 0·56 (0·47 to 0·66) at counts of 310 cells per μL or more; the corresponding rate ratio for fluticasone furoate-vilanterol versus umeclidinium-vilanterol was 1·09 (0·91 to 1·29) and 0·56 (0·47 to 0·66), respectively. Similar results were observed for FEV1, Transition Dyspnoea Index, and SGRQ total score; however, the relationship with FEV1 was less marked. At blood eosinophil counts less than 90 cells per μL and at counts of 310 cells per μL or more, the triple therapy versus umeclidinium-vilanterol treatment difference was 40 mL (95% CI 10 to 70) and 60 mL (20 to 100) for trough FEV1, -0·01 (-0·68 to 0·66) and 0·30 (-0·37 to 0·97) for Transition Dyspnoea Index score, and -0·01 (-1·81 to 1·78) and -2·78 (-4·64 to -0·92) for SGRQ total score, respectively. Smoking status modified the relationship between observed efficacy and blood eosinophil count for moderate or severe exacerbations, Transition Dyspnoea Index, and FEV1, with former smokers being more corticosteroid responsive at any eosinophil count than current smokers.
Interpretation:
This analysis of the IMPACT trial shows that assessment of blood eosinophil count and smoking status has the potential to optimise ICS use in clinical practice in patients with COPD and a history of exacerbations.
Funding:
GlaxoSmithKline.
Aims
Acute exacerbation is a major event that alters the natural course of chronic obstructive pulmonary disease (COPD), and recurrent exacerbation results in worse clinical outcomes and greater economic consequences. While some patients suffer frequent exacerbations, others experience no exacerbations; this study was designed to detect proteins that were differentially abundant in COPD frequent exacerbators and assess whether those expression profiles are unique among COPD patients.
Main methods
Tandem mass tag labeled quantitative proteomics combined with two-dimensional liquid chromatography-tandem mass spectrometry was used to detect the changes in the lung proteome in COPD frequent exacerbators and infrequent exacerbators. A series of bioinformatics analyses were performed to screen potential signatures of COPD frequent exacerbations. The accuracy of proteomic results was further verified by western blot studies.
Key findings
Compared with infrequent exacerbators, 23 proteins in the lung tissues from frequent exacerbators showed significant degrees of differential expression; combined bioinformatics analyses of proteome indicated that the immune network for IgA production and the phenylalanine metabolism pathway were associated with frequent exacerbations. The Western blot analysis confirmed the expression pattern of three significantly regulated proteins (HLA-DQA1, pIgR and biglycan).
Significance
These findings indicate that immune response might play a key role in the pathophysiological mechanisms of COPD frequent exacerbations. Our results make a crucial contribution to the search for a comprehensive understanding of potential pathophysiological mechanisms associated with the frequent exacerbations of COPD, and might provide guidance for treating frequent exacerbations.
Background:
Treatment with systemic corticosteroids in patients with acute exacerbations of chronic obstructive pulmonary disease (COPD) is associated with debilitating adverse effects. Therefore, strategies to reduce systemic corticosteroid exposure are urgently required and might be offered by a personalised biomarker-guided approach to treatment. The aim of this study was to determine whether an algorithm based on blood eosinophil counts could safely reduce systemic corticosteroid exposure in patients admitted to hospital with acute exacerbations of COPD.
Methods:
We did a multicentre, randomised, controlled, open-label, non-inferiority trial at the respiratory departments of three different university-affiliated hospitals in Denmark. Eligible participants were patients included within 24h of admission to the participating sites, aged at least 40 years, with known airflow limitation (defined as a post-bronchodilator FEV1/forced vital capacity [FVC] ratio ≤0·70) and a specialist-verified diagnosis of COPD, who were designated to start on systemic corticosteroids by the respiratory medicine physician on duty. We randomly assigned patients (1:1) to either eosinophil-guided therapy or standard therapy with systemic corticosteroids. Both investigators and patients were aware of the group assignment. All patients received 80 mg of intravenous methylprednisolone on the first day. The eosinophil-guided group were from the second day given 37·5 mg of prednisolone oral tablet daily (for a maximum of up to 4 days) on days when their blood eosinophil count was at least 0·3 × 109 cells per L. On days when the eosinophil count was lower, prednisolone was not administered. If a patient was discharged during the treatment period, a treatment based on the last measured eosinophil count was prescribed for the remaining days within the 5-day period (last observation carried forward). The control group received 37·5 mg of prednisolone tablets daily from the second day for 4 days. The primary outcome was the number of days alive and out of hospital within 14 days after recruitment, assessed by intention to treat (ITT). Secondary outcomes included treatment failure at day 30 (ie, recurrence of acute exacerbation of COPD resulting in emergency room visits, admission to hospital, or need to intensify pharmacological treatment), number of deaths on day 30, and duration of treatment with systemic corticosteroids. The non-inferiority margin was 1·2 days (SD 3·8). This trial is registered at ClinicalTrials.gov, number NCT02857842, and was completed in January, 2019.
Findings:
Between Aug 3, 2016, and Sept 30, 2018, 159 patients in the eosinophil-guided group and 159 patients in the control group were included in the ITT analyses. There was no between-group difference for days alive and out of hospital within 14 days after recruitment: mean 8·9 days (95% CI 8·3-9·6) in the eosinophil-guided group versus 9·3 days (8·7-9·9) in the control group (absolute difference -0·4, 95% CI -1·3 to 0·5; p=0·34). Treatment failure at 30 days occurred in 42 (26%) of 159 patients in the eosinophil-guided group and 41 (26%) of 159 in the control group (difference 0·6%, 95% CI -9·0 to 10·3; p=0·90). At 30 days nine patients (6%) of 159 in the eosinophil-guided group and six (4%) of 159 in the control group had died (difference 1·9%, 95% CI -2·8 to 6·5; p=0·43). Median duration of systemic corticosteroid therapy was lower in the eosinophil-guided group: 2 days (IQR 1·0 to 3·0) compared with 5 days (5·0 to 5·0) in the control group, p<0·0001.
Interpretation:
Eosinophil-guided therapy was non-inferior compared with standard care for the number of days alive and out of hospital, and reduced the duration of systemic corticosteroid exposure, although we could not entirely exclude harm on some secondary outcome measures. Larger studies will help to determine the full safety profile of this strategy and its role in the management of COPD exacerbations.
Funding:
The Danish Regions Medical Fund and the Danish Council for Independent Research.
Chronic obstructive pulmonary disease (COPD) which comprises the phenotypes of chronic bronchitis and emphysema is often associated with pulmonary hypertension (PH). However, currently no approved therapy exists for PH-COPD. Signalling of the nitric oxide/cyclic guanoside monophosphate (NO-cGMP) axis plays an important role in PH and COPD.
We investigated the treatment effect of riociguat, which promotes the NO-cGMP pathway, in the mouse model of smoke-induced PH and emphysema in a curative approach and retrospectively analysed the effect of riociguat treatment on PH in single patients with PH-COPD.
In mice with established PH and emphysema (after 8 months of cigarette smoke exposure) riociguat treatment for another 3 months fully reversed PH. Moreover, histological hallmarks of emphysema were decreased. Microarray analysis revealed involvement of different signalling pathways, e.g. related to matrix metalloproteinases (MMPs). MMP activity was decreased in vivo by riociguat. In PH-COPD patients treated with riociguat (n=7) the pulmonary vascular resistance, airway resistance and circulating MMP levels decreased, while oxygenation at rest was not significantly changed.
Conclusions: Riociguat may be beneficial for treatment of PH-COPD. Further long-term prospective studies are necessary to investigate the tolerability, efficacy on functional parameters and the effect specifically on pulmonary emphysema in COPD patients.
Background:
Severe emphysema is a debilitating condition with few treatment options. Lung volume reduction procedures in the treatment of severe emphysema have shown excellent results in selected patients but their exact role remains unclear with studies reporting a wide variation in outcomes. We therefore aimed to evaluate the effects of volume reduction.
Methods:
We did a systematic review and meta-analysis. We searched MEDLINE on Sept 29, 2016, for trials of lung volume reduction in patients with emphysema, and we did an updated search on Embase and PubMed on June 18, 2018. We only included randomised controlled studies published in English evaluating the intervention with either sham or standard of care. Inclusion was limited to trials of techniques in which there was sustainable volume reduction. Primary outcomes were residual volume, FEV1, St George's Respiratory Questionnaire (SGRQ), and 6-min walk distance (6MWT). Secondary outcomes were severe adverse events (including mortality), short-term mortality, and overall mortality. We extracted summary level data from the trial publications and where necessary we obtained unpublished data. A random-effects model with the I2 statistic was used to determine heterogeneity and trial weight in each analysis. The study is registered with the PROSPERO database, number CRD42016045705.
Findings:
We identified 4747 references in the search, and included 20 randomised controlled trials of lung volume reduction involving 2794 participants with emphysema. Following lung volume reduction from any of the interventions in pooled analyses (ie, surgery, endobronchial valve, endobronchial coil, or sclerosing agents), the mean differences compared with the control were reduction in residual volume of 0·58 L (95% CI -0·80 to -0·37), increase in FEV1 of 15·87% (95% CI 12·27 to 19·47), improvement in 6MWT of 43·28 m (31·36 to 55·21), and reduction in the SGRQ of 9·39 points (-10·92 to -7·86). The odds ratio for a severe adverse event, which included mortality, was 6·21 (95% CI 4·02 to 9·58) following intervention. Regression analysis showed improvements relative to the degree of volume reduction: FEV1 (r2=0·86; p<0·0001), 6MWT (r2=0·77; p<0·0001), and SGRQ (r2=0·70; p<0·0001). Most studies were at high risk of bias for lack of blinding, and heterogeneity was high for some outcomes when pooled across all interventions, but was generally lower in the subgroups by intervention type.
Interpretation:
Despite limitations of high risk of bias and heterogeneity for some analyses, our results provide support that lung volume reduction in patients with severe emphysema on maximal medical treatment has clinically meaningful benefits. These benefits should be considered alongside potential adverse events.
Funding:
None.
Background: the concept that the small conducting airways <2mm in diameter become the major site of airflow obstruction in chronic obstructive pulmonary disease (COPD) is well established in the literature. It has also been shown that the last generation of small conducting airways, terminal bronchioles, are significantly destroyed in patients with very-severe COPD. What is not known is at what stage in the development of COPD the loss of small airways occurs, or how loss of terminal and transitional (first generation of respiratory airways) bronchioles - relates to the loss alveolar surface area that characterizes emphysema.
Methods: a novel multi-resolution computed tomography (CT) imaging protocol was applied to systematically, randomly sampled whole lungs or lobes of smokers with normal lung function (n=10), mild (n=10), moderate (n=8), and very-severe COPD (n=6). The 34 lung specimens provided 262 lung tissue samples for stereological assessment of the number and morphology of terminal and transitional bronchioles, airspace size (Lm), alveolar surface area.
Findings: the new data demonstrate that 41% of terminal bronchioles, 57% of transitional bronchioles, and 37% of the alveolar surface area is lost in patients with mild and moderate COPD compared to control smokers, before any emphysematous changes can be detected by CT. We also show these pathological changes correlate with lung function decline. Importantly, we demonstrate that loss of terminal and transitional bronchioles occurs in regions of the lung that have no loss of alveolar surface area. Further, we validated using histology, that the surviving small airways have thickened walls and narrowed lumens which become more obstructed as the disease progresses.
Interpretation: these data demonstrate that small airways disease is an early pathological feature in mild and moderate COPD. Importantly, this study emphasises that early intervention in mild and moderate COPD patients is most likely required for disease modification.
Background:
Because of the rapid change in economic development and lifestyle in China, and the ageing population, concerns have grown that chronic obstructive pulmonary disease (COPD) could become epidemic. An up-to-date nationwide estimation of COPD prevalence in China is needed.
Methods:
We did a cross-sectional survey of a nationally representative sample of individuals from mainland China aged 40 years or older. The primary outcome was COPD, defined according to the 2017 Global Initiative for Chronic Obstructive Lung Disease (GOLD) lung function criteria.
Findings:
Between Dec 29, 2014, and Dec 31, 2015, 66 752 adults were recruited to the study population. The estimated standardised prevalence of COPD was 13·6% (95% CI 12·0-15·2). The prevalence of COPD differed significantly between men and women (19·0%, 95% CI 16·9-21·2 vs 8·1%, 6·8-9·3; p<0·0001), mainly because of a significant difference in smoking status between men and women (current smokers 58·2% vs 4·0%). The prevalence of COPD differed by geographic region, with the highest prevalence in southwest China (20·2%, 95% CI 14·7-25·8) and the lowest in central China (10·2%, 8·2-12·2). Among adults with COPD, 56·4% (95% CI 53·7-59·2) had mild disease (GOLD stage I), 36·3% (34·3-38·3) had moderate disease (GOLD stage II), 6·5% (5·5-7·4) had severe disease (GOLD stage III), and 0·9% (0·6-1·1) had very severe disease (GOLD stage IV).
Interpretation:
In a large, nationally representative sample of adults aged 40 years or older, the estimated overall prevalence of COPD in China in 2014-15 was 13·6%, indicating that this disease has become a major public-health problem. Strategies aimed at prevention and treatment of COPD are needed urgently.
Funding:
Chinese Central Government, the Ministry of Science and Technology of The People's Republic of China, and the National Natural Science Foundation of China.
Asthma and chronic obstructive pulmonary disease (COPD) are complex and overlapping diseases that include inflammatory phenotypes. Novel anti-eosinophilic/anti-neutrophilic strategies demand rapid inflammatory phenotyping, which might be accessible from exhaled breath.
Our objective was to capture clinical/inflammatory phenotypes in patients with chronic airway disease using an electronic nose (eNose) in a training and validation set.
This was a multicentre cross-sectional study in which exhaled breath from asthma and COPD patients (n=435; training n=321 and validation n=114) was analysed using eNose technology. Data analysis involved signal processing and statistics based on principal component analysis followed by unsupervised cluster analysis and supervised linear regression.
Clustering based on eNose resulted in five significant combined asthma and COPD clusters that differed regarding ethnicity (p=0.01), systemic eosinophilia (p=0.02) and neutrophilia (p=0.03), body mass index (p=0.04), exhaled nitric oxide fraction (p<0.01), atopy (p<0.01) and exacerbation rate (p<0.01). Significant regression models were found for the prediction of eosinophilic (R ² =0.581) and neutrophilic (R ² =0.409) blood counts based on eNose. Similar clusters and regression results were obtained in the validation set.
Phenotyping a combined sample of asthma and COPD patients using eNose provides validated clusters that are not determined by diagnosis, but rather by clinical/inflammatory characteristics. eNose identified systemic neutrophilia and/or eosinophilia in a dose-dependent manner.
Background:
The Global Initiative for Chronic Obstructive Lung Disease (GOLD) 2017 classification separates the spirometric 1-4 staging from the ABCD groups defined by symptoms and exacerbations. Little is known about how this new classification predicts mortality in patients with chronic obstructive pulmonary disease (COPD). We aimed to establish the predictive ability of the GOLD 2017 classification, compared with earlier classifications, for all-cause and respiratory mortality, both when using its main ABCD groups and when further subdividing according to spirometric 1-4 staging.
Methods:
In this nationwide cohort study, we enrolled patients with COPD with data available in the Danish registry for COPD. To be included in this registry, individuals must have been outpatients in hospital-based pulmonary clinics in Denmark. Eligible patients were aged 30 years or older; had received a primary diagnosis of COPD (International Classification of Diseases [ICD]-10 J44.X) or acute respiratory failure (ICD-10 J96.X) in combination with COPD (ICD-10 J44.X) as a secondary diagnosis; and had complete data on FEV1, body-mass index, modified Medical Research Council dyspnoea scale score, and smoking status. We categorised eligible patients with complete data according to the 2007, 2011, and 2017 GOLD classifications at the first contact with an outpatient clinic. For the GOLD 2017 classification, we further subdivided the patients by spirometry into 16 subgroups (1A to 4D). We calculated adjusted hazard ratios (HRs) for all-cause and respiratory mortality and compared the predictive ability of the three GOLD classifications (2007, 2011, and 2017) using receiver operating curves.
Findings:
We enrolled 33 765 patients with COPD, who were outpatients in Danish hospitals between Jan 1, 2008, and Nov 30, 2013, in the main cohort assessed for all-cause mortality. 22 621 of these patients had data available on cause-specific mortality (respiratory) and were included in a subcohort followed from Jan 1, 2008, to Dec 31, 2011. For the GOLD 2017 classification, 3 year mortality increased with increasing exacerbations and dyspnoea from group A (all-cause mortality 10·0%, respiratory mortality 3·0%) to group D (all-cause mortality 36·9%, respiratory mortality 18·0%). However, 3 year mortality was higher for group B patients (all-cause mortality 23·8%, respiratory mortality 9·7%) than for group C patients (all-cause mortality 17·4%, respiratory mortality 6·4%). Compared with group A, adjusted HRs for all-cause mortality ranged from 2·05 (95% CI 1·87-2·26) for group B, to 1·47 (1·31-1·65) for group C, and to 3·01 (2·75-3·30) for group D. Area under the curve for all-cause mortality was 0·61 (95% CI 0·60-0·61) for GOLD 2007, 0·61 (0·60-0·62) for GOLD 2011, and 0·63 (0·53-0·73) for GOLD 2017. Area under the curve for respiratory mortality was 0·64 (0·62-0·65) for GOLD 2007, 0·63 (0·62-0·64) for GOLD 2011, and 0·65 (0·53-0·78) for GOLD 2017. The GOLD 2017 classification based on ABCD groups only did not predict mortality better than the earlier 2007 and 2011 GOLD classifications. However, when 16 subgroups (1A to 4D) were defined, the new classification predicted mortality more accurately than the previous systems (p<0·0001).
Interpretation:
We showed that the new GOLD 2017 ABCD classification does not predict all-cause and respiratory mortality more accurately than the previous GOLD systems from 2007 and 2011.
Funding:
Danish Lung Association, Program for Clinical Research Infrastructure.
It is uncertain whether phenotypes of asthma and chronic obstructive pulmonary disease (COPD) vary between populations with different genetic and environmental characteristics. Here, our objective was to compare the phenotypes of airways disease in two separate populations.
This was a cross-sectional observational study in adult populations from New Zealand and China. Participants aged 40–75 years who reported wheeze and breathlessness in the last 12 months were randomly selected from the general population and underwent detailed characterisation. Complete data for cluster analysis were available for 345 participants. Hierarchical cluster analysis was undertaken, based on 12 variables: forced expiratory volume in 1 s (FEV 1 ), FEV 1 /forced vital capacity ratio, bronchodilator reversibility, peak expiratory flow variability, transfer coefficient of the lung for carbon monoxide, exhaled nitric oxide fraction, total IgE, C-reactive protein, age of symptom onset, body mass index, health status and cigarette smoke exposure.
Cluster analysis of the combined dataset described five phenotypes: “severe late-onset asthma/COPD overlap group”, “moderately severe early-onset asthma/COPD overlap group”, “moderate to severe asthma group with type 2 predominant disease”, and two groups with minimal airflow obstruction, differentiated by age of onset. Separate analyses by country showed similar patterns; however, a distinct obese/comorbid group was observed in the New Zealand population.
Cluster analysis of adults with symptomatic airways disease suggests the presence of similar asthma/COPD overlap phenotypes within populations with different genetic and environmental characteristics, and an obese/comorbid phenotype in a Western population.
Purpose of review:
The application of artificial intelligence in the diagnosis of obstructive lung diseases is an exciting phenomenon. Artificial intelligence algorithms work by finding patterns in data obtained from diagnostic tests, which can be used to predict clinical outcomes or to detect obstructive phenotypes. The purpose of this review is to describe the latest trends and to discuss the future potential of artificial intelligence in the diagnosis of obstructive lung diseases.
Recent findings:
Machine learning has been successfully used in automated interpretation of pulmonary function tests for differential diagnosis of obstructive lung diseases. Deep learning models such as convolutional neural network are state-of-the art for obstructive pattern recognition in computed tomography. Machine learning has also been applied in other diagnostic approaches such as forced oscillation test, breath analysis, lung sound analysis and telemedicine with promising results in small-scale studies.
Summary:
Overall, the application of artificial intelligence has produced encouraging results in the diagnosis of obstructive lung diseases. However, large-scale studies are still required to validate current findings and to boost its adoption by the medical community.
This study aimed to identify simple rules for allocating chronic obstructive pulmonary disease (COPD) patients to clinical phenotypes identified by cluster analyses.
Data from 2409 COPD patients of French/Belgian COPD cohorts were analysed using cluster analysis resulting in the identification of subgroups, for which clinical relevance was determined by comparing 3-year all-cause mortality. Classification and regression trees (CARTs) were used to develop an algorithm for allocating patients to these subgroups. This algorithm was tested in 3651 patients from the COPD Cohorts Collaborative International Assessment (3CIA) initiative.
Cluster analysis identified five subgroups of COPD patients with different clinical characteristics (especially regarding severity of respiratory disease and the presence of cardiovascular comorbidities and diabetes). The CART-based algorithm indicated that the variables relevant for patient grouping differed markedly between patients with isolated respiratory disease (FEV 1 , dyspnoea grade) and those with multi-morbidity (dyspnoea grade, age, FEV 1 and body mass index). Application of this algorithm to the 3CIA cohorts confirmed that it identified subgroups of patients with different clinical characteristics, mortality rates (median, from 4% to 27%) and age at death (median, from 68 to 76 years).
A simple algorithm, integrating respiratory characteristics and comorbidities, allowed the identification of clinically relevant COPD phenotypes.
Background:
The COPD frequent exacerbator phenotype is usually defined by at least 2 treated exacerbations per year and is associated with a huge impact on patient health. However, existence of this phenotype and corresponding threshold still need to be formally confirmed by statistical methods analysing exacerbations profiles with no specific a priori hypothesis.
Objective:
To confirm the existence of the frequent exacerbator phenotype with an innovative unbiased statistical analysis of prospectively recorded exacerbations.
Methods:
Data of COPD patients from the French cohort EXACO were analysed using the KmL method designed to cluster longitudinal data and ROC curve analysis to determine the best threshold to allocate patients to identified clusters. Univariate and multivariate analyses were performed to study characteristics associated with different clusters.
Results:
Two clusters of patients were identified based on exacerbation frequency over time with 2.89 exacerbations per year on average in the first cluster (n=348) and 0.71 in the second (n=116). The best threshold to distinguish these clusters was 2 moderate to severe exacerbations per year. Frequent exacerbators had more airflow limitation, symptoms and health related quality of life impairment. A simple clinical score was derived to help identifying patients at risk of exacerbations.
Conclusions:
These analyses confirmed the existence and clinical relevance of a frequent exacerbator subgroup of COPD patients, and the currently used threshold to define this phenotype.
Background and objective:
Cluster analysis has been utilized to explore phenotypic heterogeneity in chronic obstructive pulmonary disease (COPD). To date, little is known about the longitudinal variability of clusters in COPD patients. We aimed to evaluate the 2-year cluster variability in stable COPD patients.
Methods:
We evaluated the following variables in COPD patients at baseline and 2 years later: age, gender, pack-year history, body mass index (BMI), modified Medical Research Council (MMRC) scale, 6-min walking distance (6MWD), spirometry and COPD Assessment Test (CAT). Patient classification was performed using cluster analysis at baseline and 2 years later. Each patient's cluster variability after 2 years and its parameters associated with cluster change were explored.
Results:
A total of 521 smokers with COPD were evaluated at baseline and 2 years later. Three different clusters were consistently identified at both evaluation times: cluster A (of younger age, mild airway limitation, few symptoms), cluster B (intermediate) and cluster C (of older age, severe airway limitation and highly symptomatic). Two years later, 70% of patients were unchanged, whereas 30% changed from one cluster to another: 20% from A to B; 15% from B to A; 15% from B to C; 42% from C to B and 8% from C to A. 6MWD, forced expiratory volume in 1 s (FEV1 ) % and CAT were the principal parameters responsible for this change.
Conclusion:
After 2 years of follow-up, most of the COPD patients maintained their cluster assignment. Exercise tolerance, lung function and quality of life were the main driving parameters in those who change their cluster assignment.
BACKGROUND: Increased oxidative stress and infl ammation has a role in the pathogenesis of chronic obstructive pulmonary disease (COPD). Drugs with antioxidant and anti-infl ammatory properties, such as N-acetylcysteine, might provide a useful therapeutic approach for COPD. We aimed to assess whether N-acetylcysteine could reduce
the rate of exacerbations in patients with COPD.
METHODS: In our prospective, randomised, double-blind, placebo-controlled, parallel-group study, we enrolled patients aged 40–80 years with moderate-to-severe COPD (post-bronchodilator forced expiratory volume in 1 s [FEV1]/forced vital capacity <0·7 and FEV1 of 30–70% of predicted) at 34 hospitals in China. We stratifi ed patients according to use of inhaled corticosteroids (regular use or not) at baseline and randomly allocated them to receive N-acetylcysteine (one 600 mg tablet, twice daily) or matched placebo for 1 year. The primary endpoint was the annual exacerbation rate in patients who received at least one dose of study drug and had at least one assessment visit after randomisation.
This study is registered with the Chinese Clinical Trials Registry, ChiCTR-TRC-09000460.
FINDINGS: Between June 25, 2009, and Dec 29, 2010, we screened 1297 patients, of whom 1006 were eligible for randomisation (504 to N-acetylcysteine and 502 to placebo). After 1 year, we noted 497 acute exacerbations in 482 patients in the N-acetylcysteine group who received at least one dose and had at least one assessment visit
(1·16 exacerbations per patient-year) and 641 acute exacerbations in 482 patients in the placebo group (1·49 exacerbations per patient-year; risk ratio 0·78, 95% CI 0·67–0·90; p=0·0011). N-acetylcysteine was well tolerated: 146 (29%) of 495 patients who received at least one dose of N-acetylcysteine had adverse events (48 serious),
as did 130 (26%) of 495 patients who received at least one dose of placebo (46 serious). The most common serious adverse event was acute exacerbation of COPD, occurring in 32 (6%) of 495 patients in the N-acetylcysteine group and 36 (7%) of 495 patients in the placebo group.
Interpretation: Our findings show that in Chinese patients with moderate-to-severe COPD, long-term use of N-acetylcysteine 600 mg twice daily can prevent exacerbations, especially in disease of moderate severity. Future studies are needed to explore effi cacy in patients with mild COPD (GOLD I).
Objective:
To study the distinct clinical phenotype of chronic airway diseases by hierarchical cluster analysis and two-step cluster analysis.
Methods:
A population sample of adult patients in Donghuamen community, Dongcheng district and Qinghe community, Haidian district, Beijing from April 2012 to January 2015, who had wheeze within the last 12 months, underwent detailed investigation, including a clinical questionnaire, pulmonary function tests, total serum IgE levels, blood eosinophil level and a peak flow diary. Nine variables were chosen as evaluating parameters, including pre-salbutamol forced expired volume in one second(FEV1)/forced vital capacity(FVC) ratio, pre-salbutamol FEV1, percentage of post-salbutamol change in FEV1, residual capacity, diffusing capacity of the lung for carbon monoxide/alveolar volume adjusted for haemoglobin level, peak expiratory flow(PEF) variability, serum IgE level, cumulative tobacco cigarette consumption (pack-years) and respiratory symptoms (cough and expectoration). Subjects' different clinical phenotype by hierarchical cluster analysis and two-step cluster analysis was identified.
Results:
(1) Four clusters were identified by hierarchical cluster analysis. Cluster 1 was chronic bronchitis in smokers with normal pulmonary function. Cluster 2 was chronic bronchitis or mild chronic obstructive pulmonary disease (COPD) patients with mild airflow limitation. Cluster 3 included COPD patients with heavy smoking, poor quality of life and severe airflow limitation. Cluster 4 recognized atopic patients with mild airflow limitation, elevated serum IgE and clinical features of asthma. Significant differences were revealed regarding pre-salbutamol FEV1/FVC%, pre-salbutamol FEV1% pred, post-salbutamol change in FEV1%, maximal mid-expiratory flow curve(MMEF)% pred, carbon monoxide diffusing capacity per liter of alveolar(DLCO)/(VA)% pred, residual volume(RV)% pred, total serum IgE level, smoking history (pack-years), St.George's respiratory questionnaire(SGRQ) score, acute exacerbation in the past one year, PEF variability and allergic dermatitis (P<0.05). (2) Four clusters were also identified by two-step cluster analysis as followings, cluster 1, COPD patients with moderate to severe airflow limitation; cluster 2, asthma and COPD patients with heavy smoking, airflow limitation and increased airways reversibility; cluster 3, patients having less smoking and normal pulmonary function with wheezing but no chronic cough; cluster 4, chronic bronchitis patients with normal pulmonary function and chronic cough. Significant differences were revealed regarding gender distribution, respiratory symptoms, pre-salbutamol FEV1/FVC%, pre-salbutamol FEV1% pred, post-salbutamol change in FEV1%, MMEF% pred, DLCO/VA% pred, RV% pred, PEF variability, total serum IgE level, cumulative tobacco cigarette consumption (pack-years), and SGRQ score (P<0.05).
Conclusion:
By different cluster analyses, distinct clinical phenotypes of chronic airway diseases are identified. Thus, individualized treatments may guide doctors to provide based on different phenotypes.
Background
COPD is a heterogeneous disease, but there is little consensus on specific definitions for COPD subtypes. Unsupervised clustering offers the promise of ‘unbiased’ data-driven assessment of COPD heterogeneity. Multiple groups have identified COPD subtypes using cluster analysis, but there has been no systematic assessment of the reproducibility of these subtypes.
Objective
We performed clustering analyses across 10 cohorts in North America and Europe in order to assess the reproducibility of (1) correlation patterns of key COPD-related clinical characteristics and (2) clustering results.
Methods
We studied 17 146 individuals with COPD using identical methods and common COPD-related characteristics across cohorts (FEV1, FEV1/FVC, FVC, body mass index, Modified Medical Research Council score, asthma and cardiovascular comorbid disease). Correlation patterns between these clinical characteristics were assessed by principal components analysis (PCA). Cluster analysis was performed using k-medoids and hierarchical clustering, and concordance of clustering solutions was quantified with normalised mutual information (NMI), a metric that ranges from 0 to 1 with higher values indicating greater concordance.
Results
The reproducibility of COPD clustering subtypes across studies was modest (median NMI range 0.17–0.43). For methods that excluded individuals that did not clearly belong to any cluster, agreement was better but still suboptimal (median NMI range 0.32–0.60). Continuous representations of COPD clinical characteristics derived from PCA were much more consistent across studies.
Conclusions
Identical clustering analyses across multiple COPD cohorts showed modest reproducibility. COPD heterogeneity is better characterised by continuous disease traits coexisting in varying degrees within the same individual, rather than by mutually exclusive COPD subtypes.
Background:
COPD can be diagnosed early using spirometry, but spirometry use is only recommended in symptomatic smokers, even though early stages of COPD can be asymptomatic. We investigated the prognosis of individuals with asymptomatic and symptomatic, undiagnosed COPD in the general population in Denmark.
Methods:
In this prospective cohort study, we analysed data from 95 288 individuals aged 20-100 years from the Copenhagen General Population Study. 32 518 (34%) of these individuals were regarded as being at high risk for COPD (defined as individuals aged 40 years or older, with cumulative tobacco consumption of ten pack-years or higher, and without self-reported or a previous hospital contact for asthma). COPD was defined as FEV1/forced vital capacity (FVC) of less than 70% and less than the lower limit of normal, and FEV1 of less than 80% of the predicted normal value. Individuals were considered undiagnosed if neither a previous COPD hospital contact, nor medical treatment for COPD, was registered. We obtained information on exacerbations and pneumonia from the National Danish Patient Registry and vital status from the National Danish Civil Registration System, and cause of death from the National Danish Causes of Death Registry. We used Cox proportional hazard models to assess risk of exacerbations, pneumonia, deaths due to respiratory causes, and deaths from all causes from 2003 to 2014.
Findings:
Between Nov 26, 2003, and July 10, 2013, 95 288 individuals were screened and 32 518 (34%) were at high risk of having COPD. 3699 (11%) of these participants met the COPD criteria and 2903 (78%) were undiagnosed, of whom 2052 (71%) were symptomatic. During a median follow-up of 6·1 years (IQR 4·9), we recorded 800 exacerbations, 2038 cases of pneumonia, and 2789 deaths in the 32 518 individuals at high risk of having COPD, including 152 deaths due to respiratory disease. Compared with individuals without COPD, the age and sex adjusted hazard ratio (HR) was 5·0 (95% CI 2·8-8·9) for exacerbations, 1·7 (1·3-2·2) for pneumonia, 0·7 (0·2-3·0) for death from respiratory causes, and 1·3 (1·1-1·6) for death from all causes in individuals with undiagnosed, asymptomatic COPD. Corresponding HRs were 15·5 (11·0-21·8) for exacerbations, 2·8 (2·4-3·3) for pneumonia, 4·3 (2·8-6·7) for death from respiratory causes, and 2·0 (1·8-2·3) for death from all causes in individuals with undiagnosed, symptomatic COPD.
Interpretation:
Individuals with undiagnosed, symptomatic COPD had an increased risk of exacerbations, pneumonia, and death. Individuals with undiagnosed, asymptomatic COPD had an increased risk of exacerbations and pneumonia. These findings suggest that better initiatives for early diagnosis and treatment of COPD are needed.
Funding:
The Danish Lung Association, the Danish Cancer Society, Herlev and Gentofte Hospital, Copenhagen University Hospital, and University of Copenhagen.
Chronic obstructive pulmonary disease (COPD) is a heterogeneous disorder. COPD patients may have different clinical features, imaging characteristics and natural history. Multiple studies have investigated heterogeneity using statistical methods such as unsupervised clustering to define different subgroups of COPD based largely on clinical phenotypes. Some studies have performed clustering using genetic data or limited numbers of blood biomarkers. Our primary goal was to use proteomic data to find subtypes of COPD within clinically similar individuals. In the Treatment of Emphysema with a gamma-Selective Retinoid Agonist (TESRA) study, multiplex biomarker panels were run in serum samples collected prior to randomization. After implementing an algorithm to minimize missing values, the dataset included 396 COPD individuals and 87 biomarkers. Using hierarchical clustering, we identified 3 COPD subgroups, containing 267 (67.4%), 104 (26.3%), and 25 (6.3%) individuals, respectively. The third cluster had less emphysema on quantitative analysis of chest computed tomography scans (p=0.03) and worse disease-related quality of life based on the St. George's Respiratory Questionnaire (total score cluster 1: 45.6, cluster 2: 45.4, cluster 3: 56.6; p=0.01), despite similar levels of lung function impairment (forced expiratory volume in 1 second (49.2%, 49.2%, 48.2 % predicted, respectively). Enrichment analysis showed the biomarkers distinguishing cluster 3 mapped to platelet alpha granule and cell chemotaxis pathways. Thus, we identified a subgroup which has less emphysema but may have greater inflammation, which could be potentially targeted with anti-inflammatory therapies.