Abstract
In COPD patients, exacerbation has a detrimental influence on the quality of life, disease progression and socioeconomic burden. This study aimed to develop and validate models to predict exacerbation, frequent exacerbations and severe exacerbations in COPD patients. We conducted an observational prospective multicenter study. Clinical data of all outpatients with stable COPD were collected from Beijing Chaoyang Hospital and Beijing Renhe Hospital between January 2018 and December 2019. Patients were followed up for 1 year. The data from Chaoyang Hospital was used for modeling dataset, and that of Renhe Hospital was used for external validation dataset. The final dataset included 456 patients, with 326 patients as the model group and 130 patients as the validation group. Using LABA + ICS, frequent exacerbations in the past year and CAT score were independent risk factors for exacerbation in the next year (OR = 2.307, 2.722 and 1.147), and FVC %pred as a protective factor (OR = 0.975). Combined with chronic heart failure, frequent exacerbations in the past year, blood EOS counts and CAT score were independent risk factors for frequent exacerbations in the next year (OR = 4.818, 2.602, 1.015 and 1.342). Using LABA + ICS, combined with chronic heart failure, frequent exacerbations in the past year and CAT score were independent risk factors for severe exacerbations in the next year (OR = 1.950, 3.135, 2.980 and 1.133). Based on these prognostic models, nomograms were generated. The prediction models were simple and useful tools for predicting the risk of exacerbation, frequent exacerbations and severe exacerbations of COPD patients in North China.
Introduction
Chronic Obstructive Pulmonary Disease (COPD) is a chronic disease characterized by persistent respiratory symptoms including dyspnea, cough and/or sputum production, and airflow limitation [Citation1,Citation2]. The global prevalence of COPD was 13.6% in people aged 40 years or older in China, indicating that COPD has already become a major public health issue [Citation3].
Exacerbations are important clinical events in COPD as well as critical predictors of the prognosis of COPD [Citation4]. In COPD patients, exacerbations have a significant detrimental influence on quality of life, disease progression and socioeconomic burden [Citation5]. Therefore, early identification and standard treatment of COPD exacerbations are major and arduous medical tasks in clinical practice.
Factors associated with increased risk of exacerbations in COPD have been an area of interest for many years. A multivariate analysis based on the ECLIPSE study demonstrated that the history of previous exacerbations is the best predictor of exacerbations [Citation6]. Several large-scale studies have suggested that the severity of the airflow limitation is associated with the increased risk of hospitalization, exacerbation and mortality [Citation7]. With the development of remote technologies and telemedicine, monitoring changes in respiratory rate and respiratory sounds may play a certain role in predicting exacerbations [Citation8]. Some studies have provided clues for the ability of GOLD classification (group A-D) to predict the risk of exacerbations [Citation9,Citation10], but others have shown that it has low discriminative power regarding the exacerbation risk for patients in groups A and B [Citation11]. Several studies have shown that certain biomarkers are correlated with the exacerbation, such as blood eosinophil counts [Citation12,Citation13], C-reactive protein and procalcitonin [Citation14]. However, to date, no specific biomarker has been discovered to predict exacerbations. The occurrence of exacerbations has been linked to many variables in previous studies. Thus, it’s critical to develop a prediction model to comprehensively assess the risk of exacerbations. A prediction model for the risk of exacerbations in COPD enables clinicians to identify high-risk COPD patients who were most likely to suffer from exacerbations and provide target therapy. Amin Adibi et al. developed a model that can be used (as a decision tool) to predict the incidence and severity of COPD exacerbations [Citation15]. However, there is a paucity of large-scale studies on the assessment of risk of exacerbations in China and a robust prediction model for future exacerbation risks.
The purpose of the study was to develop a generalizable model to predict the risk of exacerbation, frequent exacerbations and severe exacerbations, and to provide a theoretical basis for the early therapy of exacerbations, thus improving the quality of life and reducing the economic burden and mortality.
Material and methods
Participants and study design
To develop the prediction model, we conducted an observational prospective multicenter study. Clinical information of all outpatients with stable COPD was collected from Beijing Chaoyang Hospital and Beijing Renhe Hospital between January 2018 and December 2019. All patients enrolled in our study were consecutive patients satisfying the diagnosis of stable COPD based on GOLD 2017 (Global Initiative for Chronic Obstructive Lung Disease (GOLD), 2017) [Citation16]. Exclusion criteria were as follows: (1) A history of exacerbation in the past 3 months; (2) Suffering from other diseases leading to the obvious destruction of lung tissue, such as severe bronchiectasis and tuberculosis; (3) Patients with serious pleural disease and/or lesions involving the sternum and rib, etc.; (4) Suffering from other serious uncontrolled diseases; (5) History of any abdominal or thoracic surgery in the past 3 months; (6) Previous ocular surgery or any other active ocular disease in the past 3 months; (7) Myocardial infarction in the past 3 months; (8) Hospitalization for heart disease within the past 3 months; (9) Ongoing anti-tuberculosis treatment; (10) Pregnancy or lactation. Baseline variables and clinical follow-up data were collected and recorded prospectively. The data of Chaoyang Hospital was used for modeling dataset, and that of Renhe Hospital was used for external validation dataset. Following the baseline visit, all participants received optimal treatment based on GOLD 2017. The research was conducted in accordance with the Declaration of Helsinki. Patients are fully informed before participating in the study and a written informed consent must be signed. The Ethics Committee of Beijing Chao-Yang Hospital (2017-KE-92) approved the collection of clinical data from the patients.
Clinical variables related to COPD
The baseline data were prospectively collected at enrollment through the hospital’s electronic health record and through a direct patient visit. The following variables were evaluated in COPD patients: anthropometry (sex, age, body mass index(BMI), smoking status), disease duration, clinical symptom (cough, expectoration, wheeze, dyspnea), comorbidities (hypertension, coronary heart disease, diabetes mellitus, chronic heart failure, gastroesophageal reflux disease), history of exacerbation in the past year, history of pneumonia in the past year, medication use in the past year (long-acting muscarinic antagonist (LAMA), long-acting β2-agonist and inhaled corticosteroid (LABA + ICS), antibiotic), ventilation in the past year (noninvasive ventilation (NIV), invasive mechanical ventilation), laboratory examinations (white blood cells (WBC), neutrophils (NEU), eosinophils (EOS), fibrinogen, C-reactive protein(CRP)), postbronchodilator pulmonary function indices (forced vital capacity (FVC), forced vital capacity as percentage of predicted (FVC%pred), forced expiratory volume in one second (FEV1), forced expiratory volume in one second as percentage of predicted (FEV1%pred), and forced vital capacity rate of one second (FEV1/FVC)), arterial blood gas (pH, partial pressure of oxygen (PaO2), partial pressure of carbon dioxide (PaCO2)), Six minute walking distance (6MWD); questionnaires (Saint George’s Respiratory Questionnaire (SGRQ), COPD Assessment Test (CAT), modified Medical Research Council (mMRC) dyspnea scale, the Gastroesophageal Reflux Disease Questionnaire (GERD Q)). GOLD stage1–4 were recorded according to FEV1%pred, and GOLD group A-D were based on symptom burden and risk of exacerbation [Citation16]. According to the 2016 ESC guidelines, diagnosis of chronic heart failure was performed by cardiologists based on the presence of typical symptoms and signs, the levels of natriuretic peptides and the left ventricular ejection fraction (LVEF) measurement by echocardiography [Citation17]. The questionnaires were completed by specially trained medical personnel.
Follow-up and outcomes
The follow-up information came from outpatient follow-up reviews or telephone follow-ups. All participants were followed up once a month for a total of 12 months after enrollment. The definition of exacerbation was based on GOLD 2017. Exacerbation was defined as an acute worsening of respiratory symptoms that necessitated additional therapy and frequent exacerbations were defined as two or more exacerbations per year. Severe exacerbations were defined as an emergency-room visit or hospitalization for COPD [Citation16].
The follow-up information included the total number of exacerbations, the number of hospitalizations, emergency-room visits and/or outpatient visits owing to exacerbation. The patients were divided into three groups based on the number of exacerbations followed up in the model group: the no exacerbation group, the one exacerbation group and the frequent exacerbations group.
Outcomes of interest included the occurrence of exacerbation, frequent exacerbations and severe exacerbations of 12 months.
Statistical analysis
Statistical analysis was performed with SPSS 26.0 and R version 4.1.0. Descriptive statistics of the median with 25–75% interquartile range were used for quantitative variables, absolute frequencies, and percentages for qualitative data. Normality was tested by the Kolmogorov–Smirnov test and Shapiro–Wilk test. One-way analyses of variance (One-Way ANOVA) and Kruskal–Wallis test were used to compare the means of multiple groups. The prediction model was developed based on the modeling set and validated on the validation set. The binary logistic regression model was used to construct a prediction model. The variables included in the multivariate logistic analysis are those found to be meaningful in our univariate analysis, as well as the variables determined based on clinical relevance. The multivariable logistic model was fitted containing the independent variables and the relevant important clinical variables in the univariate analysis. An Akaike’s information criterion (AIC) was used for parsimonious goodness of fit comparison of the models. To evaluate the performance of the model, we used the receiver operating characteristic (ROC) curve and calculated the area under the curve (AUC). A nomogram was then developed from the prediction model. In order to verify the accuracy of the nomogram, we used the calibration curve to evaluate the nomogram. A comparison-wise significance level of 5% was used.
Results
Patient characteristics
Fourty-four patients who were lost to follow-up were excluded, four of whom died. The final dataset included 456 patients with 326 from Chaoyang Hospital as the model group and 130 from Renhe Hospital as the validation group. Baseline characteristics in the model group and the validation group are summarized in . In the modeling dataset, the total mean age was 67.7 years (SD 8.5) and 271 (83.13%) were men. In the validation dataset, the total mean age was 68.8 years (SD 11.4) and 89 (68.46%) were men. The most common comorbidities included hypertension (42.64%), coronary heart disease (22.39%), and diabetes mellitus (15.95%).
Table 1. Baseline characteristics in the model and validation group.
In the modeling dataset, the no exacerbation, the one exacerbation and the frequent exacerbations group included 191, 64 and 71 patients respectively (). There were no significant differences in sex, age, BMI, disease duration, symptoms and smoking status in the three groups. The proportion of patients combined with chronic heart failure in the frequent exacerbations group was higher than that in the other two groups (11.27% VS 4.19% and 1.56%, p = 0.043, 0.035). For the history of exacerbation, the proportion of previous exacerbations in the one exacerbation and the frequent exacerbations group was higher than in the no exacerbation group. In the frequent exacerbations and the one exacerbation group, the proportion of patients who used LABA + ICS in the past year was significantly higher than in the no exacerbation group (73.24% and 71.88% VS 50.79%, p = 0.001, 0.003). Twelve patients have received ventilation previously: NIV in 8 (2.45%) patients and invasive mechanical ventilation in 4 (1.23%) patients.
Table 2. Baseline characteristics of different exacerbation groups in the model group.
The laboratory findings, arterial blood gas, and pulmonary function parameters are also summarized in . There were no differences in laboratory findings between the three groups. The pH value in arterial blood gas analysis of patients in the frequent exacerbations group was lower than that of patients in the one exacerbation group (7.40 VS 7.42, p = 0.027), however, PaO2 and PaCO2 were not comparable in the three groups. Results of postbronchodilator pulmonary function were available in all patients. The values of FVC %pred of patients in the frequent exacerbations group were lower than that of patients in the no exacerbation group (71.24 VS 80.50, p = 0.001), the values of FEV1%pred of patients in the frequent exacerbations group were lower than that in the other two groups (41.65 VS 50.84 and 50.99, p = 0.004, 0.023). And the values of FEV1/FVC in the frequent exacerbations group were lower than that in the one exacerbation group (43.59 VS 49.77, p = 0.008). The patients in the frequent exacerbations group were more likely to be in GOLD stage 4 than the other two groups (36.62% VS 23.04% and 10.94%, p = 0.027, 0.036), while the patients in the no exacerbation group were more likely to be in GOLD group B (49.74% VS 31.25% and 32.39%, p = 0.010, 0.012) and less likely to be in GOLD group D (36.13% VS 51.56% and 59.15%, p = 0.029, <0.001). The scores of the frequent exacerbations group were higher than those of the no exacerbation group on SGRQ and mMRC (38.98 VS 27.47, p < 0.001; 2.36 VS 1.71 p < 0.001). The CAT score of the frequent exacerbations group was higher than that of the other two groups (26.07 VS 19.66 and 21.98, p < 0.001, 0.007).
Independent prognostic factors for exacerbation
Independent predictors were identified by multivariable logistics regression. The prediction model was constructed to predict exacerbation, frequent exacerbations and severe exacerbations in 12 months were presented in . Using LABA + ICS, frequent exacerbations in the past year and CAT score served as independent risk factors for exacerbation in the next year (OR = 2.307 (95% CI 1.309–4.068); 2.722 (95% CI 1.407–5.262) and 1.147 (95% CI 1.088–1.210)), and FVC %pred as a protective factor (OR = 0.975 (95% CI 0.951–0.999)). Combined with chronic heart failure, frequent exacerbations in the past year, blood EOS counts and CAT score were independent risk factors for frequent exacerbations in the next year (OR = 4.818 (95% CI 1.277–18.914; 2.602 (95% CI 1.065–6.356); 1.015 (95% CI 1.002–1.028); and 1.342 (95% CI 1.208–1.490)). Using LABA + ICS, combined with chronic heart failure, frequent exacerbations in the past year, CAT score were independent risk factors for severe exacerbations in the next year (OR = 1.950 (95% CI 1.076–3.535), 3.135 (95% CI 1.021–9.626), 2.980 (95% CI 1.510–5.882) and 1.133 (95% CI 1.075–1.194)).
Table 3. Predictive model of exacerbation in the model group.
Table 4. Predictive model of frequent exacerbations in the model group.
Table 5. Predictive model of severe exacerbations in the model group.
All of the variables were used to create the nomograms for exacerbation, frequent exacerbations and severe exacerbations. The predictive nomograms are shown in . By adding up the scores related to each variable and projecting total scores to the bottom scales, we were easily able to calculate the risk of exacerbation, frequent exacerbations and severe exacerbations in one month.
Figure 1. The nomogram of predicting exacerbation (A) and calibration curves in the modeling (B) and validation set (C).
Figure 2. The nomogram of predicting frequent exacerbations (A) and calibration curves in the modeling (B) and validation set (C).
Figure 3. The nomogram of predicting severe exacerbations (A) and calibration curves in the modeling (B) and validation set (C).
Calibration and validation of prognostic nomograms
To assess their discriminative ability, we constructed the ROC of the model group and validation group and calculated AUC (). The corresponding AUC/ROC (Area Under ROC curve) values were given in . The AUC of exacerbation model group was 0.748 (sensitivity: 0.781; specificity: 0.675), and the AUC of validation group was 0.737 (sensitivity: 0.724; specificity: 0.731). The AUC of frequent exacerbations model group was 0.842 (sensitivity: 0.800; specificity: 0.820), and the AUC of validation group was 0.824 (sensitivity: 0.694; specificity: 0.880). The AUC of the severe exacerbations model group was 0.761 (sensitivity: 0.700; specificity: 0.731), and the AUC of validation group was 0.795 (sensitivity: 0.674; specificity: 0.816).
Figure 4. The ROC curves of model and validation group. (A) The AUC values for predicting exacerbation. (B) The AUC values for predicting frequent exacerbations. (C) the AUC values for predicting severe exacerbations.
Table 6. Discriminative ability of exacerbation prediction model.
Table 7. Discriminative ability of frequent exacerbations prediction model.
Table 8. Discriminative ability of severe exacerbations prediction model.
The ROC curve for the predictive power of the regression model revealed that the discriminating capability of the three models was moderately strong, with the exacerbation, frequent exacerbations and severe exacerbations prediction models outperforming the others. Further, the calibration curves might depict the relationship between actual probability and predicted probability, as shown in .
Discussion
The most important finding of the study was the development and validation of the exacerbation, frequent exacerbations and severe exacerbations predictive model, which employs simple and generally obtainable clinical and demographic variables to predict the risk of exacerbations within 12 months. The prediction model aims to discover independent risk factors and predict the risk of exacerbations to target prevention and therefore lower the burden of COPD.
In our study, we established predictive models for the risk of exacerbation, frequent exacerbations and severe exacerbations within 12 months. According to our findings, using LABA + ICS and frequent exacerbations in the past year, FVC%pred and CAT score were independent predictors for exacerbation. Furthermore, combined with chronic heart failure, frequent exacerbations in the past year, blood EOS counts and CAT score were independent predictors for frequent exacerbations. Using LABA + ICS, combined with chronic heart failure, frequent exacerbations in the past year and CAT score were independent predictors for severe exacerbations. The three prediction models and established nomograms have been demonstrated to better visualize these predictive models.
In our study, patients combined with chronic heart failure had a greater risk of frequent exacerbations and severe exacerbations in the next year. Previous studies have shown that the functional, structural, and metabolic skeletal muscle abnormalities in COPD and chronic heart failure were identical [Citation17], which may verify the conclusion from the study. In recent studies, the coexistence of COPD and chronic heart failure significantly reduces the quality of life and increases morbidity, disability, and mortality of patients [Citation18]. Thus, for patients with stable COPD, on the one hand, clinicians should pay more attention to the important indicators representing cardiac function, on the other hand, we ought to place a greater emphasis on the health management of chronic heart failure.
The ECLIPSE study showed that 71% of patients with frequent exacerbations in the first and second years would experience exacerbations more than 2 times in the third year [Citation6]. Several studies have shown that exacerbation history was a meaningful predictor of exacerbation and was a criterion for assessing the risk of exacerbation [Citation19]. In our study, the one exacerbation and the frequent exacerbations group had significantly more exacerbations in the past year than the no exacerbation group. Additionally, frequent exacerbations in the past were an independent risk factor for exacerbation, frequent exacerbations and severe exacerbations in the next year. It highlights the significance of the previous history of exacerbation in predicting exacerbation.
In the previous history of medication, patients who used LABA + ICS are at increased risk of exacerbations in the next year. Using LABA + ICS was an independent predictor for the occurrence of severe exacerbations, whereas the past use of LABA + ICS has no bearing on the occurrence of frequent exacerbations. The triple therapy of LABA/LAMA/ICS has been demonstrated to significantly lower COPD exacerbation rates [Citation20]. From the univariate analyses, the proportion of LABA + ICS utilized in the one exacerbation and the frequent exacerbations group was comparable, both being higher than the no exacerbation group. It can be estimated that the impact of the LABA + ICS on the prevention of the risk need to be investigated. In the future, it will be required to improve the medication guidance and strengthen the management of COPD patients who are in a stable stage.
The relationship between blood EOS counts and exacerbations in previous studies is not consistent. It has been demonstrated that blood EOS counts in stable COPD patients might be utilized to predict exacerbations [Citation13]. On the other hand, other researchers suggest that EOS levels have little bearing on the prognosis in patients with exacerbation [Citation21]. In this study, blood EOS counts have a significant influence in predicting frequent exacerbations but have little effect in predicting exacerbations. It means that an increase in blood EOS counts is associated with a poor prognosis. Future investigations could stratify COPD patients into different groups based on the blood EOS counts, then explore the differences between these groups, including clinical characteristics, treatment, prognosis, etc.
Pulmonary function is well established to have a significant role in many aspects, including diagnosis, differential diagnosis, prognosis evaluation, and treatment of COPD. In the baseline data, the average FEV1/FVC in the one exacerbation group was higher than that of the other two groups and considerably higher than that of the frequent exacerbations group. The FEV1/FVC was not proportional to the severity of the disease. These results are in line with recent studies indicating that pulmonary function was weakly related to clinical symptoms and quality of life and that it was no longer used to categorize COPD severity. From our study, it can be observed that pulmonary function may have little clinical significance in predicting exacerbations. We found that only a decrease in FVC %pred increased the risk of exacerbation, while FEV1 and FEV/FVC had no effect in predicting exacerbation. This result, however, has never been described before. FVC is generally considered to be related to restrictive ventilation dysfunction [Citation22]. As for the relationship between FVC %pred and COPD, Chen et al. discovered that inconsistent-FVC groups (before bronchodilator FVC < 80% and after bronchodilator FVC ≥ 80%pred) showed the best responsiveness to bronchodilator and had more exacerbation occurrences [Citation23]. Xu and colleagues demonstrated that acute exposures to air pollutants were associated with a reduction in FVC %pred in COPD patients [Citation24]. We infer that the decrease in FVC %pred might be related to air trapping, which leads to an increase in residual volume and a decrease in vital capacity and impairs the exchange of oxygen and carbon dioxide, thus increasing the risk of exacerbation. However, several studies still have shown that FEV1%pred and FEV1/FVC are related to exacerbations, even if they are not the primary contributing factors [Citation6,Citation15]. Future studies may be able to focus more on the link between FVC or other pulmonary function parameters in COPD and exacerbations.
We filled out a variety of questionnaires for this study and the CAT has an effect on exacerbation, frequent exacerbations and severe exacerbations. The CAT was a validated short and simple instrument for assessing the impact of COPD on health status that might predict COPD exacerbation, health status deterioration, depression, and mortality [Citation25]. It was a tool that focused on describing the severity of a patient’s symptoms [Citation26]. When compared to other questionnaires, CAT was the strongest predictor of exacerbation in our study. This finding could be explained by the fact that the severity of symptoms is a significant factor in predicting future exacerbations. Worsened symptoms are closely associated with an increased risk of COPD exacerbations [Citation27]. It is important to note that the questionnaire for symptoms may be a very sensitive parameter for detecting the changes of COPD patients and assessing COPD exacerbations. Thus, future research will be required to pay more attention to the evaluation of symptoms. Furthermore, our conclusions would be of great utility value to clinicians in predicting COPD exacerbations and providing targeted therapies.
Compared with the established models for predicting exacerbation risk in previous studies, this study included more different types of variables in the screening process of predictive factors. Fewer studies, for example, include the history of medical care as a predictor in the model. Further, most predictive models lack external verification to verify the generalizability [Citation28]. At last, we constructed nomograms for visualizing the risk of exacerbation which may be used in clinical practice more effectively.
There are a few limitations to the current study that should be mentioned. First, the majority of patients were followed up by phone, with only a tiny proportion of patients receiving face-to-face follow-up in outpatient clinics, making it difficult to compare the dynamic changes of numerous indicators and intuitively comprehend the disease state of patients. Secondly, there are missing values in variables of different groups, which may bias the accuracy of prediction. Furthermore, the dynamic changes such as the change in treatment and tobacco habit are also important factors that may modulate the exacerbation factors. However, we did not find the change of treatment and tobacco habit in our patients. Future studies should put emphasis on these two indicators and further develop the prediction model. Finally, we should expand the sample size of patients to improve the accuracy and practicality of the prediction model in the future.
Conclusion
In conclusion, the research created and verified predictive models for acute exacerbations based on the clinical parameters, laboratory examinations and medical therapy to predict exacerbations within 12 months. The nomogram acted as an excellent tool to integrate clinical characteristics and established simple and useful tools for predicting the risk of exacerbation, frequent exacerbations and severe exacerbations of COPD patients.
Ethics statement
All patients provided informed consent. The Ethics Committee of Beijing Chao-Yang Hospital (2017-KE-92) approved the collection of clinical data from the patients.
Authors’ contributions
Authorship and contributorship: All authors made a significant contribution to the work. Z.Y. and H.S. had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Z. Y., B. X. contributed to the conception and design of the study; H.S., W. W., W. X., L. D., C. X. contributed to the performing of the study; H.S., W. W., W. X., L. D., C. X. drafted of the manuscript for important intellectual content; Z. Y., H.S., B. X. contributed to the data analysis; Z. Y., H.S. and B. X. contributed to the writing of the paper; All coauthors revised the work critically for important intellectual content, agreed on the journal to which the article has been submitted, gave final approval for this version to be published and agreed to be accountable for all aspects of the work.
| Abbreviations | ||
| COPD | = | chronic obstructive pulmonary disease |
| LABA + ICS | = | long-acting β2-agonist and inhaled corticosteroid |
| CAT | = | COPD assessment test |
| FVC %pred | = | forced vital capacity as percentage of predicted |
| EOS | = | eosinophils |
Disclosure statement
No potential conflict of interest was reported by the author(s).
Data availability statement
All data generated or analyzed during this study are included in this article. Further inquiries can be directed to the corresponding author.
Additional information
Funding
References
- Vos T, Flaxman AD, Naghavi M, et al. Years lived with disability (YLDs) for 1160 sequelae of 289 diseases and injuries 1990-2010: a systematic analysis for the global burden of disease study 2010. Lancet. 2012;380(9859):2163–2196. doi: 10.1016/S0140-6736(12)61729-2.
- Lozano R, Naghavi M, Foreman K, et al. Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the global burden of disease study 2010. Lancet. 2012;380(9859):2095–2128. doi: 10.1016/S0140-6736(12)61728-0.
- Fang L, Gao P, Bao H, et al. Chronic obstructive pulmonary disease in China: a nationwide prevalence study. Lancet Respir Med. 2018;6(6):421–430. doi: 10.1016/S2213-2600(18)30103-6.
- Wedzicha JA, Seemungal TA. COPD exacerbations: defining their cause and prevention. Lancet. 2007;370(9589):786–796. doi: 10.1016/S0140-6736(07)61382-8.
- Seemungal TA, Donaldson GC, Paul EA, et al. Effect of exacerbation on quality of life in patients with chronic obstructive pulmonary disease. Am J Respir Crit Care Med. 1998;157(5 Pt 1):1418–1422. doi: 10.1164/ajrccm.157.5.9709032.
- Hurst JR, Vestbo J, Anzueto A, et al. Susceptibility to exacerbation in chronic obstructive pulmonary disease. N Engl J Med. 2010;363(12):1128–1138. doi: 10.1056/NEJMoa0909883.
- Müllerova H, Maselli DJ, Locantore N, et al. Hospitalized exacerbations of COPD: risk factors and outcomes in the ECLIPSE cohort. Chest. 2015;147(4):999–1007. doi: 10.1378/chest.14-0655.
- Baroi S, McNamara RJ, McKenzie DK, et al. Advances in remote respiratory assessments for people with chronic obstructive pulmonary disease: a systematic review. Telemed J E Health. 2018;24(6):415–424. doi: 10.1089/tmj.2017.0160.
- Kim J, Lee CH, Lee MG, et al. Acute exacerbation according to GOLD 2017 categories in patients with chronic obstructive pulmonary disease. Arch Bronconeumol (Engl Ed). 2019;55(8):414–420. doi: 10.1016/j.arbres.2019.02.004.
- Le LAK, Johannessen A, Hardie JA, et al. Prevalence and prognostic ability of the GOLD 2017 classification compared to the GOLD 2011 classification in a Norwegian COPD cohort. Int J Chron Obstruct Pulmon Dis. 2019;14:1639–1655. doi: 10.2147/COPD.S194019.
- Erol S, Sen E, Gizem Kilic Y, et al. Does the 2017 revision improve the ability of GOLD to predict risk of future moderate and severe exacerbation? Clin Respir J. 2018;12(8):2354–2360. doi: 10.1111/crj.12914.
- Pascoe S, Locantore N, Dransfield MT, et al. Blood eosinophil counts, exacerbations, and response to the addition of inhaled fluticasone furoate to vilanterol in patients with chronic obstructive pulmonary disease: a secondary analysis of data from two parallel randomised controlled trials. Lancet Respir Med. 2015;3(6):435–442. doi: 10.1016/S2213-2600(15)00106-X.
- Kang HS, Kim SK, Kim YH, et al. The association between eosinophilic exacerbation and eosinophilic levels in stable COPD. BMC Pulm Med. 2021;21(1):74. doi: 10.1186/s12890-021-01443-4.
- Gong C, Yang Y, Chen M, et al. Effect of procalcitonin on the prognosis of patients with COPD. Biomed Rep. 2020;12(6):313–318.
- Adibi A, Sin DD, Safari A, et al. The acute COPD exacerbation prediction tool (ACCEPT): a modelling study. Lancet Respir Med. 2020;8(10):1013–1021. doi: 10.1016/S2213-2600(19)30397-2.
- Vogelmeier CF, Criner GJ, Martinez FJ, et al. Global strategy for the diagnosis, management, and prevention of chronic obstructive lung disease 2017 report. GOLD executive summary. Am J Respir Crit Care Med. 2017;195(5):557–582. doi: 10.1164/rccm.201701-0218PP.
- P, Voors AA, Anker SD, Bueno H, et al. 2016 ESC guidelines for the diagnosis and treatment of acute and chronic heart failure: the task force for the diagnosis and treatment of acute and chronic heart failure of the European Society of Cardiology (ESC) developed with the special contribution of the Heart Failure Association (HFA) of the ESC. Eur Heart J. 2016;37(27):2129–2200. doi: 10.1093/eurheartj/ehw128.
- Vitacca M, Paneroni M. Rehabilitation of patients with coexisting COPD and heart failure. COPD. 2018;15(3):231–237. doi: 10.1080/15412555.2018.1468427.
- Global Initiative for Chronic Obstructive Lung Disease (GOLD). Global strategy for the diagnosis, management and prevention of chronic obstructive pulmonary disease. 2021 REPORT.
- Lee HW, Park J, Jo J, et al. Comparisons of exacerbations and mortality among regular inhaled therapies for patients with stable chronic obstructive pulmonary disease: systematic review and Bayesian network meta-analysis. PLoS Med. 2019;16(11):e1002958. doi: 10.1371/journal.pmed.1002958.
- Martínez-Gestoso S, García-Sanz MT, Calvo-Álvarez U, et al. Variability of blood eosinophil count and prognosis of COPD exacerbations. Ann Med. 2021;53(1):1152–1158.
- Rivero-Yeverino D. Spirometry: basic concepts. Rev Alerg Mex. 2019;66(1):76–84. doi: 10.29262/ram.v66i1.536.
- Chen C, He Y, Chen Q, et al. A specific subtype of chronic obstructive pulmonary disease classified by forced vital capacity. J Thorac Dis. 2018;10(12):6547–6556. doi: 10.21037/jtd.2018.12.06.
- Gao N, Xu W, Ji J, et al. Lung function and systemic inflammation associated with short-term air pollution exposure in chronic obstructive pulmonary disease patients in Beijing, China. Environ Health. 2020;19(1):12. doi: 10.1186/s12940-020-0568-1.
- Karloh M, Fleig Mayer A, Maurici R, et al. The COPD assessment test: what do we know so far?: a systematic review and meta-analysis about clinical outcomes prediction and classification of patients into GOLD stages. Chest. 2016;149(2):413–425. doi: 10.1378/chest.15-1752.
- Jones PW, Harding G, Berry P, et al. Development and first validation of the COPD assessment test. Eur Respir J. 2009;34(3):648–654. doi: 10.1183/09031936.00102509.
- Mackay AJ, Kostikas K, Roche N, et al. Impact of baseline symptoms and health status on COPD exacerbations in the FLAME study. Respir Res. 2020;21(1):93. doi: 10.1186/s12931-020-01354-8.
- Bellou V, Belbasis L, Konstantinidis AK, et al. Prognostic models for outcome prediction in patients with chronic obstructive pulmonary disease: systematic review and critical appraisal. BMJ. 2019;367:l5358. doi: 10.1136/bmj.l5358.