ABSTRACT
BACKGROUND: We determined the relationships between cytokine expression in sputum and clinical data to characterise and understand Chronic Obstructive Pulmonary Disease (COPD) exacerbations in COPD patients. METHODS: We measured 30 cytokines in 936 sputum samples, collected at stable state (ST) and exacerbation (EX) visits from 99 participants in the Acute Exacerbation and Respiratory InfectionS in COPD (AERIS) study (NCT01360398, www.clinicaltrials.gov). We determined their longitudinal expression and examined differential expression based on disease status or exacerbation type. RESULTS: Of the cytokines, 17 were suitable for analysis. As for disease states, in EX sputum samples, IL-17A, TNF-α, IL-1ß, and IL-10 were significantly increased compared to ST sputum samples, but a logistic mixed model could not predict disease state. As for exacerbation types, bacteria-associated exacerbations showed higher expression of IL-17A, TNF-α, IL-1ß, and IL-1α. IL-1α, IL-1ß, and TNF-α were identified as suitable biomarkers for bacteria-associated exacerbation. Bacteria-associated exacerbations also formed a cluster separate from other exacerbation types in principal component analysis. CONCLUSIONS: Measurement of cytokines in sputum from COPD patients could help identify bacteria-associated exacerbations based on increased concentrations of IL-1α, IL-1ß, or TNF-α. This finding may provide a point-of-care assessment to distinguish a bacterial exacerbation of COPD from other exacerbation types.
ABSTRACT
Background: Acute exacerbations of COPD (AECOPD) are episodes of breathlessness, cough and sputum which are associated with the risk of hospitalisation, progressive lung function decline and death. They are often missed or diagnosed late. Accurate timely intervention can improve these poor outcomes. Digital tools can be used to capture symptoms and other clinical data in COPD. This study aims to apply machine learning to the largest available real-world digital dataset to develop AECOPD Prediction tools which could be used to support early intervention and improve clinical outcomes. Objective: To create and validate a machine learning predictive model that forecasts exacerbations of COPD 1-8 days in advance. The model is based on routine patient-entered data from myCOPD self-management app. Method: Adaptations of the AdaBoost algorithm were employed as machine learning approaches. The dataset included 506 patients users between 2017 and 2021. 55,066 app records were available for stable COPD event labels and 1263 records of AECOPD event labels. The data used for training the model included COPD assessment test (CAT) scores, symptom scores, smoking history, and previous exacerbation frequency. All exacerbation records used in the model were confined to the 1-8 days preceding a self-reported exacerbation event. Results: TheEasyEnsemble Classifier resulted in a Sensitivity of 67.0 % and a Specificity of 65 % with a positive predictive value (PPV) of 5.0 % and a negative predictive value (NPV) of 98.9 %. An AdaBoost model with a cost-sensitive decision tree resulted in a a Sensitivity of 35.0 % and a Specificity of 89.0 % with a PPV of 7.08 % and NPV of 98.3 %. Conclusion: This preliminary analysis demonstrates that machine learning approaches to real-world data from a widely deployed digital therapeutic has the potential to predict AECOPD and can be used to confidently exclude the risk of exacerbations of COPD within the next 8 days.
ABSTRACT
Treatment of chronic obstructive pulmonary disease (COPD) exacerbations improves outcomes; however, responses to treatment are variable, and patients with COPD often delay presentation or fail to seek therapy. The impact on exacerbation outcomes, hospitalization, and health status of delaying or failing to seek treatment is poorly understood. We studied between 1996 and 2002 a cohort of 128 patients with COPD, mean (SD) FEV1 of 1.07 (0.43) L. Patients recorded respiratory symptoms daily and reported exacerbations to the outpatient-based study team or to their primary care physician; 1,099 exacerbations were recorded by the patients, of which 658 were reported to a physician. The time between exacerbation onset and treatment was a median (interquartile range) of 3.69 (2.05.57) days, and the exacerbation recovery time was 10.7 (7.014.0) days. Earlier treatment was associated with a faster recovery (regression coefficient 0.42 days/day delay) (confidence interval, 0.190.65; p < 0.001). Patients who reported a higher proportion of exacerbations for treatment had better health-related quality of life than those patients with more untreated exacerbations (rho = 0.22, p = 0.018). Failure to report exacerbations was associated with an increased risk of emergency hospitalization (rho = 0.21, p = 0.04). Patient recognition of exacerbation symptoms and prompt treatment improves exacerbation recovery, reduces risks of hospitalization, and is associated with a better health-related quality of life
Subject(s)
Humans , Pulmonary Disease, Chronic Obstructive/prevention & control , Pulmonary Disease, Chronic Obstructive/therapy , Drug Therapy/statistics & numerical dataABSTRACT
Study objective: Patients with COPD experience lower airway and systemic inflammation, and an accelerated decline in FEV. There is no evidence on whether this inflammation changes over time, or if it is associated with a faster decline FEV. Patient and design: a cohort of 148 COPD patients (100 men) was monitored daily for a median 2.9 years (interquartile range [IQR], 2.1 to 4.8). At recruitment median age was 68.5 years (IQR, 62.5 to 73.6) and FEV as percentage of predicted (FEV percent Pred) was 38.5 percent (IQR, 27.7 to 50.3). Results: During the study, the patients experienced 1,389 exacerbations, a median of 2.52/yr (IQR 1.48 to 3.96) and FEV declined by 40.2 mL/yr or as FEV percent Pred by 1.5 percent/yr. Concerning inflammatory markers, sputum interlukin (IL)-6 rose by 9 pg/mL, sputum neutrophil count rose by 1.64 x 10,000,000 cells per gram sputum per year, and plasma fibrinogen rose by 0.10 g/L/yr (all p, 0.05). Patients with frequent exacerbations (less than or equal to 2.52/yr) had a faster rise over time in plasma fibrinogen and sputum IL-6 of 0.063 g/L/yr (p= 0.046, n= 130) and 29.5 pg/mL/yr (p< 0.001, n=98), respectively, compared to patients with infrequent exacerbations (<2.52/yr). Using the earliest stable (nonexacerbation) measured marker, patients whose IL-6 exceeded the group median had a faster FEV percentPred decline of 0.97 percent/yr (p=0.001 and .40 percent/yr (p=0.014). respectively. Conclusions: In COPD, airway and systemic inflammatory markers increase over time; high levels of these markers are associated with a faster decline in lung function (AU)