Dataset Information

Prediction of lung emphysema in COPD by spirometry and clinical symptoms: results from COSYCONET.

ABSTRACT:

Background

Lung emphysema is an important phenotype of chronic obstructive pulmonary disease (COPD), and CT scanning is strongly recommended to establish the diagnosis. This study aimed to identify criteria by which physicians with limited technical resources can improve the diagnosis of emphysema.

Methods

We studied 436 COPD patients with prospective CT scans from the COSYCONET cohort. All items of the COPD Assessment Test (CAT) and the St George's Respiratory Questionnaire (SGRQ), the modified Medical Research Council (mMRC) scale, as well as data from spirometry and CO diffusing capacity, were used to construct binary decision trees. The importance of parameters was checked by the Random Forest and AdaBoost machine learning algorithms.

Results

When relying on questionnaires only, items CAT 1 & 7 and SGRQ 8 & 12 sub-item 3 were most important for the emphysema- versus airway-dominated phenotype, and among the spirometric measures FEV₁/FVC. The combination of CAT item 1 (≤ 2) with mMRC (> 1) and FEV₁/FVC, could raise the odds for emphysema by factor 7.7. About 50% of patients showed combinations of values that did not markedly alter the likelihood for the phenotypes, and these could be easily identified in the trees. Inclusion of CO diffusing capacity revealed the transfer coefficient as dominant measure. The results of machine learning were consistent with those of the single trees.

Conclusions

Selected items (cough, sleep, breathlessness, chest condition, slow walking) from comprehensive COPD questionnaires in combination with FEV₁/FVC could raise or lower the likelihood for lung emphysema in patients with COPD. The simple, parsimonious approach proposed by us might help if diagnostic resources regarding respiratory diseases are limited. Trial registration ClinicalTrials.gov, Identifier: NCT01245933, registered 18 November 2010, https://clinicaltrials.gov/ct2/show/record/NCT01245933 .

SUBMITTER: Kellerer C

PROVIDER: S-EPMC8427948 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:Background:A substantial prevalence of cardiovascular disease is known for COPD, but detection of its presence, relationship to functional findings and contribution to symptoms remains challenging. The present analysis focusses on the cardiovascular contribution to COPD symptoms and their relationship to the patients' diagnostic status, medication and echocardiographic findings. Methods:Patients from the COPD cohort COSYCONET with data on lung function, including FEV1, residual volume/total lung capacity (RV/TLC) ratio, diffusing capacity TLCO, and echocardiographic data on left ventricular ejection fraction (LVEF) and end-diastolic diameter (LVEDD), medical history, medication, modified British Medical Research Council dyspnea scale (mMRC) and Saint Georges Respiratory Questionnaire (SGRQ) were analyzed. Results:A total of 1591 patients (GOLD 0-4: n=230/126/614/498/123) fulfilled the inclusion criteria. Ischemic heart disease, myocardial infarction or heart failure were reported in 289 patients (18.2%); 860 patients (54%) received at least one cardiovascular medication, with more than one in many patients. LVEF<50% or LVEDD>56 mm was found in 204 patients (12.8%), of whom 74 (36.3%) had neither a cardiovascular history nor medication. Among 948 patients (59.6%) without isolated hypertension, there were 21/55 (38.2%) patients with LVEF<50% and 47/88 (53.4%) with LVEDD>56 mm, who lacked both a cardiac diagnosis and medication. LVEDD and LVEF were linked to medical history; LVEDD was dependent on RV/TLC and LVEF on FEV1. Exertional COPD symptoms were best described by mMRC and the SGRQ activity score. Beyond lung function, an independent link from LVEDD on symptoms was revealed. Conclusion:A remarkable proportion of patients with suspicious echocardiographic findings were undiagnosed and untreated, implying an increased risk for an unfavorable prognosis. Cardiac size and function were dependent on lung function and only partially linked to cardiovascular history. Although the contribution of LV size to COPD symptoms was small compared to lung function, it was detectable irrespective of all other influencing factors. However, only the mMRC and SGRQ activity component were found to be suitable for this purpose.

Project description:We studied whether in patients with stable COPD blood gases (BG), especially oxygenated hemoglobin (OxyHem) as a novel biomarker confer information on disease burden and prognosis and how this adds to the information provided by the comorbidity pattern and systemic inflammation. Data from 2137 patients (GOLD grades 1-4) of the baseline dataset of the COSYCONET COPD cohort were used. The associations with dyspnea, exacerbation history, BODE-Index (cut-off ?2) and all-cause mortality over 3 years of follow-up were determined by logistic and Cox regression analyses, with sex, age, BMI and pack years as covariates. Predictive values were evaluated by ROC curves. Capillary blood gases included SaO2, PaO2, PaCO2, pH, BE and the concentration of OxyHem [haemoglobin (Hb) x fractional SaO2, g/dL] as a simple-to-measure correlate of oxygen content. Inflammatory markers were WBC, CRP, IL-6 and -8, TNF-alpha and fibrinogen, and comorbidities comprised a broad panel including cardiac and metabolic disorders. Among BG, OxyHem was associated with dyspnoea, exacerbation history, BODE-Index and mortality. Among inflammatory markers and comorbidities, only WBC and heart failure were consistently related to all outcomes. ROC analyses indicated that OxyHem provided information of a magnitude comparable to that of WBC, with optimal cut-off values of 12.5?g/dL and 8000/µL, respectively. Regarding mortality, OxyHem also carried independent, additional information, showing a hazard ratio of 2.77 (95% CI: 1.85-4.15, p?<?0.0001) for values <12.5?g/dL. For comparison, the hazard ratio for WBC?>?8000/µL was 2.33 (95% CI: 1.60-3.39, p?<?0.0001). In stable COPD, the concentration of oxygenated hemoglobin provided additional information on disease state, especially mortality risk. OxyHem can be calculated from hemoglobin concentration and oxygen saturation without the need for the measurement of PaO2. It thus appears well suited for clinical use with minimal equipment, especially for GPs.

Project description:BACKGROUND:Forced expiratory volume in one second (FEV1) characterizes the pathophysiology of COPD and different trajectories of FEV1 decline have been observed in patients with COPD (e.g. gradual or episodic). There is limited information about the development of patient-reported health-related quality of life (HRQL) over the full range of the natural history of COPD. We examined the longitudinal association between change in FEV1 and change in disease-specific and generic HRQL. METHODS:We analysed data of 1734 patients with COPD participating in the COSYCONET cohort with up to 3 years of follow-up. Patients completed the Saint George's Respiratory Questionnaire (SGRQ) and the EQ-5D Visual Analog Scale (EQ VAS). Change score models were used to investigate the relationship between HRQL and FEV1 and to calculate mean changes in HRQL per FEV1 change categories [decrease (? -?100?ml), no change, increase (? 100?ml)] after 3 years. Applying hierarchical linear models (HLM), we estimated the cross-sectional between-subject difference and the longitudinal within-subject change of HRQL as related to a FEV1 difference or change. RESULTS:We observed a statistically significant deterioration in SGRQ (total score?+?1.3?units) after 3 years, which was completely driven by the activity component (+?4?units). No significant change was found for the generic EQ VAS. Over the same period, 58% of patients experienced a decrease in FEV1, 28% were recorded as no change in FEV1, and 13% experienced an increase. The relationship between HRQL and FEV1 was found to be approximately linear with decrease in FEV1 being statistically significantly associated with a deterioration in SGRQ (+?3.20?units). Increase in FEV1 was associated with improvements in SGRQ (-?3.81?units). The associations between change in FEV1 and the EQ VAS were similar. Results of the HLMs were consistent and highly statistically significant, indicating cross-sectional and longitudinal associations. The largest estimates were found for the association between FEV1 and the SGRQ activity domain. CONCLUSIONS:Difference and change in FEV1 over time correlate with difference and change in disease-specific and generic HRQL. We conclude, that deterioration of HRQL should induce timely re-examination of physical status and lung function and possibly reassessment of therapeutic regimes. TRIAL REGISTRATION:NCT01245933. Date of registration: 18 November 2010.