Government & Politics

BQ FACED ERJ2014 appendix

Description
BQ FACED ERJ2014 appendix
Published
of 6
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Related Documents
Share
Transcript
   1 SUPPLEMENTAL APPENDIX 1 Initial selection of variables The following easy-to-measure variables were collected from all patients (see supplemental appendix): General (age and gender); historical (smoking in packs-year and other relevant background); anthropometric (body-mass index [BMI clinical (dyspnea measured with the modified scale of the Medical Research Council [MMRC]) [1]   and the macroscopic appearance of sputum [mucous, mucopurulent or purulent functional (forced vital capacity [FVC] and forced expiratory volume in the first second [FEV1], both post-bronchodilator and as a percentage of the predicted values, as well as the presence of respiratory insufficiency, defined as a saturation of oxyhemoglobin lower than 90% when breathing room air); radiological (number of lobes affected by bronchiectasis); and microbiological (chronic colonization by potentially pathogenic microorganisms (PPM), evaluated separately by PA and multiresistant gram-negative bacilli, as well as the isolation of Staphylococcus aureus, atypical micobacteria and fungi). Lastly, the number of exacerbations and hospitalizations for exacerbation of bronchiectasis was recorded. Exacerbation was defined according to the relevant national guidelines as a sustained acute presentation of changes in the characteristics of sputum (increased volume, consistency, purulence or hemoptysis) and/or intensification of dyspnea not explainable by any possible causes that may accompany them, although not necessarily, such as increased coughing, fever, asthenia, general poor condition, anorexia, weight loss, pleuritic chest pain, changes in the   2 respiratory exploration, radiological alterations, deteriorated pulmonary function or increased peripheral concentration of systemic markers of inflammation [2]) In the first stage, we included as eligible for the final score variables that proved significant (p≤0.1) in the univariate study resulting from the comparison of survivors and non-survivors 5 years after the radiological diagnosis of bronchiectasis, as well as those variables, in the opinion of the researchers, were clinically important or previously associated in the literature with increased mortality in patients with bronchiectasis, such as the body-mass index in kg/m 2  [3-6]. Of the variables that showed a high colinearity (defined as a correlation coefficient >0.6), we selected those that were easiest to obtain and had the greatest clinical interest. We therefore chose FEV1 from the functional variables, and the number of hospitalizations in the year prior to the radiological diagnosis of bronchiectasis from the exacerbation variables. Conversely, we decided to rule out the gasometric variables, other more complex radiological and functional evaluations and the quality of life measurements as none of these are assessed in routine clinical practice. Chronic colonization was defined as isolation of the same PPM after the diagnosis of bronchiectasis in three consecutive respiratory samples taken at least one month apart within a period of six months [2]. All the variables were obtained at a maximum of 6 months after the radiological diagnosis of bronchiectasis, except for the hospitalizations, which corresponded to the year prior to the radiological diagnosis. All the variables were always measured in a phase of clinical stability of at least 4 weeks.   3 SUPPLEMENTAL APPENDIX 2 Statistical analysis The variables of the survivor and non-survivor groups (univariate analysis) were compared with a student’s t test if they followed a normal distribution, or with their corresponding non-parametric test if they did not. The chi-squared test was used for the comparison of qualitative and dichotomic variables. The univariate analysis was used for the selection of the initial set of 12 variables. The normality of the variables was confirmed with the Kolgomorov-Smirnov test. The data corresponding to the different hospitals (centers) were compared with a one-way ANOVA test. A stepwise multiple logistic regression analysis was used to identify the variables chosen from the initial set of 12 to constitute the final score. This selection was limited to those variables that showed a significant association (p entry of 0.10 and p exit of 0.05) with the dependent variable (5-year all-cause mortality). These variables were subsequently dichotomized, and in each variable the optimal cut-off point was chosen to reflect the greatest individual capacity to predict death after 5 years, calculated according to the corresponding largest areas under the receiver operating characteristic (ROC) curves (AUC-ROC). Once they were dichotomized, these variables were reintroduced into a stepwise multiple logistic regression model to calculate each one’s relative weight in the final score. To simplify the score as much as possible, and make it easy to calculate and apply, the relative weight of each variable was determined by rounding the β coefficient of each variable in the logistic regression model to the nearest whole number, following the methodology use by the published report on the construction of the well-known   4 cardiovascular score created with the Framingham series [7] .  The intensity of the predictive capacity of each variable was determined by calculating its odds ratio (OR) and the 95% confidence interval (95% CI). Both the diagnostic capacity of the constructed score and its validation were determined by tracing their corresponding AUC-ROC and 95% CI. ROC curve plots the true positives (sensitivity) versus false positives (1-specificity) for the FACED score. The perfect classification or predictive value is achieved when the sensitivity is 1 and 1-specificity (or false positives) is 0 with an AUC-ROC of 1. Any power predictive value of the FACED score in our case would draw a straight line from the lower left corner to the upper right corner), with an AUC-ROC of 0.5 (50%). The greater the AUC-ROC, the better the predictive value of the FACED score. An AUC-ROC greater than 0.8 (or 80%) has been defined as an excellent prognosis or diagnostic power of the classifier (in our case, the FACED score).   The two AUC were compared with the C-statistics test. The final index was eventually divided into 3 groups with a progressively increasing score, and the predictive capacity of mortality of each score group was calculated in both the construction cohort and the validation cohort using a Kaplan-Meier method. A patient’s loss or death during follow-up was considered as a censured data. The score group associated with the least probability of death was named as mild bronchiectasis, that associated with an intermediate probability of death as moderate bronchiectasis, and that associated with the greatest probability of death as severe bronchiectasis. The comparison between the different Kaplan-Meier curves two by two was performed with the log-rank test. Unless otherwise indicated, p<0.05 was considered significant.   5 The score’s predictive power was also calculated for deaths from respiratory causes after 5 years of follow-up from the diagnosis of bronchiectasis. The AUC-ROC were thus calculated in both the validation and construction cohorts and compared using the C-statistics. We also determined whether the two cohorts were significantly different with respect to the three aforementioned groups with different prognoses References 1. Medical Research Council Working Party. Long-term domiciliary oxygen therapy in chronic cor pulmonale complicating chronic bronchitis and emphysema. Lancet 1981; 1 : 681-5 2. Vendrell M, de Gracia J, Olveira C, et al. Diagnóstico y tratamiento de las bronquiectasias. Arch Bronconeumol 2008; 44 : 629-640 3. Chailleux E, Fauroux B, Binet F, Dautzenberg B, Polu JM. Predictors of survival in patients receiving domiciliary oxygen therapy or mechanical ventilation. A 10-year analysis of ANTADIR Observatory. Chest 1996; 109 : 741-749 4. Keistinen T, Säynäjäkangas O, Tuuponen T, Kivelä SL. Bronchiectasis: an orphan disease with a poorly-understood prognosis. Eur Respir J. 1997; 10 :2784-2787. 5. Onen ZP, Gulbay BE, Sen E, et al. Analysis of the factors related to mortality in patients with bronchiectasis. Respir Med. 2007; 101 :1390-7. 6. Loebinger MR, Wells AU, Hansell DM, et al. Mortality in bronchiectasis: a long-term study assessing the factors influencing survival. Eur Respir J 2009; 34 : 843-849
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks
SAVE OUR EARTH

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!

x