Development and validation of a prognostic index (BODEXS90) for mortality in stable chronic obstructive pulmonary disease

Abstract

Introduction Several multidimensional indices have been proposed to predict mortality in chronic obstructive pulmonary disease (COPD). The BODEX index is simple and easy to use for this purpose in all clinical settings. Only a few prognostic indices have integrated oxygenation variables, with measurement methods that are not practical for real life clinical practice in all settings.

Objectives To develop and externally validate a new prognostic index (BODEXS90) that combines the variables included in BODEX index with rest peripheral oxygen saturation measured with finger oximetry (SpO2) to predict all-cause mortality in stable COPD.

Method Observational, non-intervention, multicenter historic cohort study. The BODEXS90 index was developed in a derivation cohort and externally validated in a validation cohort. Calibration of the index was carried out using Hosmer-Lemeshow test. The discrimination capacity of BODEXS90 and BODEX were compared by means of receiver-operating characteristics curves. Modelling of the index was carried out by crude and adjusted Cox regression analysis.

Results The derivation and validation cohorts included 787 and 1179 subjects, respectively. SpO2 predicted all cause-mortality independently of BODEX index. Discrimination capacity of BODEXS90 to predict the outcome was significantly higher than that of BODEX, particularly for more severely affected patients, both in the derivation and in the validation cohorts.

Conclusions The new index is potentially useful for designing clinical decision-making algorithms in stable COPD.