Prognostic assessment of mortality and hospitalizations of outpatients with advanced chronic obstructive pulmonary disease. Usefulness of the CODEX index.

OBJECTIVES

To validate the CODEX index in outpatients with advanced chronic obstructive pulmonary disease (COPD).

PATIENTS AND METHODS

We studied all patients with COPD treated in a chronic respiratory disease unit. We calculated the BODEX and CODEX indices and their relationship with mortality, hospitalisations or both and performed an analysis by number of events (mortality and/or readmissions), using the Cox proportional hazards analysis.

RESULTS

We included 80 patients (90% men) with a mean age of 73.4 years. The mean follow-up was 656 days, with an interquartile range (25-75%) of 417-642 days. Seventeen patients died (21%) and 57 (71.3%) required hospitalisation for COPD. The CODEX index was significantly related to mortality (P<.008; HR: 1.56; 95% CI: 1.1-2.15), hospitalisations (P<.01; HR: 1.35; 95% CI: 1.13-1.62) and the combined variable (P<.03; HR: 1.27; 95% CI: 1.1-1.5). The BODEX index was not associated with mortality (P=.17) but was associated with hospitalisation (P<.001; HR: 1.4; 95% CI: 1.15-1.73) and the combined variable (P<.03; HR: 1.2; 95% CI: 1.02-1.34). There were 187 events during follow-up. Both the CODEX (P<.001; HR: 1.17; 95% CI: 1.1-1.27) and BODEX (P<.02; HR: 1.12; 95% CI: 1.02-1.23) indices were related to the number of events. However, after adjusting for the interaction between the 2 indices, only the CODEX index maintained statistical significance for the combined variable for patients (P<.03) and in the analysis by number of events (P<.001).

CONCLUSIONS

Both the CODEX and BODEX indices are useful for predicting hospitalisations, although the prognostic ability of the CODEX index is greater than that of the BODEX index, both for mortality and hospitalisations.