Rendimientos funcionales asociados con resultados clínicos adversos en pacientes con EPOC y su integración en el índice pronóstico BODE

Resum

Background: multiple studies identified and replicated that the 6MWT test is associated with adverse effects of COPD, and that is why it has been the preferred test to be included in prognostic indexes such as BODE. Other functional tests which test physical performance showed some association with mortality, and a poor association to severe exacerbation, but there have hardly been any studies that have replicated or compared with the 6MM association. Objectives: 1) To identify the association of functional tests in COPD patients with suffering a severe exacerbation in the following year and to compare their predictive ability with 6MWT; 2) To identify the association of functional tests with a higher risk of mortality within the following 5 years and compare their predictive ability with 6MWT; 3) To identify the score of previous test/s more accurately to discriminate those that will die and compare their discriminative ability with its counterpart in 6MM; 4) To determine whether a change in the BODE index is associated with risk of mortality during the following 5 years and to assess its discriminative ability. Methods: a prospective longitudinal study (2014-2019) was conducted with a sample of 137 subjects. The results measures were severe exacerbations of COPD during the following year and mortality for any cause during the following 5 years. Various functional performance tests were collected: two muscle strength tests, SPPB (consisting of a balance test, 4MGS and 5-STS) and 6MWT. For objective 1, multiple logistic regression methods were used, the probabilities of having the event were computed, and with them a ROC curve and its AUC were established; for objective 2, multiple Cox regression models were used, as well as ROC curves and AUC; for objective 3, the point with the highest discriminative accuracy was identified, and their positive and negative likelihood statistics were calculated and compared; and finally, for objective 4, a Cox multivariate regression model, survival curves and ROC curves were used. Results: 24 patients (18.9%) suffered a severe exacerbation in the following year and 37 (27.0%) passed away in the following 5 years. Regarding severe exacerbations, 5-STS was independently associated [ESTAS SIGLAS NO SÉ SI SON ESPAÑOL O INGLÉS](OR: 3.84; IC 95%: 1.14-12.94; p=0.030). In addition, the probabilities of having the event were found to have good predictive ability to discriminate (AUC: 0.793; IC 95%: 0.704-0.882), and very similar to 6MM (AUC: 0.722; IC 95%: 0.63-0.81). The score of 16 seconds in 5-STS was identified as having the highest discriminative accuracy (Se=0.703; Es=0.810). Their likelihood ratios were notably better than those calculated for the 348m in the 6MM, especially the negative one (7.69 and 2.00, respectively). Conclusions: the baseline score in 5-STS is independently associated with suffering a severe exacerbation in the following and with mortality risk during the following 5 years, having a good predictive ability (and similar to 6MWT) to discriminate subjects that may have these events in both cases. Also, performing 5-STS<16s allows to discriminate the subjects that will not die during the following 5 years and provides a similar discriminative ability to to that of the BODE index