Predicting mortality in haemodialysis patients: a comparison between lung ultrasonography, bioimpedance data and echocardiography parameters

Abstract

BACKGROUND: The use of lung ultrasonography to evaluate extravascular lung water and its consequences has received growing attention in different clinical areas, including, in recent years, end-stage renal disease patients treated by haemodialysis (HD). Lung congestion is a direct consequence of either overall overhydration and/or cardiac dysfunction, but the exact contribution of each of these tests to mortality is unknown. METHODS: In this prospective observational study, we enrolled 96 patients from a single HD unit undergoing thrice weekly HD. We used three different methods of evaluation: lung ultrasonography (pre- and post-dialysis), bioimpedance spectroscopy (pre- and post-dialysis) and echocardiography (pre-dialysis). The objective of the study was to test for the first time the prognostic value of ultrasound lung comets (ULC) combined with bioimpedance-derived data [total body water (TBW), extravascular water, hydration status-ΔHS] and several echocardiographic parameters. Mortality was analysed after a median of 405.5-day follow-up. RESULTS: Pre-dialysis lung congestion was classified as moderate (ULC = 16-30) in 19.8% of the patients and severe in 12.5% of patients (ULC > 30), while only 19.8% appear to be hyperhydrated (ΔHS > 15%). The pre-dialysis ultrasound lung congestion score correlated significantly with all of the bioimpedance-derived parameters. In a multivariate Cox model that included ULC score, demographic, ecocardiographic and bioimpedance parameters, the factors that remained significantly associated with survival time were the pre-dialysis ULC score and left ventricular mass index. The pre-HD ULC score has a significant discriminating power for survival, while the bioimpedance-derived hydration status has no discriminatory abilities in terms of survival. CONCLUSIONS: To our knowledge, this study is the first one that compares three different strategies to predict mortality in haemodialysed patients. The lung comet score emerged as the best predictor for the relationship hydration status-mortality, independently of bioimpedance-derived parameters in this population.