Clinical validation of near-patient assessment of fluid status in dialysis patients

Abstract

Dialysis patients have an extraordinarily high level of cardiovascular morbidity and mortality. Volume expansion is common among this cohort and this is linked to markers of cardiac dysfunction, inflammation and low plasma albumin, all of which predict worse survival. Fluid status deteriorates over time due to loss of residual renal function (RRF) and progressive muscle wasting and currently there is a lack of effective or validated clinical tools to detect such changes that can inform clinical decision making given that gold standard methods are not practical for routine use.

This thesis first evaluates the role of albumin and fluid status of peritoneal dialysis (PD) patients and demonstrates that hypoalbuminaemia is an important determinant of tissue overhydration that is largely extravascular given the normal plasma volume. Attempts to ‘dry out’ such patients may precipitate hypovolaemic episodes and loss of RRF. The cardiac sub-analysis suggests that cardiac dysfunction, in particular heart failure with normal ejection fraction (HFNEF) is prevalent and associated with higher intravascular volume.

Next the potential of bioimpedance analysis (BIA) in the form of vector plots to detect the changing body composition in order to improve fluid management was investigated in a multi-centre randomised controlled trial. The clinical impact of this tool was small given the overall stability of fluid status in non-anuric PD patients but deterioration in anuric subject could be ameliorated; in the secondary cardiac analysis, patients in the intervention group seems to experience less damaged hearts. This study underlined the value of RRF in maintaining normal body composition and cardiac function. Lastly, the feasibility of measuring lung water volume using breath analysis of deuterium concentration by flowing afterglow mass spectrometer (FA-MS) following mouth inhalation and intravenous injection of deuterium was explored. The study data vastly overestimated the predicted volume and alternative approaches are needed in the future. (300 words)