Every week, thousands of people on haemodialysis have a heart attack, stroke, or die—and the concentration of sodium in their dialysis fluid may be partly to blame. The problem is a hidden trade-off. Dialysis removes excess sodium from the blood, but the standard sodium level in the fluid varies between units, from low to high. A lower sodium setting pulls out more salt, reducing fluid overload and blood pressure—but it can also destabilise the circulation during treatment, causing cramps and drops in blood pressure. No large trial has ever tested which default setting actually saves more lives or prevents more cardiovascular events. This trial will settle that question. It randomly assigns dialysis units across the UK to use either a lower or a higher default sodium concentration, then tracks every patient’s outcomes using national health records. Because changing a default setting costs nothing and requires no new equipment, a clear result could be adopted immediately in every dialysis unit in the country—and globally. For patients, that could mean fewer hospitalisations and longer survival, without any change to their daily routine. For the health service, it could reduce the burden of a treatment that is both costly and gruelling.
View original technical description
Research question Does a lower unit default dialysate sodium of 137mmol/L, compared to 140mmol/L, result in fewer cardiovascular events or improve survival in haemodialysis recipients? Background People with chronic kidney failure receiving haemodialysis suffer high rates of death and cardiovascular events. Sodium retention, hypertension and fluid overload are important mediators of morbidity and mortality in this population. Most haemodialysis units use a default (or standard) dialysate sodium concentration ranging from 135 to 142mmol/L. A lower dialysate sodium concentration improves sodium removal, but at the cost of increased intradialytic haemodynamic instability. Few randomized studies have investigated this question, and none were powered for clinically relevant outcomes such as cardiovascular events, hospitalization or mortality. Aims and objectives To compare the risk of a composite of acute myocardial infarction (AMI), stroke, or all-cause mortality in adult maintenance haemodialysis patients dialyzing at units with a lower default dialysate sodium of 137mmol/L, compared to those with a higher default dialysate sodium of 140mmol/L, in a pragmatic cluster randomized trial. Methods One hundred dialysis units, managed by 35 trial sites, in the UK will be randomized to use a default dialysate sodium of 137mmol/L or to use 140mmol/L. Follow up and outcomes on all patients dialyzing at these units, or joining during the study period, will be ascertained from administrative (NHS digital) and registry (UK Renal Registry) data. The study will form part of a global cluster randomized study, enrolling a total of 414 units (globally), and an estimated 50,000 participants. The study duration will be defined by reaching 26,910 primary events, providing 90% power to detect a 10% difference in risk of the primary outcome. Patients at each unit will be offered the opportunity to opt-out of use of their data for this study. All other care will follow usual unit policies and procedures, and clinicians responsible for the participating patients will be able to vary individual participant dialysate sodium if a clinical indication arises. Timelines for delivery Ethical approval and randomisation of 80% of units by end 2022, remainder of units randomised by end 2023, follow up complete end 2024, publication of primary analysis in 2025, UK-specific substudies to follow in 2025-6. Anticipated impact and dissemination As a change in default dialysate sodium is simple to implement and essentially free, this study will have immediate impact. The impact will be global and local. Furthermore, the health economic and secondary analyses of healthcare utilization among UK dialysis patients will provide a unique insight into this burdensome and costly treatment.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know