BACKGROUND Dialysate [Na+] is often overlooked as a contributor to hypertension in patients on haemodialysis (HD). We report observational experience with a facility level decrease in dialysate [Na+] from 141 mmol/l to 138 mmol/l, in the absence of concurrent change with respect to dietary sodium regulation. METHODS The sample comprised all patients (n=52) dialysing at a single HD facility over an 8-month period flanking the change in dialysate [Na+]. Outcomes included repeated observations of blood pressure (BP), interdialytic weight gain (IDWG), pre-dialysis plasma [Na+] and adverse events. Predictors other than dialysate [Na+] included patient demographics, clinical characteristics and number of antihypertensive medications. The study used a longitudinal unbalanced panel design, and hierarchical linear and Poisson mixed models. RESULTS In multivariate analyses, the change in dialysate [Na+] was associated with a statistically significant small to medium-sized decrease in pre- and post-dialysis systolic and diastolic BP, pre-dialysis plasma [Na+], but not IDWG. Change was greatest in the patient tertile with the highest initial BP. There was no change in the frequency of adverse events. Modelling dialysate [Na+] exposure as the diffusion gradient from dialysate to blood water did not improve the strength of associations. CONCLUSIONS A facility level decrease in dialysate [Na+] from 141 mmol/l to 138 mmol/l appears to be safe and well tolerated, and a useful means of improving BP control. The lack of change in IDWG probably reflects lack of dietary salt restriction, and but does raise the issue of volume-independent effects of sodium exposure on BP.