Background: It is the object of debate whether a low or high dialysate sodium concentration (DNa(+)) should be advocated in chronic haemodialysis patients. In this paper, we aimed at evaluating benefits and harms of different DNa(+) prescriptions through a systematic review of the available literature.

Methods: MEDLINE and CENTRAL databases were searched for studies comparing low or high DNa(+) prescriptions. Outcomes of interest were mortality, blood pressure (BP), interdialytic weight gain (IDWG), plasma sodium, hospitalizations, use of anti-hypertensive agents and intradialytic complications.

Results: Twenty-three studies (76 635 subjects) were reviewed. There was high heterogeneity in the number of patients analysed, overall study quality, duration of follow-up, DNa(+) and even in the definition of 'high' or 'low' DNa(+). The only three studies looking at mortality were observational. The risk of death was related to the plasma-DNa(+) gradient, but was also shown to be confounded by indication from the dialysate sodium prescription itself. BP was not markedly affected by high or low DNa(+). Patients treated with higher DNa(+) had overall higher IDWG when compared with those with lower DNa(+). Three studies reported a significant increase in intra-dialytic hypotensive episodes in patients receiving low DNa(+). Data on hospitalizations and use of anti-hypertensive agents were sparse and inconclusive.

Conclusions: There is currently no definite evidence proving the superiority of a low or high uniform DNa(+) on hard or surrogate endpoints in maintenance haemodialysis patients. Future trials adequately powered to evaluate the impact of different DNa(+) on mortality or other patient-centred outcomes are needed.