Isonatraemic haemodialysis in the management of salt and water overload: a crossover trial at an academic hospital in Dakar, Senegal

,


INtrODUctION
Chronic accumulation of sodium and water contributes to cardiovascular risk in patients with kidney failure who are being treated with maintenance haemodialysis. The volume overload is involved in the pathogenesis of hypertension, left ventricular hypertrophy (LVH) and sudden death [1][2][3]. Several strategies have been successfully implemented to improve the salt and water status, such as sodium restriction [4,5], increasing the frequency and duration of haemodialysis sessions [6,7], volume management using bioimpedance [8,9] and, more recently, individualized prescription of the dialysis sodium concentration (Na + dialysate) [10][11][12].
Isonatraemic haemodialysis has been shown to improve haemodynamic tolerance of the dialysis procedure in patients over 70 years [13]. There are several variations, including isonatraemic, isoconductive and isotonic haemodialysis. Isonatraemic haemodialysis matches the Na + dialysate to the patient's average pre-dialysis serum Na + concentrations, the "set point" (SP) [14]. In one study of 27 patients, isonatraemic haemodialysis reduced interdialytic weight gain (IDWG). There were also improvements in thirst score, blood pressure (BP) control and decreased intradialytic hypotension [10].
To our knowledge, no previous work on isonatraemic haemodialysis or individualized Na + dialysate prescription has been conducted in Senegal. This trial was performed to assess the impact of isonatraemic haemodialysis on the reduction of IDWG and BP. We also aimed to assess its tolerability, focusing on intradialytic hypotension and cramping during haemodialysis sessions.

MEtHODS
A crossover trial was conducted at the Aristide Le Dantec University Hospital in Dakar, Senegal, from 1 April to 15 May 2020, in patients with kidney failure who were stable on treatment with chronic haemodialysis.
The study was performed in two phases: first, a "control phase", during which patients had three weeks of haemodialysis (nine sessions of treatment) with a standard Na + dialysate of 138 mmol/L. During this phase, the serum Na + set point (SP) for each patient was calculated from three predialytic mid-week values, and the intra-individual variability and the coefficient of variation of natraemia were calculated [10]. All events occurring during haemodialysis sessions and those reported by the patient during the interdialytic periods were recorded. Na + measurements used an ion-selective electrode (ARCHITECT ci410 analyser, Abbott Laboratories).
In the second phase, the "individualized phase", patients who had completed the control phase had three weeks of haemodialysis with a Na + dialysate equal to their SP. Haemodialysis parameters and intra-and interdialytic events were again recorded. Intra-individual variability in serum Na + concentrations were calculated by the difference between the minimum and the maximum concentrations for each patient. The coefficient of variation was equal to the ratio of the standard deviation to the mean of the serum Na + concentration for each patient.
Patients were included if they were over 18 years of age, stable on chronic haemodialysis for more than three months, receiving three treatment sessions per week and had signed written informed consent. Diabetic patients, anyone who had a serious cardiovascular event in the preceding three months, and those using a temporary haemodialysis catheter or who had vascular access dysfunction were not included.
Data were captured using Microsoft Excel and analysed using R software version 4.03. Numerical variables were summarised using mean and standard deviation and categorical variables were summarised using counts and proportions. The control and individualized phases were compared using paired z tests and McNemar's tests. A P value of <0.05 was considered statistically significant.
The study was approved by the Ethics Committee of the Department of Medicine of Cheikh Anta Diop University of Dakar (study registration number 033/2020/CER/ UCAD). All patients provided written, informed consent. The work was not supported by any external funding.

rESULtS
Of 100 patients on chronic haemodialysis at our centre, 60 were included in the study. Only 32 had complete measurements and could be included in the analysis ( Figure  1). Their mean age was 55.5 ± 12.1 years, with a male/ female ratio of 1.3. Anaemia was noted in 22 patients (69%); 13 patients (41%) had haemoglobin concentrations between 9 g/dL and 11 g/dL and iron deficiency was noted in 7 patients (22%). Secondary hyperparathyroidism was present in 21 patients and vitamin D deficiency in 11 patients. Cardiac ultrasound performed on 20 patients (63%) reported left ventricular hypertrophy in 18. The mean left ventricular ejection fraction was 65.3 ± 9.6%. Twenty-one patients (66%) were receiving antihypertensive treatment ( Table 1).
The mean serum Na + concentration was 135.8 ± 1.9 mmol/L, with a mean coefficient of variation of 2% and mean intra-individual variability of 4.6 ± 2.9 mmol/L. Sixty percent of predialytic measurements were between The mean ultrafiltration (UF) volume was slightly higher in the control phase, but this was not statistically significant (1.9 L vs 1.8 L; P = 0.70). The mean IDWG was 1.9 kg vs 1.8 kg in the control and individualized phases, respectively (P = 0.75). A significant reduction in postdialytic systolic BP was observed during the individualized phase (148 mmHg vs 135 mmHg; P = 0.04). This trend was also observed with the pre-and intradialytic BP without reaching statistical significance (Table 2).
Qualitative parameters documented during the two phases are summarised in Table 3. Headaches were more frequent during the individualized phase (8 vs 5 episodes; P = 0.04). There was no difference between the two phases on intraand postdialytic symptoms (cramps, hypotension and intradialytic hypertension).

DIScUSSION
Our study has revealed that isonatraemic haemodialysis had no significant effect on IDWG, pre-dialysis BP and intradialytic BP. Post-dialysis systolic BP was significantly reduced. Our results are similar to those of Thein et al. [15] but differ from other studies. Aramreddy [16] reported a significant reduction of IDWG after a decrease in Na + dialysate by 2 mmol/L, and Elshahawy et al. [17] found that the individualization of Na + dialysate was associated with a decrease of IDWG and better BP control. This effect was also reported in the DISO trial [18].
Several studies have reported reductions of predialytic systolic BP and non-significant reductions in postdialytic isonatraemic haemodialysis in the management of salt and water overload.  Abbreviations: ACEi, angiotensin conversion enzyme inhibitors; ARBs, angiotensin receptor blockers; PPIs, proton pump inhibitors; NSAIDs, non-steroidal anti-inflammatory drugs; IDWG = interdialytic weight gain; Other, sickle cell anaemia, autoimmune haemolytic anaemia and peptic ulcer disease.
systolic BP [10,17]. The reduction in BP has been reported by other studies where Na + dialysate was reduced without being isonatraemic [11,15]. The improvement in BP is linked to the reduction of IDWG. However, we did not observe a significant decrease in IDWG. This suggests that the beneficial effect on BP of isonatraemic haemodialysis may not be related solely to blood volume control.
In our study, the mean predialytic serum Na + concentration was 135.8 mmol/L. This is similar to that reported by De  Apart from the headaches that were more frequent in the individualized phase, isonatraemic haemodialysis was well tolerated. This finding is corroborated by several other studies [16,18]. In an older study, De Paula et al. noted better tolerance with a significant reduction in intradialytic hypotension [10]. This effect of isonatraemic haemodialysis might be related to better volume control with a reduction in UF volumes.
Equipping our haemodialysis units with machines able to perform online measurement of Na + concentration or conductivity would seem to be the best way to demonstrate the clinical benefits of isonatraemic haemodialysis in its isoconductive form. Isonatraemic haemodialysis must be integrated into the therapeutic arsenal, which would also include patient education on salt and water intake.
The relatively small sample population and the short duration were limitations of our study. The benefits of isonatraemic haemodialysis can be delayed, occurring months after starting the intervention.

cONcLUSIONS
IDWG as well as pre-and intradialytic BP were unaffected by isonatraemic haemodialysis, whereas postdialytic BP was significantly reduced. This is a well-tolerated modality.
Upgrading haemodialysis machines to measure conductivity in real time would provide the optimal conditions for performing isonatric haemodialysis.