Metabolic acidosis is frequent in chronic hemodialysis patients and is associated with adverse clinical outcomes. Two commonly used strategies to correct acidosis are oral sodium bicarbonate supplementation and increasing the bicarbonate concentration of the dialysate, but their comparative effectiveness and tolerance in routine care remain uncertain. This pilot, prospective, randomized, open-label, two-center trial will compare oral sodium bicarbonate versus higher dialysate bicarbonate in chronic hemodialysis patients with metabolic acidosis, using predialysis plasma bicarbonate concentrations, so-called "reserves alcalines" or "alkaline reserves" in local laboratory reports, as a pragmatic marker of acid-base status. Approximately 30 acidotic patients (serum bicarbonate < 22 mmol/L) will be randomized 1:1 to receive either oral sodium bicarbonate or an increase in dialysate bicarbonate for 4 weeks; an additional non-acidotic observational group will provide descriptive reference data. The primary outcome is the change in predialysis serum bicarbonate from baseline (Day 0) to Day 28 between the two randomized arms. Secondary outcomes include the proportion of patients reaching target serum bicarbonate levels, the weekly kinetics of correction, dialysis adequacy (Kt/V and online clearance monitoring), intradialytic tolerance (blood pressure, cramps, hypotension, symptoms), and sodium-related safety (natremia, interdialytic weight gain). Feasibility indicators such as recruitment, retention, adherence to treatment and dialysate adjustment, and data completeness will also be described to inform the design of a larger definitive trial.

Metabolic acidosis is a common complication in chronic hemodialysis patients and is associated with adverse nutritional, cardiovascular, and bone outcomes. In clinical practice, two main strategies are used to correct acidosis: oral sodium bicarbonate supplementation and increasing the bicarbonate concentration of the dialysate. Both approaches are recommended in guidelines, but their comparative effectiveness and tolerance in routine hemodialysis care, particularly in resource-limited settings, remain uncertain.

This pilot, prospective, randomized, open-label, multicenter study will compare these two strategies in adult chronic hemodialysis patients with metabolic acidosis, defined by low predialysis serum (or plasma) bicarbonate concentrations, so-called "reserves alcalines" or "alkaline reserves" in local laboratory reports. Approximately 30 acidotic patients (serum bicarbonate < 22 mmol/L) will be enrolled and randomized in a 1:1 ratio to either oral sodium bicarbonate supplementation (Arm A) or an increase in dialysate bicarbonate concentration (Arm B) for 4 weeks. In addition, a non-acidotic observational control group of hemodialysis patients with stable, adequate serum bicarbonate levels will be followed descriptively to provide reference data on acid-base status, dialysis adequacy, and tolerance.

Randomization among acidotic patients will be performed after matching them in pairs according to the severity of metabolic acidosis (baseline predialysis serum bicarbonate / "reserves alcalines") and age. Patients will be ordered from the lowest to the highest serum bicarbonate value, then matched two-by-two on similar bicarbonate level and age. Within each pair, allocation to oral sodium bicarbonate (Arm A) or increased dialysate bicarbonate (Arm B) will be determined by a computer-generated random number in a spreadsheet, corresponding to a block randomization with block size 2 after matching on acidosis severity. The non-acidotic control group will not be randomized and will receive usual care.

The primary outcome is the change in predialysis serum bicarbonate from baseline (Day 0) to Day 28, comparing the two randomized arms. Secondary outcomes include the proportion of patients achieving target serum bicarbonate at Day 28, the weekly kinetics of bicarbonate correction, changes in serum potassium, and dialysis adequacy assessed by Kt/V and online clearance monitoring. Intradialytic and interdialytic tolerance will be evaluated through blood pressure, interdialytic weight gain, cramps, hypotension, thirst, digestive symptoms, and any treatment discontinuation or dose reduction related to intolerance. Sodium-related safety will be assessed by predialysis natremia and interdialytic weight gain, given the potential impact of both oral sodium bicarbonate and higher dialysate bicarbonate on sodium load.

As a pilot trial, this study also includes predefined feasibility objectives. Feasibility outcomes will describe recruitment and retention rates, adherence to oral treatment and to the dialysate bicarbonate adjustment algorithm, and data completeness for key clinical and laboratory variables. These feasibility indicators will be used to judge the practicality of the protocol and to inform the design and assumptions of a larger, definitive randomized controlled trial.

Inclusion Criteria:

• Age ≥ 18 years.
• Chronic hemodialysis for at least 3 months.
• Stable hemodialysis prescription (no major change in dialysis prescription in the previous weeks, as per local practice).
• Predialysis low "alkaline reserves" (serum bicarbonate concentration) according to the operational threshold used in the unit (e.g., "alkaline reserves" < 22 mmol/L), consistent with KDIGO 2017 definition of metabolic acidosis.
• Ability to give written informed consent.

For the observational group:

• Age ≥ 18 years.
• Chronic hemodialysis for at least 3 months.
• Predialysis "alkaline reserves" considered stable and within the locally defined normal range (≥ 22mmol/L).
• Ability to give written informed consent.

Exclusion Criteria:

• Recent hemodynamic instability (e.g., repeated intradialytic hypotension or unstable blood pressure in the previous weeks).
• Recent hospitalization for an acute condition.
• Acute infection or major intercurrent acute event at the time of screening.
• Major change in dialysis prescription in the 2 weeks prior to inclusion (e.g., change in dialysis schedule, duration, or dialysate composition outside the study protocol).
• Severe digestive disorders limiting oral intake (e.g., persistent vomiting, severe malabsorption, or any condition preventing safe oral Bicardis administration).
• Known hypernatremia or high risk of uncontrolled sodium and fluid overload (as judged by the investigator).
• Severe uncontrolled hypercalcemia or hypocalcemia.
• Any condition that, in the investigator's judgment, would preclude safe participation or interfere with study procedures (e.g., very limited life expectancy, inability to attend scheduled visits).
• Refusal to participate or inability to provide written informed consent.

Individual participant data (IPD) from this pilot trial will not be shared publicly. The study has a small sample size in a limited number of centers, which increases the risk of re-identification of participants despite de-identification procedures. In addition, data were collected in a specific local context without prior consent for open data sharing, and current ethical approvals and institutional policies do not cover broad external data sharing.

Aggregated results and methodological details will be disseminated through scientific presentations and publications, but raw IPD will remain confidential and accessible only to the investigative team and relevant oversight bodies, in accordance with applicable regulations and ethics committee requirements.