Sodium Salts

What it is

This entry covers the inorganic sodium salts that are used as medicines, kept deliberately separate from dietary salt. Four families are in scope: sodium chloride, given intravenously as 0.9% saline or as hypertonic solutions and by mouth in oral rehydration salts; sodium bicarbonate, given intravenously as an alkalinising agent and by mouth as an antacid; sodium phosphate (monobasic plus dibasic), used by mouth as a saline bowel purgative before colonoscopy; and sodium citrate with citric acid, an oral systemic alkaliniser. Each is a different licensed product with its own trials, its own label and its own warnings, so the evidence below is kept separate salt by salt and never pooled.

What the research says

What the evidence shows depends entirely on which salt and which use. Reduced-osmolarity oral rehydration salts have the strongest evidence, from a Cochrane review of randomised trials in children with diarrhoea. Intravenous 0.9% saline reliably replaces volume and sodium, but whether it is better or worse than a balanced crystalloid is genuinely unsettled: one large trial favoured balanced fluid and three others found no difference. Sodium bicarbonate did not change the primary outcome in the one randomised trial in severe metabolic acidaemia, and a 2026 systematic review found no benefit in out-of-hospital cardiac arrest. Oral sodium phosphate cleans the colon effectively but carries a boxed warning for acute phosphate nephropathy. Sodium citrate's uses rest on the label, not on outcome trials that we could find.

Evidence grade: Disputed.

How it works

Drug class: Inorganic sodium salts used as medicines: crystalloid volume and electrolyte replacement agents (sodium chloride), systemic alkalinising agents (sodium bicarbonate, sodium citrate) and a saline osmotic bowel purgative (sodium phosphate)

Sodium is the main positively charged particle in the fluid outside the body's cells, and water follows it. Giving a sodium salt therefore expands that fluid compartment, which is why saline is used to replace lost volume. The anion decides the rest: chloride loads the body with acid-forming anion, bicarbonate and citrate are converted to buffer and raise blood and urine pH, and phosphate is poorly absorbed from the gut so it holds water inside the bowel and flushes it out. (Source 1)

Boxed warning

There have been rare, but serious reports of acute phosphate nephropathy in patients who received oral sodium phosphate products for colon cleansing prior to colonoscopy. Some cases have resulted in permanent impairment of renal function and some patients required long-term dialysis. While some cases have occurred in patients without identifiable risk factors, patients at increased risk of acute phosphate nephropathy may include those with increased age, hypovolemia, increased bowel transit time (such as bowel obstruction), active colitis, or baseline kidney disease, and those using medicines that affect renal perfusion or function (such as diuretics, angiotensin converting enzyme [ACE] inhibitors, angiotensin receptor blockers [ARBs], and possibly nonsteriodal anti-inflammatory drugs [NSAIDs]). See WARNINGS . It is important to use the dose and dosing regimen as recommended (pm/am split dose). See DOSAGE and ADMINISTRATION .

(Source 2)

What it is used for

• Saline does replace volume and sodium, and the label frames it that way. Whether it is the right crystalloid is disputed: the SMART trial found slightly fewer major adverse kidney events with balanced fluid, while SPLIT, BaSICS and PLUS found no difference in kidney injury or death. Evidence: disputed. (Source 3)
• In the COBI randomised trial, continuous 20% hypertonic saline did not improve neurological outcome at six months. The trial authors themselves said the confidence intervals were wide and the study may have been underpowered, so this is a null result rather than proof of no effect. Evidence: not-supported. (Source 4)
• The one randomised trial, BICAR-ICU, found no effect on its primary composite outcome of death by day 28 plus organ failure at day 7. A pre-specified subgroup with acute kidney injury did better, which is a signal to test, not a result to rely on. Evidence: limited. (Source 5)
• A 2026 systematic review and meta-analysis of 11 studies in out-of-hospital cardiac arrest found no significant difference in return of spontaneous circulation or survival to discharge. The product label for the dilute injection explicitly tells prescribers not to use it in cardiac arrest, for a practical reason about concentration rather than about outcomes. Evidence: not-supported. (Source 6)
• This is a regulatory position recorded on the label rather than an outcome-trial finding. We found no randomised outcome trial of urinary alkalinisation with sodium bicarbonate in poisoning in this search. Evidence: unknown. (Source 7)
• In the pivotal phase 3 trial the tablets were non-inferior to the earlier sodium phosphate tablet for colon cleansing, with overall response rates of 95% to 97%. The trial was investigator-blinded and had no placebo or polyethylene glycol comparator, and the product carries a boxed warning for acute phosphate nephropathy. Evidence: limited. (Source 8)
• The approved use rests on the label's own statement of value, with a 2026 label date. We found no randomised outcome trial of this preparation in this search, so the strength of the claim is not established from the literature. Evidence: unknown. (Source 9)
• A Cochrane review found that reduced-osmolarity oral rehydration solution, which contains less sodium than the older WHO formula, led to fewer unscheduled intravenous infusions, lower stool output and less vomiting, with no detected excess of low blood sodium. This is the strongest evidence in the group, and it is evidence for less sodium rather than more. Evidence: established. (Source 10)

Interactions

• ACE inhibitors, angiotensin receptor blockers and diuretics (with oral sodium phosphate bowel preparation) (case reports): Medicines that affect blood flow through the kidney or alter electrolytes raise the risk of acute phosphate nephropathy from an oral sodium phosphate bowel preparation. In the biopsy series, 14 of the 16 affected patients who had high blood pressure were taking an ACE inhibitor or an angiotensin receptor blocker. (Source 2)
• Aluminium-based antacids (with sodium citrate and citric acid oral solution) (label): The label states plainly that this product should not be given at the same time as an aluminium-based antacid. (Source 11)
• Corticosteroids and corticotropin (with intravenous sodium chloride) (label): Steroids make the body hold on to sodium and water, so the label asks for caution when intravenous sodium chloride is given to someone taking them. No rate or effect size is given. (Source 12)
• Diuretics that cause a low-chloride alkalosis (with sodium bicarbonate injection) (label): Sodium bicarbonate injection is contraindicated in people on such diuretics and in people losing chloride through vomiting or gastric suction, because adding bicarbonate deepens the alkalosis. (Source 13)
• Potassium-enriched salt substitutes (clinical trial): Swapping sodium chloride for a potassium-containing substitute is the one sodium intervention tested for hard outcomes in a large randomised trial. Clinical high potassium was the pre-specified safety outcome and the rate was not significantly higher than with regular salt, but the trial excluded nobody on the basis of kidney function in a way that would let this be generalised to people with advanced kidney disease. (Source 14)
• Food and timing of doses (sodium citrate and citric acid oral solution) (label): The label asks for the solution to be diluted in water and taken after meals, specifically to avoid its saline laxative effect. (Source 11)
• Clear liquids and hydration around an oral sodium phosphate bowel preparation (case reports): The preparation works by pulling water into the bowel, so inadequate fluid intake is itself a risk factor for kidney injury. The label ties hydration before, during and after the preparation to the fatalities it reports, and the biopsy series named inadequate hydration as a potential cause. (Source 15)
• Alcohol (and benzodiazepine) withdrawal during an oral sodium phosphate bowel preparation (label): The only documented alcohol-related interaction we found in this group is in the oral sodium phosphate label: people withdrawing from alcohol are listed among those at higher risk of the seizures that have been reported with sodium phosphate products. No rate is given, and we found no study of alcohol with sodium chloride, sodium bicarbonate or sodium citrate. (Source 16)

Stopping it

• None of these products has a described withdrawal syndrome, a taper or a rebound effect in the literature we read. They are given for a defined clinical purpose and stopped when that purpose ends; the sodium chloride injection label frames the question as monitoring rather than tapering, tying continued use to repeated laboratory checks of fluid, electrolyte and acid-base balance. (Source 1)
• For dietary sodium, which is the question readers usually mean, the Cochrane review's own conclusion is that reducing intake to recommended levels moves mean arterial pressure by about 0.4 mmHg in people with normal blood pressure and about 4 mmHg in people with hypertension, while hormone and lipid changes were more consistent than the blood pressure change. That is a statement about reducing intake, not about stopping a medicine. (Source 17)

What goes wrong

Oral sodium phosphate bowel purgatives have caused acute phosphate nephropathy with permanent loss of kidney function. (Source 18)

• Case series, Low certainty.
• Size: 21 patients identified among 7,349 native renal biopsies at one centre over 2000 to 2004.
• Who: people who had had a colonoscopy prepared with oral sodium phosphate solution or Visicol, with normal blood calcium.
• How long: mean follow-up 16.7 months.
• Result: mean baseline serum creatinine 1.0 mg/dl rising to a mean of 3.9 mg/dl at a median of 1 month after colonoscopy; 4 of 21 went on to permanent haemodialysis and the remaining 17 all developed chronic renal insufficiency, mean serum creatinine 2.4 mg/dl.
• Funding: not stated in the abstract.

Limit of this finding: This is a biopsy case series from a single referral centre with no denominator of exposed people, so it cannot give an incidence rate and there is no placebo comparison. The quotation uses "APhN" without ever expanding it; it is the source's own abbreviation for acute phosphate nephropathy. The denominator, 7,349, is renal biopsies at one centre and not a count of people exposed to oral sodium phosphate, so no rate of any kind can be read off these numbers.

Patients presented with ARF and a mean creatinine of 3.9 mg/dl at a median of 1 mo after colonoscopy. In a few patients, ARF was discovered within 3 d of colonoscopy, at which time hyperphosphatemia was documented. Patients had minimal proteinuria, normocalcemia, and bland urinary sediment. At follow-up (mean 16.7 mo), four patients had gone on to require permanent hemodialysis. The remaining 17 patients all have developed chronic renal insufficiency (mean serum creatinine, 2.4 mg/dl).

Oral sodium phosphate colon cleansing has been followed by deaths from fluid shifts, severe electrolyte abnormalities and cardiac arrhythmias. (Source 15)

• Official position, Certainty not rated.
• Size: not quantified in the label.
• Who: people taking oral sodium phosphate products before colonoscopy, particularly those with kidney impairment, bowel perforation, or who overdosed.
• How long: typically within days of the preparation.
• Result: no rate given; the label lists fatalities, renal failure, acute phosphate nephropathy, nephrocalcinosis, seizures and arrhythmias without frequencies.
• Funding: regulator-approved product label (a position, dated 2010)

Administration of sodium phosphate products prior to colonoscopy for colon cleansing has resulted in fatalities due to significant fluid shifts, severe electrolyte abnormalities, and cardiac arrhythmias. These fatalities have been observed in patients with renal insufficiency, in patients with bowel perforation, and in patients who misused or overdosed sodium phosphate products.

In the pivotal colon cleansing trial, the most common adverse events with oral sodium phosphate tablets were bloating, nausea, abdominal pain and vomiting, and they were common. (Source 19)

• Randomized trial, Moderate certainty.
• Size: 272, 265 and 268 patients in the three tabulated arms (805 in total); the label introduces the table as covering the colon preparation trials "(n=931)"
• Who: adults scheduled for elective colonoscopy.
• How long: single bowel preparation.
• Result: bloating 31% at 48 g and 39% at 60 g, nausea 26% and 37%, abdominal pain 23% and 24%, vomiting 4% and 10%; the comparator Visicol 60 g gave 41%, 30%, 25% and 9%.
• Funding: manufacturer trials reported in the regulator-approved label.

Limit of this finding: There was no placebo arm, and the label states that diarrhoea was treated as part of the intended effect and so was not counted as an adverse event, which means the table understates total gastrointestinal upset. The label says the table covers 931 patients but its own three column denominators (272, 265 and 268) come to 805. Elsewhere the same label uses 931 for the whole controlled colon-preparation programme, so 931 is not the denominator of these percentages and a reader should not treat it as one.

Table 2 shows the most common adverse events associated with the use of 48 grams of OsmoPrep, 60 grams of OsmoPrep, and 60 grams of Visicol in the colon preparation trials (n=931). Table 2: Frequency of Adverse Events of Any Severity Occurring in Greater Than 3% of Patients in the OsmoPrep Trials OsmoPrep 32 tabs (48 g) N=272 OsmoPrep 40 tabs (60 g) N=265 Visicol 40 tabs (60 g) N=268 Bloating 31% 39% 41% Nausea 26% 37% 30% Abdominal Pain 23% 24% 25% Vomiting 4% 10% 9%

Intravenous sodium chloride can cause fluid or solute overload, dilution of serum electrolytes, congestion and pulmonary oedema. (Source 12)

• Official position, Certainty not rated.
• Size: not quantified in the label.
• Who: people receiving intravenous sodium chloride, especially with heart failure or severe kidney impairment.
• How long: during infusion.
• Result: no rates given in the label.
• Funding: regulator-approved product label (a position, dated 2026)

The intravenous administration of Sodium Chloride Injection, USP can cause fluid and/or solute overloading resulting in dilution of serum electrolyte concentrations, overhydration, congested states or pulmonary edema.

A two-litre infusion of 0.9% saline in healthy volunteers produced sustained high blood chloride and measurably reduced kidney blood flow and cortical perfusion, which a balanced solution did not. (Source 20)

• Randomized trial, Low certainty.
• Size: 12 healthy adult male subjects, randomised double-blind crossover.
• Who: healthy adult men.
• How long: 2-litre infusion over 1 hour, imaging for 90 minutes, sampling for 4 hours.
• Result: sustained hyperchloraemia with saline but not Plasma-Lyte 148 (P < 0.0001); significant reduction in mean renal artery flow velocity (P = 0.045) and renal cortical tissue perfusion (P = 0.008) from baseline after saline but not after Plasma-Lyte 148; no difference in urinary NGAL (P = 0.917)
• Funding: not stated in the abstract.

Limit of this finding: Twelve healthy young men is a very small, physiological study with surrogate imaging outcomes and no clinical endpoints; it explains a mechanism and cannot show patient harm. An erratum to this paper exists (Ann Surg 2013;258(6):1118) and we could not obtain its content, so a reviewer should confirm it does not change these figures. The final sentence's "first human study" claim is a physiological finding in 12 healthy young men over 90 minutes of imaging and is not evidence that saline harms the kidneys of patients.

Sustained hyperchloremia was seen with saline but not with Plasma-Lyte 148 (P < 0.0001), and fall in strong ion difference was greater with the former (P = 0.025). Blood volume changes were identical (P = 0.867), but there was greater expansion of the extravascular fluid volume after saline (P = 0.029). There was a significant reduction in mean renal artery flow velocity (P = 0.045) and renal cortical tissue perfusion (P = 0.008) from baseline after saline, but not after Plasma-Lyte 148.

Overly aggressive sodium bicarbonate therapy can cause metabolic alkalosis and high blood sodium, and the salt load can cause sodium retention and volume overload. (Source 21)

• Official position, Certainty not rated.
• Size: not quantified in the label.
• Who: people receiving sodium bicarbonate injection, including children and people sensitive to a sodium load.
• How long: during administration.
• Result: no rates given in the label.
• Funding: regulator-approved product label (a position, dated 2026)

Overly aggressive therapy with sodium bicarbonate injection can result in metabolic alkalosis (associated with muscular twitchings, irritability and tetany) and hypernatremia. Monitor electrolytes periodically during administration and treat abnormalities appropriately.

What the evidence supports

Reduced-osmolarity oral rehydration solution, which contains less sodium than the older WHO formula, reduced the need for unscheduled intravenous fluid in children with acute diarrhoea. (Source 10)

• Systematic review, Moderate certainty.
• Size: 11 trials reported the primary outcome; 8 trials were meta-analysed.
• Who: children admitted to hospital with dehydration from acute diarrhoea.
• How long: acute illness episodes.
• Result: Mantel Haenzel odds ratio 0.59, 95% confidence interval 0.45 to 0.79, with no evidence of heterogeneity; the review reports no absolute risk reduction or number needed to treat.
• Funding: not stated.

Limit of this finding: The review gives only an odds ratio, so the absolute benefit and number needed to treat cannot be read off it. Three of the eleven trials had no intravenous infusions in either arm.

In a meta-analysis of 8 trials, reduced osmolarity ORS was associated with fewer unscheduled intravenous fluid infusions compared with WHO standard ORS (Mantel Haenzel odds ratio 0.59, 95% confidence interval 0.45 to 0.79) with no evidence for heterogeneity between trials.

In the SMART trial, balanced crystalloids rather than 0.9% saline were followed by a small absolute reduction in major adverse kidney events. (Source 3)

• Randomized trial, Moderate certainty.
• Size: 15,802 adults (7,942 balanced crystalloid, 7,860 saline)
• Who: adults admitted to five intensive care units at one US academic centre.
• How long: outcomes censored at hospital discharge or 30 days.
• Result: major adverse kidney event in 1,139 of 7,942 (14.3%) with balanced crystalloids versus 1,211 of 7,860 (15.4%) with saline, an absolute difference of 1.1 percentage points (number needed to treat about 91); marginal odds ratio 0.91, 95% CI 0.84 to 0.99, P=0.04; in-hospital mortality at 30 days 10.3% versus 11.1%, P=0.06.
• Funding: Funded by the Vanderbilt Institute for Clinical and Translational Research and others, as stated in the abstract.

Limit of this finding: This was a single-centre, unblinded, cluster-randomised multiple-crossover trial, and the two odds ratios reported for the same comparison (marginal 0.91 and conditional 0.90) come from different models; a reader should not treat the composite outcome as a mortality benefit, because mortality alone was not significant.

Among the 7942 patients in the balanced-crystalloids group, 1139 (14.3%) had a major adverse kidney event, as compared with 1211 of 7860 patients (15.4%) in the saline group (marginal odds ratio, 0.91; 95% confidence interval [CI], 0.84 to 0.99; conditional odds ratio, 0.90; 95% CI, 0.82 to 0.99; P=0.04).

Replacing part of dietary salt with a potassium-containing substitute reduced stroke, major cardiovascular events and death in a large randomised trial, which is evidence against high sodium intake rather than for any medicinal sodium salt. (Source 14)

• Randomized trial, Moderate certainty.
• Size: 20,995 people in 600 villages.
• Who: rural Chinese adults with a history of stroke, or aged 60 or over with high blood pressure.
• How long: mean follow-up 4.74 years.
• Result: stroke 29.14 versus 33.65 events per 1000 person-years, rate ratio 0.86, 95% CI 0.77 to 0.96, P = 0.006; major cardiovascular events 49.09 versus 56.29 per 1000 person-years, rate ratio 0.87, 95% CI 0.80 to 0.94; death 39.28 versus 44.61 per 1000 person-years, rate ratio 0.88, 95% CI 0.82 to 0.95; serious adverse events attributed to hyperkalaemia 3.35 versus 3.30 per 1000 person-years, rate ratio 1.04, 95% CI 0.80 to 1.37, P = 0.76.
• Funding: National Health and Medical Research Council of Australia, as stated in the abstract.

Limit of this finding: The trial was open-label and the substitute both lowered sodium and added potassium, so it cannot separate the two. It was conducted in a population with unusually high baseline salt intake and high stroke risk, so the absolute benefit will not transfer directly.

The rate of stroke was lower with the salt substitute than with regular salt (29.14 events vs. 33.65 events per 1000 person-years; rate ratio, 0.86; 95% confidence interval [CI], 0.77 to 0.96; P = 0.006), as were the rates of major cardiovascular events (49.09 events vs. 56.29 events per 1000 person-years; rate ratio, 0.87; 95% CI, 0.80 to 0.94; P<0.001) and death (39.28 events vs. 44.61 events per 1000 person-years; rate ratio, 0.88; 95% CI, 0.82 to 0.95; P<0.001).

What the evidence does not support

Three further large randomised trials found no difference between a balanced crystalloid and 0.9% saline for death or kidney injury. (Source 22)

• Randomized trial, High certainty.
• Size: 2,278 patients (SPLIT), 11,052 randomised (BaSICS) and 5,037 patients (PLUS)
• Who: adults in intensive care in New Zealand, Brazil, and Australia and New Zealand.
• How long: 28 to 90 days.
• Result: SPLIT: acute kidney injury 9.6% versus 9.2%, absolute difference 0.4 percentage points, 95% CI -2.1% to 2.9%, P=.77. BaSICS: death by day 90 in 1,381 of 5,230 (26.4%) versus 1,439 of 5,290 (27.2%), adjusted hazard ratio 0.97, 95% CI 0.90-1.05, P=.47. PLUS: death within 90 days 21.8% versus 22.0%, difference -0.15 percentage points, 95% CI -3.60 to 3.30, P=0.90.
• Funding: PLUS funded by the National Health and Medical Research Council of Australia and the Health Research Council of New Zealand, as stated in the abstract; funding not stated in the SPLIT and BaSICS abstracts.

Limit of this finding: Two roundings in the PLUS abstract do not reconcile on their face: 21.8% against 22.0% is a 0.2-point gap while the stated difference is -0.15 points, and the same 0.41 mg/dL creatinine rise is converted to 36.6 micromol/L in one arm and 36.1 in the other. Both follow from computing on unrounded data and neither changes the result, which is no difference.

Death within 90 days after randomization occurred in 530 of 2433 patients (21.8%) in the BMES group and in 530 of 2413 patients (22.0%) in the saline group, for a difference of -0.15 percentage points (95% confidence interval [CI], -3.60 to 3.30; P = 0.90).

BaSICS found no mortality benefit from a balanced solution and its authors said the findings do not support using it. (Source 23)

• Randomized trial, High certainty.
• Size: 11,052 randomised, 10,520 (95.2%) analysed.
• Who: critically ill adults at 75 intensive care units in Brazil with at least one risk factor for worse outcomes.
• How long: 90 days.
• Result: 1,381 of 5,230 (26.4%) died with a balanced solution versus 1,439 of 5,290 (27.2%) with saline; adjusted hazard ratio 0.97, 95% CI 0.90-1.05, P=.47; absolute difference about 0.8 percentage points, confidence interval crossing no effect.
• Funding: not stated in the abstract.

By day 90, 1381 of 5230 patients (26.4%) assigned to a balanced solution died vs 1439 of 5290 patients (27.2%) assigned to saline solution (adjusted hazard ratio, 0.97 [95% CI, 0.90-1.05]; P = .47). There were no unexpected treatment-related severe adverse events in either group. CONCLUSION AND RELEVANCE: Among critically ill patients requiring fluid challenges, use of a balanced solution compared with 0.9% saline solution did not significantly reduce 90-day mortality. The findings do not support the use of this balanced solution.

Continuous 20% hypertonic saline did not improve six-month neurological outcome after traumatic brain injury. (Source 4)

• Randomized trial, Moderate certainty.
• Size: 370 patients randomised, 359 (97%) completed.
• Who: adults with moderate to severe traumatic brain injury in nine French intensive care units.
• How long: infusion for 48 hours or longer; outcome at 6 months.
• Result: adjusted common odds ratio for the Extended Glasgow Outcome Scale 1.02, 95% CI 0.71-1.47, P=.92; intracranial hypertension 33.7% versus 36.3%, absolute difference -2.6%, 95% CI -12.3% to 7.2%; six-month mortality 15.9% versus 20.8%, absolute difference -4.9%, 95% CI -12.8% to 3.1%.
• Funding: not stated in the abstract.

Limit of this finding: Three figures in the source do not reconcile with their own denominators: 77 of 370 women is 20.8%, not the 20.2% printed; and 62 and 37 of 185 per arm give 33.5% and 20.0%, not the 33.7% and 20.8% printed, which fit slightly smaller analysed populations the abstract never states. The conclusion also says 10 of 12 secondary outcomes were not significantly different, which means two were, and neither the abstract nor this passage says which.

The adjusted common OR for the GOS-E score at 6 months was 1.02 (95% CI, 0.71-1.47; P = .92). Of the 12 secondary outcomes, 10 were not significantly different. Intracranial hypertension developed in 62 (33.7%) patients in the intervention group and 66 (36.3%) patients in the control group (absolute difference, -2.6% [95% CI, -12.3% to 7.2%]; OR, 0.80 [95% CI, 0.51-1.26]). There was no significant difference in 6-month mortality (29 [15.9%] in the intervention group vs 37 [20.8%] in the control group; absolute difference, -4.9% [95% CI, -12.8% to 3.1%]; hazard ratio, 0.79 [95% CI, 0.48-1.28]). CONCLUSIONS AND RELEVANCE: Among patients with moderate to severe traumatic brain injury, treatment with continuous infusion of 20% hypertonic saline compared with standard care did not result in a significantly better neurological status at 6 months. However, confidence intervals for the findings were wide, and the study may have had limited power to detect a clinically important difference.

Sodium bicarbonate did not change the primary outcome in the only randomised trial in severe metabolic acidaemia in intensive care. (Source 5)

• Randomized trial, Moderate certainty.
• Size: 389 patients (194 control, 195 bicarbonate)
• Who: adults admitted within 48 hours to 26 French intensive care units with pH 7.20 or below and severe illness.
• How long: 28 days.
• Result: primary composite outcome in 138 of 194 (71%) control versus 128 of 195 (66%) bicarbonate, absolute difference estimate -5.5%, 95% CI -15.2 to 4.2, p=0.24; day 28 survival 46% versus 55%, p=0.09; in the pre-specified AKIN 2-3 stratum survival 37% versus 54%, p=0.0283.
• Funding: French Ministry of Health and the Societe Francaise d'Anesthesie Reanimation, as stated in the abstract.

Limit of this finding: The trial was open-label and the subgroup result in patients with acute kidney injury comes from one of three pre-specified strata; a reader should not take it as evidence that bicarbonate saves lives in acidaemia generally. The abstract itself carries a published correction marker on that sentence. The INTERPRETATION says bicarbonate decreased the primary composite outcome and day-28 mortality in the acute kidney injury stratum, but the results quoted alongside give only a day-28 survival figure for that stratum (37% against 54%, p=0.0283) and no composite figure for it; the only primary-outcome result in the block is the overall, non-significant one.

The primary outcome occurred in 138 (71%) of 194 patients in the control group and 128 (66%) of 195 in the bicarbonate group (absolute difference estimate -5·5%, 95% CI -15·2 to 4·2; p=0·24).

Sodium bicarbonate given in out-of-hospital cardiac arrest was not associated with better return of circulation or survival to discharge. (Source 6)

• Systematic review, Low certainty.
• Size: 11 studies (2 randomised trials, 9 observational), 126,013 people.
• Who: people with out-of-hospital cardiac arrest.
• How long: to hospital discharge.
• Result: return of spontaneous circulation odds ratio 0.85, 95% CI 0.59-1.23, p = 0.38; survival to discharge odds ratio 0.63, 95% CI 0.17-2.28, p = 0.48.
• Funding: not stated in the abstract.

Limit of this finding: Nine of the eleven studies were observational, so confounding by indication is likely: sicker or longer arrests are more likely to receive bicarbonate. The review explicitly does not exclude benefit in hyperkalaemia or tricyclic antidepressant overdose.

No significant differences were observed in ROSC (OR 0.85, 95% CI 0.59-1.23; p = 0.38) or survival to discharge (OR 0.63, 95% CI 0.17-2.28; p = 0.48). CONCLUSIONS: Administration of sodium bicarbonate in patients with OHCA was not associated with improved ROSC or survival to discharge. Therefore, its routine use cannot be recommended.

Where the evidence is mixed

A Cochrane review found that in white participants sodium reduction lowers blood pressure only slightly when blood pressure is normal and substantially when it is high, and that it raises renin, aldosterone, noradrenaline, cholesterol and triglyceride. (Source 24)

• Systematic review, High certainty.
• Size: 195 included references; for example 5,982 participants across 95 trials for systolic pressure in white participants with normal blood pressure.
• Who: people randomised to low and high sodium diets, stratified by race.
• How long: trials of varying length; the review's own search ran to April 2018 with a top-up in March 2020.
• Result: white participants with normal blood pressure: systolic mean difference -1.14 mmHg, 95% CI -1.65 to -0.63; diastolic +0.01 mmHg, 95% CI -0.37 to 0.39. White participants with hypertension: systolic -5.71 mmHg, 95% CI -6.67 to -4.74; diastolic -2.87 mmHg, 95% CI -3.41 to -2.32. During sodium reduction cholesterol increased 5.19 mg/dL, 95% CI 2.1 to 8.3, and triglyceride 7.10 mg/dL, 95% CI 3.1 to 11.1.
• Funding: Cochrane review; funding not stated in the abstract.

Limit of this finding: Every figure quoted here is for white participants, which the quoted sentence states; the review reports different and less certain figures for black and Asian participants. Its race-stratified subgroups rest on as few as 253 participants in some strata, and it measures blood pressure, hormones and lipids rather than heart attacks or strokes, so it cannot settle whether sodium reduction changes outcomes. The source prints "3998 participants,88 trials" with a missing space; that missing space is the source's own and is reproduced as printed.

The effect of sodium reduction (from 203 to 65 mmol/day) on BP in white participants was as follows: Normal blood pressure: SBP: mean difference (MD) -1.14 mmHg (95% confidence interval (CI): -1.65 to -0.63), 5982 participants, 95 trials; DBP: MD + 0.01 mmHg (95% CI: -0.37 to 0.39), 6276 participants, 96 trials. Hypertension: SBP: MD -5.71 mmHg (95% CI: -6.67 to -4.74), 3998 participants,88 trials; DBP: MD -2.87 mmHg (95% CI: -3.41 to -2.32), 4032 participants, 89 trials (all high-quality evidence).

In the PURE prospective cohort, the association between estimated sodium excretion and death or cardiovascular events was J-shaped, with the lowest risk at moderate rather than low sodium, and higher potassium excretion blunted the risk seen with high sodium. (Source 25)

• Cohort study, Low certainty.
• Size: 103,570 people who provided morning fasting urine samples in 18 countries.
• Who: adults in urban and rural communities across 18 high, middle and low income countries.
• How long: median follow-up 8.2 years.
• Result: 7,884 (6.1%) died or had a major cardiovascular event; relative to moderate sodium (3-5 g/day) with higher potassium, hazard ratios were 1.23 (1.11 to 1.37) for low potassium with low sodium, 1.21 (1.11 to 1.32) for low potassium with high sodium, 1.19 (1.02 to 1.38) for low sodium with higher potassium, and 1.10 (1.02 to 1.18) for high sodium with higher potassium; higher potassium excretion attenuated the increased cardiovascular risk associated with high sodium excretion, P for interaction=0.007.
• Funding: not stated in the abstract.

Limit of this finding: This is observational, and sodium intake was estimated from a single morning fasting urine sample using a formula, a method critics say systematically distorts the shape of the curve. It cannot show that eating more salt causes lower risk, and reverse causation in sick people who eat less is not excluded. The figure that 0.002% of participants met both the low-sodium and high-potassium targets is about two people out of the 103,570 in the cohort, so it is a statement about how rare that combination is and not a risk estimate. The cohort shares quoted here total 76.0%, leaving a sixth category (moderate sodium with low potassium) that the passage does not name.

A J-shaped association was observed of sodium excretion and inverse association of potassium excretion with death and cardiovascular events. For joint sodium and potassium excretion categories, the lowest risk of death and cardiovascular events occurred in the group with moderate sodium excretion (3-5 g/day) and higher potassium excretion (21.9% of cohort).

Where the research disagrees

Whether 0.9% saline is worse than a balanced crystalloid for critically ill adults

• Semler and colleagues, SMART trial, 2018, single-centre cluster-randomised multiple-crossover trial in 15,802 adults: Among the 7942 patients in the balanced-crystalloids group, 1139 (14.3%) had a major adverse kidney event, as compared with 1211 of 7860 patients (15.4%) in the saline group (marginal odds ratio, 0.91; 95% confidence interval [CI], 0.84 to 0.99 (Source 3)
• Young and colleagues, SPLIT trial, 2015, double-blind cluster randomised double-crossover trial in 2,278 patients: In the buffered crystalloid group, 102 of 1067 patients (9.6%) developed AKI within 90 days after enrollment compared with 94 of 1025 patients (9.2%) in the saline group (absolute difference, 0.4% [95% CI, -2.1% to 2.9%]; relative risk [RR], 1.04 [95% CI, 0.80 to 1.36]; P = .77). (Source 26)
• Zampieri and colleagues, BaSICS trial, 2021, double-blind factorial randomised trial at 75 Brazilian intensive care units, 11,052 randomised: use of a balanced solution compared with 0.9% saline solution did not significantly reduce 90-day mortality. The findings do not support the use of this balanced solution. (Source 23)
• Finfer and colleagues, PLUS trial, 2022, double-blind randomised controlled trial in 5,037 patients at 53 intensive care units: We found no evidence that the risk of death or acute kidney injury among critically ill adults in the ICU was lower with the use of BMES than with saline. (Source 22)

Whether reducing dietary sodium improves health outcomes, and whether the risk curve is J-shaped

• He, Li and MacGregor, 2013, systematic review and meta-analysis of 34 randomised trials, 3,230 participants, blood pressure outcomes: Meta-analysis showed that the mean change in urinary sodium (reduced salt v usual salt) was -75 mmol/24 h (equivalent to a reduction of 4.4 g/day salt), and with this reduction in salt intake, the mean change in blood pressure was -4.18 mm Hg (95% confidence interval -5.18 to -3.18, I(2)=75%) for systolic blood pressure and -2.06 mm Hg (-2.67 to -1.45, I(2)=68%) for diastolic blood pressure. (Source 27)
• Graudal and colleagues, Cochrane review, 2020, Cochrane systematic review of 195 references, blood pressure, hormone and lipid outcomes: In white participants, sodium reduction in accordance with the public recommendations resulted in mean arterial pressure (MAP) decrease of about 0.4 mmHg in participants with normal blood pressure (Source 17)
• O'Donnell and colleagues, PURE cohort, 2019, prospective cohort of 103,570 people in 18 countries, estimated 24-hour sodium from a single morning fasting urine sample: A J-shaped association was observed of sodium excretion and inverse association of potassium excretion with death and cardiovascular events. (Source 25)
• Neal and colleagues, SSaSS trial, 2021, open-label cluster-randomised trial in 20,995 people, hard cardiovascular outcomes: The rate of stroke was lower with the salt substitute than with regular salt (29.14 events vs. 33.65 events per 1000 person-years; rate ratio, 0.86; 95% confidence interval [CI], 0.77 to 0.96; P = 0.006) (Source 14)

Whether a boxed warning applies to sodium salts as a group

• The oral sodium phosphate tablet label (OsmoPrep), 2010, regulator-approved boxed warning, LOINC section 34066-1: There have been rare, but serious reports of acute phosphate nephropathy in patients who received oral sodium phosphate products for colon cleansing prior to colonoscopy. (Source 2)
• The sodium bicarbonate injection label, 2026, regulator-approved label with no 34066-1 section; its strongest statements are a contraindication and warnings: Sodium bicarbonate injection is an alkalinizing agent indicated: •to raise plasma pH in adult and pediatric patients with clinically significant metabolic acidosis. (1.1) •to alkalinize the urine in drug poisoning in adult and pediatric patients in cases in which urinary drug elimination is enhanced by urinary alkalinaztion. (1.2) Limitation of Use Do not use sodium bicarbonate injection in settings where rapid administration of sodium bicarbonate is required, such as cardiac arrest. In such situations use a more concentrated solution. (Source 7)
• The sodium chloride injection label, 2026, regulator-approved label with no 34066-1 section; its strongest statement sits in an ordinary WARNINGS section: Sodium Chloride Injection, USP should be used with great care, if at all, in patients with congestive heart failure, severe renal insufficiency (Source 12)

How much

• Reference intake: There is no reader-facing dose for any of these products: each is prescribed or supervised, and the dose is set by the clinician for the clinical state. The sodium chloride injection label states the indication as parenteral replenishment of fluid and sodium chloride as required by the patient's clinical condition, and ties monitoring to periodic laboratory determinations. (Source 1)
• Upper limit: For oral sodium phosphate tablets there is no reader-facing ceiling and no permitted range: the label fixes one dose and one schedule. The recommended dose of OsmoPrep Tablets for colon cleansing for adult patients is 32 tablets (48 grams of sodium phosphate) taken orally with a total of 2 quarts of clear liquids. The same section sets out the split: The evening before the colonoscopy procedure: Take 4 OsmoPrep Tablets with 8 ounces of clear liquids every 15 minutes for a total of 20 tablets. On the day of the colonoscopy procedure: Starting 3-5 hours before the procedure, take 4 OsmoPrep Tablets with 8 ounces of clear liquids every 15 minutes for a total of 12 tablets. It also closes the window on repeat use: Patients should not use OsmoPrep for colon cleansing within seven days of previous administration. No upper limit for a reader is given anywhere. (Source 28)
• Studied: The BICAR-ICU trial infused 4.2% intravenous sodium bicarbonate to keep arterial pH above 7.30, with each infusion in the range of 125-250 mL in 30 min and a maximum of 1000 mL within 24 h after inclusion. (Source 5)
• Studied: The COBI trial gave continuous infusion of 20% hypertonic saline solution plus standard care for 48 hours or longer if patients remained at risk of intracranial hypertension. (Source 4)
• Studied: The pivotal colon cleansing trial gave 60 grams of sodium phosphate in split doses with 2.5 quarts of clear liquids, or 48 grams with 2 quarts of clear liquids. (Source 8)
• Studied: The SSaSS trial gave villages a salt substitute of 75% sodium chloride and 25% potassium chloride by mass, compared with regular salt of 100% sodium chloride. (Source 14)

A common belief, and what the research shows

The belief: The salt argument settles it: because dietary salt is bad for you, the saline drip in hospital must be harmful too, and because baking soda is harmless in the kitchen, sodium bicarbonate must be harmless as a medicine.

What the research shows: These are separate evidence bases and they point in different directions. For dietary sodium the randomised evidence is about blood pressure and, in one large trial of a salt substitute, about stroke: The rate of stroke was lower with the salt substitute than with regular salt (29.14 events vs. 33.65 events per 1000 person-years; rate ratio, 0.86; 95% confidence interval [CI], 0.77 to 0.96; P = 0.006). For intravenous saline the question is which crystalloid, not whether sodium is bad, and three of the four large trials found nothing: We found no evidence that the risk of death or acute kidney injury among critically ill adults in the ICU was lower with the use of BMES than with saline. And a sodium salt that is innocuous in food can be dangerous as a medicine: the oral sodium phosphate bowel preparation carries a boxed warning because Some cases have resulted in permanent impairment of renal function and some patients required long-term dialysis.

Questions and answers

What is it?

Pharmaceutical sodium salts are manufactured medicines in which sodium is paired with a chosen partner ion. Sodium chloride injection dissociates into sodium and chloride ions and is used to replace water and electrolytes. Sodium bicarbonate is given as an alkalinising agent. Sodium phosphate and sodium citrate are a bowel purgative and an oral alkaliniser respectively. They are not interchangeable with each other or with table salt. (Source 1)

What does it do in the body?

Sodium is the main cation outside the body's cells and water follows it, so a sodium salt expands the fluid around the cells. That is why saline is used to replace lost volume. The partner ion does the rest of the work: bicarbonate and citrate raise pH, phosphate holds water in the bowel. Water balance and sodium handling are controlled mainly by the kidney. (Source 29)

Is it good or bad for you?

Both, and which one depends on the salt, the amount and the person. Reduced-sodium oral rehydration salts clearly help children with diarrhoea. Intravenous saline replaces lost volume but has been questioned against balanced fluids. Oral sodium phosphate cleans the colon well and has also caused permanent kidney failure, which is why it carries a boxed warning. In people with heart failure or poor kidney function, a sodium load is a hazard in its own right. (Source 12)

How do you get more of it?

These are prescribed or supervised products, not something to seek out. The only form a person handles themselves is oral rehydration salts and over-the-counter sodium bicarbonate antacid. For dietary sodium the studied intervention has run the other way: the SSaSS trial replaced part of the sodium in household salt with potassium, and the rates of stroke, major cardiovascular events and death all fell. (Source 14)

If it is harmful, what reduces it?

When a sodium salt has done harm, the harm is managed by stopping the salt load, correcting the electrolytes and supporting the kidney. For sodium bicarbonate the label asks for electrolytes to be monitored and abnormalities treated. For oral sodium phosphate kidney injury, four of twenty-one patients in the biopsy series needed permanent dialysis, so some of the damage was not reversible. (Source 21)

Why might someone be low in it or missing it?

Low blood sodium is usually a problem of too much water rather than too little salt. The label describes exactly that mechanism: intravenous fluid can dilute serum electrolytes and cause overhydration. Clinically, losses from vomiting, gastric suction, diarrhoea and diuretics, and kidney disease, are what shift sodium and chloride. A person eating ordinary food is not short of dietary sodium. (Source 12)

Which whole foods contain it or feed it?

Dietary sodium comes overwhelmingly from salt added in food manufacture and cooking, which is a different thing from these medicines. In the SSaSS trial the food-level intervention was the household salt itself, replaced by a 75% sodium chloride and 25% potassium chloride mix. Oral rehydration salts are the one medicinal sodium salt taken as a drink with food-derived glucose. (Source 30)

What happens if you do not have it?

Withholding these medicines when they are indicated means the underlying problem goes untreated: unreplaced volume loss, uncorrected acidosis, or an inadequately prepared colon. In children with diarrhoea, the trials measured exactly this: more children needed unscheduled intravenous fluid on the higher-sodium formula than on the reduced-osmolarity one, with an odds ratio of 0.59 favouring the reduced formula. (Source 10)

How can you test for it?

Blood sodium, chloride, bicarbonate and creatinine are measured on a routine blood panel, and that is the test. The sodium chloride injection label ties safe use to repeated laboratory checks of fluid balance, electrolytes and acid-base status. For oral sodium phosphate the label asks for baseline and post-colonoscopy phosphate, calcium, potassium, sodium, creatinine and BUN in people at higher risk. Estimating a person's habitual dietary sodium is far less reliable: the PURE cohort used a formula applied to a single morning fasting urine sample, which is the main reason its J-shaped curve is disputed. (Source 15)

References

1. DailyMed / US FDA Structured Product Label (Fresenius Kabi USA, LLC). Sodium Chloride Injection, USP - CLINICAL PHARMACOLOGY section. 2026. Read the source
2. DailyMed / US FDA Structured Product Label (Physicians Total Care, Inc.). OSMOPREP (sodium phosphate, monobasic, monohydrate, sodium phosphate, dibasic anhydrous) tablet - BOXED WARNING section (LOINC 34066-1). 2010. Read the source
3. The New England journal of medicine. Balanced Crystalloids versus Saline in Critically Ill Adults.. 2018. PMID 29485925, DOI 10.1056/NEJMoa1711584. Read the source
4. JAMA. Effect of Continuous Infusion of Hypertonic Saline vs Standard Care on 6-Month Neurological Outcomes in Patients With Traumatic Brain Injury: The COBI Randomized Clinical Trial.. 2021. PMID 34032829, DOI 10.1001/jama.2021.5561. Read the source
5. Lancet (London, England). Sodium bicarbonate therapy for patients with severe metabolic acidaemia in the intensive care unit (BICAR-ICU): a multicentre, open-label, randomised controlled, phase 3 trial.. 2018. PMID 29910040, DOI 10.1016/S0140-6736(18)31080-8. Read the source
6. Prehospital emergency care. Effectiveness of Sodium Bicarbonate Administration in Out-of-Hospital Cardiac Arrests: An Updated Systematic Review and Meta-Analysis.. 2026. PMID 41091632, DOI 10.1080/10903127.2025.2575439. Read the source
7. DailyMed / US FDA Structured Product Label (Baxter Healthcare Corporation). Sodium Bicarbonate Injection, USP - Highlights of Prescribing Information, INDICATIONS AND USAGE excerpt. 2026. Read the source
8. DailyMed / US FDA Structured Product Label (Physicians Total Care, Inc.). OSMOPREP tablet - CLINICAL STUDIES section, phase 3 colon cleansing response rates (Table 1) with its footnotes. 2010. Read the source
9. DailyMed / US FDA Structured Product Label (Chartwell Governmental & Specialty Rx, LLC). Sodium Citrate and Citric Acid Oral Solution - INDICATIONS AND USAGE section. 2026. Read the source
10. The Cochrane database of systematic reviews. Reduced osmolarity oral rehydration solution for treating dehydration caused by acute diarrhoea in children.. 2002. PMID 11869639, DOI 10.1002/14651858.CD002847. Read the source
11. DailyMed / US FDA Structured Product Label (Chartwell Governmental & Specialty Rx, LLC). Sodium Citrate and Citric Acid Oral Solution - PRECAUTIONS section. 2026. Read the source
12. DailyMed / US FDA Structured Product Label (Fresenius Kabi USA, LLC). Sodium Chloride Injection, USP - WARNINGS section. 2026. Read the source
13. DailyMed / US FDA Structured Product Label (Baxter Healthcare Corporation). SODIUM BICARBONATE injection - section 4 CONTRAINDICATIONS (body text, LOINC 34070-3). 2026. Read the source
14. The New England journal of medicine. Effect of Salt Substitution on Cardiovascular Events and Death.. 2021. PMID 34459569, DOI 10.1056/NEJMoa2105675. Read the source
15. DailyMed / US FDA Structured Product Label (Physicians Total Care, Inc.). OSMOPREP tablet - WARNINGS section, opening paragraphs on fatalities, cautions and baseline laboratory testing. 2010. Read the source
16. DailyMed / US FDA Structured Product Label (Physicians Total Care, Inc.). OSMOPREP tablet - WARNINGS section, paragraphs on seizures and on arrhythmias. 2010. Read the source
17. The Cochrane database of systematic reviews. Effects of low sodium diet versus high sodium diet on blood pressure, renin, aldosterone, catecholamines, cholesterol, and triglyceride.. 2020. PMID 33314019, DOI 10.1002/14651858.CD004022.pub5. Read the source
18. Journal of the American Society of Nephrology : JASN. Acute phosphate nephropathy following oral sodium phosphate bowel purgative: an underrecognized cause of chronic renal failure.. 2005. PMID 16192415, DOI 10.1681/ASN.2005050496. Read the source
19. DailyMed / US FDA Structured Product Label (Physicians Total Care, Inc.). OSMOPREP tablet - ADVERSE REACTIONS section, Table 2 of adverse events occurring in greater than 3% of patients. 2010. Read the source
20. Annals of surgery. A randomized, controlled, double-blind crossover study on the effects of 2-L infusions of 0.9% saline and plasma-lyte® 148 on renal blood flow velocity and renal cortical tissue perfusion in healthy volunteers.. 2012. PMID 22580944, DOI 10.1097/SLA.0b013e318256be72. Read the source
21. DailyMed / US FDA Structured Product Label (Baxter Healthcare Corporation). Sodium Bicarbonate Injection, USP - section 5.2 Electrolyte Imbalances. 2026. Read the source
22. The New England journal of medicine. Balanced Multielectrolyte Solution versus Saline in Critically Ill Adults.. 2022. PMID 35041780, DOI 10.1056/NEJMoa2114464. Read the source
23. JAMA. Effect of Intravenous Fluid Treatment With a Balanced Solution vs 0.9% Saline Solution on Mortality in Critically Ill Patients: The BaSICS Randomized Clinical Trial.. 2021. PMID 34375394, DOI 10.1001/jama.2021.11684. Read the source
24. The Cochrane database of systematic reviews. Effects of low sodium diet versus high sodium diet on blood pressure, renin, aldosterone, catecholamines, cholesterol, and triglyceride.. 2020. PMID 33314019, DOI 10.1002/14651858.CD004022.pub5. Read the source
25. BMJ (Clinical research ed.). Joint association of urinary sodium and potassium excretion with cardiovascular events and mortality: prospective cohort study.. 2019. PMID 30867146, DOI 10.1136/bmj.l772. Read the source
26. JAMA. Effect of a Buffered Crystalloid Solution vs Saline on Acute Kidney Injury Among Patients in the Intensive Care Unit: The SPLIT Randomized Clinical Trial.. 2015. PMID 26444692, DOI 10.1001/jama.2015.12334. Read the source
27. BMJ (Clinical research ed.). Effect of longer term modest salt reduction on blood pressure: Cochrane systematic review and meta-analysis of randomised trials.. 2013. PMID 23558162, DOI 10.1136/bmj.f1325. Read the source
28. DailyMed / US FDA Structured Product Label (Physicians Total Care, Inc.). OSMOPREP tablet - DOSAGE AND ADMINISTRATION section (LOINC 34068-7), the recommended dose and the pm/am split regimen. 2010. Read the source
29. DailyMed / US FDA Structured Product Label (Fresenius Kabi USA, LLC). Sodium Chloride Injection, USP - CLINICAL PHARMACOLOGY section, closing paragraphs on water and sodium balance. 2026. Read the source
30. The Cochrane database of systematic reviews. Reduced osmolarity oral rehydration solution for treating dehydration caused by acute diarrhoea in children.. 2002. PMID 11869639, DOI 10.1002/14651858.CD002847. Read the source