Salt Toxicosis and Hypernatraemia in Dogs: A Clinical Reference

Bottom line

Salt toxicosis is one of the few small-animal toxicoses in which the treatment, rather than the toxicant, is the most likely thing to kill the patient. In dogs, clinical signs of toxicosis can appear after ingestion of 2-3 g/kg of salt and the acute oral lethal dose is approximately 4 g/kg, with serum and CSF sodium concentrations >160 mEq/L indicating salt toxicosis [1]. Before you write a fluid plan, establish one thing: whether the hypernatraemia arose over 24 hours or less, or has been present for more than 24 hours. That single distinction sets your ceiling, because the published target is a decrease in [Na] of 1 mmol/hr in acute cases and 0.5 mmol/hr in chronic cases [2]. Correct faster than the brain can shed the idiogenic osmoles it generated to defend itself and you convert a survivable electrolyte derangement into fatal cerebral oedema [1].

How sodium accumulates and why the brain is the organ at risk

The sodium concentration is not what injures the brain. The osmotic gradient across the neuronal membrane is, and that gradient depends on how much time the brain has had to adapt.

Rapid development of hypernatraemia results in cerebral dehydration and neuronal cell shrinkage, with the brain then pulling away from the calvarium, which disrupts the blood supply to the brain and can cause tearing of vessels and haemorrhage [1]. To limit that water loss, cells of the brain increase their intracellular osmolarity through the generation of idiogenic osmoles [1] - osmotically active chemicals produced within neurons [2].

That adaptation is protective while the hypernatraemia persists, which is why chronic hypernatraemia, defined as lasting for more than 24 hours, is often well tolerated, whereas acute hypernatraemia occurring over 24 hours or less causes cellular shrinkage that is particularly deleterious within the central nervous system [2]. It is also exactly what makes correction dangerous. Once water access is restored, a rapid decrease in serum sodium concentration drives intracellular water influx into neurons along the osmotic gradient and can lead to cerebral oedema [1]. The adapted brain is the vulnerable brain: idiogenic osmoles are cleared far more slowly than plasma sodium can be dropped with a bag of fluids.

A second mechanism compounds this in true sodium-gain toxicosis. Increased sodium retention lets sodium passively diffuse into the cerebrospinal fluid, and active transport is required to remove it; because that removal is energy-dependent, extracellular sodium concentrations usually decrease faster than sodium concentrations in the cerebrospinal fluid, so sodium trapped in the central nervous system attracts water because of the osmotic gradient, resulting in cerebral oedema [3].

The practical consequence is the single most useful sentence in this literature: the severity of the clinical signs is related more to the rapidity of the onset of hypernatraemia than the magnitude [3]. A dog at 175 mmol/L that arrived there in four hours and a dog at 175 mmol/L that arrived there over four days are different patients with different permissible correction rates, and the history is what separates them.

Common exposure sources in practice

Companion animal exposures to excess salt have included the use of salt as an emetic, which is no longer recommended, and the ingestion of various salt-containing materials including rock salt and dough-salt mixtures [1]. In practice, a handful of scenarios account for most presentations.

Homemade play dough and salt-dough ornaments. This is the classic and it is genuinely lethal. In a reported case, an eight-year-old neutered male Airedale terrier was presented for evaluation of a 45-minute episode of continuous seizure activity after ingesting a salt-flour mixture used as clay for the sculpting of small figurines; serum sodium was 211 mEq/L against a reference range of 145 to 158 mEq/L, and the dog died despite aggressive therapy directed at the hypernatraemia [4]. Ask directly about craft dough and seasonal ornament dough when the history includes baking or decorating.

Paintballs. Paintball fill is osmotically active, and the resulting hypernatraemia can climb after admission rather than peak at presentation. In a 2025 report, an eight-year-old castrated male Cavalier King Charles spaniel presented with abnormal mentation and ataxia and had only mild hypernatraemia at 157 mmol/L, yet developed tonic-clonic seizures four hours after admission, and approximately 11 hours after admission the plasma sodium had increased to 170 mmol/L, at which point the dog became comatose and required endotracheal intubation [5]. The lesson is that a reassuring admission sodium in a paintball case does not let you stop measuring.

Seawater at the beach. Dogs have been reported to develop hypernatraemia after swimming or playing in the ocean, which contains approximately 3.5% sodium, without having access to fresh water [1]. Ask specifically about fresh-water availability during a long beach session; the exposure is repeated small-volume ingestion over hours rather than a single bolus.

Table salt given as an emetic. Call this out with owners and with colleagues, because it persists in folk advice and online. It is no longer recommended [1], and the margin is far narrower than people assume: half a teaspoon is enough table salt to potentially cause hypernatraemia in a 6-kilogram dog [6]. A dog that has been given salt at home for a non-toxic ingestion can arrive with an iatrogenic toxicosis considerably worse than the original exposure.

Ice melts and de-icing salts. The most common ingredients in these products are sodium chloride, potassium chloride, magnesium chloride, calcium carbonate and calcium magnesium acetate, and a few ice melts contain urea [3]. A dose of 4 g/kg of sodium chloride can be lethal in dogs [3]. In 1998, more than 50 cases of ice-melt exposure were reported to the ASPCA Animal Poison Control Center, with vomiting the most prevalent sign, occurring in 30% of the cases [3]. Note that activated charcoal does not usually adsorb the salts in ice melts, so it is not a useful decontaminant here [3].

Sodium phosphate enemas. Application of a sodium phosphate containing enema caused life-threatening metabolic disorders in a Dachshund and two cats, characterised by dehydration and various neurological deficits including seizures, with hypernatraemia, hypocalcaemia, hyperphosphataemia and polycythaemia; despite intensive treatment the dog died, whereas the cats recovered completely, leading the authors to conclude that sodium phosphate enemas should be used with caution or not at all in cats and small dogs [7].

Water deprivation with concentrated food. Salt toxicosis is often directly related to water consumption and can be reduced notably or abolished completely in production animals by means of appropriate management of factors such as mechanical failure of waterers, overcrowding, unpalatable medicated water, new surroundings or frozen water sources [1]. In small animals this is the boarding kennel, the frozen outdoor bowl, and the dog left with dry food and an empty bucket.

Activated charcoal - the iatrogenic cause you control. Hypernatraemia has been reported in animals treated with improperly mixed oral electrolyte solutions, remedies for diarrhoea, or secondary to treatment with activated charcoal [1]. In nine healthy adult dogs given a single 2 g/kg dose of activated charcoal suspension with sorbitol, the increase in serum sodium concentration was significantly higher following administration compared with the control period, and the authors concluded that patients given a single dose are at risk of developing dehydration and secondary hypernatraemia [8]. The clinical worst case is documented: a seven-year-old spayed female Corgi treated for chocolate ingestion with activated charcoal with sorbitol at 1.9 g/kg and subsequently activated charcoal without sorbitol at 0.9 g/kg developed a serum sodium of 174 mmol/L against a reference range of 144 to 160 mmol/L, went on to neurological signs and azotaemia, and was discharged 57 hours following admission [9]. If you are giving repeated charcoal for an osmotically active toxicant, as in the protocols used for ibuprofen and other NSAID toxicoses, measure electrolytes rather than assume.

Clinical presentation

Presentation is biphasic in most sodium-gain cases: gastrointestinal first, neurological second.

Excess salt intake in dogs results in vomiting within several hours after ingestion, and clinical signs can progress to weakness, diarrhoea, muscle tremors and seizures [1]. That matters beyond patient comfort, because the gastroenteritis and diuresis that occur may combine to cause dehydration, further worsening the patient's condition [3] - the free-water losses stack on top of the sodium load.

Neurological signs track concentration imprecisely. Clinical signs are not usually noted until [Na] is greater than 170 mmol/L [2], but that threshold is not a promise: rate of onset dominates magnitude [3], and the paintball case above seized at a sodium of 157 mmol/L that was unchanged from presentation [5]. Conversely, an animal that adapted slowly may be unremarkable at a number that would obtund an acute case.

Because obtundation, tremors and seizures are the presenting complaint, salt toxicosis sits in a differential list with the other neurotoxicoses. A serum electrolyte panel separates it from lead toxicosis, bufo toad toxicosis and the neurotoxic mushroom toxicoses in minutes, which is the argument for running one on every seizing patient rather than reaching for anticonvulsants alone.

Diagnosis and the sodium thresholds that matter

Normal sodium concentrations in animals are 135 to 155 mEq/L in plasma and 135 to 150 mEq/L in cerebrospinal fluid, and serum and cerebrospinal fluid sodium concentrations above 160 mEq/L, or brain sodium concentrations above 1,800 ppm, are considered diagnostic of sodium ion toxicosis [3]. Brain sodium is necessarily a post-mortem measurement and will not help the patient in front of you; it is a confirmatory finding for a case that has already died. Serum and CSF concentrations of sodium >160 mEq/L, especially when CSF has a greater sodium concentration than serum, indicate salt toxicosis [1] - the CSF-exceeding-serum pattern is the specific fingerprint of sodium gain rather than water loss.

For triage, an [Na] measurement of 3 to 4 mmol/L above normal is of little concern, but [Na] >160 mmol/L should be specifically addressed [2]. Older guidance sets the intervention line higher, holding that treatment should be instituted if the serum sodium concentration is greater than 170 mEq per L [10]. As of August 2026, the more conservative >160 mmol/L threshold is the one reflected in current clinical review material [2], and the practical reading is that anything above 160 mmol/L deserves a plan and anything above 170 mmol/L deserves an aggressive monitoring and treatment plan - at the same correction ceiling.

The threshold that actually changes management is not a concentration at all. It is time. Acute hypernatraemia is that occurring over 24 hours or less; chronic hypernatraemia is that lasting for more than 24 hours [2]. You will frequently not know which applies, because owners do not know. When the timeline is genuinely unknown, treat the patient as chronic and use the slower ceiling - the cost of correcting a truly acute case too slowly is a longer hospitalisation, whereas the cost of correcting a chronic case too quickly is cerebral oedema.

Document a baseline sodium, the time it was drawn, and the presumed time of exposure on the treatment sheet. Every rate calculation that follows depends on those three values.

Treatment: free water deficit and the correction-rate problem

Calculate the free water deficit

Two equivalent formulations are in general use. One expresses the deficit as FWD = 0.6 x BW x (current [Na+] / desired [Na+] - 1), where FWD is free water deficit in litres, BW is body weight in kilograms, and [Na+] is serum sodium ion concentration in mEq/L [1]. The other expresses it as FWD in litres = {(Patient [Na] - Target [Na]) / Target [Na]} x TBW, where TBW is total body water, calculated as weight in kilograms multiplied by 0.6 [2]. They return the same number.

The free water deficit replaces water already lost. It does not account for ongoing losses or maintenance requirements, both of which must be added separately, and in a vomiting patient the ongoing losses are not trivial.

Choose the fluid

If other fluids are not needed, the free water deficit should be replaced using 5% dextrose [2]. Historically, 5% dextrose in water or other hypotonic fluids are given slowly intravenously [10], and either 5% dextrose in water or saline solution has been suggested [3]. The choice of fluid is less important than the rate; what matters is that it is hypotonic relative to the patient and that you titrate to a measured sodium rather than running a calculated volume blind.

The correction rate - the number to get right

This is the heart of the case. Serum sodium concentration should be lowered at a rate of 0.5-1.0 mEq/L per hour, with the slower rate recommended for cases of chronic hypernatraemia [1]. Stated as the acute-versus-chronic split, the target is a decrease in [Na] of 1 mmol/hr in acute cases and 0.5 mmol/hr in chronic cases [2].

Read that as two separate ceilings, not a range to pick from:

• Hypernatraemia that developed over 24 hours or less: up to 1 mmol/L per hour.
• Hypernatraemia present for more than 24 hours, or of unknown duration: no more than 0.5 mmol/L per hour.

The daily envelope matters as much as the hourly one. Not more than 50% of the free water deficit should be replaced in the first 24 hours, with the remaining deficit replaced in the following 24 to 48 hours [1], and for all affected animals the treatment should slowly return the animal to normal water and electrolyte balance over 2 to 3 days [1]. Consistent with that, the water deficit has long been replaced over 48 to 72 hours [10], and serum sodium concentrations should be lowered gradually over 48 to 72 hours to avoid osmotic injury to cells [3].

Too rapid correction of hypernatraemia can lead to cerebral oedema and worsening of the animal [10]. This is not a theoretical risk to be weighed casually against the risk of leaving the sodium high: in a patient whose brain has adapted, over-rapid correction is the more likely of the two to kill, and it is the failure mode this page exists to prevent. Where a genuinely acute case gives you more room, it is because the brain has not yet accumulated idiogenic osmoles - the permission comes from the timeline, never from the sodium value alone.

Adjuncts

In cases of salt intoxication, diuretics must be given in addition to slow water replacement to avoid the development of pulmonary oedema [10]. Loop diuretics such as furosemide at 2 to 4 mg/kg three or four times daily, orally, intravenously or intramuscularly, may help in acute toxicosis and may prevent the development of pulmonary oedema during fluid therapy [3]. Anticonvulsant medication may be necessary to control central nervous system signs during therapy [3]. Use sodium bicarbonate cautiously for treating acidosis because of possible worsening of hypernatraemia and hyperosmolality [3].

Monitoring

Serum [Na] should be rechecked every 4 to 6 hours [2]. In a rapidly moving case, or one where the timeline is uncertain, tighten that interval rather than loosen it - the calculation is only as good as the last measurement, and the rate you are actually achieving is frequently not the rate you prescribed.

Monitor the neurological examination alongside the chemistry, because deterioration during correction is the signal you are looking for. Clinical signs associated with too rapid a drop in [Na] and cerebral oedema include obtundation, limb rigidity and seizures [2]. A patient that was improving and then declines while the sodium is falling should prompt you to slow or pause correction and reassess, not to push harder.

Also monitor and correct abnormalities in hydration status, electrolyte concentrations and heart muscle activity [3], and weigh the patient serially - body weight is the most honest measure of whether your free-water replacement is reaching the patient.

Prognosis and what predicts a poor outcome

As of August 2026, a large retrospective review of 16,691 dogs and 4,211 cats with a measured blood or serum sodium concentration over a 60-month period found that 957 dogs (5.7%) and 338 cats (8.0%) were diagnosed with hypernatraemia [11]. Case fatality rates of dogs and cats with hypernatraemia were 20.6% and 28.1% respectively, compared with 4.4% and 4.5% in animals with a normal blood or serum sodium concentration [11].

Magnitude predicts outcome. The magnitude of hypernatraemia was linearly associated with a higher case fatality rate [11]. In that study, moderate hypernatraemia was defined as 11 to 15 mmol/L above the reference range and severe hypernatraemia as 16 mmol/L or more above the reference range [11] - note that these are increments above the laboratory reference interval, not absolute concentrations, so they must be read against your own analyser's range. Among the animals with moderate or severe hypernatraemia, 50% of dogs and 38.5% of cats presented with community-acquired hypernatraemia, and 50% of dogs and 61.5% of cats developed hospital-acquired hypernatraemia [11].

That last figure deserves to change behaviour. Half of the serious canine hypernatraemia in that population developed under veterinary care. Combined with the documented risk from activated charcoal [8][9] and from sodium phosphate enemas [7], the most actionable prognostic lever available to a clinician is often preventing the derangement in a hospitalised patient rather than correcting it.

Recovery from severe hypernatraemia can nonetheless be complete, and neurological signs at presentation are not by themselves a reason to advise euthanasia. The comatose, intubated paintball case recovered to a normal neurological examination and was discharged after 68 hours, with owners reporting no persistent clinical signs two weeks after discharge [5]. Prognosis is better framed on the trajectory during controlled correction than on the admission sodium.

Frequently Asked Questions

What is the maximum safe rate at which to correct hypernatraemia in a dog?

Serum sodium should be lowered at a rate of 0.5 to 1.0 mEq/L per hour, with the slower rate reserved for chronic cases. Expressed as the acute-versus-chronic split, the target is a decrease in sodium of 1 mmol per hour in acute cases (acute = arising over 24 hours or less) and 0.5 mmol per hour in chronic cases. Alongside the hourly ceiling, replace no more than half the free water deficit in the first 24 hours and the remainder over the following 24 to 48 hours, returning the patient to normal water and electrolyte balance over two to three days.

How do I distinguish acute from chronic hypernatraemia when the owner does not know the timeline?

Acute hypernatraemia is that occurring over 24 hours or less, and chronic hypernatraemia is that lasting more than 24 hours. When the history genuinely cannot establish which applies, default to the chronic ceiling of 0.5 mmol per hour. Correcting a truly acute case too slowly costs time in hospital; correcting a chronic case too quickly causes cerebral oedema, so the asymmetry of harm sets the default.

Why is rapid correction dangerous if the high sodium is the problem?

Because the brain adapts. To resist water loss into a hypertonic extracellular space, brain cells raise their intracellular osmolarity by generating idiogenic osmoles. Those osmoles clear slowly. If plasma sodium is dropped faster than the brain can shed them, water moves down the osmotic gradient into neurons and causes cerebral oedema. The danger is the gradient you create during treatment, not the sodium concentration itself.

Which fluid should I use to replace the free water deficit?

If no other fluid is indicated, replace the free water deficit with 5% dextrose. Hypotonic alternatives including 5% dextrose in water have long been used. The fluid choice matters less than the rate: titrate to a measured sodium rather than infusing a calculated volume without rechecking.

Is it ever appropriate to use table salt to make a dog vomit at home?

No. Salt as an emetic is no longer recommended. The margin for error is very small - half a teaspoon of table salt is enough to potentially cause hypernatraemia in a 6-kilogram dog - so a dog given salt for a non-toxic ingestion can arrive with an iatrogenic toxicosis worse than the exposure that prompted it. Worth stating explicitly to owners who have read otherwise online.

How often should I recheck sodium during correction?

Recheck serum sodium every 4 to 6 hours, and more frequently in a rapidly changing patient or when the timeline is uncertain. Track the neurological examination in parallel, because obtundation, limb rigidity and seizures are the signs of too rapid a drop in sodium with cerebral oedema, and they can appear while the chemistry still looks like progress.

Can activated charcoal cause hypernatraemia?

Yes, and it is a recognised iatrogenic cause. Healthy dogs given a single 2 g/kg dose of activated charcoal suspension containing sorbitol showed a significantly greater rise in serum sodium than during a control period, along with weight loss and haemoconcentration. A published case documents a dog reaching a serum sodium of 174 mmol/L after charcoal given for chocolate ingestion. If charcoal is used, particularly repeated doses or with an osmotically active toxicant, monitor electrolytes and hydration rather than assuming.

What predicts a poor outcome in a hypernatraemic patient?

Severity does. Case fatality rates in a large retrospective study were 20.6% for dogs and 28.1% for cats with hypernatraemia, against 4.4% and 4.5% for animals with normal sodium, and the magnitude of hypernatraemia was linearly associated with a higher case fatality rate. Severity in that study was measured as the increment above the laboratory reference range rather than an absolute value. Importantly, a high proportion of serious cases were hospital-acquired, and full neurological recovery after severe, correctly corrected hypernatraemia is well documented, so the admission sodium alone should not drive a prognosis.