Corrected Sodium Calculator

High blood sugar pulls water into your bloodstream and dilutes your sodium, so a low reading during hyperglycemia is often not a real sodium problem. This corrected sodium calculator runs all five correction equations in clinical use, because they disagree with each other, and shows you exactly where that disagreement starts to change the answer.

Enter your sodium and glucose in the calculator. Then read on, because the corrected number only means something if you have the right kind of low sodium in the first place.

Why High Blood Sugar Lowers Your Sodium

Glucose that cannot get into cells stays in the blood and acts as an osmotic pull. Water follows it out of the cells and into the bloodstream. More water, same amount of sodium, so the concentration falls.

This is real dilution, not a lab error. Your sodium genuinely is low at that moment. What the correction estimates is where it would sit once the glucose is treated and the water goes back where it belongs.

The catch is that this only describes one of three different reasons a sodium result comes back low, and the formula is wrong for the other two. More on that below.

The Corrected Sodium Formula

You will be quoted two. There are at least five in clinical use.

Katz, 1973:

Corrected sodium = measured sodium + 0.016 × (glucose in mg/dL − 100)

Hillier, 1999:

Corrected sodium = measured sodium + 0.024 × (glucose in mg/dL − 100)

Both are quoted verbatim from MDCalc's published evidence for this calculation. In plain terms, Katz says sodium drops 1.6 mEq/L for every 100 mg/dL of glucose above 100. Hillier says 2.4. The other three, and why they matter, are further down.

Where 1.6 and 2.4 came from

Katz calculated 1.6 in 1973. He did not measure it. Based on theoretical modelling he argued the rule of thumb in use at the time, 2.8, overstated the dilution, because hyperosmolality persists and less water leaves the cells than that rule predicted. His smaller number is what every textbook printed for the next twenty-six years.

Nobody tested it until 1999, when Hillier and colleagues blocked insulin secretion in 6 healthy volunteers, pushed glucose above 600 mg/dL and brought it back down while sampling sodium every ten minutes.

What they found was not a single number. The relationship was not a straight line: 1.6 fitted well up to a glucose of 400 mg/dL, and above that a factor nearer 4.0 fitted better. The 2.4 you see quoted everywhere is the mean across the whole range, and Pathology is explicit that Hillier proposed it as a simplification that worked reasonably well at all glucose values. It is a practical compromise, not a physiological constant.

Six volunteers is a small experiment. It is also still the controlled experiment the field refers back to, and Pathology's 2025 verdict is that clinical texts continue to use Katz or Hillier without a clear preference of one over the other.

The worked example where they disagree

Most of the time both formulas land in the same place and the argument is academic. Sometimes they do not.

Take a measured sodium of 125 mEq/L with a glucose of 600 mg/dL.

Same blood draw. One formula says the sodium problem persists after the sugar is fixed, the other says it resolves. That is not a rounding difference, it is a different clinical conclusion about whether anything else needs treating.

Our calculator runs both and flags it when they land on opposite sides of 135.

There Are Not Two Equations. There Are Five.

This is the part no calculator page carries, and it is the reason the number you were given may not match the number someone else calculates from the same blood draw.

Lam and Wijeratne catalogued the equations actually in clinical use in their 2025 comparison in Pathology, the journal of the Royal College of Pathologists of Australasia. There are five. Here is what each one does to the same patient: sodium 125 mEq/L, glucose 600 mg/dL (33.3 mmol/L).

One blood draw. A 6 mEq/L spread. Two equations call it hyponatremia, two call it normal, and one lands precisely on the 135 boundary.

The calculator on this page runs all five, because knowing which one produced your number is the whole point.

Why they diverge, and exactly where

Below a glucose of 30 mmol/L, about 540 mg/dL, this argument mostly does not matter. Lam and Wijeratne found the four equations still in routine use agree within ±3 mmol/L, which is the Royal College of Pathologists of Australasia's allowable performance specification for sodium.

Above 30 mmol/L, they come apart. In 1,222 real patient episodes from a hospital laboratory over five years, at glucose values above that threshold:

• Katz classified 27 times more people as hyponatremic than Hillier did
• Hillier classified 1.7 times more people as hypernatremic than Katz did

Twenty-seven times. From the same blood draws, on the same patients, using two equations that contemporary clinical texts treat as interchangeable.

Our worked example sits at 600 mg/dL, above that threshold. That is not an accident. It is the range where the choice of equation stops being academic.

Across all 1,222 episodes, correction moved 41.6% of results out of the hyponatremic range down to 4.3%. It also tripled the proportion classified as hypernatremic, from 7.6% to 22.3%. Correction is not simply reassurance, it moves people in both directions, and that matters: Chuang and colleagues found 90-day mortality in severe hyperglycemia was higher at both ends, in those who stayed hyponatremic after correction and those who became severely hypernatremic.

Above 400 mg/dL, no equation is really right

Hillier's nonlinearity finding is what sits underneath all of this. Above a glucose of 400 mg/dL a factor nearer 4.0 fitted their data, and none of the five equations above uses 4.0. Nobody has published one that does.

American Family Physician stated the direction of that error plainly. The conventional 1.6 factor underestimated sodium in most cases once glucose passed 300 mg/dL, and in almost every measurement once it passed 500 mg/dL.

So, in practice: once glucose passes roughly 300 mg/dL, read every corrected value as a floor rather than an answer, and the higher the glucose, the more true that becomes.

A note on units

Sodium is reported in mEq/L or mmol/L and the two are numerically identical, which is why MedlinePlus prints it as 136 to 144 mEq/L (136 to 144 mmol/L). Nothing to convert.

Glucose is the one that changes. The US reports mg/dL, most of the rest of the world reports mmol/L.

Here is a small detail that turns out to matter. Sources convert 100 mg/dL of glucose to either 5.5 or 5.6 mmol/L, inconsistently. American Family Physician uses 5.6. Lam and Wijeratne point out that glucose has a molecular weight of 180.156 g/mol, so the true value sits between the two at 5.55 mmol/L, and they standardise on 5.55 specifically to remove that source of variation.

We use 5.55 for the same reason. In SI units the two published forms are:

• Katz: sodium + 1.6 × (glucose − 5.55) ÷ 5.55, which clinicians shorten to (glucose − 5.55) ÷ 3.5
• Hillier: sodium + 2.4 × (glucose − 5.55) ÷ 5.55, shortened to (glucose − 5.55) ÷ 2.3

Both shortcuts come straight from the Pathology comparison and exist because they can be done at the bedside without a calculator. This calculator converts your figure and runs the full forms, so nothing gets rounded by hand. Paste straight off your report either way.

Which Kind of Low Sodium Do You Actually Have?

This is the part that decides whether the corrected number means anything, and it is missing from every calculator we checked.

A sodium below 135 has three quite different explanations, and the glucose correction only applies to one of them.

Serum osmolality is what separates them, and it is a test your doctor can order alongside the sodium.

Hyponatremia is not rare, either. StatPearls puts the prevalence at 20% to 35% among patients in the hospital.

There Is No Albumin Correction for Sodium

People search for one constantly, so here is the direct answer: it does not exist, and that is not an oversight.

Calcium has an albumin correction. The anion gap has one. Sodium does not, because the mechanism is completely different. If the reason you are here is a low albumin on the same panel, our guide to low albumin covers what it means on its own.

With calcium, albumin genuinely binds the mineral, so low albumin genuinely lowers the measured total. Arithmetic can estimate around it. With sodium, high protein or high lipid does not change the sodium at all. It changes how much of your blood sample is water, and an analyzer that dilutes the sample before measuring gets fooled by that.

StatPearls is precise about the mechanism. The indirect ion-selective electrode method dilutes the serum before testing, which is exactly what makes it vulnerable when the sample is loaded with extra solute. The direct method skips the dilution, and its result does not require a correction factor.

So the fix is not a formula. It is a different measurement, usually on a blood gas analyzer, which uses the direct method. If you have very high triglycerides, very high cholesterol, or a paraprotein disorder such as myeloma, that is the conversation to have with your doctor.

If the albumin question brought you here, the tool you were probably looking for is our corrected calcium calculator.

Sodium Ranges and What They Mean

Severity bands follow the StatPearls hyponatremia chapter and the normal range follows MedlinePlus. As always, your own laboratory's printed range wins over any of these.

The Part That Matters Most Is How Fast It Gets Corrected

Neither MDCalc's hyperglycemia tool nor Omni mentions this on the page, and it is the piece with the real consequences.

When sodium has been low for a while, the brain adapts. Raise it too quickly and you get osmotic demyelination syndrome, formerly called central pontine myelinolysis, which can cause seizures, disorientation, coma and in severe cases locked-in syndrome.

Current guidance is to raise sodium by no more than 10 mEq/L in the first 24 hours. Here is the uncomfortable part: even the recommended emergency approach of raising it 6 mEq/L to relieve severe neurological symptoms exceeds the correction limit in about 4.5% to 28% of patients.

That is why this is a hospital decision, not a home one. It matters in diabetic ketoacidosis especially: as insulin brings glucose down, sodium rises on its own, and the corrected value tells the team roughly where it is heading.

If glucose control is the underlying issue, our A1C calculator shows your three-month average, and our eGFR calculator covers the kidney function that sits underneath most sodium handling.

How This Calculator Compares to MDCalc and Omni

We opened both and checked what they actually do.

Straight assessment: MDCalc is the better bedside tool and its evidence write-up on the Katz versus Hillier history is genuinely good. It also has a dedicated Sodium Correction Rate calculator, which is the right home for that math. The gap is that a worried person reading their own lab report has to find two separate tools and still will not learn that their low sodium might be pseudohyponatremia.

Omni explains the physiology clearly and is easy to use. It just stops at the arithmetic.

We built this one for the person holding the lab report.

Why Use Our Corrected Sodium Calculator

• All five equations in clinical use, side by side, not just the two you usually get
• A flag when Katz and Hillier land on opposite sides of the diagnostic line
• A warning once glucose passes the point where the equations stop agreeing
• Glucose in mg/dL or mmol/L, sodium in either equivalent unit
• Tells you whether the correction even applies to your situation
• No sign-up, nothing stored

When to Get Help Now

Low sodium with symptoms is urgent. Get same-day care for confusion, drowsiness, seizures, persistent vomiting, severe headache or unsteadiness, particularly if blood sugar is also high.

For a mildly low sodium with no symptoms, the sensible next step is a repeat panel with glucose and serum osmolality drawn at the same time. Your doctor decides what else is needed.

Corrected Sodium Is Not a Diagnosis

This calculator is for informational and screening purposes only and does not constitute medical advice or diagnosis. Always consult a qualified healthcare professional. The correction estimates what sodium would be at a normal glucose. It does not tell you why sodium is low, it does not apply to pseudohyponatremia or hypotonic hyponatremia, and decisions about correcting sodium belong in a clinical setting because correcting too fast carries serious risk.

Browse our other health calculators for related lab tools.

Frequently Asked Questions