Calculated serum osmolality estimates the concentration of major dissolved particles in serum from routinely measured laboratory values, most commonly sodium, glucose and urea/BUN. The osmol gap is the difference between directly measured osmolality and that calculated estimate. Because both the calculation formula and laboratory measurement contribute variation, the gap is an interpretive clue rather than a diagnosis.
What is serum osmolality?
Osmolality describes the concentration of osmotically active particles per kilogram of solvent and is commonly reported as mOsm/kg. The Royal College of Pathologists of Australasia distinguishes this from osmolarity, which is expressed per litre of solution.
Measured and calculated osmolality are not the same thing
Measured osmolality
Obtained directly by the laboratory using an osmometer; freezing-point depression is a commonly used method.
Calculated osmolality
A mathematical estimate based on selected major measured osmoles such as sodium, glucose and urea/BUN.
These values are related, but they are not expected to match perfectly in every sample. The difference between them is what produces the osmol gap.
The calculated-osmolality formula used by MedMaths
MedMaths uses the simple Smithline-Gardner-associated equation supported in an Australian harmonisation review for general clinical use.
With SI inputs, sodium, glucose and urea are entered in mmol/L. With conventional US reporting, the equivalent form is:
This is a calculated estimate. It does not replace a directly measured laboratory osmolality when a measured value is clinically required.
For an osmol gap, confirm the laboratory osmolality was directly measured.
A 2023 Clinical Toxicology sentinel event described a real ethylene-glycol exposure in which a calculated laboratory osmolality was mistaken for a measured value, producing a falsely reassuring negative gap. Direct serum osmolality is commonly measured by freezing-point depression; use contemporaneous chemistry and osmometry values where possible.
Worked example: calculated serum osmolality
Sodium: 140 mmol/L
Glucose: 5 mmol/L
Urea: 5 mmol/L
2 × 140 + 5 + 5
= 290 mOsm/kg
Calculated serum osmolality: 290 mOsm/kg
For input-to-answer arithmetic, use the Serum Osmolality Calculator.
What is the osmol gap?
Measured osmolality: 304 mOsm/kg
Calculated osmolality: 290 mOsm/kg
Osmol gap: 14 mOsm/kg
A positive gap means the measured sample contains more osmotic activity than the selected equation predicts. It does not identify which substance or process caused that difference.
Why is the osmol gap not normally exactly zero?
Calculated osmolality includes only selected measured substances, while real serum contains additional solutes. The Australian harmonisation review also emphasises that osmol-gap uncertainty reflects variation in the calculated estimate, error in measured osmolality and variation in unmeasured analytes. Small positive or negative gaps can therefore occur without representing one specific disease or toxin.
For the Smithline-Gardner calculation used by MedMaths, the 2016 harmonisation work proposed 10 mOsm/kg as a method-specific reference limit. That is useful context, not a universal diagnostic threshold: the local laboratory/toxicology interval and clinical context still take priority, and one gap value cannot diagnose or exclude toxic alcohol exposure.
Why do osmolality formulas give different answers?
Many published equations exist. Some use a sodium coefficient different from 2, some include potassium, and some add constants or use different coefficients for glucose or urea. The RCPA manual gives an example using 1.86 × (Na + K) + glucose + urea + 10, while the MedMaths calculator uses the simpler Smithline-Gardner equation.
Choy and colleagues compared many published approaches and concluded that 2 × sodium + glucose + urea was fit for purpose in healthy subjects and general hospital patients and performed well across analytical platforms. The practical consequence is important: the equation used changes the calculated value and therefore changes the osmol gap.
Compare like with like.
For serial interpretation, use the calculation method and reference interval specified by the relevant laboratory, toxicology service or protocol rather than switching formulas between results.
Osmolality and osmolarity are related but technically different
- Osmolality is expressed per kilogram of solvent.
- Osmolarity is expressed per litre of solution.
The terms are sometimes used loosely in clinical discussion, but measured serum osmolality is conventionally reported in mOsm/kg. MedMaths uses osmolality and osmol gap terminology to match that laboratory convention.
What can an elevated osmol gap tell you?
An increased gap suggests that measured osmotically active substances are present that are not adequately represented by the selected equation. The RCPA lists suspected exposure to alcohols and other osmotically active substances among the major applications of measured osmolality and osmol-gap assessment.
However, the osmol gap is a screening and interpretation tool. It cannot distinguish methanol from ethylene glycol, ethanol, isopropanol or other contributors simply from the number itself.
A normal osmol gap does not rule out toxic alcohol exposure
Timing matters. The parent alcohol contributes to the osmol gap, but as methanol or ethylene glycol is metabolised, the osmol gap can fall while toxic acidic metabolites and an anion-gap metabolic acidosis become more prominent. Reviews of toxic alcohol poisoning therefore describe both the usefulness and the limitations of the osmol gap in patient management.
Do not use a normal osmol gap as a stand-alone exclusion test.
If toxic alcohol exposure is clinically suspected, urgent toxicology assessment and appropriate direct laboratory testing take priority over reassurance from one calculated gap.
How ethanol changes the osmol gap
Ethanol contributes to measured osmolality, so an ethanol-adjusted calculation depends on the coefficient used. MedMaths deliberately shows the assumption rather than hiding it.
Molecular-weight approach
ethanol mg/dL ÷ 4.6
This corresponds approximately to ethanol mmol/L × 1.0.
Purssell empirical approach
ethanol mg/dL ÷ 3.7
Purssell and colleagues derived this empirically in 2001; in SI units it is approximately ethanol mmol/L × 1.25.
The evidence is not unanimous. A 2012 validation study of 603 emergency-department patients estimated an ethanol contribution of approximately mg/dL ÷ 4.0, while a later prospective volunteer study comparing the displayed 4.6 and 3.7 coefficients found that 4.6 produced values closer to subjects' baseline osmol gap in that cohort. MedMaths therefore presents 4.6 and Purssell 3.7 as two explicit conventions, not as an exhaustive map of every published coefficient and not as universally correct alternatives.
Use values from the same clinical time point where possible
Sodium, glucose, urea/BUN, measured osmolality and ethanol can change with treatment, metabolism and time. When calculating an osmol gap, use contemporaneous laboratory values where possible rather than combining measurements collected at substantially different times.
Osmol gap and anion gap answer different questions
The osmol gap compares measured osmolality with calculated osmolality. The anion gap uses measured electrolytes to estimate unmeasured ionic charge. They can both contribute information in some toxic and metabolic presentations, but they are not interchangeable tests.
For the separate acid-base concept, see the Anion Gap Interpretation Guide or use the Anion Gap Calculator.
Total osmolality is not the same as effective osmolality / tonicity
Calculated total osmolality includes urea. Effective osmolality, often used as a tonicity estimate, excludes urea and is calculated by MedMaths as 2 × sodium + glucose with glucose in mmol/L. The calculator now shows this as a separate secondary result so it is not confused with total osmolality or the osmol gap.
The ADA 2026 hyperglycaemic-crisis criteria use effective serum osmolality >300 mOsm/kg as one possible HHS hyperosmolarity criterion, alongside the rest of the diagnostic criteria. The MedMaths effective-osmolality output is therefore a calculation aid, not an HHS diagnosis and not a substitute for serial clinical and laboratory assessment.
If the specific question is how hyperglycaemia affects interpretation of measured sodium, see Corrected Sodium in Hyperglycaemia.
Common errors to check
- Confusing directly measured osmometer osmolality with a chemistry-derived calculated estimate.
- Trying to calculate an osmol gap without a measured osmolality.
- Entering BUN in mg/dL as though it were already urea in mmol/L, or vice versa.
- Entering glucose in mg/dL without applying the required unit conversion.
- Mixing different calculated-osmolality formulas when comparing serial gaps.
- Applying one universal osmol-gap cutoff regardless of formula and laboratory method.
- Assuming an elevated gap identifies a particular toxic alcohol.
- Assuming a normal gap independently excludes toxic alcohol exposure.
- Ignoring ethanol when it materially contributes to measured osmolality, or entering an ethanol mmol/L result as the same number in mg/dL (or vice versa).
- Combining laboratory values from different clinical time points without recognising the limitation.
What the osmol gap cannot decide
Calculated serum osmolality and the osmol gap are laboratory interpretation tools. They do not diagnose or exclude toxic alcohol poisoning, determine the cause of altered mental status, or decide whether antidotal treatment, dialysis or another emergency intervention is required. Those decisions depend on the clinical presentation, exposure history, acid-base findings, direct testing where available and the relevant toxicology or emergency pathway.
Need the arithmetic?
Enter sodium, glucose, urea/BUN and optional measured osmolality or ethanol in the MedMaths calculator. The calculator keeps the formula and ethanol assumptions visible.
Open the Serum Osmolality Calculator →Sources and references
- Royal College of Pathologists of Australasia — Osmolality. Defines osmolality, osmolarity and the osmol gap; describes freezing-point measurement, formula variation and clinical applications.
- Mayo Clinic Laboratories — Osmolality, Serum. Laboratory source documenting direct serum osmolality measurement by freezing-point depression.
- Pires KD et al. Minding the osmol gap: a sentinel event and subsequent laboratory investigation. Clin Toxicol. 2023;61(11):1001-1003. Demonstrates the safety risk of mistaking a calculated laboratory osmolality for a directly measured result.
- Choy KW, Wijeratne N, Lu ZX. Harmonisation of Osmolal Gap — Can We Use a Common Formula? Clin Biochem Rev. 2016;37(3):113-119. Supports the Smithline-Gardner equation, proposes 10 mOsm/kg as a method-specific reference limit, and explains formula- and measurement-related uncertainty.
- Ross JA, Borek HA, Holstege CP, King JD. Toxic Alcohol Poisoning. Emerg Med Clin North Am. 2022;40(2):327-341. Reviews toxic-alcohol metabolism and the utility and limitations of the osmol gap.
- Lynd LD et al. An evaluation of the osmole gap as a screening test for toxic alcohol poisoning. Supports using the osmol gap as a screening clue rather than a stand-alone diagnostic or exclusion test.
- American Diabetes Association. Diabetes Care in the Hospital: Standards of Care in Diabetes—2026. Distinguishes effective osmolality (2 × Na + glucose) from total osmolality (2 × Na + glucose + urea) within the full HHS criteria.
- Garrard A et al. Validation of a pre-existing formula to calculate the contribution of ethanol to the osmolar gap. Clin Toxicol. 2012;50(7):562-566. In 603 ED patients, estimated an ethanol conversion factor of approximately mg/dL ÷4.0.
- Purssell RA et al. Derivation and validation of a formula to calculate the contribution of ethanol to the osmolal gap. Ann Emerg Med. 2001;38(6):653-659. Derives the empirical ethanol divisor of 3.7.
- Ethanol and the Limitations of the Osmol Gap. Prospective volunteer study comparing the 4.6 and 3.7 ethanol corrections and demonstrating persistent baseline variation.