Osmolar gap calculator

Short answer: The osmolar gap (also written osmolal gap or osmol gap) is measured serum osmolality minus calculated serum osmolality. A gap that is larger than expected suggests that the serum contains osmotically active substances the formula does not count. The classic ones are methanol, ethylene glycol, isopropanol, propylene glycol and ethanol [1, 2]. The gap is a screening clue, not a diagnosis. Its “normal” value depends on the formula and the laboratory. A normal gap does not exclude toxic-alcohol poisoning [2, 3, 4].

Units on this page. US conventional units come first (Na⁺ in mEq/L, glucose and BUN in mg/dL), with SI in brackets (mmol/L). The measured value is an osmolality (mOsm/kg). The calculated value is formally an osmolarity estimate (mOsm/L). The gap subtracts one from the other, which is one source of its imprecision [5].

1. The formulas

Measured osmolality

Measured on an osmometer, preferably by freezing-point depression. Vapor-pressure osmometers may not detect volatile solutes such as ethanol and methanol. Walker et al. reported a methanol-poisoned patient with no gap for exactly this reason [7]. See How Osmometers Work.

Calculated osmolality: US units (Smithline–Gardner form)

Calc Osm = 2 × Na⁺ + glucose/18 + BUN/2.8

with Na⁺ in mEq/L and glucose and BUN in mg/dL [4, 8]. (The 1976 Smithline–Gardner abstract does not print the equation [1]. The exact form used in the gap-validation literature is in Lynd et al., Table 1 [4].) The divisors convert mg/dL to mmol/L: glucose has a molar mass of about 180 g/mol (180 ÷ 10 = 18), and BUN is reported as nitrogen, 28 g of N per mole of urea (28 ÷ 10 = 2.8).

BUN is not urea. The ÷2.8 divisor applies only to BUN (blood urea nitrogen) in mg/dL. If your lab reports urea in mg/dL, divide by 6.0 (urea molar mass is about 60) to get mmol/L. Urea mg/dL ≈ BUN mg/dL × 2.14. Feeding urea mg/dL into a “/2.8” field overstates the term about 2.14-fold: urea 30 mg/dL is 5.0 mmol/L, not 10.7. European labs usually report urea in mmol/L, which goes into the SI formula as is.

Calculated osmolality: SI units

Calc Osm = 2 × Na⁺ + glucose + urea

with all values in mmol/L [5, 9]. Choy et al. reviewed decades of competing formulas and recommended this simple Smithline–Gardner form for harmonized lab reporting [5].

Alternative: Dorwart–Chalmers (1975)

Calc Osm = 1.86 × Na⁺ + glucose/18 + BUN/2.8 + 9

In 715 hospital sera, this was the most accurate of the 13 equations the authors tested [10]. Different formulas give different “normal” gaps. Report which formula you used (see Example 4).

Adding ethanol

Ethanol is a common contributor to an elevated gap in practice. Include it whenever a level is available. osmolar.com uses /3.7 by default and shows /4.6 as the labelled alternative:

Lynd et al. compared the two coefficients when screening for toxic alcohols. The 1.25 (SI) coefficient, which is equivalent to /3.7, gave higher specificity and positive predictive value without loss of sensitivity [4]. Note that “/4.6” and “1.25 × mmol/L” are not equivalent: /4.6 corresponds to 1.0 × mmol/L. Some online references mix them up. State which coefficient you used.

Osmolar gap = measured osmolality − calculated osmolality

2. What is a “normal” osmolar gap?

There is no single normal value:

• The traditional teaching is that a normal gap is below 10 [12]. Kraut and Kurtz note that measured osmolality is “usually within 10 mOsm/L” of the calculated value [2].
• In 305 consecutive ED patients with known ethanol levels, Hoffman et al. found a mean gap of −2 ± 6 mOsm (with 2Na + BUN/2.8 + Glu/18 + EtOH/4.6). Different equations shifted the mean gap anywhere from −5 to +15 mOsm. Their conclusion: “small osmol gaps should not be used to eliminate the possibility of toxic alcohol ingestion” [3].
• In Purssell’s validation cohort, residual gaps after the ethanol correction ranged from −18.40 to +17.85 mOsm/L [11].
• Lynd et al. note that the gap is non-specific (it can also rise in DKA, shock and alcoholic ketoacidosis), has a wide normal range, may run higher in patients older than 60 years, and can miss small but toxic ethylene glycol and methanol levels [4].
• The gap carries the combined error of the measured osmolality, each analyte in the formula, and any unmeasured endogenous solutes [5].

A gap of 10 or more (Merck ≥10) is a commonly used threshold, not a sharp boundary [16]. A normal gap does not exclude toxic-alcohol ingestion.

Illustration: a patient whose baseline gap is −5 can carry a substantial toxic-alcohol load and still show a “normal” gap of +8.

How big is big?

Kraut and Kurtz: ketoacidosis, lactic acidosis, renal failure and critical illness with hyponatremia can raise the gap, usually to ≤15–20 mOsm/L. A gap >20 mOsm/L strongly suggests accumulation of an alcohol (including ethanol, if it was not included in the calculation) [2]. StatPearls quotes the same ≤15–20 vs >20 split and an average gap of about 11 in lactic acidosis [6]. These are not hard cut-offs: alcoholic ketoacidosis has been reported with raw gaps above 20 before ethanol correction [15].

3. Toxic alcohols and the gap

Molar mass decides how much a given level raises osmolality

From Kraut and Kurtz, Table 2 [2]:

Light molecules (methanol) raise the gap a lot per mg/dL. Heavy ones (ethylene glycol, diethylene glycol) raise it little. Clinically dangerous ethylene glycol levels can produce only a modest gap [2].

Estimating a concentration from the gap (rough)

Standard conversion factors (multiply the gap by the factor to estimate mg/dL): methanol 3.2, isopropanol 6.0, ethylene glycol 6.2, propylene glycol 7.6, acetone 5.8 [13]. In a 2020 academic-center dataset, these estimates tended to overestimate the measured concentration [13], and they assume no other unmeasured osmoles are present. Use them as a sanity check, not as a substitute for a direct assay.

Timing: the gap falls as the anion gap rises

The parent alcohols raise osmolality. Their acid metabolites (formate from methanol, glycolate and oxalate from ethylene glycol) raise the anion gap and lower bicarbonate. Early after ingestion you may see a high osmolar gap with a normal anion gap. Late, you may see a high anion gap with a normal osmolar gap. “Either can be absent, depending on the time after exposure when blood is sampled” [2]. Isopropanol typically causes hyperosmolality without metabolic acidosis, because it is metabolized to acetone, not to an acid [2].

Other causes of a raised gap

Pseudohyponatremia from severe hyperlipidemia or hyperproteinemia (the formula’s Na⁺ is falsely low) [6, 1]. Mannitol and propylene glycol (for example from IV drug vehicles) [2]. Ketoacidosis, lactic acidosis and advanced kidney failure [2, 6].

Use the Serum Osmolality Calculator to run these numbers with a unit toggle and selectable ethanol coefficient.

4. Limitations: read before you rely on a gap

1. A normal gap does not exclude toxic-alcohol ingestion, because of baseline variability [3], heavy molecules [2], and late presentation after metabolism [2].
2. Sensitivity and specificity are threshold-dependent. With a cut-off of 10, Lynd et al. found sensitivity of 0.90 (ethanol coefficient 1.0) and 0.85 (coefficient 1.25) for patients needing antidotal therapy, with specificity of only 0.22 and 0.50 [4].
3. Formula and coefficient choice shift the result by several mOsm [3, 10, 11].
4. Method: vapor-pressure osmometry misses volatile alcohols [6, 7]. Use freezing-point depression and draw the osmolality and chemistry from the same sample at the same time.
5. Units mismatch: measured mOsm/kg vs calculated mOsm/L [5].

5. Worked examples

Example 1: Baseline gap (US units)

Na⁺ 140 mEq/L, glucose 90 mg/dL, BUN 14 mg/dL, ethanol 0. Measured osmolality 292 mOsm/kg.

Step 1. 2 × Na⁺ = 2 × 140 = 280 Step 2. Glucose/18 = 90 ÷ 18 = 5.0 Step 3. BUN/2.8 = 14 ÷ 2.8 = 5.0 Step 4. Calculated = 280 + 5 + 5 = 290 mOsm/L Step 5. Gap = 292 − 290 = +2, which is unremarkable.

Example 2: Ethanol explains the gap (default /3.7, with /4.6 for comparison)

Na⁺ 138, glucose 108 mg/dL, BUN 21 mg/dL, ethanol 230 mg/dL. Measured 345 mOsm/kg.

Step 1. Calc (no ethanol) = 2 × 138 + 108/18 + 21/2.8 = 276 + 6.0 + 7.5 = 289.5 Step 2. Raw gap = 345 − 289.5 = 55.5 Step 3a (default). Ethanol by Purssell’s term: 230 ÷ 3.7 = 62.16, leaving a residual gap of 55.5 − 62.16 = −6.7 [11] Step 3b (alternative). Ethanol by the molar term: 230 ÷ 4.6 = 50.0, leaving a residual gap of 55.5 − 50.0 = +5.5 [3]

Interpretation: under either coefficient, ethanol plausibly accounts for most of the gap. The 12-point swing between coefficients shows why you must state which one you used. A small residual gap still does not exclude a co-ingested toxic alcohol [3].

Rounding note: keep full precision through the calculation and round only the final displayed gap.

Example 3: SI units, unexplained gap, rough toxic-alcohol estimate

Na⁺ 140 mmol/L, glucose 5.5 mmol/L, urea 5.0 mmol/L, ethanol 0. Measured 325 mOsm/kg.

Step 1. Calc = 2 × 140 + 5.5 + 5.0 = 280 + 10.5 = 290.5 Step 2. Gap = 325 − 290.5 = 34.5. Kraut and Kurtz consider a gap above 20 strongly suggestive of an accumulated alcohol [2]. Step 3. If methanol were the sole cause: 34.5 × 3.2 ≈ 110 mg/dL (≈ 34.5 mmol/L) [13]. Step 4. If ethylene glycol: 34.5 × 6.2 ≈ 214 mg/dL [13].

Interpretation: these are rough estimates. In one academic-center dataset these standard factors tended to overestimate measured concentrations, and they assume no other unmeasured osmoles (ethanol, ketones, lactate) [13]. They justify an urgent specific assay and toxicology input. They do not replace them.

Example 4: The formula changes the “normal” gap

Same labs as Example 1 (Na⁺ 140, glucose 90, BUN 14, measured 292).

Dorwart–Chalmers: 1.86 × 140 + 90/18 + 14/2.8 + 9 = 260.4 + 5 + 5 + 9 = 279.4, giving a gap of 292 − 279.4 = +12.6.

The same patient has a gap of +2 with Smithline–Gardner and +12.6 with Dorwart–Chalmers. A fixed cut-off of 10 or more (Merck ≥10) cannot be portable across formulas [3, 5].

6. Practice questions

Q1. Na⁺ 135, glucose 180 mg/dL, BUN 28 mg/dL, measured osmolality 300 mOsm/kg, no ethanol. What is the gap?

Q2. Why might a patient who drank ethylene glycol 20 hours ago have a normal osmolar gap but a high anion gap?

Q3. The lab measures osmolality by vapor-pressure osmometry. A patient with suspected methanol ingestion has a gap of +3. What is the pitfall?

Q4. Which raises the serum osmolality more per 10 mg/dL: methanol or ethylene glycol? Why?

Q5. A patient’s isopropanol level is high. Would you expect a high anion gap metabolic acidosis?

Related tools and pages

• Serum Osmolality Calculator (calculated, effective, gap, ethanol)
• Normal Osmolality Range
• How Osmometers Work (freezing-point vs vapor-pressure)
• Osmolality vs Osmolarity
• How we check content · Disclaimer

Possible poisoning? In the US, call the Poison Help line at 1-800-222-1222 [14]. Elsewhere, call your local poison center or emergency number, or go to an emergency department. Do not wait for a calculation.

Educational content, not medical advice. The osmolar gap is a screening calculation. Suspected toxic-alcohol ingestion is a medical emergency. Contact a poison center or toxicologist and follow your institution’s protocols. Do not use this page to rule poisoning in or out, or to make treatment decisions.

References

1. Smithline N, Gardner KD Jr. Gaps—anionic and osmolal. JAMA. 1976;236(14):1594–1597. https://doi.org/10.1001/jama.236.14.1594 (PubMed: https://pubmed.ncbi.nlm.nih.gov/989132/)
2. Kraut JA, Kurtz I. Toxic alcohol ingestions: clinical features, diagnosis, and management. Clin J Am Soc Nephrol. 2008;3(1):208–225. https://doi.org/10.2215/CJN.03220807 (PubMed: https://pubmed.ncbi.nlm.nih.gov/18045860/)
3. Hoffman RS, Smilkstein MJ, Howland MA, Goldfrank LR. Osmol gaps revisited: normal values and limitations. J Toxicol Clin Toxicol. 1993;31(1):81–93. https://pubmed.ncbi.nlm.nih.gov/8433417/
4. Lynd LD, Richardson KJ, Purssell RA, et al. An evaluation of the osmole gap as a screening test for toxic alcohol poisoning. BMC Emerg Med. 2008;8:5. https://pmc.ncbi.nlm.nih.gov/articles/PMC2390580/
5. Choy KW, Wijeratne N, Lu ZX, Doery JC. Harmonisation of Osmolal Gap – Can We Use a Common Formula? Clin Biochem Rev. 2016;37(3):113–119. https://pmc.ncbi.nlm.nih.gov/articles/PMC5111243/
6. StatPearls. Osmometer. NCBI Bookshelf NBK589659. https://www.ncbi.nlm.nih.gov/books/NBK589659/
7. Walker JA, Schwartzbard A, Krauss EA, Sherman RA, Eisinger RP. The missing gap. A pitfall in the diagnosis of alcohol intoxication by osmometry. Arch Intern Med. 1986;146(9):1843–1844. https://pubmed.ncbi.nlm.nih.gov/3753127/ (see also Draviam EJ, Custer EM, Schoen I. Am J Clin Pathol. 1984;82(6):706–709. https://doi.org/10.1093/ajcp/82.6.706)
8. StatPearls. Serum Osmolality. NCBI Bookshelf NBK567764. https://www.ncbi.nlm.nih.gov/books/NBK567764/
9. StatPearls. Physiology, Osmoregulation and Excretion. NCBI Bookshelf NBK541108. https://www.ncbi.nlm.nih.gov/books/NBK541108/
10. Dorwart WV, Chalmers L. Comparison of methods for calculating serum osmolality from chemical concentrations, and the prognostic value of such calculations. Clin Chem. 1975;21(2):190–194. https://pubmed.ncbi.nlm.nih.gov/1112025/
11. Purssell RA, Pudek M, Brubacher J, Abu-Laban RB. Derivation and validation of a formula to calculate the contribution of ethanol to the osmolal gap. Ann Emerg Med. 2001;38(6):653–659. https://pubmed.ncbi.nlm.nih.gov/11719745/
12. StatPearls. Anion Gap and Non-Anion Gap Metabolic Acidosis. NCBI Bookshelf NBK448090. https://www.ncbi.nlm.nih.gov/books/NBK448090/
13. Greene HR, Krasowski MD. Correlation of osmolal gap with measured concentrations of acetone, ethylene glycol, isopropanol, methanol, and propylene glycol in patients at an academic medical center. Toxicol Rep. 2020;7:81–88. https://pmc.ncbi.nlm.nih.gov/articles/PMC6939068/
14. HRSA Poison Help. Contact Us. https://poisonhelp.hrsa.gov/contact-us
15. Schelling JR, Howard RL, Winter SD, Linas SL. Increased osmolal gap in alcoholic ketoacidosis and lactic acidosis. Ann Intern Med. 1990;113(8):580–582. https://pubmed.ncbi.nlm.nih.gov/2400167/
16. Merck Manual Professional Edition. Water and Sodium Balance. Accessed 2026-10-18. https://www.merckmanuals.com/professional/nephrology/fluid-metabolism/water-and-sodium-balance