Osmolal gap: what a raised gap means and its causes

What the osmolal gap measures

The osmolal gap measures solutes in the serum that the routine chemistry panel does not report. Kraut and Kurtz explain in their 2008 review that normal serum osmolality, 285 to 290 mOsm/L, comes from sodium with its counterbalancing chloride and bicarbonate, and from glucose and urea, which is why the formula 2 × Na + BUN/2.8 + glucose/18 estimates it well. The lab measures osmolality directly with an osmometer, and that value is usually within 10 mOsm/L of the calculated one. When a low molecular weight substance such as an alcohol builds up in the blood, it raises the measured osmolality but not the calculated one, and the difference between them is the gap. Smithline and Gardner described the osmolal gap in JAMA in 1976 as a bedside tool that directs attention to laboratory error, high serum protein, high lipids or unmeasured osmoles. Some authors call the calculated value an osmolarity, in mOsm/L, and the measured one an osmolality, in mOsm/kg.

The serum osmolality calculator works out both versions and switches between mg/dL and mmol/L without changing the answer.

What counts as a raised osmolal gap

A raised osmolal gap is usually defined as above 10, and a gap above 20 is the level Kraut and Kurtz tie to an alcohol in the blood. The 10 cut-off follows from their statement that measured and calculated osmolality usually agree within 10 mOsm/L. Lynd and colleagues tested it in 131 patients at two tertiary-care hospitals who had ethylene glycol and methanol measured, 20 of them with levels above the threshold for treatment. A gap above 10 had a sensitivity of 0.90 but a specificity of only 0.22 with the standard ethanol term; using Purssell's ethanol term raised specificity to 0.50, with a sensitivity of 0.85. For the patients whose levels were high enough for dialysis to be recommended, the sensitivity was 1 with both terms. The cut-off of 20 rests on a different point: Kraut and Kurtz state that ketoacidosis, lactic acidosis, renal failure and critical illness with hyponatremia can raise the gap, but usually to no more than 15 to 20.

Causes of a raised osmolal gap

The causes of a raised osmolal gap fall into three groups: alcohols, other acute illnesses, and errors in the measurement or the sample. Ethanol is the first thing to check, because its contribution can be calculated: Purssell and colleagues found that ethanol in mg/dL divided by 3.7, or 1.25 times ethanol in mmol/L, matches its effect on the gap. The toxic alcohols Kraut and Xing list are methanol, ethylene glycol, diethylene glycol, propylene glycol and isopropanol. Kraut and Kurtz note that most of these also cause a high anion gap metabolic acidosis as they are metabolized, while isopropanol raises osmolality without acidosis. Kraut and Xing add that diabetic and alcoholic ketoacidosis, acute kidney injury, chronic kidney disease and lactic acidosis can produce both a high anion gap acidosis and a raised osmolal gap, so they belong in the same differential. Smithline and Gardner listed very high protein or lipid levels and laboratory error as non-toxic causes.

For the acidosis side of the work-up, the anion gap calculator gives the albumin-corrected anion gap from the same sample.

Why a normal osmolal gap does not rule out a toxic alcohol

A normal osmolal gap does not rule out a toxic alcohol for three reasons that the sources state directly. First, the normal gap is wide: Hoffman and colleagues measured a mean of −2 ± 6 mOsm in 305 adults, so a patient whose own baseline is −8 can carry a meaningful amount of alcohol and still show a gap under 10. They concluded that small gaps should not be used to eliminate the possibility of toxic alcohol ingestion. Second, heavier alcohols add few osmoles per milligram. Kraut and Kurtz calculate that methanol at 50 mg/dL raises osmolality by about 15 mOsm/L, but ethylene glycol at the same concentration by only about 9. Third, timing matters: early after methanol ingestion the osmolal gap is largest and the anion gap smallest, and as the alcohol is metabolized the osmolal gap may return to baseline while the anion gap keeps rising. Kraut's 2015 review adds that definitive detection needs gas or liquid chromatography, which is not always available.

Worked examples: two raised osmolal gaps

Two patients with the same routine chemistry show why ethanol is counted before a raised osmolal gap is called unexplained. Both have sodium 140 mmol/L, glucose 90 mg/dL and BUN 14 mg/dL, so the calculated osmolality is 2 × 140 + 90 ÷ 18 + 14 ÷ 2.8, which is 280 + 5 + 5, or 290 mOsm/kg. The first patient has a measured osmolality of 345 and a serum ethanol of 185 mg/dL. Without ethanol the gap would be 55, far above 20. With Purssell's term, ethanol adds 185 ÷ 3.7, or 50, so the expected osmolality is 340 and the gap that is left is 5, within the usual range: ethanol explains it. The second patient has a measured osmolality of 318 and no ethanol detected. The gap is 318 − 290, or 28, above 20, and in Kraut and Kurtz's terms this points to another alcohol in the blood, so the anion gap and a specific toxic alcohol level become the next laboratory steps.

Enter both patients in the serum osmolality calculator to see the same results; in mmol/L, the inputs are glucose 5.0, urea 5.0 and ethanol 40.2, and rounding moves the first gap by less than 1.