About the Stool (Faecal) Osmotic Gap
Gap = 290 − 2 × (stool sodium + stool potassium), using liquid stool. Above 100 mOsm/kg indicates osmotic diarrhoea — an unabsorbed solute pulling water in — while below 50 mOsm/kg indicates secretory diarrhoea, where electrolytes account for essentially all the stool osmolality. Between 50 and 100 is genuinely indeterminate. The 290 is plasma osmolality, used deliberately instead of the measured stool osmolality: stool ferments in the pot, so a measured value rises after collection and produces a falsely low gap. A negative gap is not physiologically possible and points to a diluted or contaminated specimen.
Formula
Stool osmotic gap = 290 − 2 × (stool Na⁺ + stool K⁺)
> 100 mOsm/kg → osmotic
< 50 mOsm/kg → secretory
50–100 → indeterminate- 290
- Assumed plasma osmolality in mOsm/kg. Used in place of the measured stool osmolality on purpose — see the notes.
- 2 ×
- Accounts for the unmeasured anions accompanying sodium and potassium, on the assumption of electroneutrality.
- stool Na⁺ + stool K⁺
- Measured faecal electrolytes in mmol/L, on liquid stool.
- The 290 is deliberate. Stool continues to ferment after collection, generating osmotically active particles, so a measured stool osmolality is falsely high and the resulting gap falsely low. Using plasma osmolality avoids an artefact that gets worse the longer the specimen sits.
- A measured stool osmolality well BELOW 290 is itself informative — it cannot occur physiologically and indicates the sample was diluted with water or urine.
- A negative calculated gap is likewise impossible from real stool and points to contamination or dilution rather than to very brisk secretion.
- Some sources use 125 mOsm/kg rather than 100 as the osmotic threshold. This calculator uses 100, with 50 as the secretory threshold, following the experimental derivation.
- The sample must be liquid. Stool electrolytes on a formed sample reflect nothing useful.
- Both thresholds are strict: a gap of exactly 100 or exactly 50 falls into the indeterminate band rather than the neighbouring one.
Interpreting the result
Treat the result as a direction of travel rather than a diagnosis. A wide gap points to an osmotic load and should be confirmed by the simplest possible test: osmotic diarrhoea largely stops with fasting, because withdrawing the load withdraws the driver. Reviewing the diet and medication list for lactose, sugar alcohols such as sorbitol and mannitol, and magnesium-containing antacids or laxatives usually identifies the culprit, and a stool pH below 6 supports carbohydrate malabsorption since fermentation acidifies the stool. A narrow gap points to active secretion, which characteristically persists during fasting and continues overnight — both worth asking about explicitly, since a diarrhoea that stops at night is unlikely to be secretory. Bile-acid diarrhoea is common and treatable and deserves early consideration, alongside microscopic colitis, which needs random biopsies at colonoscopy because the mucosa looks normal. The indeterminate band between 50 and 100 is common in practice and often reflects a genuinely mixed picture; a fasting trial discriminates better than repeating the calculation.
| Score | Band | What it means | Action |
|---|---|---|---|
| > 100 mOsm/kg | Osmotic diarrhoea | An unmeasured osmotically active solute is drawing water into the lumen. Should largely stop with fasting | Fasting trial; review diet and drugs for lactose, sugar alcohols and magnesium; check stool pH (below 6 supports carbohydrate malabsorption) |
| 50–100 mOsm/kg | Indeterminate | Does not separate the two mechanisms. Common in practice and often a genuinely mixed picture | A fasting trial discriminates better than recalculating; confirm the sample was truly liquid |
| < 50 mOsm/kg | Secretory diarrhoea | Electrolytes account for essentially all stool osmolality. Persists with fasting and typically continues overnight | Screen for bile-acid diarrhoea; colonoscopy with random biopsies for microscopic colitis; consider laxative screen and neuroendocrine causes |
| Negative | Specimen problem | Not physiologically possible from stool alone. Indicates dilution or contamination, a recognised presentation of factitious diarrhoea | Repeat on an observed, properly collected specimen before acting on it |
What the Stool Osmotic Gap needs (2 inputs)
- Stool sodium (mmol/L)
- Measured on liquid stool. A formed or semi-formed sample gives an uninterpretable result and should not be sent for this purpose.
- Stool potassium (mmol/L)
- From the same sample. Potassium is typically the larger of the two cations in secretory diarrhoea of colonic origin.
Units. Stool sodium and potassium are entered in mmol/L, which is the same as mEq/L for both, since each is monovalent. The gap is reported in mOsm/kg. The 290 in the formula is assumed plasma osmolality and is not entered — see the notes on why the measured stool osmolality should not be substituted for it.
What it returns
- Osmotic gap
- In mOsm/kg. The difference between assumed plasma osmolality and the osmolality accounted for by measured electrolytes.
- Mechanism
- Osmotic (above 100), secretory (below 50), or indeterminate (50 to 100). A negative gap is flagged separately as a specimen problem.
How it is calculated
Stool water is osmotically in equilibrium with plasma by the time it reaches the rectum, so its total osmolality is about 290 mOsm/kg whatever the cause of the diarrhoea. What differs is which particles account for that total. In secretory diarrhoea the intestine actively pumps electrolytes and water into the lumen, so sodium, potassium and their accompanying anions make up almost the whole osmolality and the gap between 290 and the electrolyte contribution is small. In osmotic diarrhoea an unabsorbed solute — lactose in a lactase-deficient patient, sorbitol, magnesium — holds water in the lumen and contributes osmoles that the electrolyte measurement does not see, so the electrolyte contribution falls well short of 290 and the gap widens. Eherer and Fordtran demonstrated this across experimentally induced diarrhoeas of known cause, which is also where the practical warning about measured osmolality comes from: bacterial fermentation continues in vitro, so the measured value drifts upward after collection and using it in place of 290 introduces an artefact that grows with delay.
Facts & figures
| Osmotic (gap > 100) | Secretory (gap < 50) |
|---|---|
| Lactose and other carbohydrate malabsorption | Bile-acid diarrhoea |
| Sugar alcohols — sorbitol, mannitol, xylitol | Microscopic colitis |
| Magnesium-containing antacids and laxatives | Stimulant laxative use |
| Polyethylene glycol and lactulose | Neuroendocrine tumours (VIPoma, carcinoid, gastrinoma) |
| Generalised malabsorption, coeliac disease | Infection, including post-infective states |
| Pancreatic exocrine insufficiency | Hyperthyroidism |
The two columns need different investigations, which is the practical value of splitting them early rather than testing for everything at once.
| Osmotic | Secretory | |
|---|---|---|
| Response to fasting | Largely stops | Persists |
| Nocturnal symptoms | Uncommon | Characteristic |
| Typical stool volume | Smaller | Often large |
| Stool pH | Often below 6 if carbohydrate | Usually neutral |
| Osmotic gap | Above 100 mOsm/kg | Below 50 mOsm/kg |
The fasting response is the single most useful confirmatory manoeuvre and costs nothing. A diarrhoea that reliably stops overnight is unlikely to be secretory.
Evidence
Experimental derivation — Eherer and Fordtran
1992Study of faecal osmotic gap and stool pH across experimentally induced diarrhoeas of known and differing causes, establishing how the gap behaves in osmotic versus secretory mechanisms and characterising the behaviour of measured stool osmolality after collection.
Demonstrated that the osmotic gap reliably separates osmotic from secretory mechanisms, and that measured stool osmolality is unreliable because fermentation continues in vitro — the basis for using plasma osmolality of 290 in the formula rather than the measured value.
Clinical review and diagnostic algorithms — Schiller et al.
2017Review of the diagnosis and management of chronic diarrhoea, setting the osmotic gap within a structured diagnostic approach.
Positions the osmotic gap as an early triage step in chronic diarrhoea, with the fasting response as the principal confirmatory manoeuvre and the caution that intermediate values are common and unhelpful.
How it compares
Stool Osmotic Gap vs Fasting trial
The fasting response is the better confirmatory test, and the osmotic gap is the better triage test — use them in that order.
Osmotic diarrhoea largely stops when the osmotic load is withdrawn, whereas secretory diarrhoea persists. That distinction is more robust than the calculated gap, particularly in the common indeterminate band, but it takes 24–48 hours of supervised fasting to establish. The gap is available from a single specimen and narrows the differential immediately, so it comes first and the fasting trial confirms.
Stool Osmotic Gap vs Rome IV functional diarrhoea
Sequential — the Rome IV criteria are a diagnosis of exclusion, and the osmotic gap is part of the exclusion.
Rome IV functional diarrhoea requires loose stools in the majority of stools for at least three months, in the absence of an identifiable organic cause and without predominant pain. The osmotic gap belongs to the work that establishes that absence: a clearly osmotic or clearly secretory gap points towards an organic mechanism worth pursuing before a functional label is applied.
Stool Osmotic Gap vs Bristol Stool Form Scale
Complementary and prerequisite — Bristol describes stool form, and the osmotic gap requires a form (types 6–7) before it can be interpreted at all.
Bristol classifies stool consistency from 1 to 7 and is the standard way of documenting that a diarrhoea is genuinely loose. Because stool electrolytes are uninterpretable on a formed sample, recording a Bristol type is a practical precondition for sending the test rather than a competing measure.
Pearls & pitfalls
- The sample must be liquid. Stool electrolytes on a formed or semi-formed specimen produce a number that cannot be interpreted.
- Do not substitute a measured stool osmolality for the 290. Fermentation continues after collection, so the measured value is falsely high and the resulting gap falsely low — and the error grows with transit delay.
- A negative gap is not brisk secretion. It is arithmetically impossible from real stool and indicates dilution or contamination, classically added water or urine in factitious diarrhoea.
- A measured stool osmolality well below 290 confirms that dilution, and is the one situation where measuring it is worthwhile.
- Both thresholds are strict. A gap of exactly 100 or exactly 50 is indeterminate, not osmotic or secretory respectively.
- The indeterminate band is common and is not a failure of the test. Move to a fasting trial rather than repeating the calculation.
- Ask about nocturnal symptoms explicitly. Diarrhoea that reliably stops overnight is unlikely to be secretory, whatever the number says.
- Some sources use 125 rather than 100 as the osmotic threshold. Note which you are using when comparing results across settings.
- Laxative misuse can produce either pattern — magnesium-containing agents osmotic, stimulant agents secretory — so a laxative screen is worth considering in both bands.
Critical actions
- Exclude infection before investigating chronic diarrhoea mechanistically.
- Send stool electrolytes only on a genuinely liquid sample, and note the collection-to-laboratory interval.
- Confirm an osmotic result with a fasting trial, and review diet and medications for lactose, sugar alcohols and magnesium.
- Where the picture is secretory, screen for bile-acid diarrhoea early — it is common, treatable and frequently missed.
- Arrange colonoscopy with random biopsies where microscopic colitis is plausible; the mucosa looks normal and targeted biopsies will miss it.
- Repeat on an observed specimen where the gap is negative, before pursuing anything else.
- Consider a laxative screen in both osmotic and secretory pictures.
Why this score exists
The counterintuitive design choice — using an assumed plasma osmolality of 290 instead of measuring the stool osmolality directly — is the most instructive thing about this test. Measuring the real value seems obviously better, and it is worse. Stool is a living culture, and bacterial fermentation continues after the sample is collected, generating osmotically active particles and driving the measured osmolality upward by an amount that depends on how long the pot sat before it reached the laboratory. Substituting a fixed physiological constant removes a variable that is pure artefact. The measured osmolality retains exactly one use, and it is a different one: a value substantially below plasma osmolality cannot arise physiologically and is evidence that the sample was diluted, which is how the test doubles as a check for factitious diarrhoea.
About the creator
First author, 1992 experimental derivation
Characterised the faecal osmotic gap and stool pH across experimentally induced diarrhoeas of differing cause.
Senior author; foundational work on intestinal fluid and electrolyte transport
Established much of the physiology of intestinal absorption and secretion on which the interpretation of the gap rests.
Limitations
- The indeterminate band between 50 and 100 mOsm/kg is wide and frequently occupied, so a substantial proportion of cases are not resolved by the test.
- Requires liquid stool, which excludes patients whose diarrhoea is intermittent or only mildly loose.
- Assumes stool water is in osmotic equilibrium with plasma at about 290 mOsm/kg, which holds in most circumstances but not in very rapid transit.
- The doubling of measured cations assumes electroneutrality with unmeasured anions, which is an approximation.
- Mixed mechanisms are common — malabsorption with bacterial overgrowth, for instance — and produce intermediate values that reflect reality rather than measurement error.
- Thresholds vary between sources, with 125 used in place of 100 in some, making values near the boundary non-comparable across settings.
- Identifies mechanism only. It does not name a cause, and both bands still require the full differential to be worked through.
If you are the patient
When diarrhoea has gone on for weeks, doctors want to know which of two things is driving it, because that decides what to test for next. Either something in the gut is drawing water in — an unabsorbed sugar, for example, or a magnesium-containing medicine — or the bowel lining is actively pumping fluid out. To tell these apart, a sample of liquid stool is tested for its salt content, and a simple calculation gives a number called the osmotic gap. A high number suggests the first type, which usually improves when you stop eating for a while and often traces back to something in the diet or a medication. A low number suggests the second type, which keeps going even when fasting and typically wakes people at night; that points towards causes such as bile-acid diarrhoea or inflammation of the bowel lining. Numbers in the middle do not settle the question, which is common and not a problem with the test. One useful extra: a result that comes out negative is not possible from real stool, and usually means the sample had water added to it.
Frequently asked questions
What is the stool osmotic gap?#
A calculated value, 290 − 2 × (stool sodium + stool potassium), that separates osmotic from secretory diarrhoea. Above 100 mOsm/kg indicates osmotic diarrhoea, below 50 indicates secretory, and values between the two are indeterminate.
Why does the formula use 290 rather than the measured stool osmolality?#
Because stool continues to ferment after collection, generating osmotically active particles. A measured stool osmolality is therefore falsely high — by an amount that grows the longer the sample sits — which makes the calculated gap falsely low. Substituting a fixed plasma osmolality removes an artefact rather than adding an approximation.
What does a high stool osmotic gap mean?#
That an unabsorbed osmotically active solute is drawing water into the bowel. Common causes are lactose and other carbohydrate malabsorption, sugar alcohols such as sorbitol and mannitol, magnesium-containing antacids and laxatives, and generalised malabsorption. Osmotic diarrhoea should largely stop with fasting.
What does a low stool osmotic gap mean?#
That electrolytes account for essentially all the stool osmolality, indicating active secretion. Consider bile-acid diarrhoea, microscopic colitis, stimulant laxative use, neuroendocrine tumours and infection. Secretory diarrhoea persists during fasting and characteristically continues overnight.
What does a negative stool osmotic gap mean?#
It is not physiologically possible from stool alone, so it indicates a diluted or contaminated specimen — most often water or dilute urine added to the sample, which is a recognised presentation of factitious diarrhoea. Repeat on an observed, properly collected specimen.
Is the threshold 100 or 125 mOsm/kg?#
Both are in use. This calculator uses above 100 for osmotic and below 50 for secretory, following the experimental derivation; some sources set the osmotic threshold at 125. Values near the boundary are therefore not directly comparable across settings, which is worth noting when reading a result from elsewhere.
Can the test be done on a formed stool sample?#
No. Stool electrolytes are only interpretable on genuinely liquid stool. A formed or semi-formed sample produces a number that means nothing, so documenting stool form — for instance with the Bristol scale — before sending the test is worthwhile.
What if the gap comes back in the indeterminate range?#
That is common and does not indicate a failed test — many cases are genuinely mixed. A supervised fasting trial discriminates better than repeating the calculation, and the common treatable causes are usually worth pursuing in parallel rather than waiting for the mechanism to declare itself.
References
Original / primary reference
Clinical application
- Schiller LR, Pardi DS, Sellin JH. Chronic Diarrhea: Diagnosis and Management. Clin Gastroenterol Hepatol. 2017;15(2):182-193.e3.
- Smalley W, Falck-Ytter C, Carrasco-Labra A, Wani S, Lytvyn L, Falck-Ytter Y. AGA Clinical Practice Guidelines on the Laboratory Evaluation of Functional Diarrhea and Diarrhea-Predominant Irritable Bowel Syndrome in Adults (IBS-D). Gastroenterology. 2019;157(3):851-854.