Anion Gap in the ICU
In the intensive care unit the anion gap is a trending tool, not a one-off number. Serial measurements track the rise and fall of unmeasured anions in shock, sepsis, and multi-organ failure — and the gap that fails to close carries an independent mortality signal.
Why the ICU demands serial anion gap monitoring
A single anion gap in a critically ill patient is a snapshot; the trend is the diagnostic. Serial anion gap measurement — typically every 6 to 12 hours in shock — reveals whether the underlying acidosis is worsening, resolving, or plateauing. In septic shock the gap tracks lactate clearance: a falling gap parallels resolution of tissue hypoperfusion, while a gap that climbs despite resuscitation signals ongoing ischaemia, uncontrolled source, or a second insult such as abdominal compartment syndrome.
This makes the anion gap an inexpensive, rapidly available marker of resuscitation adequacy, available wherever a basic metabolic panel is run. Use the anion gap calculator at each time point and chart the trend alongside arterial lactate — the two move together in pure lactic lactic acidosis but diverge when other unmeasured anions (ketones, toxins, uremic acids) are contributing.
An elevated anion gap that persists or rises after 24 hours of apparently adequate resuscitation is a red flag. Reassess for ongoing sources: un-drained infection, ischaemic bowel, continued seizure, occult toxic alcohol ingestion, or adrenal insufficiency. Persistent unexplained acidosis independently predicts ICU mortality.
Unmeasured anions and mortality
Multiple ICU studies have shown that an unexplained anion gap elevation — one that cannot be fully accounted for by lactate, albumin, phosphate, or known toxins — is associated with increased mortality independent of illness severity scores (Kellum, Critical Care Medicine; Surviving Sepsis Campaign). The "unmeasured anions" responsible are incompletely characterised but include organic acid intermediates from hypoperfused tissue, sulfates, and other metabolic products that the failing circulation cannot clear and the failing kidney cannot excrete.
The practical implication is that in a critically ill patient, an anion gap of 20 mEq/L with a normal lactate should not be dismissed. It is a real signal of unmeasured acids and a marker of physiological reserve exhaustion. Worked through a full case of sepsis with lactic acidosis to see how the gap and lactate are interpreted together at the bedside.
The strong ion gap (SIG)
Because the standard anion gap lumps together albumin, phosphate, and a residual "strong ion difference" that it cannot resolve, intensivists interested in a more sensitive acid-base tool use the strong ion gap (SIG). The SIG is derived from the Stewart quantitative approach, which treats plasma pH as a function of three independent variables: the strong ion difference (SID, dominated by Na⁺, K⁺, Cl⁻), the total weak acid concentration (albumin and phosphate), and PaCO₂ (Stewart, 1981; Kellum, Critical Care Medicine).
The SIG is the difference between the apparent SID (from measured strong ions) and the effective SID (from charges on albumin and phosphate). In health it is near zero; in critical illness it rises, reflecting unmeasured anions that the conventional anion gap misses — particularly when albumin is low. The SIG therefore corrects for the two principal confounders of the standard anion gap (hypoalbuminemia and hyperphosphatemia) and is a more sensitive detector of occult metabolic acidosis in ICU patients.
Whether the extra sensitivity translates into better outcomes remains debated. Most intensivists use the standard albumin-corrected anion gap as the everyday tool — it is faster, requires fewer inputs, and is familiar — and reserve the full Stewart/SIG calculation for the small fraction of patients whose acid-base disturbance is complex or unexplained by the gap alone.
Practical ICU workflow
At every arterial or venous blood gas, calculate the anion gap (corrected for albumin), compute the delta gap calculator to expose mixed disorders, and plot the trend. The most common ICU pattern is a mixed metabolic acidosis — a high anion gap component (lactate, ketones, uremia) sitting on top of a hyperchloremic component from aggressive normal saline resuscitation. The delta ratio sorts these out: a ratio in the 0.4–0.8 band is the classic mixed pattern of sepsis resuscitated with chloride-rich fluids.
| ICU scenario | Anion gap pattern | Implication |
|---|---|---|
| Septic shock, early | Rising rapidly with lactate | Ongoing hypoperfusion — escalate resuscitation. |
| Septic shock, post-resuscitation | Falling slowly | Lactate clearance — favourable sign. |
| Large-volume saline resuscitation | Normal or low with hyperchloremia | Hyperchloremic (NAGMA) component — consider balanced crystalloid. |
| Renal failure in multi-organ dysfunction | Progressive rise | Uremic acid retention — consider renal replacement therapy. |
| Persistent unexplained elevation | High despite normal lactate | Unmeasured anions — independent mortality risk. |