Anion Gap: What It Means on Your Blood Test and What It Means for Athletes
Table of Contents
Key Takeaways: Anion Gap
- The anion gap is calculated, not measured: sodium minus the sum of chloride and bicarbonate. Normally about 80% of it reflects the negative charge on blood proteins, chiefly albumin.[1]
- There is no universal normal value. Reference texts list ranges from 3–9 mEq/L (ion-selective electrode methods) to 8–16 mEq/L, so a result must be read against the reporting laboratory’s own range.[2]
- A high value most often signals metabolic acidosis, but laboratory error, sample handling and a raised albumin can also push it up.[1]
- After a 2-minute all-out cycling test, plasma lactate rose from 1.4 to 20.0 mmol/L, and lactate and bicarbonate had still not returned to baseline an hour later.[3]
- The gap is a weak lactate detector. A single measurement is not recommended as a stand-alone risk marker in critically ill patients, and an individual baseline has been proposed as a potentially useful reference, although its added diagnostic value remains uncertain.[2][4][5]
- In my clinical experience, the anion gap is not a routine test and is used mainly in acute and specialist care, while for acid–base assessment I usually rely on a blood gas analysis. It is a supporting finding, interpreted together with other results and the patient’s symptoms, and being an athlete is not in itself a reason to measure it.
Introduction: A Calculated Number With an Athletic Twist
The anion gap can appear on chemistry reports, sometimes without explanation. It is calculated from three electrolytes that a metabolic panel can provide, and clinicians use it to classify acid–base disorders, catch some laboratory errors, and raise suspicion of conditions such as multiple myeloma or bromide and lithium poisoning.[1]
Many patients have probably never heard of the anion gap, and on a laboratory report it can look odd or even alarming. In my experience it is not usually requested as a routine test, so it is not an everyday test for clinicians either. It is closer to an acute, emergency-type test, used mainly in specialist care, while in non-urgent medicine, primary care or occupational health I see it used rarely. The picture is different in an emergency department or on an internal medicine ward, where an elevated gap is common among hospital patients.[1] In practice, it tends to come up when the cause of an acidosis is being worked out.
In my clinical experience, I usually rely on a blood gas analysis (Astrup) when assessing acute acid–base disturbances, because it shows pH and carbon dioxide directly. The anion gap serves a different purpose in characterising a metabolic acidosis, and in the workplaces where I have worked it has usually not appeared directly on the blood gas report. This may partly explain why I see it so rarely, and why it can be an unfamiliar derived quantity even to many clinicians.
For athletes there is a twist. Intense exercise produces large amounts of lactate, exactly the kind of “unmeasured” anion that widens the gap,[1][3] and exercise also concentrates the blood and raises plasma albumin.[3] A raised value in an athlete can therefore reflect lactate from a recent hard session.
This article covers how the value is calculated, what high and low results usually mean, where the calculation falls short, and how exercise, hydration and diet influence it.
What Is the Anion Gap?
Blood is electrically neutral, but a standard panel measures only some of the ions. The anion gap is the difference between the main measured cation and the two main measured anions:
Anion gap = Na⁺ − (Cl⁻ + HCO₃⁻)
Under normal conditions about 80% of the gap comes from negative charges on circulating proteins, of which albumin is the most abundant.[1] Potassium can be added to the sodium side, but because its concentration is small, it is often left out.[1]
In a healthy person much of the gap therefore reflects plasma proteins, chiefly albumin; it can rise when something adds organic anions such as lactate or ketoacids.[1] (For these single-charged ions, mEq/L and mmol/L are numerically identical.)
Normal Anion Gap Range: Why There Is No Single Number
A review of major textbooks and clinical references found definitions ranging from about 12 mEq/L with a range of 8–16 mEq/L, to 3–9 or 3–11 mEq/L when sodium and chloride are measured with ion-selective electrodes.[2] Electrode methods give lower averages mainly because they report higher chloride values, and whatever the method, the normal range within a population can span 8–10 mEq/L.[2]
Two consequences follow. First, a result must be read against the range of the laboratory that produced it.[1] Second, because healthy people differ so much, a real change in one person can stay inside the reference interval: in the authors’ illustrative example, someone whose baseline sits at the bottom of the normal range could rise by about 8 mEq/L before being flagged.[2]
For a population reference point, the mean albumin-adjusted value among 2,714 US adults aged 20–49 years in a national health survey was 10.26 mEq/L.[6]
What Does a High Anion Gap Mean?
An elevated value is common among hospital patients, and its most frequent cause is metabolic acidosis from excess acid production or reduced acid excretion.[1] When an acid accumulates whose anion is not chloride, such as lactate in lactic acidosis or β-hydroxybutyrate and acetoacetate in ketoacidosis, its hydrogen ions consume bicarbonate while the anion stays in the blood, so the gap widens.[1] The same principle applies when failing kidneys retain anions, and in methanol or ethylene glycol poisoning.[1]
Not every metabolic acidosis widens the gap. When the problem is bicarbonate loss, through the bowel or through the urine as in some forms of renal tubular acidosis, chloride replaces bicarbonate and the gap stays normal: a normal anion gap, or hyperchloremic, acidosis.[1]
In my clinical experience, the situations in which the anion gap may be used are largely the same ones in which a blood gas analysis is often needed. These are mostly emergency presentations involving metabolic disturbances: the workup of diabetic ketoacidosis, acidosis of any cause and intoxications. In these cases the gap may be used alongside the blood gas result. The physiology fits this pattern, since ketoacidosis and methanol or ethylene glycol poisoning lead to accumulation of acids whose anions the standard panel does not measure.[1]
A high value without acidosis also occurs. Recognised causes include laboratory error, severe volume depletion with a raised albumin, metabolic alkalosis, severe hyperphosphatemia and anionic paraproteins.[1] In metabolic alkalosis from vomiting or diuretics, the gap can rise by about 4–6 mEq/L, mostly because albumin increases.[1] Handling matters too: serum left exposed to air for more than an hour loses bicarbonate artificially, which raises the calculated gap.[1]
What Does a Low Value Mean?
A low gap is relatively uncommon. Laboratory error is the most frequent explanation and low albumin the second.[1] Less often it reflects an excess of positively charged IgG antibodies, as in IgG myeloma or polyclonal gammopathy, or intoxication with lithium, bromide or iodide.[1]
If blood is not separated promptly, white cells keep producing carbon dioxide in the tube, falsely raising bicarbonate and lowering the gap by as much as 2–3 mEq/L.[1] If a low gap persists after laboratory error and low albumin have been ruled out, IgG accumulation is one possible explanation to consider.[1]
Where the Calculation Falls Short
Albumin shifts the baseline. Each 1 g/dL fall in albumin lowers the gap by about 2.3–2.5 mEq/L, and a rise raises it by a similar amount.[2] The usual correction is: corrected gap = gap + 2.5 × (normal albumin − measured albumin, in g/dL).[2] It is not a cure-all. In 639 sets of hospital laboratory values, an uncorrected gap of 12 or more identified lactate of at least 2.5 mmol/L with 39% sensitivity and 89% specificity; the corrected gap reached 75% sensitivity but only 59% specificity, and discrimination was practically identical (AUC 0.757 versus 0.750).[7]
It is a poor lactate test. A systematic review of nine studies covering 4,504 samples from 2,111 critically ill patients found that the gap had high specificity but low sensitivity for lactate above 2.5 mmol/L: in the three studies using a 12 mEq/L cut-off, sensitivity was 0.39–0.57 and specificity above 0.8.[4] Discrimination was better for lactate above 4 or 5 mmol/L (summary AUC 0.87), yet the authors concluded that the gap should not replace lactate measurement.[4] Kraut and Nagami agree: when lactic acidosis is suspected, measure lactate directly.[2]
A single value is a weak risk marker. A systematic review of 19 studies including 12,497 critically ill patients concluded that one measurement cannot be recommended for risk stratification, noting that intravenous fluids may affect how the gap performs as a prognostic marker.[5]
These data come from hospital populations, not athletes. Within them, the gap discriminated severe hyperlactatemia better than mild elevations.[4]
This is also how the gap is used in practice. From a clinical perspective, it works as a supporting finding rather than a stand-alone test: in my experience it does not usually lead to clinical decisions by itself, and it is typically interpreted together with other laboratory results and in relation to the patient’s symptoms. That approach is consistent with the evidence above, which does not support relying on a single measurement for risk stratification in critically ill patients.[5]
Anion Gap in Athletes: What Training Does to the Number
In my clinical practice, being an athlete is not in itself a reason to measure the anion gap. The indication is clinical and, in my experience, the same for athletes and for sedentary people: as a rule it follows the clinical assessment and the patient’s symptoms. Recent or regular exercise is part of the history and is taken into account. Hard exercise can transiently load the blood with lactate and lower bicarbonate, as the data below show,[3] but in my view this seldom changes the clinical decision.
Hard exercise adds lactate, an “unmeasured” anion
The standard formula does not subtract lactate, so when lactate accumulates and consumes bicarbonate, the conventional gap widens.[1] In athletes the quantities are large. When seven healthy young men cycled to exhaustion over roughly two minutes, arterial plasma lactate climbed from 1.4 to 20.0 mmol/L while plasma bicarbonate fell from 24.0 to 14.1 mmol/L; plasma sodium rose by about 10 mmol/L, mainly because the blood became more concentrated.[3] In another study, five healthy men performing four 30-second maximal cycling bouts reached plasma lactate of 21 meq/L by the end of the fourth bout.[8]
To make this concrete, Medbø and colleagues report arterial plasma values of sodium 143.9, chloride 107.9 and bicarbonate 14.1 mmol/L in samples drawn 20–30 seconds after their 2-minute test, against 134.5, 101.8 and 24.0 mmol/L before.[3] The authors’ own gap is a different quantity from the conventional gap above, because it includes potassium and subtracts measured lactate; it fell in arterial plasma from 11.1 to 8.1 mmol/L and was restored one hour after exercise.[3]
Recovery takes time. One hour after the all-out effort, every plasma electrolyte except lactate and bicarbonate was back to its pre-exercise value, and arterial pH was still below its pre-exercise value.[3] After a maximal incremental cycling test, eight moderately trained men still had plasma lactate of 9.2 mEq/L after 20 minutes of passive rest.[9]
Dehydration and albumin
Because albumin carries most of the unmeasured negative charge, anything that concentrates it raises the gap, by roughly 2.3–2.5 mEq/L per 1 g/dL, and severe volume depletion is a recognised cause of an elevated value.[1][2] Even a 2-minute maximal effort concentrated the blood measurably: haemoglobin rose by 8%, and arterial plasma albumin increased from 0.640 to 0.714 mmol/L.[3] An athlete who arrives for blood work dehydrated, after a hot session or during a weight cut, may therefore have a higher albumin concentration, which could contribute to a higher gap, although the exercise study cited here did not examine heat exposure or weight cutting.
Low-carbohydrate and ketogenic diets
Ketoacids are classic gap-widening anions,[1] which raises the question of acid–base balance in athletes on ketogenic diets. In 24 elite race walkers, 21 days of a ketogenic low-carbohydrate, high-fat diet produced no significant differences in blood pH, bicarbonate or lactate compared with a high-carbohydrate diet, before or after a graded exercise test, despite a higher calculated dietary acid load.[10] Two caveats: the study did not report the anion gap itself, and the authors note that shorter exposures of up to three days to major macronutrient shifts do appear to alter acid–base status.[10] These findings come from elite race walkers after 21 days on the diet.
Fitness and the resting value
At rest, the gap may carry a faint signal about aerobic fitness. In the 2,714 adults aged 20–49 years mentioned earlier, each standard-deviation (2.36 mEq/L) higher albumin-adjusted value was associated with an odds ratio for low cardiorespiratory fitness of 1.30 (95% CI 1.15–1.48) and a 0.97 mL/kg/min lower estimated VO₂max after multivariable adjustment.[6] VO₂max was estimated from a submaximal treadmill test, the design was cross-sectional, and the authors state that causation cannot be inferred.[6] It is a population association, not a fitness test.
Conclusion: Anion Gap
The anion gap is a simple subtraction that in a healthy person largely reflects plasma proteins, chiefly albumin, and that can rise when acids accumulate in disease.[1] Its range varies by laboratory method, it can stay “normal” despite a meaningful individual change, and it is a poor substitute for measuring lactate.[2][4] In my experience it is rarely used and mostly in acute and specialist care, where I usually rely on a blood gas analysis for acid–base assessment, so it can be unfamiliar even to many clinicians. In my clinical experience the gap works as a supporting finding rather than a stand-alone test: it is interpreted together with other laboratory results and the patient’s symptoms, and the indication to measure it is clinical, the same for athletes and for sedentary people. For athletes, a sample drawn soon after intense exercise may reflect transient changes in lactate and electrolytes rather than resting values.[3] In my view, recent exercise seldom changes the clinical decision.
References
[1] https://doi.org/10.2215/CJN.03020906
[2] https://doi.org/10.2215/CJN.04040413
[3] https://doi.org/10.12697/akut.2009.14.06
[4] https://doi.org/10.1186/s13104-017-2853-9
[5] https://doi.org/10.1186/s12871-016-0241-y
[6] https://pmc.ncbi.nlm.nih.gov/articles/PMC3340439/
[7] https://doi.org/10.1136/emj.2005.031898
[8] https://doi.org/10.1152/ajpregu.1992.262.1.R126
