Strong Ion Difference Calculator
Strong Ion Difference Calculator
Apparent strong ion difference by the Stewart physicochemical approach — the account of acid-base balance that explains why the bicarbonate moved rather than recording that it did, and that predicts the acidosis large-volume saline causes. A second lens, not a replacement for the anion gap and base excess.
Apparent strong ion difference
6 inputs → SIDaNa⁺ 140, K⁺ 4.0, ionised Ca²⁺ 1.15, Mg²⁺ 0.9, Cl⁻ 105, lactate 1.0 mmol/L
Apparent strong ion difference
Effective SID = HCO₃⁻ + charge on albumin + charge on phosphate
Strong ion gap = SIDa − SIDe
- strong ions
- ions that are fully dissociated at physiological pH, so their concentration is fixed by how much is there rather than by the pH. Sodium, potassium, calcium, magnesium, chloride and lactate are the ones that matter in plasma
- × 2
- calcium and magnesium are divalent, so a concentration in mmol/L is doubled to give mEq/L of charge. Ionised calcium of 1.15 mmol/L is 2.30 mEq/L. Use IONISED calcium — bound calcium carries no free charge
- normal SIDa
- about 40–42 mEq/L in normal plasma. A fall is an acidosis and a rise an alkalosis, in this account
- urate
- some published versions subtract urate as well. It contributes roughly 0.3 mEq/L at a normal concentration, it is not reported in mEq/L on any report the reader will be holding, and it is omitted here — deliberately, and it is stated rather than hidden
- why SID changes pH
- plasma must stay electrically neutral. If the excess of strong cations over strong anions narrows, water dissociates further to supply the missing positive charge, and the hydrogen ion concentration rises. In Stewart’s account SID, pCO₂ and the total weak acid concentration are the three independent variables, and bicarbonate and pH are consequences of them
- the saline effect
- 0.9% sodium chloride contains 154 mmol/L of sodium and 154 mmol/L of chloride, so its own SID is zero. Infusing it lowers the plasma SID towards zero and causes an acidosis, without any acid having been given. This is the prediction the traditional approach cannot make as cleanly
- not computed here
- effective SID and the strong ion gap. The charge on albumin and phosphate is pH-dependent, the published equations differ, and a normal strong ion gap has been reported anywhere between 0 and 8 mEq/L. A number with that much disagreement behind it does not belong in a headline
Worked example
Na⁺ 140, K⁺ 4.0, ionised Ca²⁺ 1.15, Mg²⁺ 0.9, Cl⁻ 105, lactate 1.0 mmol/L
Strong cations: 140 + 4.0 + (2 × 1.15) + (2 × 0.9) = 140 + 4.0 + 2.30 + 1.80 = 148.1 mEq/L
Strong anions: 105 + 1.0 = 106.0 mEq/L
148.1 − 106.0 = 42.1 mEq/L — within the 40–42 mEq/L usually quoted as normal
Notice how little the divalent ions matter: calcium and magnesium together contribute 4.1 mEq/L of 148.1. Sodium and chloride do almost all the work, and a Stewart reading is largely a reading of the sodium-chloride difference
Give this patient two litres of 0.9% sodium chloride and the chloride might rise to 112 while the sodium rises to 142. The SID becomes 142 + 4.0 + 2.30 + 1.80 − 112 − 1.0 = 37.1 mEq/L — an acidosis, produced by fluid alone, with no acid administered and no unmeasured anion present
The same patient with a lactate of 6.0 instead of 1.0 gives 148.1 − 111.0 = 37.1 mEq/L. The identical number, an entirely different problem — which is why the SID is a starting point and not a diagnosis
The same disorders, described two ways
| Clinical situation | Traditional reading | Stewart reading |
|---|---|---|
| Large-volume 0.9% sodium chloride | Normal-anion-gap hyperchloraemic acidosis; the mechanism is often left vague | The fluid has an SID of zero, so it drags plasma SID down and water dissociates further. Predicted before it happens |
| Lactic acidosis | Raised anion gap from an unmeasured anion | Lactate is a strong anion; the SID falls directly |
| Vomiting or nasogastric loss | Chloride-responsive metabolic alkalosis | Chloride lost without sodium, so the SID widens |
| Hypoalbuminaemia | Lowers the anion gap; correct the gap before interpreting it | Albumin is a weak acid, so losing it is alkalinising. Handled in the effective SID, not the apparent one |
| Renal failure | Raised gap from retained phosphate, sulfate and urate | Unmeasured strong anions lower the SID; sulfate appears in the strong ion gap |
| Chronic respiratory acidosis | Bicarbonate rises as renal compensation | The kidney excretes chloride, which raises the SID. Same event, different vocabulary |
Fluid composition and what it does to the SID
| Fluid | Na⁺ (mmol/L) | Cl⁻ (mmol/L) | Effective SID of the fluid | Effect on plasma |
|---|---|---|---|---|
| Plasma (for comparison) | 140 | 100–105 | about 40 mEq/L | — |
| 0.9% sodium chloride | 154 | 154 | 0 | Lowers plasma SID — acidifying |
| Hartmann’s / compound sodium lactate | 131 | 111 | about 28 mEq/L (lactate is metabolised) | Close to neutral |
| Plasma-Lyte 148 | 140 | 98 | about 50 mEq/L (acetate and gluconate metabolised) | Slightly alkalinising |
| 5% glucose | 0 | 0 | 0, but it adds no strong ions | Dilutional; lowers SID by dilution |
| 8.4% sodium bicarbonate | 1000 | 0 | very high | Strongly alkalinising |
What this page does not calculate, and why
| Quantity | Why it is absent |
|---|---|
| Effective SID (SIDe) | Needs the charge on albumin and phosphate, which is pH-dependent and given by competing published equations |
| Strong ion gap (SIG) | Follows from SIDa − SIDe, and published normal values range from 0 to about 8 mEq/L depending on the equations used. A headline number cannot carry that much disagreement |
| Urate | Contributes about 0.3 mEq/L and is not reported in mEq/L anywhere the reader will be looking |
| Unmeasured cations | Lithium, and cationic paraproteins in myeloma, raise the SID and are invisible to this calculation, exactly as they distort the anion gap |
A different account of the same physiology
The traditional approach to acid-base balance treats bicarbonate as a variable in its own right: it falls, so an acidosis is present. Peter Stewart’s objection, published in 1983, was that bicarbonate cannot be an independent variable, because its concentration is fixed by the equilibria it sits in. In his account only three things are independent — the strong ion difference, the pCO₂, and the total concentration of weak acid, chiefly albumin and phosphate — and bicarbonate and pH are consequences of those three. The mechanism is electroneutrality: plasma must carry no net charge, so if the excess of strong cations over strong anions narrows, water dissociates further to supply the missing positive charge, and the hydrogen ion concentration rises. The acidosis is not caused by acid arriving; it is caused by the ionic environment changing.
Where that framing earns its keep is intravenous fluid. A litre of 0.9% sodium chloride contains 154 millimoles each of sodium and chloride, so its own strong ion difference is zero. Infusing it pulls the plasma strong ion difference down towards zero and produces an acidosis — reliably, predictably, and without a single molecule of acid having been given. The traditional approach can describe the result afterwards as a normal-anion-gap hyperchloraemic acidosis, but it does not predict it from the bag, and generations of clinicians have been puzzled by a patient who became more acidotic on being resuscitated. Stewart’s account predicts it from the label, and it is the reason balanced crystalloids exist.
The honest position on the rest of it is that the approach is contested. Critics argue that the mechanistic claim is not chemically correct, that the full calculation accumulates measurement error across six or more analytes, that the effective strong ion difference depends on albumin and phosphate charge equations that different authors write differently, and that no study has shown patients do better when it is used. Comparative work has repeatedly found that the Stewart method identifies the same disorders as the albumin-corrected anion gap and standard base excess, in different words. Most clinicians manage acid-base problems perfectly well without it, and nothing on this page suggests otherwise.
So it is best used as a second lens. When the anion gap and the base excess agree and the story is clear, the strong ion difference adds vocabulary and not information. When they do not — a patient who is acidotic with a normal gap, a patient whose acidosis deepened during resuscitation, a patient with a profoundly low albumin whose gap cannot be trusted — looking at the sodium, the chloride and the lactate as a set of charges often makes the mechanism obvious. As with every calculation in this category, the number supports a clinician’s judgement rather than replacing it, and the blood gas is read alongside the patient rather than instead of them.
Frequently asked questions
What is a normal strong ion difference?
About 40 to 42 mEq/L for the apparent strong ion difference in normal plasma. A lower value indicates a metabolic acidosis in the Stewart account and a higher value a metabolic alkalosis. Because sodium and chloride dominate the calculation, a quick approximation is the sodium-chloride difference, which is normally about 35 mmol/L.
Why does normal saline cause an acidosis?
Because 0.9% sodium chloride contains 154 mmol/L each of sodium and chloride, so its own strong ion difference is zero. Infusing it lowers the plasma strong ion difference, and electroneutrality forces water to dissociate further, raising the hydrogen ion concentration. No acid is given; the ionic environment changes. This is the prediction the Stewart approach makes that the traditional one does not.
Should I use the Stewart approach instead of the anion gap?
No. The approach is contested, the full calculation depends on charge equations that different authors write differently, and comparative studies have found it identifies the same disorders as the albumin-corrected anion gap and standard base excess. Use it as a second lens — particularly for fluid-induced acidosis — alongside the methods that already work.
What is the strong ion gap, and why is it not calculated here?
It is the apparent strong ion difference minus the effective strong ion difference, and it represents unmeasured strong anions — the Stewart equivalent of a raised anion gap. It is not calculated here because the effective SID needs pH-dependent charge equations for albumin and phosphate on which published sources differ, and reported normal values for the gap range from 0 to about 8 mEq/L.
Should I use total or ionised calcium?
Ionised. Only the free ion carries charge in this account; calcium bound to albumin does not. Enter it in mmol/L — the calculator doubles it, because calcium is divalent and a concentration of 1.15 mmol/L is 2.30 mEq/L of charge. In practice calcium and magnesium together contribute only about 4 mEq/L of a total near 148, so the sodium and chloride dominate the result.
Related calculators
References
- Stewart PA. Modern quantitative acid-base chemistry. Can J Physiol Pharmacol. 1983;61(12):1444–61.
- Kellum JA. Clinical review: reunification of acid-base physiology. Crit Care. 2005;9(5):500–7.
- Figge J, Mydosh T, Fencl V. Serum proteins and acid-base equilibria: a follow-up. J Lab Clin Med. 1992;120(5):713–19.
- Morgan TJ. The Stewart approach — one clinician’s perspective. Clin Biochem Rev. 2009;30(2):41–54.
- Dubin A, Menises MM, Masevicius FD, et al. Comparison of three different methods of evaluation of metabolic acid-base disorders. Crit Care Med. 2007;35(5):1264–70.
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.