Arterial Blood Gas Interpreter
Arterial Blood Gas Interpreter
Read a whole arterial blood gas rather than one number from it: pH, pCO₂ and bicarbonate together name the primary disorder, say whether it is acute or chronic, and test the compensation against the published expected range — where a value outside that range means a second disorder, not over-compensation.
Arterial blood gas
pH + pCO₂ + HCO₃⁻ → disorderpH 7.28, pCO₂ 30 mmHg, bicarbonate 14 mmol/L, acute
Expected compensation, and where each formula comes from
| Primary disorder | Expected compensation (Boston rules) | Tolerance | Stated as a slope |
|---|---|---|---|
| Metabolic acidosis | pCO₂ = 1.5 × HCO₃⁻ + 8 | ± 2 mmHg | pCO₂ falls about 1.2 mmHg per 1 mmol/L fall in HCO₃⁻ |
| Metabolic alkalosis | pCO₂ = 0.7 × HCO₃⁻ + 20 | ± 5 mmHg | pCO₂ rises 0.6–0.75 mmHg per 1 mmol/L rise in HCO₃⁻, and should not exceed 55 mmHg |
| Acute respiratory acidosis | HCO₃⁻ = 24 + (pCO₂ − 40) ÷ 10 | ± 2 mmol/L | HCO₃⁻ rises 1–2 mmol/L per 10 mmHg rise in pCO₂ |
| Chronic respiratory acidosis | HCO₃⁻ = 24 + 4 × (pCO₂ − 40) ÷ 10 | ± 2 mmol/L | HCO₃⁻ rises 3–4 mmol/L per 10 mmHg rise in pCO₂ |
| Acute respiratory alkalosis | HCO₃⁻ = 24 − 2 × (40 − pCO₂) ÷ 10 | ± 2 mmol/L | HCO₃⁻ falls 1–2 mmol/L per 10 mmHg fall in pCO₂ |
| Chronic respiratory alkalosis | HCO₃⁻ = 24 − 5 × (40 − pCO₂) ÷ 10 | ± 2 mmol/L | HCO₃⁻ falls 4–5 mmol/L per 10 mmHg fall in pCO₂ |
Reading a gas in the order that finds mixed disorders
| Step | Question | Why this order |
|---|---|---|
| 1 | Do the three numbers fit Henderson-Hasselbalch? | An analyser calculates bicarbonate from pH and pCO₂, so a gas is always self-consistent. If it is not, one value is mistyped, the pCO₂ unit is wrong, or the bicarbonate is a total CO₂ from a different sample |
| 2 | Is the pH high, low, or normal? | Names the direction, and nothing else. A normal pH does not mean a normal acid-base state |
| 3 | Which of pCO₂ and HCO₃⁻ moved in the direction that would cause that pH? | That one is the primary disorder. The other is the compensation, or a second disorder |
| 4 | If the primary disorder is respiratory, is it acute or chronic? | Renal compensation takes 2–5 days. Applying the chronic expectation to an acute gas invents a metabolic acidosis; applying the acute one to a chronic gas invents a metabolic alkalosis |
| 5 | Is the compensation inside the expected range? | Outside means a second disorder. It never means over-compensation |
| 6 | What is the anion gap? | A raised gap is a metabolic acidosis whether or not the bicarbonate is low, and it is the only way to find an acidosis hidden behind an alkalosis. Correct it for albumin |
| 7 | In a raised-gap acidosis, what is the delta ratio? | Separates a pure high-gap acidosis from one with a hyperchloraemic or alkalotic process alongside it |
| 8 | What does the patient look like? | The gas has now been read. The diagnosis has not been made |
The patterns this page is designed to catch
| Gas | Naive reading | What it actually is |
|---|---|---|
| pH 7.31, pCO₂ 38, HCO₃⁻ 18 | Metabolic acidosis, partly compensated | Winter's predicts 35 ± 2. A pCO₂ of 38 is at the ceiling — the respiratory response is inadequate and a second respiratory acidosis is present |
| pH 7.40, pCO₂ 40, HCO₃⁻ 24 | Entirely normal | Possibly. Also possible: a metabolic acidosis and a metabolic alkalosis of matched size. The anion gap is the only way to tell |
| pH 7.33, pCO₂ 60, HCO₃⁻ 31 | Acute respiratory acidosis | The bicarbonate is far above an acute rise. This is chronic, or acute on chronic |
| pH 7.47, pCO₂ 25, HCO₃⁻ 18 | Respiratory alkalosis | Acute expectation is 21 mmol/L. A bicarbonate of 18 means a metabolic acidosis too — think salicylate, think sepsis |
| pH 7.52, pCO₂ 48, HCO₃⁻ 38 | Over-compensated metabolic alkalosis | There is no such thing. Expected pCO₂ is 46.6 ± 5, so 48 is appropriate — this is a simple metabolic alkalosis |
Why a compensation outside the expected range is a second disorder
Compensation is not a correction. When a metabolic acidosis develops, chemoreceptors sense the fall in pH and increase minute ventilation, which lowers the pCO₂ and pulls the pH back towards normal. The stimulus for that response is the pH displacement itself, so as the pH is restored the stimulus weakens. The response therefore stops short — it never returns the pH to 7.40 and it never carries it past. That single physiological fact is what makes every expected-compensation formula on this page useful: each one predicts a narrow band, and a measured value outside that band cannot be explained by the primary disorder at all.
The practical consequence is the most important thing on this page. A pCO₂ higher than Winter's formula predicts is not a metabolic acidosis that is only partly compensated; it is a metabolic acidosis with a respiratory acidosis alongside it. A pCO₂ lower than predicted is not enthusiastic compensation; it is a second, respiratory alkalosis. The phrase over-compensation describes something the body does not do, and using it retires the second diagnosis before anybody looks for it. In a patient with a metabolic acidosis who is tiring, that second diagnosis is impending respiratory failure, and the gas is the thing that announces it.
The second thing this page insists on is chronicity. The kidney takes two to five days to change the bicarbonate meaningfully, so the expected bicarbonate for a given pCO₂ depends entirely on how long the pCO₂ has been there. At a pCO₂ of 70 mmHg the acute expectation is a bicarbonate of 27 and the chronic expectation is 36 — a nine-millimole gap that is the difference between a normal metabolic state and a substantial metabolic acidosis or alkalosis. Nothing in the three numbers reveals which applies. It comes from the history, from previous results, and from the patient, which is why the selector on this calculator is a clinical question rather than a formality.
Finally, three numbers are not an acid-base assessment. A pH inside the reference interval is produced by no disorder, by a fully compensated one, or by two opposing disorders of matched severity, and the last of those is invisible here. Only the anion gap finds it, only the albumin-corrected gap finds it in a hypoalbuminaemic patient, and only the delta ratio says whether a raised gap accounts for the whole fall in bicarbonate. None of those needs a pO₂, and none of them replaces looking at the patient: this page reads a gas, and a clinician diagnoses a person.
Frequently asked questions
What does it mean if compensation is outside the expected range?
It means a second acid-base disorder is present. It does not mean the compensation has overshot — the body does not over-compensate, because the stimulus for compensation is the pH displacement itself and it fades as the pH is restored. A pCO₂ above the range Winter's formula predicts for a metabolic acidosis is a concurrent respiratory acidosis; a pCO₂ below it is a concurrent respiratory alkalosis.
How do I tell an acute respiratory acidosis from a chronic one?
By the bicarbonate, and by the history. Acutely, bicarbonate rises only about 1 mmol/L per 10 mmHg of pCO₂ above 40, because that is all immediate buffering can do. After two to five days of renal compensation it rises about 4 mmol/L per 10 mmHg. At a pCO₂ of 70 those expectations are 27 and 36 mmol/L. The gas cannot tell you which the patient is; previous results and the clinical history can.
Why does this calculator check whether the numbers are consistent?
Because a blood gas analyser does not measure bicarbonate — it calculates it from the measured pH and pCO₂ by the Henderson-Hasselbalch equation, so three values from one gas always agree. If they do not, a value has been mistyped, the pCO₂ has been entered in the wrong unit, or the bicarbonate has come from a separate venous biochemistry panel as a total CO₂. Interpreting an inconsistent trio interprets a patient who does not exist.
Can a normal pH mean there is no acid-base disorder?
No. A normal pH is produced by no disorder, by a fully compensated single disorder, or by two opposing disorders of matched severity. The third is common and invisible on pH, pCO₂ and bicarbonate alone. The anion gap is what finds it, which is why a normal-looking gas in an unwell patient is a reason to calculate the gap rather than to stop.
Does this page tell me about oxygenation?
No. pH, pCO₂ and bicarbonate describe acid-base state only. Oxygenation is a separate question answered by the pO₂ read against the inspired oxygen fraction, by the A-a gradient and by the P/F ratio. A patient can have a perfectly normal acid-base picture and be profoundly hypoxaemic.
Should I enter pCO₂ in kPa or mmHg?
Either — set the selector to match your analyser. Every compensation formula in use was published in mmHg, so a kPa entry is multiplied by 7.50062 before the arithmetic. A normal arterial pCO₂ is 35–45 mmHg or 4.7–6.0 kPa, and entering 40 kPa where 40 mmHg was meant is caught by the consistency check.
Related calculators
References
- Albert MS, Dell RB, Winters RW. Quantitative displacement of acid-base equilibrium in metabolic acidosis. Ann Intern Med. 1967;66(2):312–22.
- Schwartz WB, Relman AS. A critique of the parameters used in the evaluation of acid-base disorders. N Engl J Med. 1963;268:1382–8.
- Berend K, de Vries APJ, Gans ROB. Physiological approach to assessment of acid-base disturbances. N Engl J Med. 2014;371(15):1434–45.
- Adrogué HJ, Madias NE. Management of life-threatening acid-base disorders. N Engl J Med. 1998;338(1):26–34 and 338(2):107–11.
- Merck Manual Professional Edition. Acid-Base Disorders: Primary Changes and Compensations in Simple Acid-Base Disorders. Rahway, NJ: Merck & Co.
- Seifter JL. Integration of acid-base and electrolyte disorders. N Engl J Med. 2014;371(19):1821–31.
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.