01Core principlesThe concepts and mechanisms needed to understand the subject.
A blood gas describes interacting respiratory and metabolic influences on hydrogen-ion concentration. Carbon dioxide reflects alveolar ventilation relative to production, while bicarbonate represents the metabolic component and renal adaptation. A label such as metabolic acidosis is therefore only the beginning of interpretation: it does not identify the source of acid, the adequacy of compensation or the required treatment.
The clinical setting controls the meaning of the numbers. Oxygenation must be judged against inspired oxygen, and arterial and venous oxygen tensions are not interchangeable. Compensation usually moves pH toward normal but cannot safely be assumed from direction alone. Two primary disturbances may oppose each other and produce a deceptively ordinary pH.
Key points
- Check whether the sample is arterial or venous and record oxygen delivery before interpreting it.
- Acidaemia means pH below 7.35; alkalinaemia means pH above 7.45.
- Examine carbon dioxide and bicarbonate even when the pH falls within the reference range.
- Calculate the anion gap as sodium minus the sum of chloride and bicarbonate, using consistent units and laboratory conventions.
- For metabolic acidosis, expected PaCO2 in mmHg is approximately 1.5 times bicarbonate plus eight, with a two-mmHg range.
- Treat the cause and verify the response; bicarbonate administration is not an automatic response to a low pH.
02Mechanisms and patternsImportant relationships and how to distinguish them.
Acidaemia with elevated carbon dioxide suggests respiratory acidosis; alkalinaemia with reduced carbon dioxide suggests respiratory alkalosis. Consider airway disease, respiratory drive, neuromuscular function and ventilator settings.
Reduced bicarbonate with acidaemia supports metabolic acidosis; elevated bicarbonate with alkalinaemia suggests metabolic alkalosis. Vomiting, diuretics, renal dysfunction, lactate and ketones provide different explanatory pathways.
A pH near normal with substantially abnormal carbon dioxide and bicarbonate may represent compensation or opposing primary processes. The magnitude and duration of change help distinguish them.
Severe acidaemia, exhaustion, shock, dysrhythmia or altered consciousness requires physiological support while interpretation proceeds. A technically elegant calculation must not defer treatment of respiratory failure.
03Interpreting evidenceInformation, measurements and their limitations.
Consider the information, its meaning and its limitations before deciding what follows.
- 01
Sample and analyser checks - Why
- Establish whether the values answer the intended clinical question.
- Interpretation and limitations
- Check patient identity, sampling time, oxygen delivery and arterial versus venous origin. Air contamination and delayed analysis can distort results and warrant repeat sampling if inconsistent.
- 02
Electrolytes and albumin - Why
- Calculate and contextualise the measured anion gap.
- Interpretation and limitations
- Using mmol/L, anion gap equals sodium minus the sum of chloride and bicarbonate. Hypoalbuminaemia lowers the usual gap, so an apparently normal result may conceal unmeasured acids.
- 03
Lactate, glucose and blood ketones - Why
- Identify common causes of raised-gap metabolic acidosis.
- Interpretation and limitations
- Measure rather than infer ketones from glucose alone. Lactate and ketones can coexist, and a falling glucose does not by itself demonstrate resolution of ketoacidosis.
- 04
Compensation calculation and serial gases - Why
- Test whether an additional primary disturbance is present.
- Interpretation and limitations
- Use the correct carbon-dioxide units. A measured PaCO2 above the expected range during metabolic acidosis suggests additional respiratory acidosis, such as fatigue or depressed drive.
04Applied reasoningWorked examples connecting principles to decisions.
01Worked exampleRaised anion gap with inadequate ventilationAn acutely ill adult has pH 7.10, bicarbonate 12 mmol/L, PaCO2 5.3 kPa, sodium 140 mmol/L and chloride 100 mmol/L; albumin is normal.+
- 1Identify acidaemia with low bicarbonate, making metabolic acidosis a major component. Calculate the anion gap as 140 − (100 + 12), giving 28 mmol/L.
- 2Calculate expected PaCO2 as 1.5 multiplied by 12 plus 8, giving 26 mmHg with an approximate range of 24–28 mmHg, equivalent to about 3.2–3.7 kPa.
- 3Convert the measured 5.3 kPa to approximately 40 mmHg. This is above the expected compensatory range, identifying an additional respiratory acidosis rather than adequate compensation.
- 4Assess and support ventilation urgently while testing lactate, ketones and renal function to explain the raised gap; investigate toxic exposure when the history or results support it.
- 5Verify the arithmetic and repeat clinical and gas assessment after intervention. Improved pH alone is insufficient unless ventilation, perfusion and the identified underlying process also improve.
02Respiratory patternDistinguishing acute from chronic carbon-dioxide retentionCarbon dioxide is high, but the severity of acidaemia and bicarbonate elevation vary between successive samples.+
- 1Compare with previous gases and the known respiratory history; substantial renal bicarbonate retention requires time and can indicate chronic adaptation.
- 2Look for an acute deterioration superimposed on chronic retention when pH falls, conscious level worsens or carbon dioxide rises above the usual baseline.
- 3Direct treatment at ventilation and the precipitating cause, using serial gases to avoid both persistent failure and inappropriate overcorrection.
03Normal pH trapOpposing abnormalities in the same patientA patient with vomiting and a systemic illness has a near-normal pH despite abnormal bicarbonate, chloride or carbon dioxide.+
- 1Review each variable independently rather than ending interpretation at the pH; calculate the gap and consider gastric acid loss alongside lactate or ketone accumulation.
- 2Check potassium, renal function, fluid status and medication exposure to identify processes that may require different simultaneous treatments.
- 3Reassess the proposed combination against the history and serial measurements, and document both disturbances when the evidence supports a mixed disorder.
05Checking understandingVerify the reasoning, revisit uncertainties and apply feedback.
- Repeat gases at a frequency driven by clinical instability and the intervention, ensuring oxygen delivery and ventilatory support are recorded each time.
- Track lactate or blood ketones when those abnormalities explain the acidosis; pH improvement alone may conceal a persisting pathological process.
- Check potassium during treatment because insulin, changing pH and recovery of renal function can change the extracellular concentration rapidly.
- Review chloride and fluid balance after substantial saline exposure; a persisting low bicarbonate may reflect a new normal-gap acidosis rather than treatment failure.
06Special situationsVariants, exceptions and circumstances that change the usual approach.
Unit conversion discipline
PaCO2 equations commonly use mmHg although UK gases report kPa. Multiply kPa by approximately 7.5 before substituting into a mmHg-based compensation equation.
Bicarbonate measurement differences
Blood-gas bicarbonate is generally calculated from measured pH and carbon dioxide, whereas laboratory total carbon dioxide is measured; discordance needs contextual and sample-quality review.
Anion gap convention
Some laboratories include potassium in the calculation and therefore use a different reference interval. State the equation used instead of comparing mismatched values.
Ketoacidosis resolution
Persistent hyperchloraemic acidosis can follow treatment of ketoacidosis. Blood-ketone clearance and the wider clinical picture help distinguish this from ongoing ketone production.
07Common pitfallsFrequent interpretation and management errors.
- 01
Do not confuse a low bicarbonate due to respiratory compensation with proof of a primary metabolic acidosis.
- 02
Do not call a PaCO2 normal when it is inappropriately high for the degree of metabolic acidosis.
- 03
Do not use venous oxygen tension to assess whether arterial oxygenation is adequate.
- 04
Do not infer that routine IV bicarbonate is beneficial simply because a blood-gas printout shows acidaemia.