Arterial blood gas and acid-base interpretation

Why examiners ask this

A blood gas is the examiner's favorite way to see whether you think in a sequence or guess at a pattern. They hand you four numbers mid-case and watch whether you identify the primary disorder, check compensation, find a hidden second process, and then do something useful for the patient. It tests application of knowledge and organization, and the follow-up ("so what will you do?") tests judgment.

The 60-second answer

Decision points and the "why"

• Winter's formula for metabolic acidosis — expected PaCO2 = 1.5 × HCO3 + 8 ± 2.¹ If the measured PaCO2 is higher, the patient has a concurrent respiratory acidosis; in an intubated patient that means I am under-ventilating, and in an extubated one it means fatigue. That changes my extubation decision.
• Albumin-corrected anion gap — the gap falls about 2.5 mEq/L for every 1 g/dL fall in albumin below 4.¹ A cirrhotic or septic patient with an albumin of 2 and a "normal" gap of 12 really has a gap near 17, which hides a lactic acidosis.
• Delta-delta — if the rise in gap exceeds the fall in bicarbonate, a metabolic alkalosis coexists (vomiting, nasogastric suction, diuretics). If the bicarbonate falls more than the gap rises, a non-gap acidosis coexists, most often hyperchloremia from normal saline.¹
• Lactate is a perfusion signal until proven otherwise — I treat a rising lactate with volume, blood, or pressor as indicated, aiming to bring it down,² and look for the hidden cause: bowel ischemia, compartment syndrome, unrecognized bleeding. Epinephrine and metformin raise lactate without hypoperfusion; I name them but do not assume them.
• Bicarbonate only for specific indications — the Surviving Sepsis Campaign suggests against it to improve hemodynamics or cut vasopressor dose, and suggests it in septic shock with a pH of 7.2 or less and acute kidney injury (AKIN stage 2–3).² I also give it for hyperkalemia or a non-gap acidosis from bicarbonate loss. It generates CO2, which crosses cells and worsens intracellular acidosis unless I raise minute ventilation to clear it.
• Respiratory acidosis in the OR — I look at the circuit and the patient: rising CO2 production (MH, thyroid storm, sepsis, CO2 insufflation) versus falling elimination (hypoventilation, exhausted absorbent, incompetent valves, dead space).

The complication they will inject

Three hours into an open bowel resection the gas reads pH 7.18, PaCO2 32, HCO3 12, sodium 140, chloride 115, albumin 2.5, lactate 6. The gap is 13, corrected to about 17, so this is a mixed gap and non-gap metabolic acidosis with partial respiratory compensation (Winter's predicts 24–28, so the 32 means I am behind on ventilation). I tell the surgeon the lactate is 6 and rising and ask about bowel viability and bleeding. I raise minute ventilation, switch from saline to a balanced crystalloid, give blood if the hemoglobin supports it, start norepinephrine if the patient is fluid-unresponsive, and recheck the gas in 30 minutes. I do not give bicarbonate to fix the blood pressure; if the pH stays at or below 7.2 and the kidneys are failing, I give 50–100 mEq while ventilating to clear the CO2.²

Pitfalls that cost points

• Naming the disorder and stopping; the examiner wants the cause and the plan.
• Skipping compensation, which is where the second disorder hides.
• Calculating an uncorrected anion gap in a hypoalbuminemic patient.
• Reaching for bicarbonate before fixing perfusion or ventilation.
• Forgetting that large-volume normal saline causes a hyperchloremic acidosis that is iatrogenic and avoidable.

Rapid-why

Q: Why does normal saline cause acidosis? A: Its chloride of 154 mEq/L narrows the strong ion difference, lowering bicarbonate; a balanced crystalloid avoids it, and in the SMART trial balanced fluid also lowered major adverse kidney events (14.3% vs 15.4%).³

Q: What is the expected compensation in acute respiratory acidosis? A: Bicarbonate rises about 1 mEq/L per 10 mmHg rise in PaCO2 acutely and about 4 mEq/L chronically.¹

Q: Why can a COPD patient be hard to wean after you normalize the CO2? A: Hyperventilating a chronic retainer to 40 leaves a high bicarbonate and a metabolic alkalosis that suppresses the respiratory drive; I ventilate to his baseline.

Q: Why does hypothermia matter to the gas? A: Gases are analyzed at 37 °C; I use alpha-stat management in adults, interpreting the uncorrected values, because it preserves enzyme function and cerebral autoregulation.

Numbers to know

Links

• Sepsis and septic shock, dka, Massive hemorrhage and transfusion, hyperkalemia, Fluid management: crystalloid, colloid, goal-directed
• Surviving Sepsis Campaign international guidelines (2021)