A-a Gradient Calculator

A-a Gradient Calculator for Pulmonology. A normal gradient on room air is age-dependent, not a fixed number: a common bedside estimate is (age/4) + 4 mmHg, so about 10 mmHg in a young adult rising to roughly 20-25 mmHg past age 70. A normal gradient with hypoxemia points to hypoventilation or low inspired oxygen (altitude). A widened gradient localizes the defect to the alveolar-capillary unit and, combined with the response to supplemental oxygen, separates the causes: V/Q mismatch and diffusion limitation correct substantially with oxygen, whereas a true right-to-left shunt (for example pneumonia, ARDS, atelectasis, or intracardiac shunt) corrects poorly. Trending the gradient across serial gases tracks the course of parenchymal lung disease.

How this calculator works

The alveolar-arterial (A-a) oxygen gradient is the difference between calculated alveolar oxygen tension (PAO2) and measured arterial oxygen tension (PaO2) from an arterial blood gas. PAO2 is derived from the alveolar gas equation, PAO2 = FiO2 x (Patm - PH2O) - (PaCO2 / R), where at sea level on room air this reduces to PAO2 = 0.21 x (760 - 47) - (PaCO2 / 0.8), roughly 150 - 1.25 x PaCO2. The gradient (PAO2 - PaCO2) quantifies how efficiently oxygen crosses from alveolus to blood and isolates gas-exchange failure from pure hypoventilation.

When to use this calculator

Use the A-a gradient to work up any hypoxemia or unexplained dyspnea once an ABG is available, and specifically to distinguish the five mechanisms of hypoxemia. It is most useful when PaO2 or SpO2 is low and you need to know whether the lung parenchyma is the problem or whether hypoventilation alone (a raised PaCO2 with a normal gradient) explains it. It requires a room-air (or precisely known FiO2) ABG; the calculation is unreliable and difficult to interpret on high supplemental oxygen, where the simplified constants and assumed respiratory quotient break down. It is not a screening test for well patients and adds nothing when the ABG is normal.

Inputs used

  • FiO2
  • Barometric pressure
  • PaCO2
  • PaO2
  • Respiratory quotient assumption
  • Patient age when comparing expected gradient

Clinical interpretation

A normal gradient on room air is age-dependent, not a fixed number: a common bedside estimate is (age/4) + 4 mmHg, so about 10 mmHg in a young adult rising to roughly 20-25 mmHg past age 70. A normal gradient with hypoxemia points to hypoventilation or low inspired oxygen (altitude). A widened gradient localizes the defect to the alveolar-capillary unit and, combined with the response to supplemental oxygen, separates the causes: V/Q mismatch and diffusion limitation correct substantially with oxygen, whereas a true right-to-left shunt (for example pneumonia, ARDS, atelectasis, or intracardiac shunt) corrects poorly. Trending the gradient across serial gases tracks the course of parenchymal lung disease.

Worked example

A 70-year-old on room air at sea level has an ABG with PaO2 62 mmHg and PaCO2 55 mmHg. PAO2 = 150 - (55 / 0.8) = 150 - 69 = 81 mmHg, so the A-a gradient = 81 - 62 = 19 mmHg. The age-predicted upper limit is roughly (70/4) + 4 = 21.5 mmHg. Because the gradient is normal for age despite clear hypoxemia and hypercapnia, the picture is pure hypoventilation (for example opioid overdose or a COPD exacerbation with CO2 retention) rather than a parenchymal gas-exchange defect. Had the gradient been 40 mmHg, a V/Q mismatch, shunt, or diffusion process would be implicated.

Limitations and safety notes

The bedside formula assumes sea-level barometric pressure, water vapor pressure near 47 mmHg, and a respiratory quotient of about 0.8, so it degrades at altitude, on supplemental oxygen, and with markedly abnormal diets or metabolism. On high FiO2 the gradient inflates and loses discriminatory value, so the arterial/alveolar (a/A) ratio or PaO2/FiO2 ratio is preferred there. It requires an arterial sample; venous or capillary gases do not substitute. The age-adjusted upper limit is an approximation with wide inter-individual scatter, and a normal gradient does not exclude lung disease that is compensated at rest.

Frequently asked questions

Why is my patient hypoxemic but the A-a gradient is normal?

A normal gradient with a low PaO2 means the alveolar-capillary interface is working; the oxygen never reaches the alveolus in adequate amount. The two causes are hypoventilation (look for a raised PaCO2) and a low inspired oxygen tension (high altitude). Treat the ventilation problem rather than searching for parenchymal disease.

Do I have to adjust the normal value for age?

Yes. The gradient widens with age as V/Q matching becomes less uniform. A gradient of 20 mmHg is abnormal in a 25-year-old but normal in an 80-year-old. Use age/4 + 4 (or roughly 2.5 + 0.21 x age) as the upper limit of normal rather than a single fixed cutoff.

How does the A-a gradient separate shunt from V/Q mismatch?

Both widen the gradient, but they respond differently to oxygen. Give 100% oxygen: V/Q mismatch and diffusion limitation raise the PaO2 substantially and narrow the gradient, while a true shunt shows a blunted rise because deoxygenated blood bypasses ventilated alveoli entirely.

Can I use the A-a gradient on a patient receiving supplemental oxygen?

It becomes unreliable. On elevated FiO2 the calculated PAO2 rises steeply and the gradient widens even in healthy lungs, so the number is hard to interpret. Use the PaO2/FiO2 ratio or the a/A ratio instead, or obtain a room-air gas if the patient tolerates it.

References

  • Mellemgaard K. The alveolar-arterial oxygen difference: its size and components in normal man. Acta Physiol Scand. 1966;67(1):10-20. PMID: 5963295.
  • Kanber GJ, King FW, Eshchar YR, Sharp JT. The alveolar-arterial oxygen gradient in young and elderly men during air and oxygen breathing. Am Rev Respir Dis. 1968;97(3):376-81. PMID: 5638492.

Editorial review and citation methodology

Reviewed by the Quick Medical Calculator Editorial Team. Last reviewed: April 10, 2026. The review checks calculator inputs, intended population, interpretation, limitations, and source alignment.

  • Prefer original validation studies for scoring systems and prediction tools.
  • Use current specialty society guidance, transplant allocation policy, public health guidance, or regulator resources when they govern clinical use.
  • Include limitations and safety notes when a calculator is population-specific, context-dependent, or unsuitable as a standalone decision tool.

Related reviewed calculators