Clinical
The A-a gradient: reading oxygenation beyond the oxygen number
Why a normal blood oxygen level can still hide a lung problem, how the alveolar gas equation separates the lung from everything else, and why the expected gradient rises with age.
Free online tool
A-a Gradient
The question the oxygen number cannot answer
A blood gas gives the oxygen tension in arterial blood (PaO₂). When it is low, the patient is hypoxaemic — but that single number does not say why. Someone breathing shallowly with perfectly healthy lungs and someone with a pulmonary embolism can both arrive at the same low PaO₂ by completely different routes.
The alveolar-arterial gradient separates the two. It compares the oxygen that should be in the alveoli with the oxygen that actually made it into the blood. A wide gap points at the lung itself; a normal gap says the lung is transferring oxygen fine and the problem lies upstream, in how much air is arriving.
How the calculation works
The alveolar oxygen tension is not measured directly, it is derived. The alveolar gas equation estimates it from the inspired oxygen fraction, the barometric pressure, the water vapour pressure of fully humidified air, the measured carbon dioxide and the respiratory quotient.
What goes into the equation
FiO₂
0.21 at room air
The fraction of inspired oxygen. Rises with supplemental oxygen.
Barometric pressure
760 mmHg
At sea level. Falls with altitude, and the alveolar oxygen falls with it.
Water vapour
47 mmHg
Air is fully humidified by the time it reaches the alveoli, and that water displaces oxygen.
PaCO₂
measured
Carbon dioxide occupies alveolar space: the more there is, the less room for oxygen.
Respiratory quotient
0.8
The usual ratio of CO₂ produced to O₂ consumed on a mixed diet.
The gradient is then simply the alveolar value minus the measured arterial value. Everything hinges on those inputs being right: an FiO₂ entered as room air when the patient is on oxygen produces a gradient that means nothing.
Why the normal value depends on age
There is no single normal gradient. In healthy people it widens steadily across the decades, as the match between ventilation and perfusion becomes less uniform. Mellemgaard measured this in normal subjects in 1966, and the age dependence has been part of the interpretation ever since.
The expected value rises with the years
A common bedside estimate is (age + 10) ÷ 4. At 20 that is about 7.5 mmHg; at 70 it is about 20. Judging a 70-year-old against a young adult's threshold manufactures a lung problem that is not there.
What a wide or normal gradient points to
The value of the gradient is that it splits the causes of low oxygen into two groups that need different responses.
- Normal gradient with low oxygen → the lung is transferring oxygen normally. Think hypoventilation (sedation, opioids, neuromuscular weakness) or low inspired oxygen, as at altitude.
- Wide gradient with low oxygen → the problem is in the lung or the circulation through it: ventilation-perfusion mismatch, shunt, or impaired diffusion.
- Wide gradient with normal oxygen → can appear when someone compensates by breathing harder. The normal PaO₂ is being bought at a cost, and it is worth noticing.
The PaO₂/FiO₂ ratio and the ARDS grades
The gradient is sensitive to the inspired oxygen fraction, which makes it awkward to compare across patients on different oxygen settings. For that, the PaO₂/FiO₂ ratio is used instead. It is the basis of the Berlin definition of acute respiratory distress syndrome, published in 2012.
Berlin severity grades
Mild
200–300
PaO₂/FiO₂ in mmHg, with PEEP or CPAP of at least 5 cmH₂O.
Moderate
100–200
Same ventilation requirement.
Severe
≤ 100
The most severe band.
A ratio alone is not a diagnosis of ARDS
The Berlin definition also requires an acute onset within a week, bilateral opacities on imaging, and respiratory failure that is not explained by cardiac failure or fluid overload. The number is one of four criteria, not the whole thing.
Where the number misleads
- At altitude the barometric pressure is lower, so the alveolar oxygen is lower too. Using 760 mmHg in Bogotá or Mexico City inflates the gradient.
- On supplemental oxygen the gradient widens even in healthy lungs, which is exactly why the PaO₂/FiO₂ ratio exists.
- A venous sample instead of arterial invalidates the whole calculation.
- The respiratory quotient of 0.8 assumes a mixed diet and a steady state; it is an approximation, not a constant.
For clinical use, not self-assessment
This calculation requires an arterial blood gas and is interpreted alongside the clinical picture. It is a tool for health professionals, not a way to assess your own breathing at home.
Sources
