Bedside Snapshot

Normal (Adult)
P/F about 380–480 on room air; A–a gradient under 10 mmHg in a young adult. Always read with the FiO₂ and PEEP.
Critical Values
No universal critical value. Severe ARDS band: P/F ≤100 mmHg on PEEP ≥5 cm H₂O (typical; not a laboratory panic value).
High Suggests
  • Wide A–a gradient: shunt, V/Q mismatch, diffusion limitation
  • Gradient widens as FiO₂ rises, even with healthy lungs
Low Suggests
  • ARDS, pneumonia, pulmonary edema, pulmonary embolism
  • Hypoventilation if the A–a gradient is normal
Act Now If
P/F ≤300 with PEEP ≥5 cm H₂O, or S/F ≤315 with SpO₂ ≤97%: check FiO₂ and PEEP, oxygen to target, lung-protective ventilation if intubated, escalate support
Watch Out For
Pulse oximetry can overestimate SpO₂ in darker skin, so an S/F can look better than the PaO₂ shows. Confirm with a blood gas when it matters.

Reference & Critical Ranges

Lab-specific ranges: Every lab sets its own intervals, and analyzer and population change them too. Read the range on your own report. The numbers below are typical.

Reference ranges

PopulationConventionalSINotes
Adult, room air (FiO₂ 0.21)
P/F about 380–480 (PaO₂ 80–100 mmHg ÷ 0.21)
Same ratio (PaO₂ 10.7–13.3 kPa)
Calculated, not measured. P/F falls as FiO₂ rises, so compare only at a stated FiO₂.
Adult A–a gradient, young, room air
Under 10 mmHg
Under about 1.3 kPa
Rises with age and with supplemental oxygen. A wide gradient on high FiO₂ is expected.
Older adult
Wider than in young adults; no single cut-off
Wider than in young adults
Age-related V/Q mismatch lowers PaO₂ with a larger gradient. Judge against the patient's own baseline.
Pregnancy
Normal PaO₂ runs higher (see PaO₂ page)
Higher than non-pregnant range
Use the same ARDS thresholds; compare with the pregnancy-adjusted normal PaO₂ rather than the non-pregnant range.
Children (ARDS indices)
PALICC-2 cut-offs differ from adult cut-offs
Same
Ventilated children: OI and OSI. Non-invasive ventilation: P/F or S/F with PALICC-2 cut-offs (see table below).

Severity & critical values

LevelConventionalSIWhat it means / typical findings
Normal
P/F about 380–480 on room air
Same ratio
Normal oxygen exchange for a young adult at FiO₂ 0.21.
Mild ARDS (intubated)
>200 to ≤300 with PEEP ≥5 cm H₂O
Same ratio
Bilateral opacities and timing met. S/F >235 to ≤315 when SpO₂ ≤97%. Severity bands apply to intubated patients only.
Non-intubated ARDS (one category)
P/F ≤300 on HFNO ≥30 L/min or NIV/CPAP ≥5 cm H₂O
Same ratio
No mild, moderate or severe band. S/F ≤315 when SpO₂ ≤97% also qualifies.
Moderate ARDS (intubated)
>100 to ≤200
Same ratio
Needs PEEP ≥5 cm H₂O in ventilated patients. S/F >148 to ≤235 when SpO₂ ≤97%.
Critical (severe ARDS, intubated)
P/F ≤100 with PEEP ≥5 cm H₂O
Same ratio
Severe ARDS: consider ECMO referral and prone positioning. S/F ≤148 when SpO₂ ≤97%.
Critical values: P/F ≤100 with PEEP ≥5 cm H₂O is the severe ARDS band. It is typical, not a laboratory panic value. No universal critical value exists for the A–a gradient or the oxygenation index.

Conversion: mmHg × 0.133 = kPa. A ratio has no units, so the number stays the same in both systems.

PaO₂/FiO₂ (P/F) ratio
P/F = PaO₂ (mmHg) ÷ FiO₂ (as a fraction, 0.40 for 40%)

Try it: PaO₂ 80 mmHg on FiO₂ 0.40. The P/F is 200.

SpO₂/FiO₂ (S/F) ratio
S/F = SpO₂ (%) ÷ FiO₂ (as a fraction)

Only use it when SpO₂ is 97% or lower. S/F 315 matches P/F 300, and S/F 235 matches P/F 200 (ARDS Network derivation, 2007). Try it: SpO₂ 94% on FiO₂ 0.40 gives S/F 235.

Alveolar gas equation (PAO₂)
PAO₂ = FiO₂ × (barometric pressure − 47) − PaCO₂ ÷ 0.8

At sea level, barometric pressure is 760 mmHg, so the term becomes FiO₂ × 713. At altitude, use the measured local barometric pressure. The 47 mmHg is water vapour pressure at body temperature. The 0.8 is the usual respiratory quotient.

Alveolar–arterial (A–a) gradient
A–a gradient = PAO₂ − PaO₂

Worked example at sea level, FiO₂ 0.40, PaCO₂ 40 mmHg, PaO₂ 80 mmHg: PAO₂ = 0.40 × 713 − 40 ÷ 0.8 = 235. The A–a gradient is 155 mmHg.

Altitude correction (above 1,000 m)
Corrected P/F or S/F = measured ratio × (barometric pressure ÷ 760)

Apply this before you use the ARDS thresholds. When barometric pressure is below 760 mmHg, the corrected ratio comes out lower than the measured one. The same measured ratio can therefore meet an ARDS band at altitude (Matthay 2024, Table 1 footnote).

Oxygenation index (OI), invasive ventilation
OI = MAP (cm H₂O) × FiO₂ (fraction) × 100 ÷ PaO₂ (mmHg)

Example: MAP 20 cm H₂O, FiO₂ 0.60, PaO₂ 60 mmHg gives OI 20. MAP is the mean airway pressure on the ventilator, from an arterial line.

Oxygen saturation index (OSI), invasive ventilation
OSI = MAP (cm H₂O) × FiO₂ (fraction) × 100 ÷ SpO₂ (%)

OSI only works when SpO₂ is 97% or lower, so skip it above that. Example: MAP 20, FiO₂ 0.60, SpO₂ 92% gives OSI 13.0.

Causes of Abnormal Results

Low P/F or S/F (hypoxemia)

MechanismCommon causesBedside clue
Shunt (blood passes unventilated alveoli)
ARDS, pneumonia with consolidation, atelectasis, pulmonary edema
Low P/F that does not improve much with more oxygen; wide A–a gradient
V/Q mismatch
Asthma, COPD, pneumonia, pulmonary embolism, early pulmonary edema
Wide A–a gradient; improves as the underlying airway or vessel problem is treated
Hypoventilation
Opioids, sedatives, neuromuscular disease, obesity hypoventilation, chest wall injury
Rising PaCO₂ with a normal A–a gradient; low P/F from low alveolar O₂
Low inspired PO₂
High altitude or a low FiO₂ setting
Normal A–a gradient; low P/F matches low FiO₂ or altitude
Diffusion limitation (uncommon)
Interstitial fibrosis, some severe lung diseases
Wide A–a gradient
Ratio alone misleads: A low P/F alongside a normal A–a gradient is the clue for hypoventilation or low inspired oxygen. Treat the breathing, and don't stop at the lungs.

Wide A–a gradient (above the age-expected value)

  • V/Q mismatch, shunt or diffusion limitation: alveolar and arterial PO₂ split apart, so the gradient widens.
  • Higher FiO₂: more FiO₂ lifts PAO₂, which widens the gradient even when the lungs are healthy.
  • Age: the gradient normally climbs with age. Compare it with the value expected for that patient.

High P/F or S/F

A high value on supplemental oxygen is what you should expect. The oxygen is doing its job. In ventilated patients, though, that same number can mean too much oxygen (hyperoxemia). Wean FiO₂ back to the target range.

Interpretation & Bedside Approach

Step-by-step approach

  1. Is it real? Confirm the sample is arterial, free of air bubbles and analysed promptly. Record the FiO₂ and device, and the PEEP or CPAP setting, with the value.
  2. How sick is the patient? Compare the SpO₂ with the target, and check work of breathing, mental state and hemodynamics.
  3. Acute or chronic, and what is the trend? Compare with the previous value at the same FiO₂ and PEEP. Trend beats a single value.
  4. Calculate the derived values: P/F or S/F, then the A–a gradient if an ABG is available. Use OI or OSI only in ventilated children.
  5. Classify with named criteria: the 2023 Global ARDS definition for adults, or PALICC-2 for children.
  6. Escalate or disposition: act on the patient's trajectory, not only on the index. Escalate support early when the index falls.

Pattern and what it suggests

PatternSuggests
Low P/F, normal A–a gradient, rising PaCO₂
Hypoventilation (opioids, sedation, neuromuscular weakness)
Low P/F, wide A–a gradient
V/Q mismatch, shunt or diffusion limitation
S/F ≤315 with SpO₂ ≤97%, bilateral opacities, and PEEP ≥5 cm H₂O, HFNO ≥30 L/min or NIV/CPAP ≥5 cm H₂O
Can meet the adult oxygenation criterion without an ABG (timing and origin criteria still apply)
OI ≥16 or OSI ≥12 in a ventilated child
Severe PARDS (PALICC-2 stratification)

ARDS criteria (Global definition, 2023)

  1. Timing: onset or worsening of hypoxemic respiratory failure within 1 week of a known risk factor or new respiratory symptoms.
  2. Chest imaging: bilateral opacities on radiography or CT, or bilateral B-lines or consolidation on ultrasound, not fully explained by effusion, atelectasis or nodules.
  3. Origin: not fully explained by cardiac failure or fluid overload. ARDS can still be diagnosed if a risk factor is also present.
  4. Oxygenation: P/F ≤300, or S/F ≤315 when SpO₂ ≤97%, on high-flow nasal oxygen at 30 L/min or more, on NIV or CPAP with PEEP 5 cm H₂O or more, or when intubated with PEEP 5 cm H₂O or more. Above 1,000 m, correct the ratio first.

In resource-limited settings, the modified definition uses S/F ≤315 alone, with no PEEP or minimum flow requirement. Where you can measure PEEP and blood gases, go with the standard definition.

Ventilated children (PALICC-2)

  • Diagnosis on invasive ventilation: OI ≥4 or OSI ≥5 (OSI requires SpO₂ ≤97%). When using SpO₂-based criteria, titrate oxygen to SpO₂ 88–97% (PALICC-2 definition statement 1.4.5).
  • Severity, invasive ventilation: mild or moderate OI <16 or OSI <12; severe OI ≥16 or OSI ≥12. Stratify at least 4 hours after diagnosis.
  • Non-invasive ventilation: diagnosis at P/F ≤300 or S/F ≤250; severe at P/F ≤100 or S/F ≤150, with a full face mask and CPAP or PEEP of 5 cm H₂O or more.

Management Priorities

Emergency: Hypoxemia with P/F ≤300 or S/F ≤315 (SpO₂ ≤97%)

  1. Confirm the value: repeat the blood gas or SpO₂ at a stated FiO₂ and PEEP. Check that the device and the sample are correct.
  2. Oxygen to target: SpO₂ 94–98% for most adults; 88–92% with hypercapnia risk (COPD, obesity hypoventilation, neuromuscular disease) until a blood gas guides you.
  3. Support the breathing: high-flow nasal oxygen at 30 L/min or more, NIV or CPAP per protocol; intubate for failing oxygenation, exhaustion or falling consciousness.
  4. Treat the cause: reverse opioid hypoventilation, decompress a tension pneumothorax, treat bronchospasm, pulmonary edema or infection.
  5. Monitor & repeat: continuous SpO₂ and EtCO₂; repeat the blood gas after each change, recording FiO₂ and support.

Emergency: Ventilated patient with severe ARDS (P/F ≤100 or refractory hypoxemia)

  1. Lung-protective ventilation: tidal volume 4–8 mL/kg predicted body weight and plateau pressure below 30 cm H₂O (ATS 2017, retained in the 2024 update; strong recommendation).
  2. Prone positioning: more than 12 hours per day for severe ARDS (ATS 2017, retained in the 2024 update; strong recommendation, moderate certainty).
  3. Adjuncts: corticosteroids (ATS 2024, conditional, moderate certainty); neuromuscular blockers in early severe ARDS (ATS 2024, conditional, low certainty).
  4. PEEP and recruitment: higher PEEP without recruitment maneuvers (ATS 2024, conditional, low to moderate certainty). Avoid prolonged recruitment maneuvers (ATS 2024, strong, moderate certainty) and routine high-frequency oscillatory ventilation (ATS 2017, retained; strong, high certainty).
  5. Rescue referral: in selected patients with reversible disease and few futility risk factors, after optimal conventional management, consider early VV-ECMO referral when P/F is below 80 mmHg (under 7 days), or pH below 7.25 with PaCO₂ ≥60 mmHg (ATS 2024, conditional, low certainty).

Medications

Doses vary by protocol; follow local protocols and medical direction.

DrugAdult dosePediatric dose (max)Onset / key notes
Oxygen
Titrate to SpO₂ 94–98% (88–92% with hypercapnia risk), INH; no fixed max
Titrate to the target set for the child's condition, INH; no fixed max
Seconds. Recheck SpO₂ 5 minutes after each change. Record the FiO₂ used for every ratio.

Definitive care & disposition

  • ICU: need for high-flow oxygen, NIV or intubation; rising oxygen requirement; rising PaCO₂ with acidosis.
  • ECMO center: early discussion for selected patients with reversible severe ARDS in its first 7 days and few futility risk factors, with P/F below 80 mmHg or pH below 7.25 with PaCO₂ ≥60 mmHg, despite optimal ventilation and prone positioning.
  • Post-discharge: chronic hypoxemia needs outpatient assessment for long-term oxygen.

Doses are typical references. Follow local protocols, medical direction and institutional policy.

Field & Transport Considerations

Point-of-care testing

  • Pulse oximetry (S/F): you can use it in every setting. It is only valid while SpO₂ is 97% or lower. Skin pigmentation reduces accuracy.
  • Blood gas analyzer (P/F, A–a): it needs an arterial sample. Stay inside the temperature and sample limits in the device manual.
  • FiO₂ at the bedside: a ventilator or high-flow device gives you a known FiO₂. A nasal cannula or simple mask gives only an estimate, and that estimate depends on how the patient breathes.

Before and during interfacility transport

  • Obtain: FiO₂ and device, PEEP or CPAP, SpO₂ and the time of the most recent blood gas.
  • Recheck: after every change in oxygen, position or ventilator settings, and at the cadence set by protocol.
  • Falling index despite support: stabilize first, then confirm the destination and level of care. Don't hold up a transport that is itself the escalation (for example, to an ECMO or ICU center).

Altitude and flight

Altitude and cabin pressure both pull barometric pressure down, and that drop shows up in PAO₂, PaO₂ and SpO₂. Build the local or cabin barometric pressure into the alveolar equation. Once you're above 1,000 m, multiply P/F and S/F by barometric pressure ÷ 760 before you apply the ARDS thresholds (formula in the Reference section). Plan your oxygen to the target, with cylinder reserve.

Special Populations

PopulationWhat changesClinical implication
Neonates & infants
Neonatal lung disease uses separate targets and indices
These adult and PALICC-2 cut-offs are not validated for neonates. Follow neonatal protocols.
Children & adolescents
OI and OSI (invasive) or P/F and S/F with PALICC-2 cut-offs (non-invasive)
Use OSI only when SpO₂ is 97% or lower. Set the oxygen target for the child's condition.
Pregnancy
Normal PaO₂ is higher (see PaO₂ page)
Use the same ARDS thresholds; a near-normal P/F is less reassuring in pregnancy. Avoid the supine position after 20 weeks.
Older adults
A–a gradient rises with age
A mildly wide gradient can be normal for age. Judge against the patient's own baseline.
CKD / dialysis
No clinically meaningful difference for these indices
Interpret as for other adults.
Other
Darker skin pigmentation; chronic lung disease
Pulse oximetry may overestimate arterial saturation in darker skin, so confirm with a blood gas when it matters. In chronic lung disease, judge acute change from baseline (PALICC-2).

Interfering Factors & Pitfalls

Collection, handling and measurement

FactorEffectWhyAvoid / fix
Air bubble or delayed analysis
↑ or ↓ PaO₂ (unreliable)
Gas exchange continues in the syringe; air changes PO₂
Remove bubbles; analyse promptly; record the time
FiO₂ not recorded or estimated
Ratio cannot be compared
P/F depends on the FiO₂ at sampling
Record the device, flow and FiO₂ with each value
PEEP or CPAP not recorded
False − for the ARDS definition
Criteria require PEEP or CPAP at 5 cm H₂O or more
Record PEEP with each blood gas
SpO₂ above 97%
S/F invalid for ARDS
The S/F threshold was derived for SpO₂ 97% or less
Use a blood gas
Darker skin pigmentation
False − for hypoxemia (SpO₂ overestimates)
Pulse oximeter calibration bias
Confirm with PaO₂ when a decision depends on it
Carbon monoxide poisoning
False − (SpO₂ and PaO₂ look normal)
COHb is read as oxyhemoglobin
Measure COHb by co-oximetry (see carbon monoxide page)

Common misinterpretations

  • A–a gradient is a difference, not a diffusion gradient: it reflects several mechanisms, and diffusion is only one of them.
  • P/F is not a measure of lung severity alone: the number moves when FiO₂ and PEEP change.
  • OI and OSI are for ventilated children: adult ARDS runs on P/F or S/F under the 2023 definition.
Pulse oximetry bias: In darker skin, SpO₂ can read higher than the true arterial saturation. That means an S/F may understate hypoxemia. Confirm with a blood gas whenever the result would change management.

Physiology & Test Basics

What it measures

Each of these is a derived index. It pairs the oxygen you deliver (FiO₂) with the oxygen in the blood (PaO₂ or SpO₂) to grade gas exchange. This page covers the P/F ratio, the S/F ratio, the A–a gradient, and the OI and OSI used in ventilated children. A single PaO₂ value has its own page.

Physiology & regulation

  • Alveolar PO₂ rests on three inputs: FiO₂, barometric pressure and PaCO₂.
  • The A–a gradient is the gap between alveolar and arterial PO₂. That gap reflects V/Q mismatch, shunt and diffusion limitation.
  • Hypoventilation pushes PaCO₂ up and PAO₂ down, but the gradient stays the same width.
  • Ratios shift with FiO₂, so only compare values taken at the same FiO₂ and PEEP.

How it is measured

An arterial blood gas on a heparinized syringe gives the P/F and the A–a gradient, and the sample must be analysed promptly. Continuous pulse oximetry at a stated FiO₂ gives the S/F. The OI and OSI need an arterial line for PaO₂ or SpO₂, plus the mean airway pressure from the ventilator.

Why it matters

These indices grade hypoxemia and ARDS severity. They set when to escalate support and when to refer for ECMO. Read every ratio next to the FiO₂ and PEEP that produced it, because those settings steer each change to oxygen or the ventilator.

Clinical Pearls

Prehospital S/F: You can work out an S/F from SpO₂ and the device FiO₂, with no ABG. It only holds while SpO₂ is 97% or lower. Above that, get a blood gas.
Record the settings: Write the FiO₂ and PEEP next to every P/F. Change the PEEP and the ratio moves, even when the oxygen stays the same.
Nursing titration: Titrate FiO₂ to the target, then wean. Recheck the ratio after every change. A P/F that drops after weaning is a warning sign.
Hypoventilation check: Opioids or sedatives and a falling P/F? Check the breathing and the EtCO₂ before you turn up the oxygen by itself.
Transport: Before you move the patient, get the FiO₂, the PEEP and the time of the last gas. Cabin pressure lowers the expected PO₂. Recheck SpO₂ after takeoff or any altitude change.
Severe ARDS referral: Early in ARDS (under 7 days), a P/F below 80 mmHg is the ATS 2024 threshold to consider VV-ECMO in selected patients. Those patients have reversible disease and few futility risk factors, and the referral comes after optimal conventional management.

References

  1. Matthay, M. A., Arabi, Y., Arroliga, A. C., et al. (2024). A new global definition of acute respiratory distress syndrome. American Journal of Respiratory and Critical Care Medicine, 209(1), 37–47. https://doi.org/10.1164/rccm.202303-0558WS
  2. Qadir, N., Sahetya, S., Munshi, L., et al. (2024). An update on management of adult patients with acute respiratory distress syndrome: an official American Thoracic Society clinical practice guideline. American Journal of Respiratory and Critical Care Medicine, 209(1), 24–36. https://doi.org/10.1164/rccm.202311-2011ST
  3. Emeriaud, G., López-Fernández, Y. M., Iyer, N. P., et al. (2023). Executive summary of the second international guidelines for the diagnosis and management of pediatric acute respiratory distress syndrome (PALICC-2). Pediatric Critical Care Medicine, 24(2), 143–168. https://doi.org/10.1097/PCC.0000000000003147
  4. Rice, T. W., Wheeler, A. P., Bernard, G. R., et al. (2007). Comparison of the SpO₂/FIO₂ ratio and the PaO₂/FIO₂ ratio in patients with acute lung injury or ARDS. Chest, 132(2), 410–417. https://doi.org/10.1378/chest.07-0617
  5. Sarkar, M., Niranjan, N., & Banyal, P. K. (2017). Mechanisms of hypoxemia. Lung India, 34(1), 47–60. https://doi.org/10.4103/0970-2113.197116
  6. Sjoding, M. W., Dickson, R. P., Iwashyna, T. J., Gay, S. E., & Valley, T. S. (2020). Racial bias in pulse oximetry measurement [research letter]. New England Journal of Medicine, 383(25), 2477–2478. https://doi.org/10.1056/NEJMc2029240

Last updated Spotted something wrong? Report an error and include the page URL.

Medical Disclaimer
  • For Educational Purposes Only: This content is intended for educational reference and should not be used for clinical decision-making.
  • Not a Substitute for Professional Judgment: Always consult your local protocols, institutional guidelines, and supervising physicians.
  • Verify Before Acting: Users are responsible for verifying information through authoritative sources before any clinical application.
AI Assistance Notice
AI was used to assist in organizing, paraphrasing, and formatting this information.