Calculate PaO₂/FiO₂ ratio for ARDS severity classification using the Berlin Definition. Includes S/F ratio surrogate, oxygenation index, FiO₂ delivery reference, and EMR-ready respiratory documentation.
If no ABG available, enter SpO₂ to calculate the S/F ratio surrogate. Most accurate when SpO₂ ≤97%.
%
Berlin Definition requires PEEP ≥5 cmH₂O for ARDS diagnosis. Ensure patient is on at least 5 cmH₂O PEEP or CPAP when interpreting results for formal ARDS classification.
P/F Ratio—mmHg
S/F Ratio—(SpO₂/FiO₂)
Enter PaO₂ and FiO₂ above
P/F Ratio (PaO₂/FiO₂)—mmHg
S/F Ratio—
—
Berlin Definition Classification
Severe <100Moderate 100–200Mild 200–300Normal >300
PaO₂—mmHg
÷
FiO₂—decimal
=
P/F Ratio—mmHg
Oxygenation Index (OI)
OI = (FiO₂ × MAP) / PaO₂OI requires Mean Airway Pressure (MAP) from ventilator — enter MAP below for full OI calculation
Ventilator Strategy Guidance
Clinical Interpretation
Respiratory Assessment Note
Berlin Definition — ARDS Severity Classification
ARDS Level
P/F Ratio
S/F Surrogate
PEEP
Approx. Mortality
Management
Normal
> 300 mmHg
> 315
N/A
~5%
Standard oxygen therapy; investigate cause of hypoxemia
Mild ARDS
200–300 mmHg
235–315
≥5 cmH2O
~27%
NIV/CPAP consideration; lung-protective ventilation; prone if worsening
Moderate ARDS
100–200 mmHg
148–235
≥5 cmH2O
~32%
Intubation + lung-protective ventilation; PEEP optimization; prone positioning
Mortality estimates from the ARDS Definition Task Force (JAMA 2012) — Berlin Definition validation cohort of 4,188 patients. Individual outcomes vary significantly based on underlying cause, comorbidities, and institution.
FiO₂ by Oxygen Delivery Device — Reference
Delivery Device
Approximate FiO₂
Notes
Room Air
21%
0.21 — baseline
Nasal Cannula 1L
~24%
~4% per litre
Nasal Cannula 2L
~28%
~4% per litre
Nasal Cannula 4L
~36%
~4% per litre
Nasal Cannula 6L
~44%
Max flow for NC
Simple Face Mask
~35–50%
Flow-dependent
Venturi Mask
24–60%
Colour-coded valves
Non-Rebreather
~60–80%
15 L/min flow
High-Flow NC (HFNC)
21–100%
Titrated precisely
Mechanical Ventilator
21–100%
Set exactly on vent
FiO₂ for non-rebreather masks and nasal cannulas are estimates — actual delivered FiO₂ varies with patient breathing pattern, mask fit, and flow rate. Only mechanical ventilators and HFNC deliver a precisely controlled FiO₂.
P/F Ratio — Clinical Reference for ICU and Respiratory Medicine
The PaO₂/FiO₂ ratio (P/F ratio) is the most widely used index of oxygenation impairment in the ICU.
It normalizes the arterial oxygen level for the amount of supplemental oxygen being delivered, allowing direct comparison of oxygenation severity regardless of the FiO₂ being used.
On room air (FiO₂ 0.21), a normal PaO₂ of 95 mmHg produces a P/F ratio of 452 — well within the normal range.
A patient requiring 80% FiO₂ to maintain the same PaO₂ would have a P/F ratio of 95/0.80 = 119 — classified as Moderate ARDS by the Berlin Definition.
Berlin Definition — Full Diagnostic Criteria for ARDS
Timing: Acute onset within 1 week of a known clinical insult or new/worsening respiratory symptoms
Imaging: Bilateral opacities on chest X-ray or CT not fully explained by effusions, collapse, or nodules
Origin of edema: Respiratory failure not fully explained by cardiac failure or fluid overload (requires objective assessment if no risk factor present)
Oxygenation (P/F ratio): Measured with PEEP or CPAP ≥5 cmH₂O:
Tidal Volume: 6 mL/kg Ideal Body Weight (IBW) — not actual body weight. IBW = 50 + 0.91×(height in cm − 152.4) for men; 45.5 + 0.91×(height − 152.4) for women.
Plateau Pressure: Keep ≤30 cmH₂O — reduce tidal volume by 1 mL/kg if exceeded
PEEP: Titrate using ARDSNet PEEP-FiO₂ table to achieve target SpO₂ 88–95%
Target oxygenation: PaO₂ 55–80 mmHg or SpO₂ 88–95%
Severe ARDS (P/F <150): Prone positioning ≥16 hours/day (PROSEVA trial — 28-day mortality 16% vs 33%); neuromuscular blockade
Refractory ARDS (P/F <80): ECMO referral at experienced centre
Why does the Berlin Definition require PEEP ≥5 cmH₂O?
The PEEP requirement distinguishes true ARDS (diffuse alveolar damage causing non-cardiogenic pulmonary edema) from hypoxemia caused by atelectasis, which can be rapidly reversed with PEEP. If a patient has a P/F ratio of 180 with PEEP 0 but the ratio improves to 350 with PEEP 8 cmH₂O, they do not have ARDS by Berlin criteria. The PEEP requirement also ensures more consistency across institutions in how the ratio is measured.
What is the Oxygenation Index (OI) and how does it differ from P/F ratio?
The Oxygenation Index (OI) = (FiO₂ × Mean Airway Pressure × 100) / PaO₂. Unlike the P/F ratio, OI incorporates mean airway pressure (MAP), which reflects the total ventilatory burden placed on the lungs. OI is more accurate than P/F ratio for patients on mechanical ventilation because two patients can have the same P/F ratio but very different mean airway pressures — indicating different degrees of lung injury severity. OI >25 generally corresponds to severe ARDS. OI is especially used in pediatric ARDS (PARDS) criteria and ECMO candidacy assessment.
What causes ARDS and which are most common in the ICU?
ARDS causes are divided into direct pulmonary (primary) and indirect (secondary/extrapulmonary) causes. Direct causes: Pneumonia (most common — bacterial, viral, fungal), aspiration of gastric contents, inhalation injury, lung contusion, drowning. Indirect causes: Sepsis (most common indirect cause — up to 30–40% of ARDS), pancreatitis, major trauma/transfusions (TRALI), drug overdose, burns. COVID-19 pneumonia became a major cause of ARDS globally from 2020 onward, with a distinct pathophysiology (type H vs type L phenotypes described by Gattinoni et al.).
When should ECMO be considered in ARDS?
Veno-venous ECMO (VV-ECMO) is considered in severe refractory ARDS when optimal lung-protective ventilation has failed. Standard ECMO candidacy criteria include: P/F ratio <80 mmHg despite FiO₂ 1.0 and PEEP ≥10 cmH₂O; pH <7.25 with PaCO₂ ≥60 mmHg (CO₂ retention); Oxygenation Index >25. The Murray Score ≥3.0 is also used for ECMO candidacy assessment. VV-ECMO should only be performed at experienced centres — the EOLIA trial showed a trend toward benefit but did not reach statistical significance, with significant crossover (28% of control patients were rescued by ECMO).