PF ratio

• related: Pulmonology

OVERVIEW

PaO2/FiO2 ratio is the ratio of arterial oxygen partial pressure (PaO2 in mmHg) to fractional inspired oxygen (FiO2 expressed as a fraction, not a percentage)

• also known as the Horowitz index, the Carrico index, and (most conveniently) the P/F ratio
• at sea level, the normal PaO2/FiO2 ratio is ~ 400-500 mmHg (~55-65 kPa)
• MD Calc is an example of an online P/F ratio calculator – however it is quite easy to do “in your head”

P/F ratio is a widely used clinical indicator of hypoxaemia, though its diagnostic utility is controversial.

ALTERNATIVES TO P/F RATIO

Alternative indices of oxygenation include:

• Oxygen saturations in arterial blood (SpO2 and SaO2)
• S/F ratio (SpO2 to FiO2 ratio)
• PaO2 (arterial oxygen tension)
• A-a gradient (difference between alveolar oxygen tension (PAO2) and PaO2)
• Oxygenation index (OI) (the reciprocal of P/F times mean airway pressure (MAP): OI = (FiO2×MAP)/PaO2)
• P/FP Ratio (PaO2/(FiO2 X PEEP)
• a/A ratio (ratio of PaO2 and PAO2)
• Respiratory index (RI) (RI = pO2(A-a)/_p_O2(a), ie. the A-a gradient divided by the PaO2; normal RI is <0.4)
• Shunt fraction

---

ADVANTAGES OF P/F RATIO

• Quick and simple (probably the main reason for it’s widespread use)
• can be used as a rough guide to whether there is a significant A-a gradient present:
  • PaO2 should = FiO2 x 500 (e.g. 0.21 x 500 = 105 mmHg)
  • see caveats below
• More practical than the a/A ratio, as measurement of alveolar oxygen tension (PAO2) is not required
• Used in severity scoring systems
  • e.g. APACHE IV, SOFA, SAPS-II and SAPS-III
  • e.g. SMART-COP risk score for intensive respiratory or vasopressor support in community-acquired pneumonia (P/F ratio <333 mmHg if age <50y or PF ratio <250mmHg if age >50y)
  • e.g. part of the Berlin definition of Acute Respiratory Distress Syndrome (ARDS) (P/F ratio <300mmHg), and correlates with mortality (see below)

Note that the ratios for the Berlin definitions was with a PEEP setting of 5 cmH20 or more

DISADVANTAGES OF P/F RATIO

• A better P/F ratio may not be associated with better outcomes

  • e.g. In the ARDSNet ARMA study the high tidal volume strategy had better P/F ratios, but worse outcomes

• P/F ratio is dependent on barometric pressure (it is a as a “tension-based index”)

  • normal lungs (with a normal A-a gradient) will have lower PF ratios at high altitude and higher PF ratios at supra-atmospheric pressures

• P/F ratio alone cannot distinguish hypoxaemia due to alveolar hypoventilation (high PACO2) from other causes such as V/Q mismatch and shunt

  • whereas A-a gradient based indices (e.g. a/A ratio and RI) can exclude alveolar hypoventilation
  • as shunt increases, the PaO2 tends to become less and less sensitive to the PAO2 and to the FIO2, and more dependent on the mixed venous O2 content and saturation

• markedly dependent on FiO2

  • May be unreliable unless FiO2 > 0.5 and PaO2 < 100 mmHg

  • variation occurs with both right-to-left shunt (e.g. ARDS) and with widespread V/Q scatter (e.g. COPD)

  • varies with degree of shunt present – increasing the FIO2 causes the PaO2/FIO2 ratio to rise if intrapulmonary shunt is small, but to drop if the shunt is large

  • as a result, P/F ratio will vary according to the chosen SpO2 (and hence PaO2) target, as the required FiO2 will vary

• does not account for mean airway pressure or PEEP

  • The Oxygenation Index (OI) may be a more accurate measure of oxygenation dysfunction in ventilated patients as it accounts for mean airway pressure
  • P/FP Ratio adjusts the P/F ratio for the set PEEP

• requires and arterial blood gas measurement

  • S/F ratio tends to correlates with P/F ratio and is non-invasive

• highly dependent on CaO2-CvO2 (oxygen extraction)

  • arterial blood may appear well oxygenated despite lung dysfunction if mixed venous oxygen tension is high due to poor oxygen extraction by tissues, e.g. sepsis
  • P/F ratio may appear worse due to high oxygen extraction ratio (e.g. cardiogenic shock)

• does not indicate oxygen content of the blood (dependent on haemoglobin) or oxygen delivery to tissues ( dependent on cardiac output and oxygen content)

---

RULE OF THUMB FOR USE OF P/F RATIO

P/F ratio should only be used as a rule of thumb for detecting an A-a gradient when:

• the PaCO2 is normal, and
• shunt is not suspected

The FiO2 used should always be specified.

A quick comparison of the patient’s PaO2 to the product of “500 x FiO2” is a magic trick for estimating a-A gradient that should be in the arsenal of every intensivist!