PaO2/FiO2 Ratio

Physiological Basis

The PaO2/FiO2 ratio reflects the efficiency of oxygen transfer from the alveoli to the blood. It is used to assess the degree of hypoxemia and the severity of ARDS. The idea of monitoring P/F ratios was first used in the 1970's as a means of monitoring 'shock lung', preceding what we commonly know as acute respiratory distress syndrome.

While P/F ratio has predominantly been used to in ARDS, its principle holds true for all forms of parenchymal lung injury as a way of quantifying how well gas exchange is occurring at the alveolar/capillary level. In ARDS, gas exchange is impaired by a complex process of inflammatory cytokines, neutrophilic infiltration, microthrombi formation and fibrosis. In pneumonia or diffuse alveolar hemorrhage, while it is the presence of purulence and blood that impairs O2 delivery and CO2 removal, the P/F ratio holds true as a means of assessing severity of illness and response to therapy.

Clinical Application

P/F ratio helps guide treatment decisions, such as the need for mechanical ventilation or changes in oxygen therapy. Clinically, this is most relevant in acute respiratory distress syndrome (ARDS) as a component to the diagnosis, and a guide towards evidence based therapies.

Remember that ARDS is not just a reduced (<300) P/F ratio, but also includes other criteria as outlined by the Berlin Criteria:

• Onset within 7 days of identifiable trigger (direct lung injury or systemic injury)

• Bilateral patchy opacities as identified on imaging (CXR, CT, ultrasound)

• Opacities must not be related to cardiogenic pulmonary edema or intravascular volume overload

• P/F ratio <300 with PEEP greater than/equal to 5cmH20 or high flow nasal cannula >30L/min

Identifying ARDS based on the above criteria is useful, serving as a reminder to clinicians that certain ventilatory strategies and treatments will benefit patients, while others will cause harm.

Evidence Base

Numerous studies have validated the use of the PaO2/FiO2 ratio in assessing ARDS severity and guiding clinical management.

Lung Protective Ventilation

ARDS can significantly reduce the amount of lung available for gas exchange, meaning that conventional volumes on a ventilator of >6-8ml/kg (ideal body weight) can cause alveolar over-distension and risk causing barotrauma. This barotrauma can be detrimental to already fragile lung function, making the adoption of lung protective strategies a cornerstone of ARDS treatment. The first landmark study on low tidal volume ventilation was published in 2000 by the ARDS Network group, comparing 4-6ml/kg tidal volumes to traditional ventilation of ~12ml/kg. This trial was stopped early due to an almost 9% mortality difference in favour of low tidal volumes.

Since the ARDSNet trial was published, low tidal volume ventilation has been verified in multiple meta-analyses including a Cochrane Review, with additional ventilation strategies discussed below.

High PEEP

ARDS not only limits the amount of lung amenable to gas exchange, but also creates a stiff lung, meaning that once alveoli collapse it takes a significant amount of pressure to re-recruit alveoli and for gas exchange. Not only does this put the lung at risk of barotrauma, but the repeated opening and closing of alveolar units during respiration creates a new form of lung injury - atelectrauma. This has led to the study of an 'open lung' strategy, utilizing high PEEP to keep alveoli open during expiration and prevent atelectrauma.

Multiple large RCTs have shown no mortality benefit in all-comers with ARDS to high PEEP strategies, but subgroup and Bayesian analysis of these trials have shown a significant benefit in moderate-to-severe ARDS (P/F <200). There are ongoing trials looking at additional ventilation strategies - including newer parameters like driving pressure (Plateau pressure - PEEP) as a more sensitive marker of mortality in ARDS. This is not as easily adjusted on a ventilator like PEEP or tidal volume, but targeting a driving pressure <15cmH20 by optimizing PEEP, tidal volume and recruitment provide a ventilatory foundation along with the strategies outlined below in optimizing patients with ARDS.

Deep sedation/paralysis

Deeply sedating and paralyzing patients with paralysis has been variably and weakly recommended in ARDS guidelines due to inconsistent evidence. Its theoretical benefit is in reducing ventilator dyssynchrony, which can contribute to greater than recommended tidal volumes and pulmonary pressures, increasing the risk of barotrauma. The ACURASYS trial was the first to suggest a clinical benefit to paralysis, comparing this to deep sedation alone. In those with a P/F ratio <150 undergoing lung protective ventilation, using paralysis led to a significantly lower adjusted 90 day mortality. Subsequently, the ROSE trial compared deep sedation and paralysis with light sedation (RASS 0 to -1), showing no mortality benefit when comparing the two strategies. Overall, deep sedation and paralysis are still widely used in the treatment of ARDS on a case selective basis, and should be considered in the overall clinical picture when ventilator dyssynchrony is apparent or there is refractory hypoxemia.

Proning

Since PROSEVA (proning for 16 hours in patients with ARDS and P/F ratio <150), proning has become a standard part of management of moderate to severe ARDS. The benefits of proning are numerous - increased secretion clearance, homogenous distribution of lung strain, increased dorsal lung recruitment and improvement in V/Q mismatching. In considering the harms of proning, centres who prone less frequently may run in to issues with endotracheal tube dislodgement and pressure sores. Proning protocols extensively employed during the COVID-19 pandemic have standardized proning practices to allow for more efficient proning and less complications.

Remember that proning is not limited to intubated patients - COVID-19 also showed us that we can successfully prone awake patients with similar theoretical benefit. This can be limited by patient comfort, often making the duration of proning shorter and bringing in to question its efficacy, but if patient comfort allows it may be a beneficial intervention in the awake patient with moderate to severe ARDS.

ECMO

VV-ECMO can be employed as a rescue strategy in severe ARDS refractory to the above cornerstones of management. The first study to show a mortality benefit to ECMO referral was the CESAR trial, where patients with severe respiratory failure were randomized to ECMO centre referral vs best supportive ventilatory management. In CESAR, quality adjusted life at 6-months was significantly improved in the ECMO group, but notably 24% of these patients never actually received ECMO!

Nonetheless, this paved the way for further studies of ECMO in ARDS, most notably the EOLIA trial. EOLIA took almost 250 patients with severe ARDS and randomized them to receive ECMO vs best ventilatory management. Everyone randomized to ECMO was started on ECMO, unlike CESAR, but ultimately this trial showed no difference in its primary outcome of 60-day mortality. 28% of patients in the control group underwent crossover to the ECMO group as a rescue therapy, with some secondary endpoints favoring ECMO including treatment failure and duration of renal replacement therapy. EOLIA has led to some recognized benchmarks for considering ECMO, including P/F ratio <80 for 6 hours, <50 for 3 hours, and refractory respiratory acidosis (pH <7.25 and pCO2 >60). Despite the clear mortality benefit in EOLIA, subsequent systematic reviews and meta-analyses have demonstrated that ECMO confers a mortality benefit when EOLIA, CESAR, and other observational studies looking at ECMO are combined.

Faced with refractory ARDS, in appropriate candidates I will continue to consider ECMO as a bridge to recovery.

What not to do routinely

Routine prolonged lung recruitment maneuvers are not recommended - they can cause further barotrauma and harm. Lung recruitment maneuvers more generally are variable, they rely on the assumption that there is a meaningful amount of de-recruited lung amenable to recruitment. There is no routinely available way to assess for recruitability, with CT being the gold standard but not a cost-effective routine way of serial assessment.