Beyond the PaO₂/FiO₂ ratio: Rethinking ARDS severity through the Lens of physiology

Acute Respiratory Distress Syndrome (ARDS) is a heterogeneous clinical syndrome encompassing distinct physiological and biological patterns of lung injury. Despite this heterogeneity, the ratio of arterial oxygen partial pressure to inspired oxygen fraction (PaO₂/FiO₂) remains the cornerstone of ARDS definitions, severity classification, and clinical decision-making. While its simplicity has facilitated widespread use, the PaO₂/FiO₂ ratio incompletely reflects the underlying physiological mechanisms of hypoxemia and should not be interpreted as a stand-alone marker of disease severity. The PaO₂/FiO₂ ratio is highly sensitive to ventilator settings, particularly positive end-expiratory pressure (PEEP), exhibits nonlinear behavior at high inspired oxygen fractions, and provides only a static assessment of gas-exchange. Consequently, it fails to capture key dimensions of ARDS pathophysiology, including lung recruitability, mechanical heterogeneity, and the temporal evolution of injury and response to therapy. These limitations are increasingly relevant in contemporary intensive care, where ventilatory strategies and adjunctive therapies actively modify oxygenation independent of structural lung injury. In this narrative review, we critically re-examine the physiological assumptions underlying the PaO₂/FiO₂ ratio and evaluate its role in current ARDS practice. We synthesize evidence supporting alternative and complementary oxygenation metrics, such as PEEP-adjusted indices, the oxygenation index, and composite measures including the ROX index (SpO₂/FiO₂ adjusted for respiratory rate), emphasizing their physiological rationale, clinical interpretability, and practical limitations at the bedside. These metrics are discussed not as replacements, but as tools that may refine the contextual interpretation of hypoxemia. Beyond static oxygenation measures, we explore emerging paradigms that conceptualize ARDS severity as a dynamic, multidimensional construct, integrating longitudinal oxygenation trajectories with respiratory mechanics, imaging-based assessment of lung aeration, and biomarker-informed biological subphenotypes. Repositioning the PaO₂/FiO₂ ratio within this integrated physiological and biological framework may improve patient stratification, enhance the coherence of therapeutic decision-making, in line with the translational goals of modern intensive care.