Relationships between dyspnea, oxygenation and prognosis in hypoxemic respiratory failure Authors: Yanfei Shen, Xinyuan Ding Published: 2024 Dear Editor, We read with great interest the recent study [1], which explored the associations between dyspnea and intubation/mortality in patients with hypoxemic respiratory failure. In this study, dyspnea was measured using a visual analog scale (dyspnea-VAS) with a range from 0 to 100 mm. A total of 300 patients were included. Among ICU patients admitted with acute hypoxemic respiratory failure, dyspnea was found to be associated with an increased risk of intubation, but not with higher mortality. While this study provides valuable insights into respiratory failure, it also uncovers the complex relationships between dyspnea, oxygenation, intubation, and mortality. In the current study, discordance was observed between subjective dyspnea severity and objective oxygenation parameters (PaO2/FiO2). For instance, patients with higher dyspnea-VAS scores paradoxically exhibited higher PaO2/FiO2 ratios compared to those without dyspnea (PaO2/FiO2: No dyspnea: 129 [98–171] vs. Mild dyspnea: 135 [103–182] vs. Moderate dyspnea: 154 [92–197]). This discordance underscores the need to explore the interplay between respiratory dyspnea and hypoxemia from a pathophysiological perspective, which could provide valuable insights for clinical decision-making. In clinical practice, severe dyspnea is generally associated with an increased risk of intubation, while higher PaO2/FiO2 ratios, which indicate better oxygenation, are linked to a lower intubation risk. However, the discordance observed in this study suggests that dyspnea may be a multidimensional symptom, influenced by factors beyond oxygenation alone. Contributing factors may include increased respiratory effort, ventilation-perfusion mismatch, reduced lung compliance, psychological distress, etc. However, clinicians often view severe dyspnea as an indicator of impending respiratory failure, which typically prompts decisions to intubate. This study, however, raises an important clinical question: For patients with high dyspnea scores but relatively preserved oxygenation, could noninvasive strategies, such as high-flow oxygen therapy, be used to delay or even avoid intubation? The study also found that moderate-to-severe dyspnea (VAS ≥ 40 mm) was independently associated with a higher risk of intubation, but not with mortality. This finding also emphasizes the clinical question: For patients with high dyspnea scores but relatively preserved oxygenation, is early intubation truly necessary, and does it improve long-term outcomes? In conclusion, this study highlights the complexity of dyspnea as both a clinical symptom and a predictor of outcomes. The observed discordance between dyspnea severity and oxygenation reinforces the need for a comprehensive approach to patient assessment, balancing subjective symptoms with objective physiological measures, which may help optimize respiratory management and improve clinical outcomes in critically ill patients. Additionally, there are some areas in the study that lack clarity. For example, in the multivariable regression analysis, when dealing with dummy variables such as VAS or oxygenation status, only one p-value is reported. Finally, we extend our gratitude to Dr. Demoule and colleagues for their valuable work, and we hope our perspectives contribute to a deeper understanding of these findings.