Relationship between SpO2/FiO2 and PaO2/FiO2 in patients with acute respiratory distress syndrome: a single-center, retrospective cohort study from Xining, China.

BACKGROUND

The updated global criteria for diagnosing acute respiratory distress syndrome (ARDS) include SpO2/FiO2 as a substitute for PaO2/FiO2 to enhance practicality in settings with limited resources and reduce reliance on arterial blood gas testing. However, the applicability of these SpO2/FiO2 thresholds in high-altitude regions remains unvalidated. This study investigated the relationship between the SpO2/FiO2 and PaO2/FiO2 in patients with ARDS in the Xining region of Qinghai and validated whether the critical values for hypoxemia severity classification based on SpO2/FiO2 are applicable in this population. A retrospective analysis was conducted on the clinical data of 276 patients with ARDS admitted to the intensive care unit (ICU) of Qinghai Provincial People's Hospital from October 2020 to October 2024. All patients who met the inclusion criteria were classified into mild, moderate, and severe ARDS as defined by the Berlin Definition. Spearman's correlation analysis was used to assess the correlation between SpO2/FiO2 (corrected) and PaO2/FiO2 (corrected). Curve estimation was used to derive the linear and nonlinear relationships between SpO2/FiO2 and PaO2/FiO2, as well as between SpO2/FiO2 (corrected) and PaO2/FiO2 (corrected). Receiver operating characteristic curve analysis was used to validate whether the cutoff values proposed by the new global definition for SpO2/FiO2-based hypoxemia severity stratification apply to Xining.

RESULTS

A significant nonlinear relationship was observed between the SpO2/FiO2 and PaO2/FiO2 ratios. The optimal fitting model was a power function model: SpO2/FiO2 (corrected) = 5.661 × [PaO2/FiO2 (corrected)]0·71. The area under the curve (AUC) of the SpO2/FiO2(corrected) for predicting mild ARDS was 0.912, with an optimal diagnostic threshold cutoff value of 360.4. For moderate ARDS, the AUC was 0.487, with a cutoff value of 238.8. For severe ARDS, the AUC was 0.956, with an optimal cutoff value of 168.

CONCLUSION

This study revealed a nonlinear relationship between the SpO2/FiO2 and PaO2/FiO2 ratios in patients with ARDS in Xining, Qinghai Province, a characteristic that may better align with the pathophysiological features of high-altitude regions. The globally redefined hypoxemia grouping thresholds based on SpO2/FiO2 may not apply to this area, suggesting that regional diagnostic criteria should be established according to altitude variations.