FiO2 requirements during general anesthesia with dual-lung ventilation: a prospective pilot study

Jing Sun, Jiabao Chen, Wei Wang, Hongfa Wang, Jintao Liu, Mei Cheng, Zhen-feng ZhouFrontiers in Medicine 2026 · DOI

TL;DR

This pilot study establishes a strong linear correlation between the fraction of inspired oxygen (FiO2) and arterial oxygen partial pressure (PaO2) during general anesthesia with dual-lung ventilation. The findings suggest that an FiO2 of 40% is generally sufficient to maintain adequate oxygenation (PaO2 of 100-200 mmHg) in most patients, and exceeding 50% is often unnecessary.

What Problem Does This Paper Address?

Intraoperative hyperoxemia (excessively high oxygen levels) and prolonged oxygen exposure are common during general anesthesia. While guidelines suggest titrating FiO2 to maintain adequate oxygen saturation, the precise relationship between FiO2 and PaO2 in patients undergoing dual-lung ventilation is not well-defined. This lack of clarity can lead to either insufficient oxygenation or potentially harmful hyperoxemia, which has been linked to increased risks of respiratory complications and mortality. Existing models for predicting PaO2 from FiO2 are often developed for ICU settings and may not be directly applicable to the dynamic environment of intraoperative anesthesia.

Methods

This was a self-controlled, prospective pilot study involving 50 patients undergoing general anesthesia with a standardized volume-controlled ventilation strategy. After tracheal intubation, FiO2 was incrementally increased in steps of 0.1, starting from 0.3 up to 0.6, and then to 0.8. Each FiO2 level was maintained for at least 30 minutes before arterial blood gas samples were collected to measure PaO2 and other parameters. Statistical analysis primarily used repeated-measures correlation (rmcorr) to assess the relationship between FiO2 and PaO2, and linear mixed-effects models to further quantify this relationship while accounting for individual patient variability.

Datasets

Not discussed in the paper. The study collected data from 50 patients undergoing general anesthesia.

Key Insights and Ideas

  • A Clear Linear Relationship: The most significant finding is the robust, linear correlation between FiO2 and PaO2. This suggests that as FiO2 increases, PaO2 increases in a predictable manner within the studied range.
  • The "Sweet Spot" for Oxygenation: The study identifies a specific FiO2 range (25%-46%) that appears to be optimal for maintaining PaO2 within the clinically recommended 100-200 mmHg range.
  • Over-Oxygenation is Common: The data strongly suggests that many patients receive higher FiO2 levels than necessary, with a significant portion exceeding 200 mmHg PaO2 at 50% FiO2. This highlights the potential for routine over-oxygenation.
  • SpO2 Limitations: The study implicitly points out the limitations of SpO2 monitoring, as it remained high and clustered near saturation even when PaO2 levels were potentially suboptimal or excessively high, underscoring the need for blood gas analysis in certain situations.
  • FiO2/PaO2 Ratio Nuances: While FiO2 and PaO2 are strongly correlated, the relationship between FiO2 and the PaO2/FiO2 ratio is weaker. This ratio is more informative for assessing pulmonary gas exchange efficiency, especially in cases of V/Q mismatch or diffusion impairment.

Key Results

  • A strong linear correlation between FiO2 and PaO2 was observed (rmcorr r = 0.967, p < 0.001).
  • At an FiO2 of 40%, 78.0% of patients achieved a PaO2 between 100 and 200 mmHg.
  • At an FiO2 of 50%, over 68.0% of patients exceeded 200 mmHg PaO2.
  • The study suggests that an FiO2 range of 25%-46% is generally sufficient to maintain PaO2 within the 100-200 mmHg range.
  • A weak but statistically significant correlation was found between FiO2 and the PaO2/FiO2 ratio (r = 0.290, p < 0.001).

Limitations and Future Work

The authors acknowledge several limitations:

  • Single-Center Study: The study was conducted at a single center with a specific patient population (ASA I-III, elective surgery, no lung disease), limiting the generalizability of findings to other settings or patient groups.
  • Exclusion of Pulmonary Disease: Patients with pre-existing lung conditions were excluded, meaning the results may not apply to this vulnerable population.
  • Limited Ventilation Modes: The study focused solely on volume-controlled ventilation, and results might differ with other modes.
  • FiO2 Range: The study explored FiO2 from 0.3 to 0.8, so the relationship beyond this range is not characterized.
  • Lack of Outcome Data: The study focused on physiological correlations and did not assess clinical outcomes like postoperative complications.

Future work should involve multi-center studies with larger cohorts to confirm these findings. Incorporating clinical outcomes and exploring the relationship in patients with pulmonary disease are also crucial next steps. The authors also suggest developing real-time, physiology-based FiO2 titration strategies that integrate continuous gas exchange monitoring.

Summary updated

Abstract

Background Hyperoxemia and prolonged oxygen exposure were common during general anesthesia. However, the relationship between FiO 2 and PaO 2 in patients undergoing general anesthesia with dual lung ventilation remained unclear. This prospective pilot study aimed to explore this relationship. Methods A cohort of 50 patients was recruited for this self-controlled, prospective pilot study. A standardized volume-controlled ventilation strategy was applied, with FiO 2 initially set to 0.3 immediately after tracheal intubation. FiO 2 was then increased in steps of 0.1 until it reached 0.6, followed by an increase to 0.8. Each FiO 2 step was maintained for at least 30 min before blood samples were drawn for blood gas analysis at each point. Results Rmcorr analyses revealed a significant correlation between FiO 2 and PaO 2 ( p &amp;lt; 0.001). The correlation coefficient ( r ) was 0.967 and Model convergence was robust, with a gradient value of 0.00003. At an FiO 2 of 40, 78.0% of patients maintained PaO 2 between 100 and 200 mmHg, while more than 68.0% exceeded 200 mmHg at 50% FiO 2 levels. Rmcorr analysis revealed a weak but statistically significant correlation between FiO 2 and PaO 2 /FiO 2 ( r = 0.290; p &amp;lt; 0.001). Conclusion A significant linear correlation was identified between FiO 2 and PaO 2 during general anesthesia with dual-lung ventilation. In this prospective pilot study, our findings suggested that maintaining 40% FiO 2 was generally sufficient for most patients to achieve PaO 2 levels of 100–200 mmHg, unless specific clinical conditions require otherwise. Importantly, FiO 2 need not exceed 50% in most patients undergoing general anesthesia with dual-lung ventilation. Clinical trial registration http://www.chictr.org.cn , identifier ChiCTR2500095383.
BibTeX
@inproceedings{sun2026fio,
  author = {Jing Sun and Jiabao Chen and Wei Wang and Hongfa Wang and Jintao Liu and Mei Cheng and Zhen-feng Zhou},
  title = {{FiO2 requirements during general anesthesia with dual-lung ventilation: a prospective pilot study}},
  booktitle = {Frontiers in Medicine},
  year = {2026},
  doi = {10.3389/fmed.2026.1780890},
}

Loading related content...

Figure