Hyperoxemia and Oxygen Reserve Index During Neurointerventional Anesthesia

Background and rationale:

Oxygen is routinely given during general anesthesia to prevent hypoxemia. However, oxygen given beyond the patient's needs may cause arterial hyperoxemia, which has been linked to vasoconstriction, oxidative stress, and absorption atelectasis. Normobaric hyperoxia has been shown to reduce cerebral blood flow, and cumulative hyperoxemia exposure has been associated with unfavorable outcomes in ventilated patients with aneurysmal subarachnoid hemorrhage. Although these findings come from critically ill populations, they provide a physiological rationale for describing arterial oxygen levels in patients undergoing elective neurointerventional procedures, in whom maintaining cerebral perfusion is essential.

Standard pulse oximetry (SpO2) plateaus at high arterial oxygen levels and cannot detect hyperoxemia. The Oxygen Reserve Index (ORi) is a continuous, non-invasive, dimensionless index (0 to 1) derived from multi-wavelength pulse co-oximetry, reported to reflect oxygenation changes in the moderately hyperoxic range (approximately PaO2 100-200 mmHg). The ORi-PaO2 relationship varies across populations and clinical settings, and it has not been described in stable adults undergoing elective neurointerventional procedures under general anesthesia.

Study design:

This is a single-center, prospective, observational study conducted at the Department of Anesthesiology and Reanimation and the Interventional Radiology Unit of University of Health Sciences Fatih Sultan Mehmet Training and Research Hospital, Istanbul, Türkiye. No randomization, group allocation, or clinical intervention will be performed.

Procedures:

Anesthesia induction and maintenance, ventilator settings, FiO2, fluids, vasoactive drugs, and all other management will follow the attending anesthesiologist's routine practice. In addition to standard monitoring, an ORi-capable pulse co-oximetry sensor will be applied. ORi values will be recorded by the investigator but will not be used for clinical decision-making, and no changes in FiO2, ventilation, or treatment will be made based on ORi.

ORi, SpO2, FiO2, heart rate, mean arterial pressure, end-tidal carbon dioxide, body temperature, perfusion index, and signal quality will be recorded at the following time points: 5 minutes after intubation and initiation of mechanical ventilation (T1), 30 minutes (T2) and 60 minutes (T3) after the start of the procedure, every 60 minutes thereafter starting at 120 minutes if the procedure continues (T4 onward), and at the end of the procedure before FiO2 is increased for emergence (Tend).

Whenever an arterial blood gas sample is drawn by the clinical team for routine care, PaO2, PaCO2, pH, SaO2, hemoglobin, and lactate will be recorded and matched with ORi, SpO2, FiO2, perfusion index, and hemodynamic values recorded at the time of sampling or within 1 minute. No additional blood samples will be drawn for research purposes, and the timing and number of routine samples will not be altered.

Patients without a routine arterial blood gas will remain in the cohort for descriptive analyses; PaO2-based analyses will include all valid matched ORi-PaO2 pairs.

Sample size and statistical analysis:

Assuming an expected hyperoxemia proportion of 50%, a 95% confidence level, and a precision of +/-10%, the minimum required sample size is 96 patients; 100 patients will be enrolled to account for missing data. The proportion of matched measurements with PaO2 >=150 mmHg will be reported with 95% confidence intervals. Repeated measurements will be analyzed with mixed-effects models. The ORi-PaO2 relationship, including possible nonlinearity and plateau behavior, will be evaluated with appropriate regression models. The ability of ORi to discriminate PaO2 >=150 and >=200 mmHg will be explored with ROC analysis, accounting for within-patient clustering. A p value <0.05 will be considered statistically significant.

Data protection:

All data will be de-identified, recorded on study case report forms, and used by the study team for scientific purposes only.