Imputation of arterial oxygen partial pressures using pulse oximetry in surgical patients under general anaesthesia: A prospective cohort study.

Abstract

Background and aims: Monitoring oxygen saturation (SpO₂) during general anaesthesia using pulse oximetry is mandatory, though it may not always reflect the actual oxygenation status. Arterial blood gas (ABG) analysis remains the gold standard for measuring the partial pressure of oxygen (PaO₂), but it is invasive and limited by complications. This study aims to validate five existing equations for predicting PaO₂ from SpO₂ and, if applicable, to derive and validate a new equation in adult patients undergoing surgeries under general anaesthesia.

Methods: A prospective cohort study was conducted on adult patients undergoing general anaesthesia and requiring ABG measurements at several time points intraoperatively. Using SPSS software, a Bland-Altman analysis was performed to assess the agreement between derived and measured PaO2 values using five existing equations. Regression analysis was then performed to develop a new equation for predicting PaO₂, which was validated in a second cohort.

Results: During the initial phase, 150 ABG samples were collected from 83 patients. Bland-Altman analysis revealed weak agreements with all existing equations. A new logarithmic equation, El-Khatib's equation (PaO₂ = 10^-25.6 × SpO₂ ^13.9), was derived. The validation phase involved 150 ABG samples from 65 patients, demonstrating strong agreement with El-Khatib's equation (systematic bias of 13 mmHg, limits of agreement: -189 to 214 mmHg).

Conclusion: Five existing equations for predicting PaO₂ from SpO₂ revealed weak agreement in adult patients under general anaesthesia. El-Khatib's equation demonstrated strong potential for predicting PaO₂, providing a reliable non-invasive alternative for determining PaO₂ in clinical practice.