A procedure for determining subject-specific pulse oxygen saturation response

• Published in: Medical & biological engineering & computing Vol. 58; no. 4; pp. 753 - 761
• Main Authors: Burk, Kyle M., Orr, Joseph A.
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.04.2020; Springer Nature B.V
• Subjects: Biomedical and Life Sciences; Biomedical Engineering and Bioengineering; Biomedicine; Computer Applications; Correlation analysis; Hemoglobin; Human Physiology; Humans; Imaging; Iterative methods; Models, Biological; Original Article; Oximetry - methods; Oxygen; Oxygen - blood; Oxygen content; Oxygen Inhalation Therapy - methods; Oxygen saturation; Oxygen therapy; Oxyhemoglobin; Oxyhemoglobins - analysis; Parameter estimation; Partial pressure; Patients; Physical factors; Precision Medicine; Preoperative Care; Pulse oximetry; Radiology; Surgery; Oxygen saturation response; The Hill equation; Non-linear least-squares; Subject-specific modeling
• ISSN: 0140-0118; 1741-0444; 1741-0444
• DOI: 10.1007/s11517-019-02105-8

The oxyhemoglobin dissociation curve describes the relationship between the partial pressure of oxygen and the percent of hemoglobin saturated with oxygen and varies with chemical and physical factors that differ for every patient. If variability could be determined, patient-specific oxygen therapy could be administered. We have developed a procedure for characterizing variations in the oxygen dissociation curve. The purpose of this study was to validate this procedure in surgical patients. The procedure uses an automated system to alter oxygen therapy during surgery, within safe operational levels, and fit to Hill’s equation non-invasive measurements of end-tidal oxygen and peripheral pulse oxygen saturation. The best-fit parameters for the Hill equation, estimated by iterative least squares, provide an apparent dissociation curve, meaningful of the patient-specific pulse oximeter response. Thirty-nine patients participated in this study. Using patient-specific parameter values increases correlation when compared with standard values. The procedure improved the model fit of patient saturation values significantly in 19 patients. This paper has demonstrated a procedure for determining patient-specific pulse oximeter response. This procedure determined best-fit parameters resulting in a significantly improved fit when compared with standard values. These best-fit parameters increased the coefficient of determination R 2 in all cases. Graphical Abstract This patient-specific procedure improves fit significantly compared to standard estimates