Pulse oximetry is a vital, non-invasive tool used extensively in anesthesia to monitor oxygen saturation (SpO2) in arterial blood. It works on the principle of differential light absorption by oxygenated and deoxygenated hemoglobin. The pulse oximeter probe emits two wavelengths of light—red (660 nm) and infrared (940 nm)—through a pulsatile arterial bed, typically placed on the patient’s finger, toe, or ear lobe. Oxygenated hemoglobin (HbO2) absorbs more infrared light, while deoxygenated hemoglobin (Hb) absorbs more red light. Both species absorb equally at the isosbestic point near 805 nm. At each wavelength the signal has a pulsatile (AC) component from arterial inflow and a steady (DC) component from venous blood, tissue, and bone; the device divides AC by DC at each wavelength and forms the ratio of ratios, R = (AC660/DC660) / (AC940/DC940). R is converted to SpO2 with an empirical calibration curve built from healthy volunteers desaturated to about 70%: an R near 0.4 reads 100%, an R of 1.0 reads about 85%, and values below 70% are extrapolated, which is why the stated accuracy (about plus or minus 2-3% between 70% and 100%) worsens at low saturations. Takuo Aoyagi described this pulsatile principle in 1974, and commercial devices followed in the early 1980s (Severinghaus and Honda, 1987).
SpO2 reports saturation, not oxygen tension. The two are linked by the sigmoid oxyhemoglobin dissociation curve: at the normal adult P50 of 26.6 mmHg half of the hemoglobin is saturated, a PaO2 of 60 mmHg corresponds to roughly 90% saturation, 40 mmHg (mixed venous) to about 75%, and 100 mmHg to about 97-98%. Because the curve is flat above a PaO2 of 100 mmHg, an SpO2 of 100% cannot distinguish a PaO2 of 100 from one of 500 mmHg, and because it is steep below 60 mmHg, saturation falls quickly once it drops under 90%. Acidosis, hypercarbia, fever, and raised 2,3-DPG shift the curve to the right (more unloading at a given PaO2); alkalosis, hypocarbia, hypothermia, fetal hemoglobin, carboxyhemoglobin, and methemoglobin shift it to the left.
The utility of pulse oximetry in perioperative monitoring lies in its ability to provide continuous, real-time data on oxygenation status. Anesthesia providers rely on SpO2 readings to assess whether a patient is adequately oxygenated during anesthesia and surgery, where hypoxia can occur due to factors like airway obstruction, hypoventilation, or impaired gas exchange. Pulse oximetry, therefore, serves as an early warning system, enabling the rapid identification and correction of potentially life-threatening events such as hypoxia.
Additionally, pulse oximetry is critical for monitoring patients under regional or general anesthesia, during sedation, and in post-anesthesia care units (PACUs). Its non-invasive nature makes it preferable over arterial blood gas (ABG) sampling in many settings. Moreover, it is useful for detecting trends in oxygenation, allowing anesthesia providers to intervene before clinical hypoxemia develops.
However, while pulse oximetry has become a cornerstone in patient safety, it is not without limitations. Accuracy can be compromised by external factors such as poor perfusion (vasoconstriction, hypothermia, hypotension, vasopressors), motion artifact, ambient light, venous pulsation (as in severe tricuspid regurgitation, which reads low), dark nail polish or acrylic nails, and intravenous dyes: methylene blue produces a transient fall in SpO2 to as low as 65% for several minutes, and indocyanine green and indigo carmine cause smaller dips. Skin pigmentation matters too: in a 2020 cohort, occult hypoxemia (arterial saturation below 88% despite an SpO2 of 92-96%) was found in 11.7% of Black patients versus 3.6% of White patients, roughly three times as often (Sjoding et al., 2020), and the FDA's January 2025 draft guidance asks manufacturers to validate new devices in 150 or more participants spanning the Monk Skin Tone scale, with at least a quarter in each tone group. Clinically, treat a borderline SpO2 in a darkly pigmented patient as possibly lower than displayed and confirm with an arterial sample when the decision matters. Moreover, the standard pulse oximeter is calibrated against blood containing only oxyhemoglobin and deoxyhemoglobin, so any hemoglobin species with a different absorbance spectrum introduces error, a key issue discussed in subsequent sections.
