Pulse Oximetry
Basic Principle
Pulse oximetry is a noninvasive method for continuously estimating the oxygen saturation of arterial blood (SaO2, displayed as SpO2). It works by dual-wavelength spectrophotometry:
- A probe (typically on a finger, earlobe, or nose) shines light at two wavelengths - commonly 660 nm (red) and 940 nm (infrared) - through the tissue to a sensor on the other side.
- Oxyhemoglobin and deoxyhemoglobin have different light absorption spectra: deoxyhemoglobin absorbs more red light, while oxyhemoglobin absorbs more infrared light.
- The device measures the ratio of absorbance at the two wavelengths and calculates the percent saturation of hemoglobin.
(Costanzo Physiology, 7th Ed., p. 226; Schwartz's Principles of Surgery, 11th Ed., p. 474)
Why It Measures Only Arterial Saturation
Skin, venous blood, and capillary blood also absorb light, but they don't pulse. Arterial blood pulses with each heartbeat, causing a small, rhythmic change in the light absorbance signal. The oximeter isolates this pulsatile component and subtracts out the constant "background" absorbance from tissue, venous, and capillary blood - which is how it selectively reports arterial saturation rather than a mixed venous-arterial value. This is why a pulse oximeter needs a detectable pulse (poor peripheral perfusion, hypotension, or vasoconstriction can give inaccurate or unreadable results).
(Costanzo Physiology, 7th Ed., p. 226)
Important Point: It Does NOT Measure PaO2 Directly
Pulse oximetry reports % saturation, not the partial pressure of oxygen (PaO2). You can estimate PaO2 from the SpO2 using the oxygen-hemoglobin dissociation curve, but the relationship is not linear - the curve is sigmoidal, so at saturations above ~90%, large changes in PaO2 produce only small changes in SpO2. This means pulse oximetry can miss significant drops in PaO2 while SpO2 still looks reassuring, especially on the flat upper part of the curve.
Clinical Utility
- One of the most widely used continuous, noninvasive monitoring tools in anesthesia, critical care, and general ward settings.
- Used routinely in intubated/mechanically ventilated patients, during sedation, and in patients with respiratory disease.
- Helps titrate FiO2 and PEEP, and guides weaning from mechanical ventilation.
- Has reduced the need for frequent arterial blood gas sampling, and continuous monitoring in surgical patients is associated with reduced unrecognized deterioration, rescue events, and ICU transfers.
(Schwartz's Principles of Surgery, 11th Ed., p. 474)
Key Limitations and Sources of Error
- Carboxyhemoglobin (CO poisoning): The device cannot distinguish carboxyhemoglobin from oxyhemoglobin, so it falsely reads SpO2 as normal/high even when true oxygen-carrying capacity is severely reduced. Co-oximetry (multi-wavelength) is needed to detect this.
- Methemoglobinemia: When methemoglobin is markedly elevated, the oximeter characteristically reads a fixed ~85% regardless of true saturation.
- Accuracy declines at low saturations: Accuracy starts to degrade below SaO2 of ~92% and becomes unreliable below ~85%.
- Poor perfusion states: Hypotension, peripheral vasoconstriction, cold extremities, or motion artifact can prevent an adequate pulsatile signal.
- Nail polish, dyes (e.g., methylene blue), and ambient light can interfere with the signal.
- Skin pigmentation: This is an active area of concern - recent systematic reviews and meta-analyses have found that pulse oximeters can overestimate true arterial oxygen saturation in patients with darker skin pigmentation, increasing the risk of missed hypoxemia in these patients. This is a clinically important and ongoing safety issue that is not covered in older textbook editions.
(Schwartz's Principles of Surgery, 11th Ed., p. 474)
Recent Evidence Worth Flagging
Two recent systematic reviews/meta-analyses reinforce the skin-pigmentation accuracy problem and should inform clinical interpretation of SpO2 in darker-skinned patients:
- Martin et al. 2024, British Journal of Anaesthesia - systematic review on the effect of skin tone on pulse oximetry accuracy (PMID: 38368234)
- Singh et al. 2024, JMIR - systematic review/meta-analysis on skin pigmentation's impact on pulse oximetry and wearable pulse rate accuracy (PMID: 39388258)
Both support using clinical judgment (and arterial blood gas confirmation when in doubt) rather than relying on SpO2 alone in patients with darker skin, particularly near borderline saturation thresholds.