Pulse oximetry relies on the principle of spectral analysis, which is the method of analyzing physiochemical properties of matter based on their unique light-absorption characteristics. For blood, absorbance of transmitted light is dependent on the concentration of hemoglobin species.

Oximeters are made up of a light source, photodetector, and microprocessor. Light-emitting diodes (LED) emit high-frequency signals at 660 nm (red) and 940 nm (infrared) wavelengths. When positioned to traverse (transmittance) or reflect from (reflectance) a cutaneous vascular bed, the opposed photodetector measures light intensity of each transmitted signal. Signal processing exploits the pulsatile nature of arterial blood to isolate arterial saturation. The microprocessor averages these data over several pulse cycles and compares the measured absorption to a reference standard curve to determine hemoglobin saturation, which is displayed as percentage of oxyhemoglobin (SpO₂). SpO₂ and SaO₂ correlation vary with manufacturer but exhibit high accuracy (±2%) within normal physiologic range and circumstances.

Two oximetry techniques are used in clinical practice. Transmission oximetry deploys the LED and photodetector on opposite sides of a tissue bed (e.g., digit, nares, and ear lobe) such that the signal must traverse tissue. Reflectance oximeters position the LED and photodetector side by side on a single surface and can be placed in anatomic locations without an interposed vascular bed (e.g., forehead). This facilitates more proximal sensor placement with improved response time relative to core body SaO₂.

Pulse oximeters have a number of important physiologic and technical limitations that influence bedside use and interpretation (Table 8-1).

Proper pulse oximetry requires pulse detection to distinguish light absorption from arterial blood relative to the background of other tissues. Abnormal peripheral circulation as a consequence of shock, vasoconstriction, and hypothermia may prevent pulsatile flow detection. Heart rate and plethysmographic waveform display verify arterial sensing, and SpO₂ should be considered inaccurate unless corroborated by these markers. Varying pulse amplitude is easily recognized on the monitor and represents the measure of arterial pulsality at the sampled vascular bed. Quantification in the form of perfusion index is being incorporated into some software to verify signal reliability and gauge microvascular flow.

Even with verified signal detection, measurement bias limits SpO₂ reliability during physiologic extremes. Reliability deteriorates with progressive hypotension below systolic BP of 80 mm Hg. Readings generally underestimate true SaO₂. Severe hypoxemia with SaO₂ <75% is also associated with increased measurement error. However, patients with this severity of hypoxemia are typically receiving maximized intervention, and closer discrimination in this range rarely imparts new information that alters management.

A number of physical factors affect pulse oximetry accuracy. Signal reliability is influenced by sensor exposure to extraneous light, excessive movement, synthetic fingernails, nail polish, intravenous dyes, severe anemia, and abnormal hemoglobin species. Diligent probe placement and shielding the probe from extraneous light should be routine. Surface extremity warming may improve local perfusion to enable arterial pulse sensing, but SpO₂ accuracy using this technique is not confirmed. Transverse digital sensor orientation overcomes limitations resulting from nail abnormalities.

Carboxyhemoglobin (CO-Hgb) and methemoglobin (Met-Hgb) absorb light at different wavelengths. Co-oximeters (and some new generation pulse oximeters) use four wavelengths of light stimulus to selectively discriminate these species. However, CO-Hgb absorbance is close to oxyhemoglobin such that most conventional pulse oximeters sum their measurement and give artifactually high SpO₂ reading. Met-Hgb produces variable error, depending on the true oxy- and Met-Hgb levels. SpO₂ classically approximates 85% in severe toxicity.

Pulse oximetry readings lag the patient’s physiologic state. Signal averaging of 4 to 20 seconds is typical of most monitors. Delay because of sensor anatomic location and abnormal cardiac performance compound the lag relative to central SaO₂. Forehead and ear probes are closer to the heart and respond more quickly than distal extremity probes. Response difference compared to central SaO₂
is also compounded by hypoxemia (i.e., starting on the steep portion of the oxyhemoglobin dissociation curve) and slower peripheral circulation such as low cardiac output states. As such, forehead reflectance probes are often preferred in critically ill patients. All of these response delays become more clinically important during rapid desaturation such as may occur during airway management.

Hemoglobin saturation is just one part of the assessment of systemic oxygenation. Although monitoring is continuous, SpO₂ provides momentary information on arterial saturation without detailing insight into systemic oxygenation and respiratory reserve. The physiologic context of oximetry is critical for appropriate interpretation and assists estimation of a patient’s cardiopulmonary reserve for planning and execution of an airway management plan.

Oximetry measures arterial hemoglobin saturation but not the arterial oxygen tension or oxygen content of blood. The oxyhemoglobin dissociation curve describes the relationship of oxygen partial pressure (PaO₂) and saturation (SaO₂). Its sigmoidal shape hinges on varying hemoglobin affinity with successive oxygen binding. It is important to note that SpO₂ provides poor correlation with PaO₂ in the normal range. Normal SaO₂ is associated with a wide range of PaO₂ (80 to 400 mm Hg), which includes two extremes of oxygen reserve. Similarly, oximetry is insensitive at detecting progressive hypoxemia in patients with high-baseline PaO₂ (Table 8-1). Correlation is established in the hypoxemic range at and below the upper inflection point of the oxyhemoglobin curve (PaO₂ <60 mm Hg approximating SaO₂ 90% at normal pH) where desaturation is rapid with declining PaO₂.

Hemoglobin saturation must also be interpreted in the context of inspired oxygen fraction (FiO₂) to provide insight into gas exchange and physiologic reserve. Simple observation at the bedside provides qualitative assessment. More formal calculation of the SpO₂/FiO₂ (SF) ratio is advocated. For the same reasons previously discussed, SF ratio correlates with PaO₂/FiO₂ (PF) ratio in the hypoxemic range (SpO₂ <90%) but not in the normal range. As such, observation of the patient’s condition before supplemental oxygen escalation or preoxygenation provides more insight into the physiologic state. Correct interpretation of SpO₂ relative to FiO₂ is also important in assessing for failure of noninvasive ventilation. Hypoxemia and/or requirement of oxygen escalation above FiO₂ >70% leaves a thin margin of physiologic reserve for preoxygenation and execution of safe, uncomplicated endotracheal intubation.

Although PaO₂ (with or without conscious calculation of PF ratio) is a traditional and reliable gauge of pulmonary gas exchange and reserve, measurement of PaO₂ through arterial blood gas sampling before airway management is not generally helpful. The aim to maximize preoxygenation in all patients supersedes this strategy. However, knowledge of these principles and relationships provides insight into physiologic events and the fallibility of current technology during the management of critical illness.

The context of cardiac performance is also vital to interpretation of oximetry data. Although saturated hemoglobin accounts for the majority of blood oxygen content, systemic oxygen delivery is largely regulated (and limited) by cardiac performance. Pertinent to airway management, rapid desaturation and delayed response to pulmonary oxygenation should be anticipated in the setting of low cardiac output.

Finally, oxygen saturation is an unreliable gauge of ventilation, PaCO₂ level or acid-base status. Normal arterial saturation does not ensure appropriate ventilation. Oxygenation often is adequate with minimal volume of gas exchange, whereas carbon dioxide (CO₂) removal relies on pulmonary ventilation. Arterial blood gas analysis is the traditional means to measure PaCO₂, but alternative noninvasive CO₂ monitoring provides additional insight.

CO₂

Stay updated, free articles. Join our Telegram channel

Join