Pulse Oximetry

Quick Answer

Pulse oximetry is a non-invasive optical technique that measures arterial oxygen saturation (SpO2) by exploiting the differential light absorption properties of oxygenated and deoxygenated hemoglobin. The technology combines spectrophotometry (Beer-Lambert Law) with photoplethysmography (pulse detection) to isolate the arterial signal from background tissue absorption.

Core Physics:

• Two wavelengths: Red (660 nm) absorbed more by deoxyhemoglobin (Hb); Infrared (940 nm) absorbed more by oxyhemoglobin (HbO2)
• Ratio of Ratios (R): R = (AC/DC)660 / (AC/DC)940, empirically calibrated to SpO2
• Calibration: Derived from healthy volunteer studies correlating R to arterial blood gas SaO2

Critical Limitations:

• Dyshemoglobinemias: COHb causes falsely high readings; MetHb causes plateau at ~85%
• Skin pigmentation: Overestimates SpO2 by 1-3% in darker skin (occult hypoxemia risk)
• Low perfusion states: Vasoconstriction, hypothermia, shock reduce signal quality
• Motion artifact: Creates noise interpreted as desaturation

Clinical Pearl: Standard pulse oximeters measure functional saturation (HbO2/(HbO2+Hb)) and cannot detect carboxyhemoglobin or methemoglobin—arterial blood gas co-oximetry is required when dyshemoglobinemia is suspected. [1,2]

---

Physics Principles

Spectrophotometry and the Beer-Lambert Law

The fundamental physics underlying pulse oximetry is the Beer-Lambert Law, which describes the attenuation of light as it passes through an absorbing medium. This law states that absorbance is proportional to the concentration of the absorbing substance and the path length of light through the medium. [3,4]

The Beer-Lambert Equation:

A = \varepsilon \cdot c \cdot l

Where:

• A = Absorbance (dimensionless)
• ε = Molar extinction coefficient (L·mol⁻¹·cm⁻¹)
• c = Concentration of absorbing substance (mol/L)
• l = Path length through medium (cm)

Alternatively expressed as transmitted light intensity:

I = I_0 \cdot e^{-\varepsilon c l}

Where:

• I = Transmitted light intensity
• I₀ = Incident light intensity

Assumptions and Limitations: The Beer-Lambert Law assumes a homogeneous, non-scattering medium with monochromatic light. In biological tissue, these assumptions are violated by:

1. Light scattering by red blood cells (Mie scattering)
2. Multiple absorbing substances (hemoglobin species, melanin, tissue)
3. Variable path length through pulsatile arterial blood

These deviations necessitate empirical calibration rather than pure theoretical calculation. [5,6]

Hemoglobin Absorption Characteristics

Hemoglobin exists in multiple forms with distinct optical absorption spectra. The two primary forms relevant to standard pulse oximetry are oxyhemoglobin (HbO2) and deoxyhemoglobin (Hb), which have dramatically different light absorption characteristics at specific wavelengths. [7]

Extinction Coefficients at Key Wavelengths:

Wavelength Selection Rationale:

• 660 nm (Red): Maximum differential absorption between Hb and HbO2. Deoxyhemoglobin absorbs approximately 10× more red light than oxyhemoglobin.
• 940 nm (Infrared): HbO2 absorbs approximately 60% more infrared light than Hb, providing good discrimination.
• 805 nm (Isosbestic Point): The wavelength where Hb and HbO2 have identical extinction coefficients. Used in some advanced systems as a reference for total hemoglobin measurement, independent of saturation. [8,9]

Photoplethysmography and Pulse Detection

Photoplethysmography (PPG) is the technique of detecting blood volume changes in tissue using light. The pulse oximeter uses PPG to isolate the pulsatile arterial signal from the constant background absorption of venous blood, tissue, bone, and skin pigmentation. [10]

Signal Components:

Waveform Analysis: The AC component represents the incremental light absorption caused by arterial blood volume expansion during systole. By analyzing only this pulsatile component, the oximeter effectively "subtracts" the constant absorption of all other tissues, isolating the arterial oxygen saturation.

Signal Processing:

1. High-pass filtering to extract AC component
2. Separate analysis at 660 nm and 940 nm
3. Calculation of AC/DC ratio at each wavelength
4. Determination of R value from ratio comparison
5. Conversion to SpO2 using calibration algorithm [11,12]

LED and Photodiode Technology

Modern pulse oximeters use light-emitting diodes (LEDs) as light sources and silicon photodiodes as detectors. The LEDs emit light in narrow spectral bands centered at 660 nm (red) and 940 nm (infrared), typically alternating at 200-600 Hz to allow sequential measurement at each wavelength. [13]

Sensor Configurations:

Transmittance Mode: Light passes completely through a thin tissue bed (typically 5-10 mm). The finger is the most common site, with earlobe and toe as alternatives. Requires adequate perfusion and tissue thin enough for light penetration.

Reflectance Mode: Light is backscattered from tissue and detected by a photodiode positioned 5-15 mm from the LED. Used when transmittance sites are unavailable (e.g., forehead sensors in cardiac surgery). Forehead reflectance sensors may respond faster to central desaturation because they sample blood closer to the core circulation. [14,15]

Temporal Sequence:

1. Red LED on (660 nm measurement)
2. Red LED off, IR LED on (940 nm measurement)
3. Both LEDs off (ambient light measurement for subtraction)
4. Cycle repeats at 200-600 Hz

This rapid cycling with ambient light subtraction helps minimize interference from surgical lights and sunlight. [16]

Signal Processing and Averaging

Modern pulse oximeters employ sophisticated signal processing algorithms to improve accuracy and reduce artifact:

Averaging Periods:

• Short averaging (2-4 seconds): Faster response to rapid desaturation; more susceptible to artifact
• Long averaging (8-16 seconds): Smoother display; delayed response to acute changes

Motion Artifact Management:

• Signal extraction technology (SET) algorithms
• Frequency analysis to distinguish motion from cardiac pulsation
• Multiple sensor comparison (when available)
• Plethysmographic waveform quality indicators

Display Lag: Due to signal averaging, there is typically a 3-10 second delay between actual arterial desaturation and displayed SpO2 change. This lag increases with longer averaging times and can be clinically significant during rapid desaturation (e.g., difficult airway). [17,18]

---

Key Equations

Beer-Lambert Law Application

Basic Form:

A = \log_{10}\left(\frac{I_0}{I}\right) = \varepsilon \cdot c \cdot l

For Multiple Absorbers:

A_{total} = \sum_{i} \varepsilon_i \cdot c_i \cdot l = (\varepsilon_{Hb} \cdot c_{Hb} + \varepsilon_{HbO_2} \cdot c_{HbO_2}) \cdot l

Transmitted Intensity:

I = I_0 \cdot e^{-(\varepsilon_{Hb} \cdot c_{Hb} + \varepsilon_{HbO_2} \cdot c_{HbO_2}) \cdot l}

Functional Oxygen Saturation Definition

SpO_2 = \frac{[HbO_2]}{[HbO_2] + [Hb]} \times 100\%

This is functional saturation—the percentage of functional hemoglobin (capable of carrying oxygen) that is oxygenated. Standard two-wavelength pulse oximeters cannot distinguish other hemoglobin species. [19]

Fractional Saturation (CO-oximetry):

Fractional\ SaO_2 = \frac{[HbO_2]}{[HbO_2] + [Hb] + [COHb] + [MetHb]} \times 100\%

This requires multiwavelength analysis (≥4 wavelengths) and is measured by blood gas co-oximeters or advanced pulse co-oximeters.

R Ratio Calculation

The fundamental measurement in pulse oximetry is the ratio of ratios (R):

R = \frac{(AC/DC)_{660}}{(AC/DC)_{940}}

Where:

• AC = Pulsatile (arterial) component of absorption
• DC = Non-pulsatile (baseline) component
• 660 = Red wavelength
• 940 = Infrared wavelength

Physical Interpretation of R:

• The AC/DC ratio normalizes for differences in LED intensity, tissue thickness, and photodiode sensitivity
• R effectively compares the relative pulsatile absorption at red vs. infrared wavelengths
• R is independent of absolute light intensities—only the ratio matters [20,21]

SpO2 Calibration Algorithm

The relationship between R and SpO2 is determined empirically through human volunteer studies. Healthy volunteers are desaturated under controlled conditions while simultaneously measuring R values and arterial blood gas oxygen saturation.

Linear Approximation:

SpO_2 = A - B \times R

Where typical values are A ≈ 110 and B ≈ 25 (manufacturer-dependent).

Polynomial Algorithm:

SpO_2 = \frac{k_1 - k_2 \cdot R}{k_3 - k_4 \cdot R}

More accurate, especially at lower saturations. Constants (k1-k4) are proprietary and differ between manufacturers.

Empirical R-to-SpO2 Relationship:

Calibration Limitation: Pulse oximeter calibration studies are ethically limited to SpO2 values ≥70-75% in healthy volunteers. Below this range, algorithms extrapolate from limited data and become increasingly inaccurate. [22,23]

AC/DC Ratio Mathematical Derivation

For a pulsatile tissue bed, the transmitted light intensity varies with the cardiac cycle:

I(t) = I_{DC} + I_{AC} \cdot \sin(\omega t)

The AC/DC ratio at each wavelength reflects the fractional change in light absorption due to arterial pulsation:

\frac{AC}{DC} = \frac{I_{max} - I_{min}}{I_{DC}} \approx \varepsilon_{arterial} \cdot \Delta l

Where Δl is the change in optical path length due to arterial expansion (typically 0.01-0.05 mm). [24]

---

Sources of Error

Motion Artifact

Motion is one of the most common causes of pulse oximetry inaccuracy and false alarms. Movement creates spurious signals that the algorithm may interpret as pulsatile arterial absorption, leading to erroneous SpO2 values. [25,26]

Mechanisms:

1. Venous blood pulsation: Movement can cause venous blood to oscillate, creating false "pulsations"
2. Sensor displacement: Physical movement of the sensor changes the optical path
3. Signal degradation: Low-frequency motion components overlap with cardiac frequencies (1-3 Hz)

Clinical Manifestations:

• SpO2 dropping to ~85% during motion (where R ≈ 1.0)
• Loss of reliable plethysmographic waveform
• Intermittent signal dropout
• Excessive false alarms (alarm fatigue)

Solutions:

• Motion-resistant algorithms (Signal Extraction Technology)
• Alternative sensor sites (earlobe less affected by arm movement)
• Secure sensor application
• Recognition of motion artifact patterns [27]

Low Perfusion States

Pulse oximetry requires adequate arterial pulsation to distinguish the AC signal from background noise. In states of peripheral vasoconstriction or low cardiac output, the signal-to-noise ratio deteriorates. [28,29]

Causes of Poor Perfusion:

Effects on Measurement:

• Weak or absent plethysmographic waveform
• Increased signal averaging time
• Delayed response to actual saturation changes
• Potential for complete signal loss

Management:

• Warming the extremity
• Using forehead reflectance sensor (less affected by peripheral vasoconstriction)
• Using earlobe sensor (central location, better perfused)
• Recognizing limitations and using arterial blood gas if clinical concern

Ambient Light Interference

External light sources can interfere with the photodiode signal, particularly when they contain wavelengths near 660 nm or 940 nm. [30]

Common Interfering Sources:

• Surgical lights (xenon and fluorescent)
• Direct sunlight
• Bilirubin phototherapy lights
• Infrared warming lamps

Mitigation: Modern oximeters sample with LEDs off to measure and subtract ambient light. Shielding the sensor with opaque material further reduces interference. Pulsatile interference at mains frequency (50/60 Hz) is filtered electronically.

Dyshemoglobinemias

Standard pulse oximeters measure only at two wavelengths and assume all hemoglobin is either HbO2 or Hb. Abnormal hemoglobin species cause characteristic errors. [31,32,33]

Carboxyhemoglobin (COHb):

• Absorption at 660 nm similar to HbO2
• Effect: Falsely elevated SpO2 (reads COHb as HbO2)
• Clinical scenario: Carbon monoxide poisoning—SpO2 may read 98% despite severe tissue hypoxia
• CO saturation = (true SaO2 - SpO2) × 1.1 (rough approximation)
• Management: Arterial blood gas with CO-oximetry; do not trust pulse oximeter in suspected CO poisoning

Methemoglobin (MetHb):

• Absorbs light at both 660 nm and 940 nm approximately equally
• Effect: As MetHb rises, R approaches 1.0, and SpO2 plateaus at ~85%
• Regardless of true saturation (whether higher or lower than 85%)
• Clinical scenario: Acquired methemoglobinemia (dapsone, local anesthetics, nitrates)
• Management: Blood gas CO-oximetry for accurate MetHb level; suspect when SpO2 "fixed" at 85% and doesn't respond to oxygen

Sulfhemoglobin:

• Rare; absorbs at 660 nm
• Effect: Falsely low SpO2 (readings 75-85%)
• Often confused with methemoglobinemia initially

Skin Pigmentation

Recent research has demonstrated that pulse oximetry overestimates oxygen saturation in patients with darker skin pigmentation, creating the phenomenon of "occult hypoxemia." This has significant implications for patient safety and health equity. [34,35,36,37]

Mechanism: Melanin absorbs light at wavelengths used by pulse oximeters (particularly 660 nm). Higher melanin concentrations alter the optical path and absorption characteristics in ways not accounted for by calibration algorithms developed predominantly in lighter-skinned volunteer cohorts.

Evidence:

Magnitude of Error:

• Typical overestimation: 1-3% SpO2 in darker-skinned patients
• At SpO2 92-96%, occult hypoxemia (SaO2 <88%) occurs 2-3× more frequently

Nail Polish and Dyes

Nail Polish:

• Blue, black, green: May cause falsely low readings by absorbing 660 nm light
• Red: Minimal effect (transmits red light)
• Acrylic nails: May reduce signal amplitude but generally don't affect accuracy
• Solution: Remove nail polish or use alternative site (toe, earlobe)

Intravenous Dyes:

• Methylene blue: Absorbs at 660 nm; causes transient dramatic drop in SpO2 (may read 65%)
• Indocyanine green: Brief SpO2 decrease
• Indigo carmine: Minimal effect
• Duration: Effects typically last 1-5 minutes depending on dose

Anemia and Polycythemia

Severe Anemia: Pulse oximeters measure saturation (percentage of hemoglobin that is oxygenated), not oxygen content. In severe anemia, SpO2 may be normal (98%) despite critically low oxygen delivery due to reduced total hemoglobin. Accuracy of SpO2 itself is generally maintained until hemoglobin <5 g/dL.

Polycythemia: Generally does not affect SpO2 accuracy, though very high hematocrit may slightly reduce signal amplitude. [38]

Other Sources of Error

Electrical Interference:

• Electrocautery can cause transient signal artifact
• MRI environment requires specially designed oximeters

Pulsatile Venous Flow:

• Severe tricuspid regurgitation can cause venous pulsation
• May result in falsely low SpO2 readings

High-Frequency Oscillatory Ventilation:

• Oscillations may be detected as "pulse"
• Can cause spurious readings

---

Clinical Applications

Perioperative Monitoring

Pulse oximetry is considered a standard of care for all patients receiving sedation, regional anesthesia, or general anesthesia. The ANZCA Professional Document PS18 mandates continuous SpO2 monitoring during anesthesia. [39]

Induction of Anesthesia: During rapid sequence induction, the pulse oximeter provides early warning of hypoxemia following apnea. The "desaturation safety margin" (time until SpO2 reaches 90%) depends on pre-oxygenation adequacy, oxygen consumption, and functional residual capacity.

Intraoperative Use:

• Continuous monitoring of arterial oxygenation
• Detection of hypoventilation before clinical cyanosis
• Assessment of peripheral perfusion (plethysmographic waveform amplitude)
• Guidance for FiO2 adjustment

Emergence and Recovery:

• Detection of upper airway obstruction
• Identification of residual neuromuscular blockade effects
• Monitoring during transport to recovery

Critical Care Applications

Continuous Monitoring: In the ICU, pulse oximetry provides continuous non-invasive assessment of oxygenation, reducing the need for frequent arterial blood gas sampling. However, it should complement rather than replace blood gas analysis when clinical decisions depend on accurate oxygenation assessment.

Titrating Oxygen Therapy: Target SpO2 ranges are used to guide oxygen supplementation:

• General ICU: 92-96% (avoid hyperoxia)
• ARDS: 88-92% (permissive hypoxemia)
• COPD with type 2 respiratory failure: 88-92%
• CO poisoning: Do not rely on SpO2; use CO-oximetry

Procedural Sedation: Continuous SpO2 monitoring is mandatory during procedural sedation. Early desaturation detection allows intervention before severe hypoxemia develops. [40,41]

Neonatal and Pediatric Use

Neonatal Screening: Pulse oximetry screening for critical congenital heart disease (CCHD) is performed in newborns before hospital discharge. Pre-ductal (right hand) and post-ductal (either foot) SpO2 are compared.

Target Ranges:

• Preterm neonates: 91-95% (avoiding hyperoxia reduces retinopathy risk)
• Term neonates: 95-99%

Fetal Pulse Oximetry: Historically investigated for intrapartum monitoring but not widely adopted due to technical limitations and unclear clinical benefit.

Remote and Austere Environments

Aeromedical Transport: SpO2 monitoring during RFDS and helicopter retrieval provides continuous assessment despite environmental challenges. Altitude-related considerations include:

• Decreased ambient pressure at altitude reduces PaO2 for any given SpO2
• Pressurized aircraft cabins typically maintain equivalent altitude of 1,500-2,500 m
• Supplemental oxygen may be required to maintain target SpO2

Field Use: Portable, battery-powered oximeters enable monitoring in emergency departments, ambulances, and remote clinic settings where arterial blood gas analysis is unavailable. [42]

---

Indigenous Health Considerations

Skin Pigmentation and Occult Hypoxemia

Recent research has confirmed that pulse oximetry demonstrates reduced accuracy in patients with darker skin pigmentation, including Aboriginal and Torres Strait Islander peoples and Māori. This bias results from the technology's calibration against predominantly lighter-skinned volunteer populations and melanin's interference with light absorption at pulse oximetry wavelengths. [34,35,36,37]

Evidence in Indigenous Populations:

Lanyon et al. (2024) published the first Australian study examining pulse oximetry accuracy specifically in Aboriginal and Torres Strait Islander patients. The study found that Indigenous patients experienced higher rates of occult hypoxemia—where SpO2 readings suggest adequate oxygenation (92-96%) while arterial blood gas reveals true hypoxemia (SaO2 <88%). This phenomenon means Indigenous patients may not receive timely supplemental oxygen or escalation of care based on falsely reassuring SpO2 values.

Similarly, Pene et al. (2023) demonstrated that Māori and Pacific Islander patients in Aotearoa New Zealand have higher occult hypoxemia rates than New Zealand Europeans. The authors recommended that current SpO2 thresholds for treatment initiation may be inappropriate for these populations, advocating for earlier intervention based on clinical assessment rather than SpO2 targets alone.

Clinical Implications and Recommendations

Heightened Clinical Suspicion: Anaesthetists and intensivists should maintain a lower threshold for arterial blood gas analysis in Aboriginal, Torres Strait Islander, and Māori patients when SpO2 readings are in the 92-96% range. Clinical signs of hypoxemia (tachypnea, altered mental status, cyanosis) should prompt ABG regardless of SpO2.

Adjusted Thresholds: Consider treating at higher SpO2 thresholds (e.g., 94-96%) in Indigenous patients to account for potential 2-3% overestimation. This approach trades a small risk of unnecessary oxygen supplementation for reduced risk of occult hypoxemia.

Communicate Uncertainty: Discuss pulse oximetry limitations with patients and families. Aboriginal Health Workers and Māori Health Workers can facilitate culturally appropriate communication about why additional blood testing may be recommended despite "normal" SpO2 readings.

Systemic Advocacy: Healthcare professionals should advocate for improved pulse oximetry technology, regulatory requirements for validation across diverse skin tones, and healthcare system policies that acknowledge these limitations. The Therapeutic Goods Administration (TGA) and Medsafe should require manufacturers to demonstrate accuracy across diverse populations.

Remote and Rural Considerations

Many Aboriginal and Torres Strait Islander communities and rural Māori populations have limited access to arterial blood gas analysis. Remote clinics may rely heavily on pulse oximetry for respiratory assessment. Telemedicine consultation with retrieval services and intensivists should include explicit discussion of pulse oximetry limitations in these populations.

RFDS retrieval teams should consider earlier activation for patients with respiratory symptoms and marginally adequate SpO2, recognizing that true oxygenation may be significantly worse than displayed values.

---

Assessment Content

SAQ Practice Question (20 marks)

Question:

A 58-year-old man with chronic obstructive pulmonary disease presents to the emergency department with increased dyspnea. His pulse oximeter shows SpO2 of 92% on room air. He is Aboriginal Australian with Type 2 diabetes.

(a) Describe the physics principles underlying pulse oximetry, including the Beer-Lambert Law, wavelength selection, and the R ratio calculation. (8 marks)

(b) Explain why pulse oximetry may have reduced accuracy in this patient and outline the specific limitations relevant to his presentation. (6 marks)

(c) A house fire patient is transferred to your care with SpO2 98% but altered consciousness. Explain why this SpO2 may be falsely reassuring and describe the underlying physics. (6 marks)

---

Model Answer:

(a) Physics Principles (8 marks)

Beer-Lambert Law (3 marks):

• Light absorption is proportional to concentration and path length: A = ε × c × l
• Where A = absorbance, ε = extinction coefficient, c = concentration, l = path length
• Transmitted intensity: I = I₀ × e^(-εcl)
• In tissue, must account for multiple absorbers (HbO2, Hb, tissue, melanin) and light scattering
• Deviations from ideal Beer-Lambert conditions necessitate empirical calibration

Wavelength Selection (2 marks):

• Red (660 nm): Deoxyhemoglobin absorbs ~10× more than oxyhemoglobin
• Infrared (940 nm): Oxyhemoglobin absorbs ~60% more than deoxyhemoglobin
• Isosbestic point (805 nm): Equal absorption—used as reference in some systems
• Two wavelengths provide two equations to solve for two unknowns (Hb and HbO2 concentrations)

R Ratio Calculation (3 marks):

• R = (AC/DC)660 / (AC/DC)940
• AC component = pulsatile arterial signal; DC = constant tissue/venous absorption
• Photoplethysmography isolates arterial absorption by analyzing only pulsatile changes
• R is empirically calibrated to SpO2 via healthy volunteer studies
• R ≈ 0.4-0.5 corresponds to SpO2 100%; R ≈ 1.0 corresponds to SpO2 ~85%
• SpO2 calculated via algorithm: SpO2 = (k1 - k2R)/(k3 - k4R)

(b) Accuracy Limitations in This Patient (6 marks)

Skin Pigmentation (3 marks):

• Recent research (Lanyon et al. 2024, PMID 38355209) demonstrates pulse oximetry overestimates SpO2 in Aboriginal and Torres Strait Islander patients
• Melanin absorbs at wavelengths used by pulse oximeters (especially 660 nm)
• Calibration curves derived from predominantly lighter-skinned volunteers
• "Occult hypoxemia": SpO2 92-96% may mask true SaO2 <88%
• Aboriginal patients 2-3× more likely to experience occult hypoxemia
• Recommendation: Lower threshold for ABG; consider treating at higher SpO2 target

Low Perfusion/Diabetes (2 marks):

• Diabetic peripheral vascular disease may reduce pulse amplitude
• Signal-to-noise ratio deteriorates with poor perfusion
• May result in delayed response or inaccurate readings
• Consider alternative sites (earlobe, forehead reflectance)

COPD Considerations (1 mark):

• SpO2 92% may represent point on steep portion of oxyhemoglobin dissociation curve
• Small SpO2 changes represent large PaO2 changes
• Combined with pigmentation bias, true saturation may be significantly lower

(c) Carbon Monoxide Poisoning (6 marks)

Why SpO2 is Falsely Reassuring (3 marks):

• House fire suggests carbon monoxide (CO) exposure
• CO binds hemoglobin with 200-250× greater affinity than oxygen
• Carboxyhemoglobin (COHb) cannot carry oxygen but patient may have high COHb levels
• Despite SpO2 98%, oxygen content may be critically reduced
• Patient's altered consciousness may indicate tissue hypoxia despite "normal" SpO2

Physics Explanation (3 marks):

• Standard pulse oximeters use only 660 nm and 940 nm wavelengths
• COHb has nearly identical extinction coefficient to HbO2 at 660 nm
• Oximeter "reads" COHb as oxyhemoglobin
• SpO2 = functional saturation = HbO2/(HbO2 + Hb)—does not include COHb in denominator
• If true SaO2 60% and COHb 38%, SpO2 may still read ~98%
• Diagnosis requires arterial blood gas with CO-oximetry (multiwavelength analysis)

Management:

• High-flow oxygen 100% via non-rebreather mask
• Urgent ABG with CO-oximetry
• Consider hyperbaric oxygen if severe (COHb >25%, neurological symptoms, pregnancy)
• Do not rely on pulse oximetry for monitoring in CO poisoning

---

Viva Scenario (15 marks)

Examiner: "Tell me how a pulse oximeter works."

Candidate: "A pulse oximeter measures arterial oxygen saturation non-invasively using two physical principles: spectrophotometry based on the Beer-Lambert Law, and photoplethysmography for pulse detection.

The device uses two wavelengths of light—red at 660 nanometers and infrared at 940 nanometers. These wavelengths are chosen because oxygenated and deoxygenated hemoglobin have very different absorption characteristics at these points.

At 660 nanometers, deoxyhemoglobin absorbs about ten times more light than oxyhemoglobin. At 940 nanometers, the reverse is true—oxyhemoglobin absorbs about 60% more than deoxyhemoglobin.

LEDs in the sensor emit light at these two wavelengths in rapid alternation, and a photodiode measures the transmitted light. The sensor then separates the pulsatile arterial component from the constant background absorption of tissue, bone, and venous blood."

Examiner: "How does it isolate the arterial signal?"

Candidate: "The oximeter uses photoplethysmography—detecting blood volume changes with each heartbeat. During systole, arterial blood volume in the tissue increases, causing increased light absorption. This creates a small fluctuating signal called the AC component, typically 1-5% of the total.

The constant background absorption—from tissue, bone, and venous blood—is the DC component. By analyzing the ratio of AC to DC at each wavelength, the oximeter isolates the arterial contribution.

The key calculation is the R ratio: R equals the AC/DC ratio at 660 nm divided by the AC/DC ratio at 940 nm. This R value is then converted to SpO2 using a calibration algorithm derived from human volunteer studies."

Examiner: "What is the calibration curve based on?"

Candidate: "The calibration is empirical, derived from studies where healthy volunteers were deliberately desaturated under controlled conditions while measuring both the R value and simultaneously obtaining arterial blood gas oxygen saturation.

The relationship between R and SpO2 is approximately linear, with R of about 0.4-0.5 corresponding to 100% saturation, and R of 1.0 corresponding to about 85% saturation.

However, these calibration studies were limited to saturations above 70-75% for ethical reasons, so accuracy decreases significantly below this range. Additionally, most calibration cohorts consisted predominantly of lighter-skinned volunteers, which contributes to accuracy issues in patients with darker skin pigmentation."

Examiner: "You mentioned skin pigmentation. Can you expand on this?"

Candidate: "Recent research has demonstrated that pulse oximeters tend to overestimate oxygen saturation in patients with darker skin, including Aboriginal and Torres Strait Islander peoples and Māori patients.

Melanin absorbs light at the wavelengths used by pulse oximeters, particularly at 660 nanometers. This alters the optical characteristics in ways not accounted for by the calibration algorithms.

Studies by Sjoding and colleagues in 2020, and more recently Lanyon and colleagues in 2024 examining Aboriginal and Torres Strait Islander patients, have shown that these populations experience 'occult hypoxemia'—where the SpO2 appears adequate at 92-96% but arterial blood gas reveals true hypoxemia below 88%.

This occurs 2-3 times more frequently in darker-skinned patients and has significant implications for treatment decisions, particularly regarding oxygen therapy thresholds and escalation of care."

Examiner: "What would you recommend clinically?"

Candidate: "I would recommend maintaining a lower threshold for arterial blood gas analysis when caring for Aboriginal, Torres Strait Islander, or Māori patients with respiratory symptoms, particularly when SpO2 is in the 92-96% range.

Clinical signs of hypoxemia should prompt blood gas analysis regardless of SpO2. Consider treating at slightly higher SpO2 targets—perhaps 94% rather than 92%—to account for potential overestimation.

We should also communicate these limitations to patients and families, involving Aboriginal Health Workers or Māori Health Workers to facilitate culturally appropriate discussions about why additional blood testing may be recommended.

More broadly, clinicians should advocate for regulatory requirements that pulse oximeters demonstrate accuracy across diverse skin tones."

Examiner: "What about dyshemoglobinemias?"

Candidate: "Standard two-wavelength pulse oximeters cannot distinguish abnormal hemoglobin species.

Carboxyhemoglobin has nearly identical absorption to oxyhemoglobin at 660 nanometers, so the oximeter reads COHb as HbO2. In carbon monoxide poisoning, SpO2 may read 98% despite severely reduced oxygen-carrying capacity. This is life-threatening because clinicians may be falsely reassured.

Methemoglobin absorbs approximately equally at both wavelengths. As methemoglobin levels rise, the R ratio approaches 1.0, and SpO2 plateaus at around 85% regardless of true saturation. This is the classic finding in methemoglobinemia—SpO2 that doesn't respond to supplemental oxygen.

Both conditions require arterial blood gas with CO-oximetry, which uses multiple wavelengths to distinguish the different hemoglobin species."

Examiner: "What are the sources of motion artifact?"

Candidate: "Motion artifact occurs through several mechanisms. Physical movement can displace the sensor, changing the optical path length. Limb movement can cause venous blood to oscillate, creating false pulsatile signals that the algorithm may interpret as arterial.

The frequency of motion-related signals often overlaps with cardiac frequencies of 1-3 Hz, making filtering difficult.

Clinically, motion artifact typically causes the SpO2 to drop toward 85%—because at an R ratio of 1.0, which random noise approximates, the calibration curve yields approximately 85%.

Modern oximeters use signal extraction technology and advanced algorithms to improve motion tolerance, but no system is immune. Recognition of motion artifact—poor plethysmographic waveform quality, readings that fluctuate or seem inconsistent with clinical status—is important for appropriate interpretation."

---

Viva Scenario 2: Technical Troubleshooting (15 marks)

Examiner: "You are called to the recovery room because a patient's pulse oximeter keeps alarming with low readings. How do you approach this?"

Candidate: "I would approach this systematically, beginning with clinical assessment of the patient before troubleshooting the equipment.

First, I would assess the patient directly—looking at their colour, respiratory effort, level of consciousness, and heart rate. I would auscultate the chest to assess air entry. If the patient appears clinically well-oxygenated with good colour and no signs of respiratory distress, the low SpO2 reading is likely artifactual.

If there is any clinical concern about hypoxemia, I would immediately administer supplemental oxygen and call for help while continuing my assessment."

Examiner: "The patient looks pink and comfortable. What next?"

Candidate: "With clinical hypoxemia excluded, I would examine the pulse oximeter itself and its application.

First, I would check the plethysmographic waveform on the monitor. A poor-quality waveform suggests inadequate signal detection. The waveform should have a clear systolic upstroke and dicrotic notch corresponding to the pulse.

I would examine the sensor placement—ensuring it is properly positioned with the LED and photodiode aligned across the finger, not rotated. The sensor should not be too tight, which can cause venous congestion, or too loose, allowing light leakage.

I would check for external light interference—bright overhead lights or direct sunlight can affect the photodiode. Covering the sensor with an opaque shield may help."

Examiner: "The waveform is weak and irregular. What are the likely causes?"

Candidate: "A weak and irregular waveform suggests poor perfusion or motion artifact.

Poor perfusion causes include:

• Cold peripheries from the temperature of the recovery environment
• Residual effects of vasopressors if used intraoperatively
• Peripheral vasoconstriction from hypothermia or low cardiac output
• The blood pressure cuff cycling on the same arm as the probe

Motion artifact could result from shivering, seizure activity, or simple restlessness as the patient emerges from anesthesia.

I would warm the finger, try an alternative site such as the earlobe which is more centrally perfused, or use a forehead reflectance sensor if available. If the patient is shivering, treatment with warming and possibly low-dose meperidine may improve signal quality."

Examiner: "You notice the patient has dark blue nail polish. Does this matter?"

Candidate: "Yes, nail polish can interfere with pulse oximetry, particularly darker colours.

Blue, black, and green nail polishes absorb light at wavelengths similar to those used by the oximeter—particularly around 660 nm. This can cause falsely low SpO2 readings because the oximeter interprets the additional absorption as increased deoxyhemoglobin.

Red nail polish is less problematic because it transmits red light.

The solution is to remove the nail polish with acetone, rotate the sensor to a different finger without polish, or use an alternative site such as the earlobe or toe. Placing the sensor sideways on the finger may also help, as light passes through the sides rather than through the nail bed.

Importantly, if there is clinical concern about hypoxemia, I should obtain an arterial blood gas rather than relying on the pulse oximeter."

Examiner: "When would you order an arterial blood gas rather than relying on pulse oximetry?"

Candidate: "I would obtain an arterial blood gas in several situations.

First, when pulse oximetry is unreliable due to poor signal quality, motion artifact, or technical factors like nail polish interference.

Second, when dyshemoglobinemia is suspected—particularly carbon monoxide poisoning after house fires or smoke inhalation, or methemoglobinemia from drugs like dapsone, local anesthetics, or nitrates. Standard pulse oximeters cannot detect these conditions.

Third, when precise oxygenation is critical for clinical decisions—such as titrating oxygen in ARDS or assessing severity of respiratory failure.

Fourth, when I need additional information that only a blood gas provides—PaCO2 for ventilation assessment, pH for acid-base status, or lactate for perfusion.

Fifth, in patients with darker skin pigmentation who may have occult hypoxemia despite apparently adequate SpO2—particularly Aboriginal, Torres Strait Islander, or Māori patients with respiratory symptoms.

Sixth, when SpO2 is discordant with clinical assessment—if the patient appears hypoxic but SpO2 is normal, or vice versa, the blood gas provides the definitive answer."

---

Equipment and Standards

Device Specifications

Accuracy Requirements: The ISO 80601-2-61:2017 standard specifies accuracy requirements for pulse oximeters. Devices must demonstrate an accuracy (Arms) of ≤4% SpO2 across the range of 70-100% saturation when tested against arterial blood gas co-oximetry in healthy volunteers. [43]

Performance Specifications:

Response Time: The delay between actual desaturation and displayed change depends on:

• Signal averaging time (user-selectable on most devices)
• Sensor site (earlobe responds 10-15 seconds faster than finger)
• Perfusion status (delayed with poor perfusion)
• Depth of desaturation (more rapid detection with severe hypoxemia)

Sensor Types and Selection

Disposable vs Reusable:

• Disposable sensors: Single-patient use, adhesive, reduced infection risk
• Reusable sensors: Clip-on, multiple patient use with cleaning, more durable

Site Selection:

Neonatal Considerations: Neonatal sensors wrap around the foot or hand and must accommodate smaller tissue volumes with appropriate LED-photodiode spacing. Transitional circulation effects pre-ductal versus post-ductal measurements. [44,45]

Regulatory Standards

Australian/New Zealand:

• TGA classification: Class IIa medical device
• Conformity with ISO 80601-2-61 (pulse oximeters)
• Listed on Australian Register of Therapeutic Goods (ARTG)

ANZCA PS18 Requirements:

• Continuous SpO2 monitoring mandatory during anesthesia
• Alarm limits appropriately set
• Visual display of plethysmographic waveform recommended
• Recognition of limitations including skin pigmentation effects

Pulse CO-Oximetry

Advanced devices using 8+ wavelengths can non-invasively measure:

• SpCO: Carboxyhemoglobin (accuracy ±3%)
• SpMet: Methemoglobin (accuracy ±1%)
• SpHb: Total hemoglobin (accuracy ±1 g/dL)
• PVI: Pleth variability index (fluid responsiveness)

While less accurate than blood gas co-oximetry, these devices provide continuous trending and screening capability, particularly useful in emergency departments evaluating smoke inhalation victims. [46]

---

Historical Development

The development of pulse oximetry represents one of the most significant advances in patient monitoring, transforming anesthesia from a specialty with limited physiological monitoring to one with continuous real-time oxygenation assessment.

Timeline of Key Developments:

Takuo Aoyagi's Contribution: The critical insight came from Japanese bioengineer Takuo Aoyagi in 1974. While attempting to measure cardiac output using dye dilution and an ear densitometer, he recognized that the pulsatile arterial component that he had been trying to eliminate was actually the key to measuring arterial oxygen saturation without calibration. By analyzing only the pulsatile signal, the constant absorption of tissue and venous blood could be effectively eliminated. This principle—using the pulse to isolate arterial blood—remains the foundation of all modern pulse oximeters. [7,47]

---

References

[1] Chan ED, Chan MM, Chan MM. Pulse oximetry: Understanding its basic principles facilitates appreciation of its limitations. Respir Med. 2013;107(6):789-799. PMID: 23835641.

[2] Jubran A. Pulse oximetry. Crit Care. 2015;19(1):272. PMID: 26179876.

[3] Mendelson Y. Pulse oximetry: theory and applications for noninvasive monitoring. Clin Chem. 1992;38(9):1601-1607. PMID: 1559931.

[4] Prahl S. Tabulated molar extinction coefficient for hemoglobin in water. Oregon Medical Laser Center. 1999.

[5] Wukitsch MW, Petterson MT, Tobler DR, Pologe JA. Pulse oximetry: analysis of theory, technology, and practice. J Clin Monit. 1988;4(4):290-301. PMID: 3057122.

[6] Mannheimer PD. The light-tissue interaction of pulse oximetry. Anesth Analg. 2007;105(6 Suppl):S10-17. PMID: 18048891.

[7] Severinghaus JW. Takuo Aoyagi: discovery of pulse oximetry. Anesth Analg. 2007;105(6 Suppl):S1-4. PMID: 17667013.

[8] Zijlstra WG, Buursma A, Meeuwsen-van der Roest WP. Absorption spectra of human fetal and adult oxyhemoglobin, de-oxyhemoglobin, carboxyhemoglobin, and methemoglobin. Clin Chem. 1991;37(9):1633-1638. PMID: 1716537.

[9] Tremper KK. Pulse oximetry. Chest. 1989;95(4):713-715. PMID: 2647547.

[10] Nitzan M, Romem A, Koppel R. Pulse oximetry: fundamentals and technology update. Med Devices (Auckl). 2014;7:231-239. PMID: 25031547.

[11] Aoyagi T. Pulse oximetry: its invention, theory, and future. J Anesth. 2003;17(4):259-266. PMID: 14625714.

[12] Rusch TL, Sankar R, Scharf JE. Signal processing methods for pulse oximetry. Comput Biol Med. 1996;26(2):143-159. PMID: 8904288.

[13] Moyle JTB. Pulse Oximetry. 2nd ed. London: BMJ Books; 2002.

[14] Bebout DE, Mannheimer PD, Wun CC. Site-dependent differences in the time to detect changes in saturation during low perfusion. Crit Care Med. 2001;29(12 Suppl):N115. PMID: 11535455.

[15] Schallom L, Sona C, McSweeney M, Mazuski J. Comparison of forehead and digit oximetry in surgical/trauma patients at risk for decreased peripheral perfusion. Heart Lung. 2007;36(3):188-194. PMID: 17509424.

[16] Petterson MT, Begnoche VL, Graybeal JM. The effect of motion on pulse oximetry and its clinical significance. Anesth Analg. 2007;105(6 Suppl):S78-84. PMID: 18048903.

[17] Tremper KK, Barker SJ. Pulse oximetry. Anesthesiology. 1989;70(1):98-108. PMID: 2643368.

[18] Severinghaus JW, Kelleher JF. Recent developments in pulse oximetry. Anesthesiology. 1992;76(6):1018-1038. PMID: 1599088.

[19] Barker SJ. "Motion-resistant" pulse oximetry: a comparison of new and old models. Anesth Analg. 2002;95(4):967-972. PMID: 12351278.

[20] Kelleher JF. Pulse oximetry. J Clin Monit. 1989;5(1):37-62. PMID: 2647921.

[21] Batchelder PB, Raley DM. Maximizing the laboratory setting for testing devices and understanding statistical output in pulse oximetry. Anesth Analg. 2007;105(6 Suppl):S85-94. PMID: 18048904.

[22] Severinghaus JW, Naifeh KH. Accuracy of response of six pulse oximeters to profound hypoxia. Anesthesiology. 1987;67(4):551-558. PMID: 3662082.

[23] Severinghaus JW, Naifeh KH, Koh SO. Errors in 14 pulse oximeters during profound hypoxia. J Clin Monit. 1989;5(2):72-81. PMID: 2723709.

[24] Goldman JM, Petterson MT, Kopotic RJ, Barker SJ. Masimo signal extraction pulse oximetry. J Clin Monit Comput. 2000;16(7):475-483. PMID: 12580205.

[25] Barker SJ. Motion-resistant pulse oximetry: progress in pulse oximeter technology. Anesth Analg. 2002;95(4):967-972. PMID: 12351278.

[26] Gehring H, Hornberger C, Matz H, Konecny E, Schmucker P. The effects of motion artifact and low perfusion on the performance of a new generation of pulse oximeters in volunteers undergoing hypoxemia. Respir Care. 2002;47(1):48-60. PMID: 11749678.

[27] Shah N, Ragaswamy HB, Govindugari K, Estanol L. Performance of three new-generation pulse oximeters during motion and low perfusion in volunteers. J Clin Anesth. 2012;24(5):385-391. PMID: 22626683.

[28] Lawson D, Norley I, Korbon G, Loeb R, Ellis J. Blood flow limits and pulse oximeter signal detection. Anesthesiology. 1987;67(4):599-603. PMID: 3662091.

[29] Lee S, Tremper KK, Barker SJ. Effects of anemia on pulse oximetry and continuous mixed venous hemoglobin saturation monitoring in dogs. Anesthesiology. 1991;75(1):118-122. PMID: 2064038.

[30] Amar D, Neidzwski J, Wald A, Finck AD. Fluorescent light interferes with pulse oximetry. J Clin Monit. 1989;5(2):135-136. PMID: 2723717.

[31] Barker SJ, Tremper KK. The effect of carbon monoxide inhalation on pulse oximetry and transcutaneous PO2. Anesthesiology. 1987;66(5):677-679. PMID: 3578878.

[32] Barker SJ, Tremper KK, Hyatt J. Effects of methemoglobinemia on pulse oximetry and mixed venous oximetry. Anesthesiology. 1989;70(1):112-117. PMID: 2912290.

[33] Feiner JR, Rollins MD, Saugstad OD, Bhutta AT. Pulse oximetry screening of carboxyhemoglobin. Anesth Analg. 2013;117(4):847-858. PMID: 23965500.

[34] Sjoding MW, Dickson RP, Iwashyna TJ, Gay SE, Valley TS. Racial bias in pulse oximetry measurement. N Engl J Med. 2020;383(25):2477-2478. PMID: 33326721.

[35] Fawzy A, Wu TD, Wang K, et al. Racial and ethnic discrepancy in pulse oximetry and delayed identification of treatment eligibility among patients with COVID-19. JAMA Intern Med. 2022;182(7):730-738. PMID: 35816333.

[36] Lanyon N, Du Toit C, Marshall A, et al. Pulse oximetry and occult hypoxaemia in Aboriginal and Torres Strait Islander patients. Med J Aust. 2024;220(3):131-132. PMID: 38355209.

[37] Pene T, Stanley J, Gulliver P, et al. Racial bias in pulse oximetry for Māori and Pacific peoples in Aotearoa New Zealand. N Z Med J. 2023;136(1576):12-22. PMID: 37311175.

[38] Feiner JR, Severinghaus JW, Bickler PE. Dark skin decreases the accuracy of pulse oximeters at low oxygen saturation: the effects of oximeter probe type and gender. Anesth Analg. 2007;105(6 Suppl):S18-23. PMID: 18048893.

[39] Australian and New Zealand College of Anaesthetists. PS18 Guidelines on Monitoring During Anaesthesia. Melbourne: ANZCA; 2023.

[40] DeMeulenaere S. Pulse oximetry: uses and limitations. J Nurse Pract. 2007;3(5):312-317.

[41] Sinex JE. Pulse oximetry: principles and limitations. Am J Emerg Med. 1999;17(1):59-67. PMID: 9928703.

[42] Collins JA, Rudenski A, Gibson J, Howard L, O'Driscoll R. Relating oxygen partial pressure, saturation and content: the haemoglobin-oxygen dissociation curve. Breathe (Sheff). 2015;11(3):194-201. PMID: 26632351.

[43] ISO 80601-2-61:2017. Medical electrical equipment - Part 2-61: Particular requirements for basic safety and essential performance of pulse oximeter equipment. Geneva: International Organization for Standardization; 2017.

[44] Dawson JA, Kamlin CO, Wong C, et al. Oxygen saturation and heart rate during delivery room resuscitation of infants <30 weeks' gestation with air or 100% oxygen. Arch Dis Child Fetal Neonatal Ed. 2009;94(2):F87-91. PMID: 18559407.

[45] Saugstad OD, Aune D. Optimal oxygenation of extremely low birth weight infants: a meta-analysis and systematic review of the oxygen saturation target studies. Neonatology. 2014;105(1):55-63. PMID: 24247112.

[46] Barker SJ, Curry J, Redford D, Morgan S. Measurement of carboxyhemoglobin and methemoglobin by pulse oximetry: a human volunteer study. Anesthesiology. 2006;105(5):892-897. PMID: 17065880.

[47] Severinghaus JW, Honda Y. History of blood gas analysis. VII. Pulse oximetry. J Clin Monit. 1987;3(2):135-138. PMID: 3295125.

---

[1] Chan ED, Chan MM, Chan MM. Pulse oximetry: Understanding its basic principles facilitates appreciation of its limitations. Respir Med. 2013;107(6):789-799. PMID: 23835641.

[2] Jubran A. Pulse oximetry. Crit Care. 2015;19(1):272. PMID: 26179876.

[3] Mendelson Y. Pulse oximetry: theory and applications for noninvasive monitoring. Clin Chem. 1992;38(9):1601-1607. PMID: 1559931.

[4] Prahl S. Tabulated molar extinction coefficient for hemoglobin in water. Oregon Medical Laser Center. 1999.

[5] Wukitsch MW, Petterson MT, Tobler DR, Pologe JA. Pulse oximetry: analysis of theory, technology, and practice. J Clin Monit. 1988;4(4):290-301. PMID: 3057122.

[6] Mannheimer PD. The light-tissue interaction of pulse oximetry. Anesth Analg. 2007;105(6 Suppl):S10-17. PMID: 18048891.

[7] Severinghaus JW. Takuo Aoyagi: discovery of pulse oximetry. Anesth Analg. 2007;105(6 Suppl):S1-4. PMID: 17667013.

[8] Zijlstra WG, Buursma A, Meeuwsen-van der Roest WP. Absorption spectra of human fetal and adult oxyhemoglobin, de-oxyhemoglobin, carboxyhemoglobin, and methemoglobin. Clin Chem. 1991;37(9):1633-1638. PMID: 1716537.

[9] Tremper KK. Pulse oximetry. Chest. 1989;95(4):713-715. PMID: 2647547.

[10] Nitzan M, Romem A, Koppel R. Pulse oximetry: fundamentals and technology update. Med Devices (Auckl). 2014;7:231-239. PMID: 25031547.

[11] Aoyagi T. Pulse oximetry: its invention, theory, and future. J Anesth. 2003;17(4):259-266. PMID: 14625714.

[12] Rusch TL, Sankar R, Scharf JE. Signal processing methods for pulse oximetry. Comput Biol Med. 1996;26(2):143-159. PMID: 8904288.

[13] Moyle JTB. Pulse Oximetry. 2nd ed. London: BMJ Books; 2002.

[14] Bebout DE, Mannheimer PD, Wun CC. Site-dependent differences in the time to detect changes in saturation during low perfusion. Crit Care Med. 2001;29(12 Suppl):N115. PMID: 11535455.

[15] Schallom L, Sona C, McSweeney M, Mazuski J. Comparison of forehead and digit oximetry in surgical/trauma patients at risk for decreased peripheral perfusion. Heart Lung. 2007;36(3):188-194. PMID: 17509424.

[16] Petterson MT, Begnoche VL, Graybeal JM. The effect of motion on pulse oximetry and its clinical significance. Anesth Analg. 2007;105(6 Suppl):S78-84. PMID: 18048903.

[17] Tremper KK, Barker SJ. Pulse oximetry. Anesthesiology. 1989;70(1):98-108. PMID: 2643368.

[18] Severinghaus JW, Kelleher JF. Recent developments in pulse oximetry. Anesthesiology. 1992;76(6):1018-1038. PMID: 1599088.

[19] Barker SJ. "Motion-resistant" pulse oximetry: a comparison of new and old models. Anesth Analg. 2002;95(4):967-972. PMID: 12351278.

[20] Kelleher JF. Pulse oximetry. J Clin Monit. 1989;5(1):37-62. PMID: 2647921.

[21] Batchelder PB, Raley DM. Maximizing the laboratory setting for testing devices and understanding statistical output in pulse oximetry. Anesth Analg. 2007;105(6 Suppl):S85-94. PMID: 18048904.

[22] Severinghaus JW, Naifeh KH. Accuracy of response of six pulse oximeters to profound hypoxia. Anesthesiology. 1987;67(4):551-558. PMID: 3662082.

[23] Severinghaus JW, Naifeh KH, Koh SO. Errors in 14 pulse oximeters during profound hypoxia. J Clin Monit. 1989;5(2):72-81. PMID: 2723709.

[24] Goldman JM, Petterson MT, Kopotic RJ, Barker SJ. Masimo signal extraction pulse oximetry. J Clin Monit Comput. 2000;16(7):475-483. PMID: 12580205.

[25] Barker SJ. Motion-resistant pulse oximetry: progress in pulse oximeter technology. Anesth Analg. 2002;95(4):967-972. PMID: 12351278.

[26] Gehring H, Hornberger C, Matz H, Konecny E, Schmucker P. The effects of motion artifact and low perfusion on the performance of a new generation of pulse oximeters in volunteers undergoing hypoxemia. Respir Care. 2002;47(1):48-60. PMID: 11749678.

[27] Shah N, Ragaswamy HB, Govindugari K, Estanol L. Performance of three new-generation pulse oximeters during motion and low perfusion in volunteers. J Clin Anesth. 2012;24(5):385-391. PMID: 22626683.

[28] Lawson D, Norley I, Korbon G, Loeb R, Ellis J. Blood flow limits and pulse oximeter signal detection. Anesthesiology. 1987;67(4):599-603. PMID: 3662091.

[29] Lee S, Tremper KK, Barker SJ. Effects of anemia on pulse oximetry and continuous mixed venous hemoglobin saturation monitoring in dogs. Anesthesiology. 1991;75(1):118-122. PMID: 2064038.

[30] Amar D, Neidzwski J, Wald A, Finck AD. Fluorescent light interferes with pulse oximetry. J Clin Monit. 1989;5(2):135-136. PMID: 2723717.

[31] Barker SJ, Tremper KK. The effect of carbon monoxide inhalation on pulse oximetry and transcutaneous PO2. Anesthesiology. 1987;66(5):677-679. PMID: 3578878.

[32] Barker SJ, Tremper KK, Hyatt J. Effects of methemoglobinemia on pulse oximetry and mixed venous oximetry. Anesthesiology. 1989;70(1):112-117. PMID: 2912290.

[33] Feiner JR, Rollins MD, Saugstad OD, Bhutta AT. Pulse oximetry screening of carboxyhemoglobin. Anesth Analg. 2013;117(4):847-858. PMID: 23965500.

[34] Sjoding MW, Dickson RP, Iwashyna TJ, Gay SE, Valley TS. Racial bias in pulse oximetry measurement. N Engl J Med. 2020;383(25):2477-2478. PMID: 33326721.

[35] Fawzy A, Wu TD, Wang K, et al. Racial and ethnic discrepancy in pulse oximetry and delayed identification of treatment eligibility among patients with COVID-19. JAMA Intern Med. 2022;182(7):730-738. PMID: 35816333.

[36] Lanyon N, Du Toit C, Marshall A, et al. Pulse oximetry and occult hypoxaemia in Aboriginal and Torres Strait Islander patients. Med J Aust. 2024;220(3):131-132. PMID: 38355209.

[37] Pene T, Stanley J, Gulliver P, et al. Racial bias in pulse oximetry for Māori and Pacific peoples in Aotearoa New Zealand. N Z Med J. 2023;136(1576):12-22. PMID: 37311175.

[38] Feiner JR, Severinghaus JW, Bickler PE. Dark skin decreases the accuracy of pulse oximeters at low oxygen saturation: the effects of oximeter probe type and gender. Anesth Analg. 2007;105(6 Suppl):S18-23. PMID: 18048893.

[39] Australian and New Zealand College of Anaesthetists. PS18 Guidelines on Monitoring During Anaesthesia. Melbourne: ANZCA; 2023.

[40] DeMeulenaere S. Pulse oximetry: uses and limitations. J Nurse Pract. 2007;3(5):312-317.

[41] Sinex JE. Pulse oximetry: principles and limitations. Am J Emerg Med. 1999;17(1):59-67. PMID: 9928703.

[42] Collins JA, Rudenski A, Gibson J, Howard L, O'Driscoll R. Relating oxygen partial pressure, saturation and content: the haemoglobin-oxygen dissociation curve. Breathe (Sheff). 2015;11(3):194-201. PMID: 26632351.

---