# Pulse Oximetry, Capnography, and Gas Analysis

## Introduction

Pulse oximetry, capnography, and respiratory gas analysis form the triad of essential respiratory monitoring in anesthesia. These technologies provide real-time, continuous, noninvasive data on oxygenation, ventilation, and volatile agent delivery. Understanding their principles, capabilities, and limitations is fundamental to safe anesthetic practice.

## Pulse Oximetry

### Principles of Operation

Pulse oximetry is based on the Beer-Lambert law, which states that the concentration of a substance in solution can be determined by its absorption of light at specific wavelengths. The device uses two wavelengths of light: red (660 nm) and infrared (940 nm). Oxyhemoglobin absorbs more infrared light while deoxyhemoglobin absorbs more red light. The ratio of absorption (R/IR) is calibrated against volunteer desaturation data to calculate SpO2. Conventional pulse oximeters are two-wavelength devices and can only distinguish two species of hemoglobin. The device identifies the pulsatile (arterial) component of the signal to isolate arterial saturation from venous and tissue absorption.

### Clinical Applications

Pulse oximetry is an ASA standard monitor, mandatory for every anesthetic. Normal SpO2 is 95 to 100% on room air. The device provides an audible pitch change with desaturation that serves as an early warning system. The plethysmographic waveform provides information about perfusion and pulse rate. The perfusion index (PI), which is the ratio of pulsatile to non-pulsatile signal, indicates signal quality, with a low PI suggesting poor perfusion or an unreliable reading.

### Limitations and Sources of Error

Dyshemoglobins are an important source of error. Carboxyhemoglobin causes falsely high readings, with SpO2 reading approximately 100% despite true SaO2 values as low as 60%. Methemoglobin causes SpO2 to trend toward 85% regardless of the true saturation.

Other sources of error include motion artifact, which creates a pulsatile signal mimicking arterial blood flow. Low perfusion states from vasoconstriction, hypothermia, shock, or vasopressor use reduce signal quality. Nail polish (particularly blue, black, and green) can interfere with readings and should be removed or the probe placed sideways. Ambient light interference can be addressed by covering the probe with an opaque shield. Skin pigmentation may cause overestimation of SpO2 in deeply pigmented skin, particularly in the 85 to 95% range. There is an inherent lag time of approximately 30 to 60 seconds from a real physiologic change to the SpO2 display. The penumbra effect from venous pulsation in right heart failure or tricuspid regurgitation can also cause errors.

### Multi-Wavelength Oximetry

CO-oximeters such as the Masimo Rainbow SET use 8 or more wavelengths and can measure carboxyhemoglobin, methemoglobin, total hemoglobin (SpHb), and oxygen content. They are more accurate in the presence of dyshemoglobins but are still not a replacement for arterial blood gas co-oximetry.

![Two-wavelength pulse oximetry principle showing light absorption spectra of oxy- and deoxyhemoglobin](images/pulse-oximetry-principle.png)

## Capnography

### Principles of Operation

Capnography measures carbon dioxide concentration in respiratory gases using infrared absorption spectroscopy. CO2 absorbs infrared light at a wavelength of 4.26 micrometers. Two configurations exist: mainstream, in which the sensor sits directly on the airway circuit at the Y-piece (faster response, no gas sampling), and sidestream, in which gas is aspirated through a sampling line to a remote sensor (slight delay of 1 to 3 seconds, risk of water condensation and line obstruction).

### The Normal Capnogram

The normal capnogram consists of several phases. Phase I is the inspiratory baseline (0 mmHg CO2, representing dead space gas). Phase II is the rapid upstroke as alveolar gas reaches the sensor. Phase III is the alveolar plateau, with the highest point representing the end-tidal CO2 (EtCO2). Phase 0 (the descending limb) occurs as inspiration begins and CO2 drops to zero. Normal EtCO2 is 35 to 45 mmHg. The PaCO2-EtCO2 gradient is normally 2 to 5 mmHg, and an increased gradient indicates increased dead space ventilation.

### Abnormal Capnography Patterns

| Pattern | Diagnosis |
|---------|-----------|
| **Sudden drop to zero** | Esophageal intubation, circuit disconnect, apnea, cardiac arrest |
| **Exponential decline** | Pulmonary embolism, sudden hypotension, cardiac arrest |
| **Gradually decreasing EtCO2** | Hyperventilation, decreasing cardiac output, hypothermia |
| **Gradually increasing EtCO2** | Hypoventilation, increased CO2 production (MH, fever, tourniquet release) |
| **Upsloping Phase III** | Bronchospasm, COPD (uneven alveolar emptying) |
| **Elevated baseline** | Rebreathing (exhausted CO2 absorbent, incompetent inspiratory valve) |
| **Cleft in plateau** | Spontaneous breathing effort during mechanical ventilation, surgical compression of chest |
| **Curare cleft** | Partial recovery from neuromuscular blockade |

### Clinical Applications Beyond Ventilation

A sustained EtCO2 waveform for 6 or more breaths is the gold standard for confirmation of tracheal intubation. During CPR, an EtCO2 above 10 mmHg during chest compressions indicates adequate cardiac output, and a sudden rise may indicate return of spontaneous circulation. Rapidly rising EtCO2 is often the earliest sign of malignant hyperthermia. A sudden drop in EtCO2 with a widened PaCO2-EtCO2 gradient suggests air embolism.

![Normal capnogram phases and common abnormal waveform patterns](images/capnogram-patterns.png)

## Gas Analysis

### Infrared Spectroscopy

Infrared spectroscopy is used for CO2, N2O, and volatile anesthetic agents. Each gas absorbs infrared light at characteristic wavelengths. It cannot measure O2 or N2 because symmetric diatomic molecules do not absorb infrared light. Collision broadening from N2O can interfere with CO2 measurement, but modern analyzers compensate automatically.

### Oxygen Analysis

The paramagnetic analyzer exploits the strong paramagnetic property of O2, is highly accurate, and is used in sidestream analyzers. The galvanic (fuel cell) analyzer is an electrochemical cell that generates current proportional to PO2 and is used in the inspiratory limb of the circle system. The polarographic (Clark electrode) operates on a similar principle to the galvanic cell but requires external voltage and is used in blood gas analyzers.

### Agent Identification and Quantification

Modern gas analyzers identify the specific volatile agent and display its concentration in volume percent. This is important for detecting agent contamination (wrong agent in the vaporizer) or mixed agents (residual agent in the circuit). End-tidal agent concentration correlates with brain partial pressure at steady state and serves as a real-time measure of anesthetic depth, enabling MAC monitoring.

### Raman Spectroscopy and Mass Spectrometry

Raman spectroscopy uses laser light scattering and can measure all respiratory gases including O2 and N2, but is largely obsolete in clinical practice. Mass spectrometry separates gases by molecular weight and is highly accurate but expensive; it was previously used for shared multi-OR monitoring.

![Infrared absorption spectra of common anesthetic gases](images/gas-analysis-ir-spectra.png)

## Clinical Pearls

Pulse oximetry measures oxygen saturation, not oxygen delivery; a patient can have a normal SpO2 with dangerously low cardiac output and severely impaired tissue oxygenation. A sudden EtCO2 drop to zero means one of three things: esophageal intubation, complete circuit disconnect, or cardiac arrest, and the clinician should check the patient, not just the monitor. Capnography is more valuable than pulse oximetry for detecting hypoventilation, especially in patients receiving supplemental oxygen, because SpO2 desaturation is delayed by the oxygen reserve. In carbon monoxide poisoning, standard two-wavelength pulse oximetry is dangerously misleading, and a co-oximetry panel from the blood gas laboratory should be requested.

## References

1. Tremper KK, Barker SJ. Pulse oximetry. *Anesthesiology*. 1989;70(1):98-108.
2. Ortega R, Connor C, Kim S, Djang R, Patel K. Monitoring ventilation with capnography. *N Engl J Med*. 2012;367(19):e27.
3. Butterworth JF, Mackey DC, Wasnick JD. *Morgan & Mikhail's Clinical Anesthesiology*. 7th ed. McGraw-Hill; 2022. Chapter 6: Patient Monitors.
4. Jubran A. Pulse oximetry. *Crit Care*. 2015;19(1):272.