The most commonly used capnometers are of the infrared (IR) light absorption type. The analysis can occur at the machine or in the ventilator circuit. In the first instance, airway gases can be sampled along a narrow tube to the machine. Alternatively, an adaptor placed in the airway circuit uses IR light absorption to measure the CO₂ concentration directly in the gas stream. This second method provides a very fast response time and avoids problems with clogged tubing and water vapor condensation. It has fallen out of favor because of the extra weight near the endotracheal tube, a concern about reusing a device within the circuit, and the fact that the externalization of the sensor makes it more prone to damage during usage.

Capnometry is the determination of the end-tidal partial pressure of CO₂. Capnography is the graphic display of instantaneous CO₂ partial pressure versus time during the respiratory cycle. This relation is displayed as a CO₂ waveform or capnogram. The simplest use of measured end-tidal CO₂ (ETCO₂) is to follow changes in PaCO₂ level. Although the ETCO₂ level may be lower than the actual PaCO₂ level, in the absence of severe dead-space effects, alterations in PaCO₂ level will be mirrored by changes in ETCO₂ level, and these changes can be continuously monitored. This may be particularly useful when changes in minute ventilation are made and PaCO₂ levels are changing.

The waveform produced by the monitor as a capnogram gives further information in addition to the numeric value of the ETCO₂.

Dead space (ventilated but unperfused airway) can be divided into apparatus and anatomic (or conductive) dead space, which is in series with the alveoli, and alveolar (or physiologic) dead space, which is in parallel with the alveoli.

Apparatus and anatomic dead space causes the initial expired gas to have a CO₂ level of zero (phase I, Fig 87.1.). As expiration continues, alveolar gas is detected and the capnogram rises rapidly (phase II). In phase III, known as the alveolar plateau, the apparatus and anatomic dead-space gases have been expired and the gas sampled is from alveoli. If all these alveoli were perfused, then the CO₂ level would be that of “ideal” alveolar gas, i.e., very close to the PaCO₂ level. However, the CO₂ level is lower than the PaCO₂ level during the plateau because of admixture of gas from unperfused alveoli with a CO₂ level of zero (alveolar dead space). This means that the ETCO₂ level will be close to the PaCO₂ level in patients with low alveolar dead space; however, if alveolar dead space increases (e.g., as in emphysema), then the ETCO₂ will be reduced, compared to PaCO₂. The difference between end-tidal and arterial CO₂ level depends on alveolar dead space and is not altered by changes in apparatus or anatomic dead space.

Table 87.1. lists the factors influencing alveolar dead space.

Ventilation-perfusion mismatch increases the arterial-end-tidal CO₂ gradient, and studies have shown this increase to be significantly higher in patients with morbid obesity (body mass index >40 kg/m²) and in patients receiving anesthesia in the lateral position. Increased gradients have also been reported in older patients; in patients with a high American Society of Anesthesiologists’ physical status classification system (ASA) score; and in patients having episodes of hemodynamic instability, during such episodes.