Hypoxia is defined as reduction of oxygen supply to tissue below physiologic levels. Hypoxemia is defined as deficient oxygenation of the blood. From the anesthesiologist’s perspective, hypoxia is usually considered to be decreased oxygen tension (PO₂) inside the body at the tissue level or outside the body (e.g., hypoxic gas mixture), and hypoxemia is decreased oxygen tension in the arterial blood (PaO₂).

Yes. There is an age-dependent decrease in the PaO₂. In 1972, Marshall and Wyche suggested the following relationship in subjects breathing room air:

mean PaO₂ in mm Hg = 102 – 0.33 (age in years) mm Hg

About this regression line, there are 95% confidence limits (two standard deviations [SDs]) of 10 mm Hg.

Cerveri et al. studied PaO₂ in normal nonsmoking subjects ages 40 to 90 years. They found that PaO₂ was related to age in subjects between 40 and 74 years and constructed the following reference equation:

PaO₂ (mm Hg) = 143.6 – (0.39 × age) – (0.56 × BMI) – (0.57 × PaCO₂)

For subjects ≥75 years old, there was no correlation with age.

No. An SpO₂ reading of 94% does not signify hypoxemia. This patient is five decades older than a 20-year-old subject; therefore, his mean PaO₂ should be 5 × 5 mm Hg per decade, which is 25 mm Hg, less than that (95 mm Hg) expected in a 20-year-old (see section A.2, Sorbini et al.).

95 – 25 = 70 mm Hg

or, using the Marshall and Whyche equation (see section A.2),

102 – 0.33 (70) = 79 mm Hg

Assuming normal adult hemoglobin, temperature, and pH, this PaO₂ corresponds to a hemoglobin oxygen saturation of approximately 94% on the normal hemoglobin oxygen saturation versus PO₂ curve.

Although the definitions of hypoxemia given in sections A.1 and A.2 must take age into consideration, from a practical point of view, hypoxemia (in the absence of anemia) is generally considered to exist when the PaO₂ is less than 60 mm Hg, which is equivalent to a hemoglobin oxygen saturation of 90%.

The pulse oximeter is a noninvasive device that provides a real-time estimate (designated the SpO₂%) of the arterial hemoglobin saturation with oxygen. It is a transmissive oximeter in which the patient’s fingertip (or other probe site; e.g., earlobe) serves as an in vivo cuvette through which light at two different wavelengths is transmitted.

If one requires an accurate determination of the arterial hemoglobin saturation with oxygen, an arterial blood sample must be drawn and analyzed in a laboratory co-oximeter, sometimes called a hemoximeter. The laboratory co-oximeter is a transmissive oximeter that uses six or more wavelengths of light to measure total hemoglobin (HbTOT), oxygenated hemoglobin (HbO₂), deoxygenated (“reduced”) hemoglobin (RHb), metHb, HbCO, and other dyshemoglobins, such as sulfhemoglobin. Because each species of hemoglobin has a characteristic absorbance spectrum (i.e., absorbance vs. wavelength), examination of a sample of blood using these six or eight wavelengths permits identification and quantification of each hemoglobin species. Conventional pulse oximeters use only two wavelengths of light; therefore, they are unable to determine all of the different hemoglobin species and assume that only HbO₂ and RHb are present. Fractional saturation (HbO₂%) is defined as HbO₂ / (HbO₂ + RHb + HbCO + metHb)—that is, HbO₂ per total Hb. Functional saturation (SaO₂) is defined as HbO₂ / (HbO₂ + RHb). Note that dyshemoglobins are absent from the denominator in the definition of SaO₂.

The pulse oximeter combines the technologies of spectrophotometry and optical plethysmography. In the pulse oximeter probe, light-emitting diodes transmit red light at wavelength 660 nm and infrared light at wavelength 940 nm through the fingertip or other probe site. Light that passes through the probe site is sensed by a single photodetector and expressed as absorbance at each wavelength. It can be shown that the ratio of absorbances, 660/940 nm, is related to the hemoglobin saturation with oxygen. This is spectrophotometry. Detection of pulsatile flow is by optical plethysmography. With each pulse of arterial blood, the probe site (e.g., fingertip) increases in volume, the path length of the transmitted light increases,
and the absorbance of light at 660 nm and 940 nm increases. This pulse-added absorbance is considered to be due to the pulsatile flow of arterial blood at the probe site, so that the ratio of pulse-added absorbances, 660/940 nm, can be used to provide an estimate of hemoglobin oxygen saturation in arterial blood.

Each pulse oximeter manufacturer creates an empiric calibration algorithm, that is, one based on observations in human volunteers. The algorithm relates the average ratio of pulseadded absorbances, 660/990 nm, to the actual hemoglobin oxygen saturation in arterial blood samples drawn simultaneously and analyzed in a laboratory hemoximeter. Observations are made with the volunteers breathing varying FIO₂ values so that data for saturations as low as 70% are obtained. The calibration algorithm that is created relates the ratio (R) of pulseadded absorbances, 660/940 nm, to the laboratory hemoximeter readings of HbO₂ or SaO₂%, depending on the pulse oximeter manufacturer. The calibration algorithm is then stored in the software of the pulse oximeter as a “lookup” table. Therefore, the pulse oximeter merely measures R, consults its lookup table, and displays the corresponding saturation reading as SpO₂. The pulse oximeter does not actually measure saturation; it infers it from R, and, therefore, it predicts what the laboratory hemoximeter would read if an arterial sample drawn at that moment were analyzed.

A high value for R corresponds to a low SpO₂ reading; R = 1 corresponds to an SpO₂ reading of 85%, and a low value for R corresponds to a high SpO₂ reading. Whenever the SpO₂ reading is approximately 85%, one should consider the possibility of a spurious reading as a result of the “R = 1” phenomenon. This is commonly due to malpositioning of the probe on the fingertip.

The pulse oximeter (SpO₂) reading is usually specified by the manufacturer to have an SD of ±2%. Therefore, assuming a normal distribution, if the SpO₂ reading is 96%, there is a 68% likelihood that the true saturation (as measured by a hemoximeter) is 96 ± 2% (i.e., ±1 SD) and a 95% likelihood that the true saturation is 96 ± 4% (i.e., ±2 SD).

The two-wavelength pulse oximeter is most accurate when the conditions of its clinical use most closely resemble those of its calibration. Spurious readings or failure may occur during patient movement (e.g., shivering, peripheral nerve stimulation, “twitching”), presence of intense ambient light (low signal/noise ratio), electrocautery use, administration of intravenous dyes with absorbance peaks at 660 nm (e.g., methylene blue), presence of dyshemoglobins (e.g., metHb, HbCO, and sulfhemoglobin), certain colors of nail polish, poor pulsatile flow at the probe site (e.g., hypotension, vasoconstriction, Raynaud disease), and venous pulsations (e.g., tricuspid regurgitation or earlobe placement of the probe in a patient who is in a headdown position).

Recent improvements in pulse oximetry technology include paradigms to increase the signal/noise ratio, thereby permitting motion artifact reduction and increased sensitivity to pulsatile flow. Such pulse oximeters have significantly lower failure rates during patient movement (such as in the postanesthesia care unit) and during conditions of low flow (e.g., hypothermia).

In metHb, iron in the heme moiety is oxidized (such as by dapsone, benzocaine, nitric oxide, prilocaine) to the Fe³⁺ state rather than being in the normal Fe²⁺ state. metHb cannot carry oxygen and creates a physiologic anemia. The absorbance spectrum for metHb shows it to have similar absorbances to light at 660 nm and 940 nm. Therefore, the more metHb that is present, the more R tends toward 1 and the SpO₂ reading toward 85%. In the presence of metHb, the SpO₂ reading overestimates the fractional saturation and underestimates the functional saturation.

In carbon monoxide (CO) poisoning, CO combines with hemoglobin to create a physiologic hypoxemia. HbCO has a similar absorbance to HbO₂ at 660 nm but a very low absorbance at 940 nm. In the presence of HbCO, the SpO₂ overestimates fractional saturation and underestimates functional saturation. Although blood that is poisoned with CO appears “cherry red” to the naked eye, and therefore may look like it is fully saturated with oxygen, it is important to recognize that the ratio of absorbances 660/940 nm will suggest that the blood hemoglobin oxygen saturation is in the 90s. In a subject breathing FIO₂ of 1 (when functional saturation is 100%), as the level of HbCO increases, SpO₂ decreases from close to 100% to approximately 91%.

If the presence of dyshemoglobins is suspected, arterial blood must be drawn and analyzed in a laboratory hemoximeter to obtain accurate readings of saturation.

Pulse oximetry technology has evolved rapidly, and multiwavelength pulse oximeters, called pulse co-oximeters, are clinically available. They use more than seven wavelengths of light (Masimo Rainbow SET Pulse CO-Oximetry, Masimo Company, Irvine, CA), similar to the laboratory hemoximeter. This technology makes it possible to measure CO (SpCO), methemoglobin (SpMet), and total hemoglobin (SpHbt), all noninvasively and in real time.

A capnometer is a device that measures the tension (in units of millimeters of mercury [mm Hg] or kilopascals [kPa]) or concentration (volumes %) of CO₂ in the gas near the patient’s airway throughout the respiratory cycle. Capnography is the graphic display of the CO₂ concentration on the y-axis against time on the x-axis.

Capnometry may be achieved using different technologies that can monitor CO₂. Most capnometers now use infrared spectroscopy to measure the PCO₂ (i.e., CO₂ partial pressure in millimeters of mercury). A built-in barometer measures barometric pressure (P_B) so that CO₂ can also be displayed as a percentage (PCO₂ × 100/P_B = %CO₂).

Capnography is one of the American Society of Anesthesiologists’ standards for basic anesthetic monitoring (i.e., “Continual monitoring for the presence of expired CO₂ shall be performed unless invalidated by the nature of the patient, procedure or equipment.”). It is the “gold standard” for establishing the presence of ventilation.

PETCO₂ is the tension of CO₂ in the exhaled gas at the end of an exhalation. Because this gas originates from the alveoli, it is considered to represent the CO₂ tension in the alveolar gas (PACO₂). The PACO₂ results from the combination of gases coming from ideal alveoli, where ventilation and perfusion are perfectly matched and the CO₂ concentration is the same as
arterial (PACO₂), and alveolar dead space (alveoli that are ventilated but not perfused) where CO₂ concentration is the same as inspired (PICO₂) and is normally zero.

The presence of a normal capnogram and PETCO₂ depends on (1) production of CO₂ by the tissues, (2) cardiac output and pulmonary blood flow to carry CO₂ to the lungs, and (3) ventilation. Monitoring of PETCO₂ can be used to adjust the setting of a mechanical ventilator and monitor ventilation as well as to detect ventilator malfunctions and breathing system problems, such as leaks, incompetent unidirectional valves, exhausted CO₂ absorbent, metabolic problems (e.g., malignant hyperthermia), and gas embolism.

A typical normal capnogram (PCO₂ vs. time) is shown in Figure 59.1. Four phases have been described.

Phase I. This is the expiratory baseline. During exhalation, CO₂-free gas from the mechanical (apparatus) dead space (e.g., flexible or “goose-neck” connector) and anatomic dead space (large conducting airways) flows past the CO₂ sampling port.

Phase II. This is the expiratory upstroke. It is due to CO₂ from alveoli mixed with anatomic dead space gas passing the gas sampling port and is usually a steep upslope.

Phase III. This is the expiratory or alveolar plateau and is due to alveolar CO₂.

In patients with healthy lungs, it is nearly horizontal, but in patients with obstructive airway disease, there is a more pronounced upward slope. The maximum expired CO₂ is usually considered as the PETCO₂.

Phase IV. This is the inspiratory downstroke. During the commencement of inspiration, fresh (i.e., CO₂-free) gas is inhaled past the sensor, and the capnogram falls sharply to baseline.

The angle between the upstroke of phase II and the plateau of phase III is referred to as the α angle. The angle between phase III and IV is the β angle. An increase in the α angle and slope of phase III is seen commonly in acute bronchospasm. The angle usually decreases with treatment of the bronchospasm.

In a mainstream or nondiverting type of analyzer, the gas analyzer sensor is brought to the airway itself, and no gas is removed from the breathing system. A cuvette with a quartz window is placed between the breathing system Y-piece and the tracheal tube connector. As respired gases flow past the window in the cuvette, a beam of infrared radiation (wavelength 4.3 µm) is directed through the window. Absorbance of infrared radiation at 4.3 µm is used to measure PCO₂. Advantages of the nondiverting analyzer are rapid response time, accurate waveform, and no need to scavenge gases. Disadvantages are that this additional device in the airway can be a site for a disconnect and needs sterilization (or replacement of a disposable single-use cuvette) between cases, needs to be heated because water vapor condensing on the window causes error, and can be blocked by secretions or blood on its window. A very lightweight (25 g) mainstream analyzer (IRMA, Masimo, Danderyd, Sweden) capable of measuring CO₂, nitrous oxide, and all five potent inhaled anesthetics is available.

A sidestream or diverting gas analyzer aspirates gas from the breathing circuit through a sampling adapter placed by the airway. The sampled gas passes through a sampling catheter (usually 6 to 10 ft in length) to reach the gas analyzer. Advantages of diverting analyzers are that the breathing circuit adapter is disposable. The analyzer, being remote from the airway, can be more versatile and incorporate several technologies in one “box” to analyze multiple gases (e.g., infrared analysis for CO₂, nitrous oxide, anesthetic agents; paramagnetic sensor or fuel cell for oxygen). Disadvantages of sidestream sampling analysis are slower response times, the possibility of sampling leaks, errors resulting from the length of the gas sampling tube, the need to scavenge gases from the analyzer after analysis, and susceptibility of the sampling tube to becoming clogged by water or secretions. Because of their versatility in regard to gas monitoring, most anesthetizing locations now use sidestream sampling analyzers.

Elevated baseline equals the amount of CO₂ present in inspired gas (Fig. 59.2A)

Elevated expiratory plateau (Fig. 59.2D)