1 What is pulse oximetry and how does it work?

Pulse oximetry is the continuous noninvasive estimation of arterial hemoglobin-oxygen saturation. It is used routinely to monitor oxygenation in diverse clinical settings, including the operating room, emergency department, and intensive care unit. Clinical use of pulse oximetry falls into two main categories:

Pulse oximeters function by transmitting red light (660 nm, absorbed by oxyhemoglobin [O₂Hb]) and infrared light (940 nm, absorbed by deoxyhemoglobin [deoxyHb]) from two light-emitting diodes (LEDs) through tissue containing pulsatile blood. The saturation of hemoglobin with oxygen is a function of the ratio of red to infrared light absorption from the pulsatile and nonpulsatile components of the signals. Thus the saturation (SpO₂) is a function of the ratio of two ratios, cancelling out most differences caused by finger thickness, pigmentation, and other factors. A microprocessor algorithm is used to calculate the arterial saturation on the basis of calibration studies done by comparing true saturation measured on arterial blood with a CO-oximeter with the pulse oximeter reading. This calibration is factory set and is not adjustable.

Pulse oximeter probes can be applied to any site that allows orientation of the LED and photodetector opposite one another across a vascular bed. If the tissue is too thick, the signal is attenuated before reaching the detector and the oximeter cannot function. Oximeters can be applied to fingers, toes, earlobes, lips, cheeks, and the bridge of the nose. Esophageal and oral probes are also in development. Several manufacturers offer reflectance oximeter probes that can be applied to flat tissue surfaces such as the forehead or chest. Recently introduced earlobe-mounted sensors combine a pulse oximeter and a transcutaneous CO₂ electrode. Many pulse oximeters now include noise and artifact rejection software. This refinement aids the determination of SpO₂ in patients with low perfusion or motion (e.g., tremor).

2 How accurate are pulse oximeters?

Pulse oximetry is accurate within 2% to 3% of the true O₂Hb levels as measured in vitro with multiwavelength oximeters. The U.S. Food and Drug Administration requires manufacturers to demonstrate that their instruments confirm to this degree of accuracy in human subjects at between 70% and 100% oxygen saturation. Because the principle of measurement is based on a ratio of absorbance ratios, no calibration of the instrument by the user is needed or possible.

3 What factors interfere with pulse oximetry?

Most errors in oximetry measurement are the result of poor signal quality (i.e., hypoperfusion, vasoconstriction) or excessive noise (i.e., motion artifact). Optical interference may be introduced by extraneous light from fluorescent sources or infrared surgical navigation systems. Intravenous dyes (methylene blue, indocyanine green) and nail polishes (especially green, blue, or black) absorb at the wavelengths used by the oximeter and can produce artificially low measurements. Contamination from venous pulsations caused by dependent venous pooling or valvular insufficiency may also cause low readings. For a similar reason, the simultaneous arterial and venous pulsation during cardiopulmonary resuscitation (CPR) make oximeter data unreliable. Extreme hyperbilirubinemia has been reported to have variable effects on SpO₂ values. The presence of dysfunctional hemoglobin species can alter the ability of oximetry to reflect the true oxygen saturation. Studies show that darkly pigmented skin can falsely increase saturation estimates derived by some widely used pulse oximeters by up to 7% in the range of 70% to 80% saturation.

4 What effects does dyshemoglobinemia have on pulse oximetry?

Because pulse oximeters use two wavelengths of light, they are capable of differentiating only two species of hemoglobin: Hb and O₂Hb. Given that abnormal hemoglobin species such as carboxyhemoglobin (CoHb) or methemoglobin (MetHb) also absorb red and infrared light, their presence affects the SpO₂ measurement, and their quantitative contribution cannot be determined. The pulse oximeter assumes that only functional hemoglobin is present (O₂Hb or Hb), and the oxygen saturation is calculated on the basis of these amounts.

For example, CoHb is read by the limited wavelength analysis of a pulse oximeter as O₂Hb (CoHb is scarlet red), which will falsely elevate the SpO₂ reading. The absorption pattern of MetHb is interpreted by the pulse oximeter as 85% saturation; thus, progressively higher levels of MetHb cause the SpO₂ value to converge on 85% regardless of the actual SaO₂. When the presence of significant amounts of dysfunctional hemoglobin is suspected, a CO-oximeter should be used to determine O₂Hb saturation. A multiwavelength laboratory CO-oximeter determines SaO₂ more accurately in the presence of dysfunctional hemoglobins because it possesses wavelengths of light that can be used to detect the presence of CoHb and MetHb.

The presence of fetal hemoglobin has not been shown to significantly affect the accuracy of SpO₂ measurements because its light absorption properties are similar to those of adult hemoglobin.

5 What is capnography, and how does it work?

Capnography is the continuous measurement and graphic display of exhaled carbon dioxide. It is a noninvasive method to assess both ventilation and cardiac output. Most commonly, infrared light absorption by CO₂ is the method used to determine the CO₂ concentration. Sampling usually occurs in one of two ways. In a mainstream capnograph, CO₂ levels are measured with a sensor (light source and detector) placed directly in the patient’s breathing circuit. With sidestream capnography, a continuous sample of airway gas is diverted from the patient’s breathing circuit or airway to the capnograph for analysis and display. The mainstream method has a very rapid response time, but, because the sensor must be placed near the patient, long-term monitoring may be cumbersome. The sidestream method, because it uses a thin plastic sampling tube, is lighter and allows for greater flexibility, but, because transit time is unavoidable, a slower response time (approximately 3-5 seconds) results. Because of mixing of gases in the sample stream, the absolute values of the plateau and baseline may also be attenuated. The sidestream device can also be used with a modified nasal cannula or face mask to monitor CO₂ concentrations in the breath of patients who do not have endotracheal tubes in place.

The most commonly used method for measuring carbon dioxide in expired gases is infrared light absorbance. In addition, technologies such as Raman spectrometry and mass spectrometry are reliable, accurate, and responsive but generally more expensive. However, these options also offer detection of a variety of other gases and anesthetic vapors. Colorimetric detectors that attach to endotracheal tubes are available to help assess endotracheal tube placement. The colorimetric detector uses a pH-sensitive indicator strip to semiquantitatively detect exhaled CO₂. Although portable and convenient, these devices yield results that are often more difficult to interpret than conventional capnographs, and they do not provide continuous measurement of CO₂.

6 What does the capnogram reveal about a patient’s condition?

A capnograph provides a continuous display of the CO₂ concentration of gases in the airways. A normal capnogram is shown in Figure 5-1. The CO₂ partial pressure at the end of normal exhalation (phase III in the figure, end-tidal CO₂ [PETCO₂]) is a reflection of gas leaving alveoli and is an estimate of the alveolar CO₂ partial pressure (PACO₂). When ventilation and perfusion are well matched, the PACO₂ approximates the arterial PCO₂ (PaCO₂), and thus PaCO₂ equals PACO₂ plus PETCO₂. The presence of cyclical exhaled CO₂ is useful in confirming airway patency, verifying endotracheal tube placement, and verifying the adequacy of pulmonary ventilation. In addition, decreases in cardiac output caused by hypovolemia or cardiac dysfunction result in decreased pulmonary perfusion. This causes an increased alveolar dead space, which dilutes PETCO₂. Animal studies show that a 20% decrease in cardiac output causes a 15% decrease in PETCO₂.

Figure 5-1 A capnogram is a plot of airway CO₂ versus time. Phase I (inspiratory baseline) refers to inspired gas devoid of CO_2. Phase II (beginning expiration) represents expiration of anatomic dead space followed by gas from respiratory bronchioles and alveoli. Phase III (alveolar plateau) corresponds with exhalation of alveolar gases. The last portion of the plateau is termed end-tidal (PETCO₂). Phase IV (inspiratory downslope) is the beginning of the next inspiration.

Alterations in the shape of the capnogram in a patient with an endotracheal tube and mechanical ventilation often provide clues to alterations in pulmonary pathologic condition and malfunction of ventilation equipment. For example, a staircase pattern in phase II may indicate sequential emptying of the lung, which may occur in main stem partial bronchial obstruction. An upward sloping plateau during expiration is a classic indication of late emptying of poorly ventilated alveolar spaces with elevated PCO₂, which may occur with expiratory obstruction at the level of smaller airways, as seen in chronic obstructive pulmonary disease (COPD), bronchospasm, and other forms of ventilation-perfusion mismatching. A pulmonary embolus is another cause of a decrease in end-tidal CO₂.

7 What factors affect the arterial–end-tidal CO₂ gradient?

The normal gradient between PaCO₂ and PETCO₂, P(a − ET)CO₂, is < 6 mm Hg. The gradient between PaCO₂ and PETCO₂ increases when pulmonary perfusion is reduced or ventilation is maldistributed. This occurs with the development of high ventilation-perfusion alveolar units (increased dead space ventilation). The result is an increased P(a − ET)CO₂ and decreased PETCO₂. Examples of this state are COPD, pulmonary hypoperfusion, and pulmonary emboli. Technical failures such as accidental extubation and endotracheal tube cuff leak can also cause a decrease in PETCO₂.

8 Can capnography assist in CPR efforts?

In the course of CPR, capnography not only can help verify tracheal intubation, it can also monitor the adequacy of circulatory assistance. The presence of exhaled CO₂ during CPR is useful in that it provides evidence of enough circulation to produce CO₂ transport from tissues to the lungs. Thus, assuming constant minute ventilation and CO₂ production, changes in PETCO₂ reflect the status of overall circulation. Some investigators have suggested that if the PETCO₂ is > 15 mm Hg at the beginning of CPR, it may predict successful resuscitation. Failure to achieve any recovery of PETCO₂ should prompt consideration of a diagnosis of inadequate cardiac filling. This may be caused by hypovolemia, tamponade, pneumothorax, or pulmonary embolism. Low PETCO₂ also may be caused by alveolar hyperventilation, or it may indicate ineffective CPR. The return of spontaneous circulation is heralded with a substantial increase in PETCO₂, primarily because of increased flow of hypercarbic blood to the lungs.

9 What do arterial blood gas (ABG) instruments measure?

ABG instruments measure pH, PCO₂, and PO₂ using three separate electrodes. The blood gas samples are placed through an inlet into a temperature-controlled chamber (usually 37° C) where the blood is exposed to these three electrodes. The information from an ABG is essential in evaluating gas transport in blood, because the shape of the O₂Hb dissociation curve depends on blood pH (Fig. 5-2).

Figure 5-2 The oxyhemoglobin dissociation curve. DPG, Diphosphoglycerate.

10 When should ABGs be obtained?

Analysis of ABGs is used extensively in critical care to evaluate both acid-base status and oxygen and carbon dioxide gas exchange. Analysis of ABGs is indicated in virtually all cardiopulmonary conditions.

11 How is an ABG used to calculate the A − a gradient?

The alveolar gas equation is a formula used to approximate the partial pressure of oxygen in the alveolus (PAO₂).

where PB is the barometric pressure, PH₂O is the water vapor pressure (usually 47 mm Hg), FiO₂ is the fractional concentration of inspired oxygen, and R is the gas exchange ratio. (The rate of CO₂ production to O₂ utilization is usually approximately 0.8 at rest). For example, at sea level:

After calculation of the PAO₂ from the alveolar gas equation, the A − aO₂ gradient can be obtained by subtracting the PaO₂ measured from the ABGs. For example, at sea level (with PCO₂ = 40, PO₂ = 92):

The normal A − aO₂ gradient is dependent on age, body position, and nutritional status. The A − aO₂ gradient is widened under normal conditions by age, obesity, fasting, supine position, and heavy exercise. One predictive equation for estimating PO₂ (at PB = 760 mm Hg) in relation to age is as follows: PO₂ = 109 − 0.43 × Age (in years). The A − aO₂ gradient may be increased by any significant cardiopulmonary condition that results in hypoxemia and/or hypocarbia. Hypoxemia caused by simple alveolar hypoventilation will not widen the A − aO₂ gradient. Most pulmonary emboli, on the other hand, will widen the A − aO₂ gradient, because, even if hypoxemia does not occur, hypocarbia is very common.

12 Given that pulse oximetry is painless and accurate, why is ABG analysis even necessary?

Oximetry and the newer technology that made it more accessible, affordable, and accurate have decreased the need for ABG analysis in monitoring oxygen saturation. In fact, in many hospitals the number of ABGs done has decreased with the influx of oximeters into virtually every department in a hospital. However, relying on oximetry alone can lead to misdiagnosis, increased cost, and potentially fatal respiratory arrest. Consider the following examples of pitfalls of using oximetry alone that have been observed in practice.