Although the preceding description is only a sketch of how pulse oximetry is actually accomplished, it provides an adequate basis for beginning to understand what pulse oximetry can and cannot measure as well as the limitations inherent in the technique.

The most obvious limitation of pulse oximetry is based on the fact that it is only a two-channel system (two wavelength bands) and therefore can only resolve two components in the blood. In pulse oximetry it is assumed that only oxyhemoglobin and reduced hemoglobin are present in the arterial blood. Any additional chromophores in the arterial blood which absorb light in the wavelength bands used by the instrument will lead to unpredictable results. These additional chromophores include other hemoglobin species and dyes that may be injected into the blood for diagnostic purposes. The two most common species of hemoglobin, after oxyhemoglobin and reduced hemoglobin, are carboxyhemoglobin and methemoglobin. If either of these dyshemoglobins are present in concentrations above normal, the pulse oximeter will give a falsely elevated reading for the arterial oxygen saturation.

The term 'arterial oxygen saturation' has different meanings in different texts. Here, it is defined as the concentration of oxyhemoglobin divided by the concentration of total hemoglobin, expressed as a percentage. This is also known as fractional oxygen saturation. (Functional saturation is defined as the ratio of the oxyhemoglobin concentration to the sum of the oxyhemoglobin concentration and the reduced hemoglobin concentration, expressed as a percentage.) Thus, if 80 per cent of a patient's arterial hemoglobin is oxygenated and the remaining 20 per cent is deoxygenated (or reduced), the pulse oximeter will read 80 per cent. However, if, instead of deoxygenated hemoglobin, the patient had 20 per cent of his or her hemoglobin bound to carbon monoxide, the pulse oximeter would read almost 97 per cent, thus deceiving the clinician into thinking that the patient's blood was adequately oxygenated. A similar problem would occur in the presence of 20 per cent methemoglobin and 80 per cent oxyhemoglobin; in this case the instrument would read approximately 90 per cent saturation, again indicating a falsely elevated saturation. Note that these pulse oximeter readings in the presence of dyshemoglobins are neither the fractional nor the functional saturation. This is a severe and potentially dangerous limitation of current pulse oximetry.

Some manufacturers calibrate their pulse oximeters to functional saturation and some calibrate to fractional saturation. This is done by slowly desaturating a number of healthy volunteers by having them breathe hypoxic gas mixtures. At various saturation levels, arterial blood samples are drawn for invasive analysis of oxygen saturation. These measurements are paired with the readings made by the pulse oximeter at the same saturation levels, and these pairs of data points are used to generate the pulse oximeter's calibration curves. These are the curves that the instrument will use to relate the measurements made by the pulse oximeter (the ratio of the differential absorbances) to the arterial oxygen saturation to be displayed. It is important to recognize that, while the manufacturer may calibrate to either functional or fractional saturation, the oximeter can read neither. Its measurements are accurate only if the dyshemoglobins present in the blood are at normal levels (1-3 per cent) and all the remaining hemoglobin is either oxygenated or reduced. The difference in the calibration methodologies simply means that an instrument calibrated to functional saturation will always read slightly higher (approximately 1-3 per cent) than the same, or any other, pulse oximeter that is calibrated to fractional saturation.

Another limitation of the pulse oximeter stems from the fact that it needs an arterial pulse in the tissue under test to generate the signal necessary to measure oxygen saturation. When the pulse in the tissue under test, such as a finger in the case of a finger probe, is small or absent, the instrument is apt to give erroneous information. The pulse at the probe site may be small because of poor perfusion of the patient in general or poor perfusion at the probe site (e.g. because of cold hands). When perfusion is low, the pulse oximeter will generally try to increase the amplification of whatever signal it receives in an attempt to continue to provide readings of oxygen saturation. Thus any noise present, say from patient motion or from room lights that may be pulsating, will make a larger contribution to the signal and may cause the oximeter to give readings that have very little to do with the arterial oxygen saturation. Although this is a problem with current oximeters, it need not necessarily mislead the clinician. If the pulse oximeter waveform displays a reasonably clean pressure waveform for approximately 10 s, the oximeter is probably reading correctly. If the instrument does not have a waveform display, the clinician can attempt to correlate the pulse rate with the ECG and observe the stability of the saturation readings as an indication of accuracy. However, these techniques are not as reliable as being able to 'see' the data that the oximeter is actually using to make its calculations.

It would be easy to make a pulse oximeter that rarely reads incorrectly in a low-perfusion situation by not allowing readings at all below some specific trigger level. In fact, oximeters do stop making readings below some preset signal level, but the manufacturers attempt to set this point as low as possible to allow the clinician to obtain saturation readings on patients who are very sick. While this is a good strategy most of the time, it does lead to some additional problems. Sometimes pulse oximeters appear to be reading satisfactorily on a patient whose heart has stopped beating. This phenomenon is due to attempts by the instrument to read the smallest possible signal. Anything that looks like a pulsatile signal will be interpreted by the oximeter as a pulse, and the instrument will make every effort to calculate and display an oxygen saturation reading. This pulsatile signal could be generated by a ventilator inducing a pressure waveform into the venous system, or just by physically moving or slightly jostling the patient and therefore the probe. Any motion of the probe on the patient will introduce a false signal in the pulse oximeter. Again, the pulse oximeter's waveform display allows easy identification of artifactual signals.

A few other potential sources of error should also be mentioned. Excessive patient motion, particularly at the probe site, can generate signals that dwarf those caused by the arterial pulsation. Even when the tissue under test is well perfused, motion can cause false readings which may be either high or low. Electrical noise such as that generated by electrosurgery can also diminish the accuracy of pulse oximeter readings. Because the magnitude of these effects on the pulse oximeter's ability to make accurate readings depends on the perfusion level of the tissue under test, it is always best to try to find a probe site that provides the maximum possible signal level. Perfusion can be enhanced by warming the patient, rubbing the tissue at the probe site, or using vasodilator creams. Finally, high ambient light levels can overwhelm the signal detection circuitry so that the oximeter is unable to make any measurements whatsoever.

Healthy Fat Loss For A Longer Life

Healthy Fat Loss For A Longer Life

What will this book do for me? A growing number of books for laymen on the subject of health have appeared in the past decade. Never before has there been such widespread popular interest in medical science. Learn more within this guide today and download your copy now.

Get My Free Ebook

Post a comment