Facts and fiction about pulse oximetry (Proceedings)

The gold standard for determining arterial blood gas levels is actually measuring arterial blood samples with a blood gas machine to get information on the arterial oxygen partial pressure (PaO2), pH and carbon dioxide partial pressure (PaCO2).  Although ideal, this technology is expensive and not necessarily warranted for every anesthesia or critical case.  The use of a pulse oximeter to measure oxygen saturation has become common practice in veterinary medicine during anesthesia and in critical care patients.  The pulse oximeter is a valuable, non-invasive monitoring tool that can provide useful information to the anesthetist about the patient's respiratory function.  Pulse oximeters are relatively inexpensive and easy to use, although there are some limitations that should be understood in order to best interpret the data they provide.

Hemoglobin-oxygen saturation (SO2) measures the percent oxygen saturation of hemoglobin and is related to PaO2 (the measure of the oxygenating efficiency of the lungs) by a sigmoid curve called the oxyhemoglobin dissociation curve.  In general, although slightly different for each species, a PO2 of 100 mm Hg is equivalent to an SO2 of 98%; a PO2 of 80 mm Hg to an SO2 of 95%; a PO2 of 60 mm Hg to an SO2 of 90%; a PO2 of 40 mm Hg to an SO2 of 75%.  Having a general knowledge of these correlations can be helpful in using pulse oximetry data to determine an estimate of ventilation adequacy.  Most patients under anesthesia are breathing 100% and their corresponding PaO2 should be around 500 mm Hg with normal lung function.

The principle of pulse oximetry is based on the red and infared light absorption characteristics of oxygenated and deoxygenated hemoglobin.  Oxygenated hemoglobin absorbs more infared light and allows more red light to pass through it.  Deoxygenated hemoglobin absorbs more red light and allows more infared light to pass through.  Pulse oximetry uses a light emitter with red and infared LEDs that shines through a reasonably translucent site with good blood flow.  Basically, oxyhemoglobin and deoxyhemoglobin have different absorption spectra. Pulse oximetry relies on the presence of a pulsatile signal generated by arterial blood which is relatively independent of non-pulsatile arterial blood, venous and capillary blood and other tissues. The pulse oximeter has a digital microprocessor within it that stores a calibration algorithm curve and a pre-programmed empirical formula which is subsequently used to estimate the patient's arterial saturation as a percent known as SpO2.

The percentage is oxygenated hemoglobin compared to total hemoglobin.  Each manufacturer uses their own calibration method to program into the pulse oximeter and therefore different brands of pulse oximeter may display slightly different saturation values on the same patient.  Data is obtained by the oximeter several times a second and is then weighted against the manufacturer's specific algorithms.  The display may then show an averaged value over the previous 3-5 seconds, being updated every 0.5-1.0 second. For clinical purposes and monitoring, most pulse oximeters provide sufficiently accurate approximations of oxyhemoglobin saturation.  Again, blood gas analysis is the gold standard and should be used to verify accuracy if necessary.

There are two methods of sending light through the measuring site: transmission and reflectance.  In the transmitter method the emitter and photodetector are opposite of each other with the measuring site in between.  The transmission probes are usually some form of clip or C-clamp. The light can then be passed through the site and measured.  Typical sites for transmission pulse oximetry measurement in veterinary patients include the tongue, pinna, toe webbing and toes, vulva, cheek or lip folds and nostrils in horses and other large animals.  Generally dark or pigmented skin areas will not work well or at all because pigmented skin absorbs light.  Sites with minimal fur will work best.

In the reflectance method, the light and the photodetector are next to each other on top of the measuring site.  The transmission method is most commonly used, although the reflectance method can be useful in small exotics and reptiles.  The probe used for the reflectance method is often referred to as a “rat probe.”  Rat probes can be secured directly to a fur free or shaved, unpigmented appendage or they can sometimes be placed on the roof of the mouth, against the interior cheek or closed into the mouth over a tiny tongue with good results.

Many pulse oximeters provide a heart rate reading and audible pulse tone.  Some provide a waveform display called a plethsymograph.This graph is a waveform that coincides with the cardiac rate or pulse rate.  The plethsymograph is useful in assessing the quality of the signal and the effects of potential artifacts.  Some pulse oximeters have a signal strength indicator instead of a waveform.  These can also be useful in determining the strength of the signal and potential pulse quality.  If there is a big difference in the heart rate determined by the pulse oximeter and actual heart rate, it is likely that there is poor pulsatile blood flow at the testing site.

Pulse oximeters are generally accurate within 2-6% in the 60-100% saturation range.  At an oxygen saturation of greater than 90%, SpO2 tends to be slightly lower than SaO2.  At a saturation of less than 70% SpO2 tends to be slightly higher than SaO2.  The accuracy of SpO2 may deteriorate substantially as SaO2 continues to decrease.

The reliability of pulse oximetry data can be affected by intrinsic and extrinsic factors.  Intrinsic factors include the use of intravascular dyes that can affect light absorption and the presence of different forms of hemoglobin.  Pulse oximetry cannot differentiate between different forms of hemoglobin so the presence of carboxyhemoglobin or methemoglobin, the only forms which fail to transport oxygen, can alter results regardless of true saturation. 

Extrinsic factors that can affect reliability include patient movement and ambient light.  Patient movement in the form of shivering, tremors or actual recovery movement can confuse the monitor and render it either unable to find a pulse or delivering an inaccurate heart rate. Bright surgical lights can interfere with the photosensor and cause erroneous readings.  Simply shielding the probe from the lights can eliminate this problem.  Normal indoor lighting does not usually affect the pulse oximeter.

Physiologic factors that can affect the accuracy and reliabilty of the pulse oximeter include hypotension, hypothermia, hypovolemia and vasoconstriction.  These things can affect perfusion and so it is important to remember that usually a peripheral site is used for probe placement and these are most at risk for hypoperfusion during anesthetic complications.  Some anesthetic drugs, most notably the alpha 2 agonists, can cause severe vasoconstriction in patients.  Hypothermic patients lose heat in the periphery first.  Hypovolemia can cause low pulse pressure.  Although these factors can cause poor intrinsic perfusion at the measurement site, it does not necessarily mean that overall perfusion is poor.  The clamp type probes can cause perfusion restrictions at the site by applying excessive pressure to the site over time.  If there is no obvious reason for desaturation to occur it is worthwhile to readjust the probe, massage the area and re-apply, or to re-apply to a new site all together.  Wetting the tongue has become a popular fix, but this probably only works because the probe is removed and perfusion improves in the area or the warm water helps improve perfusion to the site.

Pulse oximetry data should be interpreted in conjunction with the status of the entire animal as well as with other monitored patient parameters.  Anemia may cause a low SpO2 reading despite a normal PaO2 and shocky animals may have poor perfusion which can cause a falsely low SpO2.  For a patient under anesthesia breathing 100% oxygen, saturation should be between 95-100%.  If there is sudden desaturation rapid response by the anesthetist is warranted.  If the saturation level begins trending downward during an otherwise uneventful anesthesia, investigate. 

Ensure that the patient is ventilating well with an adequate tidal volume.  Make sure there are no equipment malfunctions (stuck valves, disconnect, severely exhausted soda lime, excessive dead space) and make sure the patient has a patent airway (no tube occlusion or extubation).  To treat low oxygen saturation the anesthetist can institute positive pressure ventilation with or without a mechanical ventilator. The use of positive end expiratory pressure (PEEP) can be extremely useful in improving refractory de-saturation.

If de-saturation is determined to be because of anesthetic complications such as hypovolemia, hypothermia or poor perfusion these should be addressed by treating with fluids and providing additional heat sources.  Poor perfusion should be treated by improving cardiac output and blood pressure.  This may be done by lightening the anesthetic depth and providing support with inotropes and pressors if necessary.

It is essential to understand what a pulse oximeter can measure and what information it does not provide.  Pulse oximeters measure the oxygen saturation of the hemoglobin in arterial blood and provide a pulse rate.  They can be very useful during thoracic surgery and when one lung is being collapsed to determine that the oxygenation via the remaining lung is adequate or whether increased ventilation is necessary.  Pulse oximetry is useful in attempting to wean patients from ventilation and oxygen support.  Its value as an ongoing monitor in detecting hypoxemia has been established.

The pulse oximeter does not provide information regarding carbon dioxide (CO2) levels, the content of oxygen in the blood or the partial pressure of oxygen.  Pulse oximetry does not give the anesthetist any information regarding respiratory rate or tidal volume (i.e. ventilation).  With high inspired oxygen levels the pulse oximeter may not warn the anesthetist of impending hypoxic events.  Be aware that normal oxygen saturations can be achieved in hypoventilating patients breathing high inspired oxygen content.  The pulse oximeter is not discriminating for high PaO2 values where the oxyhemoglobin curve is flat.  The adequacy of ventilation should be assessed separately from the adequacy of saturation.        

References

Haskins, Steve C. Monitoring Anesthetized Patients. In: Tranquilli, W., Thurmon, JC., Grimm, KA, editors.  Lumb and Jones' Veterinary Anesthesia and Analgesia, Fourth Ed., Ames, Iowa.  Blackwell Publishing; 2007. pp. 548-551

Grubb, T.  Pulse Oximetry. In: Greene, SA, editor. Veterinary Anesthesia and Pain Management Secrets. Philadelphia. Hanley & Belfus, Inc.; 2002.  pp 121-126.

Levansaler, AR. Monitoring: Pulse Oximetry and Temperature and Hands On.  In: Bryant, S., editor.  Anesthesia for Veterinary Technicians. Ames, Iowa.  Blackwell Publishing; 2010.  pp. 95-97