Pulse oximetry has taken clinical anaesthesia, respirology and now critical care by storm over the past few years. It has many of the characteristics of an ideal monitoring technique: portability, noninvasiveness (a cutaneous sensor is employed), ease of use (calibration is not required) and the capability for continuous on-line monitoring of arterial oxygen saturation (SaO2).
The pulse oximeter is a spectrophotometric device that detects and calculates the differential absorption of light by oxygenated and reduced hemoglobin to produce a measurement called SpO2, an estimate of SaO2. A light source and a photodetector are contained within an ear or finger probe for easy application. Two wavelengths of monochromatic light -- red (660 nm) and infrared (940 nm) -- are used to gauge the presence of oxygenated and reduced hemoglobin in the arterial blood of the capillary bed being monitored. With each pulse beat the device interprets the ratio of the pulse-added red absorbance to the pulse-added infrared absorbance. The calculation requires previously determined calibration curves that relate transcutaneous light absorption to direct SaO2.
To estimate saturation over a wide range of pulse volumes the pulse oximeter automatically increases its amplification as the pulse signal decreases. When perfusion in the capillary bed is too low to produce a pulse strong enough to allow an accurate reading, the instrument displays a "no SpO2 value" message.
Noninvasive and continuous monitoring of the SaO2 in adult critical care units may be particularly beneficial when a patient's arterial oxygenation is precarious and hemoglobin desaturation occurs rapidly. Restlessness, agitation, confusion, cyanosis, hypotension and tachycardia are all delayed manifestations of hypoxemia and may be missed or inappropriately managed. Pulse oximetry can warn of a decrease in PaO2 before clinical signs indicate the need for ABG analysis. Hence, pulse oximetry may be useful in the management of patients with pulmonary edema or pneumonia who are undergoing ventilation, endotracheal intubation or suction, bronchoscopy, position changes, transport or weaning from mechanical ventilation.
Pulse oximetry has become a part of the regular monitoring of most, if not all, patients undergoing general anaesthesia, especially children, and it is expected that the same enthusiasm will extend to the adult critical care unit if it has not already done so. Hospital committees will have to decide whether pulse oximetry at each bedside in these units should become standard.
Clinical effectiveness
Pulse oximetry is considered a safe procedure, but because of device limitations, false-negative results for hypoxemia and/or false-positive results for normoxemia or hyperoxemia may lead to inappropriate treatment of the patient. In addition, tissue injury may occur at the measuring site as a result of probe misuse (eg, pressure sores from prolonged application or electrical shock and burns from the substitution of incompatible probes between instruments. Although the oximeter measures arterial saturation, not arterial oxygen tension (PaO2), the two are related through the oxyhemoglobin dissociation curve. Oximetry is not a sensitive guide to changes in oxygenation when the PaO2 is high. When disparity exists between SpO2, SaO2 readings, and the clinical presentation of the patient, possible causes should be explored before results are reported. Discrepancies may be reduced by monitoring at alternate sites or appropriate substitution of instruments or probes.If such steps do not remedy the disparity, results of pulse oximetry should not be reported; instead, a statement describing the corrective action should be included in the patient's medical record, and direct measurement of arterial blood gas values should be requested. The absolute limits that constitute unacceptable disparity vary with patient condition and specific device. Clinical judgment must be exercised.
When direct measurement of SaO2 is not available or accessible in a timely fashion, a SpO2 measurement may temporarily suffice if the limitations of the data are appreciated. SpO2 is appropriate for continuous and prolonged monitoring (eg, during sleep, exercise, bronchoscopy). SpO2 may be adequate when assessment of acid-base status and/or PaO2 is not required.
Considerations
Equipment: pulse oximeter and related accessories (probe of appropriate size)--the oximeter should have been validated by the manufacturer by a comparison of its values (and consequently its calibration curve) with directly measured oxyhemoglobin saturation.
Personnel: Pulse oximetry is a relatively easy procedure to perform. However, if the procedure is not properly performed or if it is performed by persons who are not cognizant of device limitations or applications, spurious results can lead to inappropriate intervention.
After agreement has been initially established between SaO2 and SpO2, the frequency of SpO2 monitoring (ie, continuous vs 'spot check') depends on the clinical status of the patient, the indications for performing the procedure and recommended guidelines. For example, continuous SpO2 monitoring may be indicated throughout a bronchoscopy for detecting episodes of desaturation, whereas a spot check may suffice for evaluating the efficacy of continued oxygen therapy in a stable postoperative patient. However, it must be emphasized that direct measurement of SaO2 is necessary whenever the SpO2 does not confirm or verify suspicions concerning the patient's clinical state.
If the device probe is intended for multiple patient use, the probe should be cleaned between patient applications according to manufacturer recommendations.
The external portion of the monitor should be cleaned according to manufacturer's recommendations whenever the device remains in a patient's room for prolonged periods, when soiled, or when it has come in contact with potentially transmissible organisms.
The precision of pulse oximeters is within + 2% to + 3% when the SaO2 is 90% or more; the precision is thus + 4% to + 6% if a 95% confidence interval (+ 2 SD) is desired. For example, if the SpO2 is 94%, then the true SaO2 may be as low as 88% or as high as 100% for 95% of the measurements. This translates into a wide range of PaO2 values. Adequate arterial oxygenation will result if a target SpO2 of 92% or more is achieved in white patients and 95% or more in black patients.
Pulse oximetry may fail to record accurately the true SaO2 during severe or rapidly produced desaturation and during physiologic extremes (e.g., hypotension, hypothermia, unstable hemodynamic factors and agitation). More information is required concerning the prevalence of signal failure with the regular use of pulse oximetry and the effect this has on nursing and medical care (e.g., Does frequent signal failure lead to greater demands on nurses to reposition the probe and to frequent, annoying alarms?).
The provision of a pulse oximeter at every bedside cannot be justified until clinical effectiveness and economic efficiency in the routine care of critically ill patients have been satisfactorily documented. However, given the present level of clinical accuracy and effectiveness in documenting arterial desaturation it would be reasonable to support limited use: a number of portable pulse oximeters should be available for use during expected arterial desaturation (e.g., during intubation and bronchoscopy) and in patients with precarious oxygenation (e.g., those with adult respiratory distress syndrome requiring a high fractional intake of oxygen or positive end-expiratory pressure). However, their limitations should be borne in mind.
