Pulse oximetry

INTRODUCTION — Pulse oximetry measures peripheral arterial oxygen saturation (SpO₂) as a surrogate marker for tissue oxygenation. It has become the standard for continuous, noninvasive assessment of oxygenation and is often considered the "fifth vital sign" [1-3]. Theoretical and clinical aspects of pulse oximetry will be reviewed here. Other measures of oxygenation, mechanisms of hypoxemia, and use of pulse oximetry in newborns for the detection of congenital heart disease are discussed separately. (See "Measures of oxygenation and mechanisms of hypoxemia" and "Newborn screening for critical congenital heart disease using pulse oximetry".)

PRINCIPLES AND EQUIPMENT — Pulse oximetry uses spectrophotometry to determine the proportion of hemoglobin that is saturated with oxygen (ie, oxygenated hemoglobin; oxyhemoglobin) in peripheral arterial blood. Light at two separate wavelengths illuminates oxygenated and deoxygenated hemoglobin in blood. The ratio of light absorbance between oxyhemoglobin and the sum of oxyhemoglobin plus deoxyhemoglobin is calculated and compared with previously calibrated direct measurements of arterial oxygen saturation (SaO₂) to establish an estimated measure of peripheral arterial oxygen saturation (SpO₂) [4].

The Beer-Lambert law — Pulse oximetry estimates peripheral SpO₂ using a variation of the Beer-Lambert law. This law states that the absorption of light of a given wavelength passing through a non-absorbing solvent, which contains an absorbing solute, is proportional to the product of the solute concentration, the light path length, and an extinction coefficient. The Beer-Lambert law can readily be applied to co-oximeters in a laboratory setting because the light path length is known and hemoglobin is in solution. However, it must be modified for pulse oximetry to overcome the obstacles associated with interference from tissue and pulsatile flow [5]. This modification involves measuring absorbance at two different wavelengths, one to detect oxyhemoglobin and the other to detect deoxyhemoglobin [1].

Probes — Pulse oximeter probes consist of two light-emitting diodes and a photodetector.

●Emitters – Deoxyhemoglobin absorbs light maximally in the red band of the spectrum (600 to 750 nm), and oxyhemoglobin absorbs maximally in the infrared band (850 to 1000 nm) [1]. Thus, the emitters emit light at 660 nm and 940 nm for optimal detection of these two substances.

●Detector – The detector (also known as sensor) detects the absorbance of light from exposed tissue. The values are processed and a saturation determined. (See 'Calibration and calculation' below.)

In general, detectors and emitters are positioned facing each other through interposed tissue (picture 1) [6]. Probes are most frequently placed on the anterior and posterior aspect of fingers, toes, or ear lobes [7]. The nasal ala is another option [8]. In infants, probes may also be placed on the palms, feet, arms, cheeks, tongue, penis, nose, or nasal septum [9]. Forehead probes have the emitter and detector adjacent to each other so that saturation is measured from light that is reflected back from (not through) exposed tissue. These sites are preferentially used since they contain a high density of vascular tissue.

Although various types of probes are available, none has shown clear superiority over another [10-13]. When clinicians cannot get a clear reading with one probe, another is often tried. (See 'Troubleshooting sources of error' below.)

The response time to changes in oxygenation varies but is generally delayed for most available probes. As an example, some studies have shown that ear probes and forehead probes respond more quickly to a change than conventional finger probes (eg, 94 versus 100 seconds for desaturation, and 23 versus 29 seconds for increases in saturation) [14,15].

Calibration and calculation — The saturation value is calculated using microprocessors that utilize absorbance readings from light-exposed oxyhemoglobin and deoxyhemoglobin. The emitters are switched on and off several hundred times per second [16]. Absorption during pulsatile flow relates to the characteristics of arterial blood plus background tissue and venous blood, whereas absorption during nonpulsatile flow is due only to the background tissue and venous blood (figure 1) [17]. Absorption at the two wavelengths during pulsatile flow is divided by absorption during nonpulsatile flow, and these ratios are fed into an algorithm in the microprocessor to yield a saturation value. The displayed value is an average based on the previous three to six seconds [6,16]. In addition to peripheral SpO₂ many pulse oximeters also display pulse rate and relative pulse amplitude [1,18].

The microprocessors of pulse oximeters are calibrated using reference tables of actual SaO₂ measurements performed using co-oximetry and compiled using data from exposing healthy volunteers exposed to decreasing fraction of inspired oxygen (FiO₂) to yield SaO₂ ranging from 100 to 75 percent. Because it would be unethical to intentionally generate lower saturations in volunteers, values for an SaO₂ less than 75 percent are obtained by extrapolation from these volunteer data. Pulse oximeter manufacturers claim that reported values between 70 and 100 percent are accurate to within 2 to 3 percent of the true value, corresponding with FDA standards [19,20]. In practice, the cut-off for acceptable accuracy is felt by many clinicians to be 80 percent (which usually reflects an arterial oxygen tension [PaO₂] of approximately 50 mmHg at a pH of 7.4), and varies depending on the model of pulse oximeter used [21].

ADVANTAGES AND DISADVANTAGES — There are many advantages of pulse oximetry over physical examination and arterial blood gas measurement:

●Rapid – Pulse oximetry is a rapid tool that accurately assesses oxygenation, particularly in emergency situations. Oxygenation is difficult to assess on the basis of physical examination alone. Frank cyanosis does not develop until the level of deoxyhemoglobin reaches 5 g/dL, which corresponds to an arterial oxygen saturation (SaO₂) of around 67 percent [4]. In addition, the threshold at which cyanosis becomes apparent is affected by multiple variables including peripheral perfusion, skin pigmentation, and hemoglobin concentration [6]. Similarly, compared with arterial blood gas analysis, pulse oximetry also provides immediate results both before and after oxygen therapy.

●Noninvasive – Blood gas analysis by arterial puncture or arterial line sampling was for many years the only available method of detecting hypoxemia, but this technique is painful and has potential complications [22]. In contrast, pulse oximetry allows noninvasive measurement of arterial hemoglobin saturation without the risks associated with arterial puncture. (See "Arterial blood gases" and "Intra-arterial catheterization for invasive monitoring: Indications, insertion techniques, and interpretation".)

●Provides continuous data – Pulse oximetry, unlike clinical examination and an indwelling arterial catheter, provides continuous data such that oxygen therapy can be easily adjusted to a target level.

Limitations include the following:

●Inability to detect hyperoxemia – Pulse oximetry is unable to detect significant hyperoxemia. This is due to the shape of the oxygen-hemoglobin dissociation curve, where large changes in the partial pressure in arterial oxygen (PaO₂) may result in no change in oxygen saturation if the saturation is already near 100 percent (eg, patient with normal lung function on mechanical ventilation receiving a fraction of inspired oxygen of 1). Therefore, in patients at risk for profound hyperoxemia, SaO₂ by pulse oximeter should be verified by arterial blood gas [23]. (See "Measures of oxygenation and mechanisms of hypoxemia" and "Adverse effects of supplemental oxygen".)

●Inability to measure arterial oxygen tension – Since pulse oximetry does not measure PaO₂, overreliance on pulse oximetry can miss detection of clinically significant hypoxemia in adults but particularly in children [24]. A large decrease in PaO₂ will not produce a significant fall in SaO₂ until the steeper portion of the oxygen hemoglobin dissociation curve is encountered at a PaO₂ of approximately 60 to 70 mmHg. This is particularly important in patients receiving supplemental oxygen. As an example, a fall in PaO₂ in such a patient from 140 to 65 mmHg would be required before a significant decrease in oxygen saturation is detected. Furthermore, pulse oximetry results are signal-averaged over several seconds. Therefore, the pulse oximeter may not detect a hypoxemic event for close to a minute after it has occurred [25,26]. This delay may be of particular significance when the device is being used for monitoring during intubation.

●Inability to measure ventilation – While pulse oximetry is a convenient way of measuring arterial oxygenation, it does not measure the arterial carbon dioxide tension (PaCO₂), which is a measure of ventilation. In addition, supplemental oxygen administered on the basis of oximetry alone in patients with hypercapnia may be harmful by worsening hypercapnia (ie, oxygen-induced hypercapnia). Therefore, when hypercapnia or hypoventilation is suspected, an arterial blood gas should be obtained or, alternatively, ventilation can be assessed with end tidal CO₂ monitoring [27-30]. As an example, one inadvertently hypoventilated patient who was administered 100 percent oxygen during hip arthroplasty and monitored with pulse oximetry alone developed a PaCO₂ of 265 mmHg and an arterial pH of 6.65, despite maintenance of oxygen saturations of 94 to 96 percent [29]. (See "The evaluation, diagnosis, and treatment of the adult patient with acute hypercapnic respiratory failure" and "The evaluation, diagnosis, and treatment of the adult patient with acute hypercapnic respiratory failure", section on 'Titration of oxygen' and "Carbon dioxide monitoring (capnography)".)

COMPLICATIONS — Because of the noninvasive aspect of pulse oximetry, local complications are extremely rare, especially when compared with arterial blood gas monitoring.

Digital injury has been reported on rare occasions in critically ill patients, although it is more likely that the injury is due to the underlying poor digital perfusion and/or use of vasopressors [31].

Burns have also been reported in patients undergoing magnetic resonance imaging (MRI), although this complication can be avoided by temporarily removing the probe during MRI [9,32]. This complication is believed to result from the generation of electrical skin currents beneath the pulse oximeter cables, which act as an antenna. Notably, no permanent adverse effects have been reported. (See "Patient evaluation for metallic or electrical implants, devices, or foreign bodies before magnetic resonance imaging", section on 'Assessing implants, devices, or foreign bodies for MRI'.)

Limitations are discussed above (See 'Advantages and disadvantages' above.)

APPLICATIONS — Pulse oximetry is indicated in any clinical setting where hypoxemia may occur. These settings include patient monitoring in emergency departments, operating rooms, emergency medical services (EMS) systems, postoperative recovery areas, endoscopy suites, sleep and exercise laboratories, oral surgery suites, cardiac catheterization suites, facilities that perform conscious sedation, labor and delivery wards, inter-facility patient transfer units, altitude facilities, aerospace medicine facilities, and patients' homes [6,27,33-41]:

Despite its widespread use, the value of oximetry has been poorly studied with no trials showing a convincing benefit on clinically meaningful outcome (eg, mortality, myocardial infarction, resource allocation [26]). Nonetheless, examples where routine use of pulse oximetry has some value include the following:

●In a pediatric intensive care unit (ICU), using pulse oximetry decreases the number of blood gases obtained and limits the duration of oxygen therapy, without jeopardizing patient outcome [35]. (See "Acute severe asthma exacerbations in children younger than 12 years: Intensive care unit management" and "Overview of neonatal respiratory distress and disorders of transition".)

●In emergency departments, initial pulse oximetry readings in children with asthma exacerbations are predictive of the need for hospitalization [42]. Several studies have shown that use of pulse oximetry also reduces the number of arterial blood gases obtained in the intensive care unit and emergency department [43,44]. (See "Acute asthma exacerbations in children younger than 12 years: Emergency department management".)

●In postoperative patients, routine pulse oximetry has been shown to decrease the need for rapid response team activation and transfer to the ICU [45]. In a randomized trial of 20,000 perioperative patients pulse oximetry use was associated with lower rates of hypoxemia when compared with patients in whom oximetry was not used (0.4 versus 8 percent) [39]. (See "Overview of post-anesthetic care for adult patients".)

●In newborns, pulse oximetry has value as a routine outpatient screen for congenital heart disease. (See "Newborn screening for critical congenital heart disease using pulse oximetry".)

●The medical response to the COVID-19 pandemic has included new applications for pulse oximetry. Because of increasingly limited healthcare resources, including hospital beds, many COVID positive patients are being managed at home. Patients can remain relatively asymptomatic despite disease progression with hypoxemia, placing them at risk for rapid deterioration. Pulse oximetry is being used both to assist in deciding which infected patients can be safely discharged from the emergency department and which patients being managed at home should present to the emergency department for further assessment and possible admission. A pulse oximetry reading in the 92 to 96 percent range is often used as an indication of adequate oxygenation in these patients. However, given the rapidly evolving understanding of the infection’s complex pathophysiology, pulse oximetry results should be considered in the broader context of the patient’s overall clinical condition [46]. Novel use of home pulse oximetry monitoring in COVID-19 patients discharged from the emergency department identifies need for hospitalization [47,48].  

INTERPRETING THE RESULTS — In most patients peripheral oxygen saturation as measured by pulse oximetry (SpO₂) provides accurate information on tissue oxygenation, allowing the clinician to assess and treat patients who are potentially hypoxemic. As a general principle, clinicians should pay attention to trends in oxygenation. When treating patients with supplemental oxygen for hypoxemia, they should target levels that are desirable for the specific etiology while simultaneously avoiding oxygen toxicity. The clinician should be aware of the limitations and errors associated with pulse oximetry and have a low threshold to obtain arterial blood for analysis. (See "Arterial blood gases" and "Venous blood gases and other alternatives to arterial blood gases" and 'Advantages and disadvantages' above and 'Calibration and calculation' above and 'Troubleshooting sources of error' below.)

Waveform analysis — Peripheral oxygen saturation can only be interpreted when the waveform is normal. A normal pulse oximeter waveform has a dicrotic notched appearance typical of an arterial waveform (figure 2) that synchronizes with a palpable or observed heart rate. Inadequate waveforms are discussed below. (See 'Inadequate waveform' below.)

Correlation with arterial oxygen saturation — Pulse oximetry provides an estimate of the proportion of hemoglobin that is saturated with oxygen (ie, SpO₂), rather than partial pressure of oxygen (PaO₂). In most patients with SpO₂ values of 90 percent or higher, the value lies within 2 to 3 percent above or below the true arterial saturation (SaO₂) reference standard [49-51]. However, the accuracy worsens when the SaO₂ is <90 percent, and especially below 80 percent [52]. Thus, SpO₂ is less reliable in critically ill patients where oxygenation can rapidly fluctuate and desaturation is common [49,53].

Important differences between arterial saturation and arterial oxygen tension should be noted:

●SpO₂ and SaO₂ reflect the main mechanism by which oxygen is carried to peripheral tissue (ie, 98 percent of arterial oxygen content is normally carried by hemoglobin).

●The PaO₂ only measures the proportion that is dissolved in plasma (ie, the remaining 2 percent), representing the minor mechanism for oxygen transport. Although arterial blood gas analysis can measure arterial saturation, the PaO₂ is commonly used to estimate arterial oxyhemoglobin saturation and oxygen content because the dissolved and hemoglobin-bound oxygen pools are in equilibrium. However, changes in pH, temperature, and the concentration of 2,3-diphosphoglycerate alter the PO₂-SaO₂ relationship, and may result in misleading calculations of oxyhemoglobin saturation (figure 3). (See "Structure and function of normal hemoglobins" and "Measures of oxygenation and mechanisms of hypoxemia".)

Optimal oxygen saturation — There is no optimal level of oxygen saturation, below which tissue hypoxia occurs because of the large number of variables that contribute to hypoxia at the tissue and cellular level (temperature, pH, tissue blood flow). As a result, there is no consensus about what constitutes normal and abnormal oximetry. Nonetheless, at sea level, we and other experts consider resting oxygen saturation ≤95 percent or exercise desaturation of ≥5 percent as abnormal [54]. However, these values should not be considered in isolation. Trends in oxygen saturation and the underlying disease process are important for interpretation. For example:

●A resting oxygen saturation of 96 percent could be abnormal if a patient previously had a resting oxygen saturation of 99 percent.

●A target level of 88 to 92 percent may be sufficient in a patient with an acute exacerbation of chronic obstructive pulmonary disease (COPD) who is chronically hypercapnic. In contrast, a target saturation of >95 percent may be considered optimal in a pregnant woman with acute respiratory distress syndrome. (See "The evaluation, diagnosis, and treatment of the adult patient with acute hypercapnic respiratory failure", section on 'Titration of oxygen' and "Critical illness during pregnancy and the peripartum period", section on 'Mechanical ventilation' and "Overview of initiating invasive mechanical ventilation in adults in the intensive care unit", section on 'Fraction of inspired oxygen' and "COPD exacerbations: Management", section on 'Oxygen therapy'.)

Similarly, there is no optimal level of oxygen saturation above which tissue hyperoxia occurs since hyperoxia cannot be assessed with oximetry or arterial blood gas analysis. Thus, clinicians should titrate oxygen according to the specific etiology, while attempting to avoid any theoretical potential for tissue hypoxia. As examples:

●In patients with an exacerbation of chronic obstructive pulmonary disease (COPD) with chronic hypercapnia, targeting levels ≥92 percent may worsen hypercapnia, while patients with carbon monoxide (CO) poisoning or air embolism should be treated with 100 percent oxygen or hyperbaric oxygen to eliminate CO or air, respectively. (See "Air embolism", section on 'Treatment' and "Carbon monoxide poisoning", section on 'Management' and "COPD exacerbations: Management", section on 'Oxygen therapy'.)

●Oxygen toxicity, particularly in premature neonates can be limited by titrating oxygen to an SpO₂ of 90 percent, which usually reflects a PaO₂ of approximately 60 mmHg at a pH of 7.4 [9,25]. (See "Adverse effects of supplemental oxygen".)

TROUBLESHOOTING SOURCES OF ERROR — Pulse oximetry is subject to artifactual and patient-related sources of error. The best defense against error is a high index of suspicion. If a saturation reading is in doubt, a health care worker can perform a quick quality assurance test by putting the probe on his or her own finger as near as possible to the original patient site [17]. This ensures that abnormal readings are not due to equipment error. The clinician should then investigate other potential sources of error, most of which are easily resolved (table 1).

Inadequate waveform — A normal pulse oximeter waveform has a dicrotic notched appearance similar to an arterial waveform. Waveforms are considered inadequate when the dicrotic notched appearance is lost (eg, low in amplitude or erratic). Because the signal is poor, many of the reasons for an inadequate waveform (eg, hypoperfusion) will also give erroneously low readings. (See 'Waveform analysis' above and 'Falsely low readings' below.)

Improper probe placement — Improper probe placement can occur in many scenarios and is often associated with loss of amplitude of the waveform (ie decreased or flattened pulse oximetry amplitude):

●Malposition or poor attachment to the skin can result in either a falsely elevated or depressed reading (light from only one of the two light-emitting diodes passes through the tissue) [55].

A similar problem can occur in infants and small children because the small size of fingers or other tissues may result in differences in the path length of one light source compared to the other.

Many of these problems can be resolved by ensuring the probe is properly attached with the light sources and detectors opposite each other in a nontangential path (eg, switch to a separate digit or use a different probe such as an ear or forehead probe) [56].

●Placement of the sensor on the same extremity as a blood pressure cuff or arterial line can cause erroneous readings and should be avoided [57].

Motion or noise artifact — A poor signal-to-noise ratio will cause signal artifact and falsely lower oximetry readings [4,18,58]. This most commonly results from motion due to shivering, seizure activity, pressure on the sensor, or transport of the patient by ambulance or helicopter. The waveform will typically appear erratic and lose its normal shape. Newer pulse oximeters appear to be less influenced by motion artifact [59,60].

Hypoperfusion — Pulse oximetry readings can be falsely low due to signal failure in the setting of hemodynamic instability or poor limb perfusion from extremity elevation, vasoconstriction, or peripheral vascular disease [49,53,61].

In adults, the accuracy of standard pulse oximeters decreases dramatically when systolic blood pressure falls below 80 mmHg, generally resulting in underestimation of the actual arterial oxygen saturation [62]. Both falsely low and falsely elevated levels have been reported in septic patients [63,64]. Repositioning the probe or using an alternate site may help.

There are conflicting data regarding superiority of forehead versus finger probes in hypoperfused states such that switching from one to another type of probe is appropriate when an inadequate tracing due to hypoperfusion is suspected [10,11,13]. When in doubt an arterial blood gas should be drawn.

Hypothermia — Hypothermia may interfere with pulse oximetry because of the associated peripheral vasoconstriction and shivering. This can contribute to a delay in the recognition of acute hypoxemia, particularly if finger probes are used [65]. Hypothermic patients should be monitored using an ear or forehead probe, which are less likely to delay recognition of acute desaturation. Warming should correct the issue. (See "Accidental hypothermia in adults".)

Falsely normal or high reading

Carboxyhemoglobin — Carboxyhemoglobin absorbs approximately the same amount of 660 nm light as oxyhemoglobin. Thus, the pulse oximetry reading represents an inexact summation of oxyhemoglobin and carboxyhemoglobin (figure 4) [4,7,66-68]. Due to the interference of high levels of carboxyhemoglobin carbon monoxide (CO) poisoning, or in chronic, heavy smokers, a falsely reassuring normal pulse oximetry reading may mask life-threatening arterial desaturation. Arterial oxygen tension (PaO₂) measurements tend to be normal because PaO₂ reflects O₂ dissolved in blood, and this process is not affected by CO. In contrast, hemoglobin-bound O₂ (which normally comprises 98 percent of arterial O₂ content) is profoundly reduced in the presence of carboxyhemoglobin. (See 'Correlation with arterial oxygen saturation' above.)

In such cases, whenever carboxyhemoglobinemia is suspected, co-oximetry (not pulse oximetry) is recommended for the measurement of carboxyhemoglobin levels. While some newer pulse oximeters may be able to detect carboxyhemoglobin, they are expensive, poorly studied, and not universally available. (See 'Co-oximetry' below and "Carbon monoxide poisoning".)

Glycohemoglobin A1c — Glycohemoglobin A1c levels greater than 7 percent in type 2 diabetics with poor glucose control have been shown to result in overestimation of arterial oxygen saturation (SaO₂) by pulse oximetry. This may be due to an increased hemoglobin oxygen affinity. However, in our experience most patients with diabetes can be monitored using pulse oximetry, but if doubt exists, arterial blood gas (ANG) analysis may be warranted [69].

Skin pigmentation — In theory, skin pigmentation (including that due to hyperbilirubinemia) should have no effect on oximetry, since it should absorb at a constant level and be subtracted out as part of the background in the SaO₂ calculation. However, several studies suggest that skin pigmentation can falsely increase pulse oximetry readings resulting in "hidden hypoxemia" [9,17,70-78]. If there is a concern that skin pigmentation is impacting pulse oximetry readings, an ABG should be obtained.

Best illustrating the impact of skin pigmentation is a multicenter study that compared pulse oximetry with arterial blood gas analysis in over 10,000 adult inpatients [74]. Values in Black individuals were compared with light-skinned individuals. In two separate cohorts, pulse oximetry ranging between 92 and 96 percent overestimated oxygen saturation as measured by ABG in 12 to 17 percent of Black individuals compared with 4 to 6 percent of White individuals. Given the importance placed on the use of pulse-oximetry to risk-stratify COVID patients, the US Food and Drug Administration and the Centers for Disease Control and Prevention highlighted the concerns raised in that study [79,80].

In other studies, an increased incidence of both signal detection errors and readings that were erroneously elevated by 4 percent or more have been described in Black patients [9,71]. A study from Singapore compared pulse oximetry readings with ABG-determined oxygen saturation in Chinese, Malay, and Indian patients in an intensive care unit. Pulse oximetry readings overestimated oxygen saturation in patients with darker skin pigmentation [81]. Similarly, in another study there was greater variability in oxygen saturation levels in patients who self-identified as Black, followed by Hispanic, Asian, and White [82]. Moreover, patients with “hidden” hypoxemia subsequently experienced higher organ dysfunction scores and in-hospital mortality.

However, a study of infants with cyanotic heart disease and baseline hypoxemia did not show a bias in pulse oximetry measurements in dark-pigmented versus light-pigmented patients [73].  

Falsely low readings

Methemoglobin — Methemoglobin absorbs at both 660 and 940 nm [7]. Methemoglobinemia should be suspected in those with cyanosis and normal PaO₂. Up to a methemoglobin level of 20 percent, SaO₂ drops by about one-half of the methemoglobin percentage. At higher methemoglobin levels, SaO₂ trends toward 85 percent regardless of the true percentage of oxyhemoglobin, thus leading to over- or underestimation of the true SaO₂ [4,9,83,84]. When methemoglobinemia is suspected, co-oximetry should be used to accurately determine the methemoglobin level. (See "Methemoglobinemia" and 'Co-oximetry' below.)

Sulfhemoglobin — Sulfhemoglobin absorbs at 660 nm, similar to oxyhemoglobin. Its absorbance at 940 nm is unknown. Sulfhemoglobinemia is most commonly caused by the ingestion of oxidizing drugs (eg, dapsone, sulfonamides, metoclopramide, nitrates). Patients present in a similar fashion to those with methemoglobinemia (cyanosis and normal PaO₂) [85-87]. Unlike methemoglobin however, sulfhemoglobin shifts the hemoglobin-oxygen dissociation curve to the right, thereby "unloading" oxygen in the periphery such that the adverse effects are not as clinically significant as with methemoglobinemia. High levels can falsely reduce the SpO_2, trending towards 85 percent, similar to methemoglobin. However, multi-wavelength co-oximetry does not accurately distinguish it from methemoglobin. If suspected, specialized biochemical testing available in a limited number of centers is required. While testing is ongoing the offending agent should be stopped, if feasible. Patients do not respond to methylene blue therapy (ie, therapy for methemoglobinemia) which may also be a diagnostic clue. There is no known antidote, however, severe cases may respond to exchange transfusion [87].

Sickle hemoglobin — Patients with sickle cell disease are at risk of hypoxemia caused by a number of pulmonary complications, which are discussed separately. (See "Overview of the pulmonary complications of sickle cell disease".).

Sickle hemoglobin generally produces pulse oximeter readings similar to normal hemoglobin. Rare cases of falsely elevated and falsely low readings have been reported especially during vaso-occlusive crises (perhaps due to hypoperfusion). However, in general, the difference between SaO₂ and SpO₂ measurements is not clinically significant [27,88-91]. When hypoxemia is suspected or doubt exists, an arterial blood gas should be drawn.

Inherited forms of abnormal hemoglobin — Inherited forms of abnormal hemoglobin (Hb) are rare but have been reported to result in falsely low SpO₂ readings (eg, Hb Lansing, Hb Bonn, Hb, Koln, Hb Hammersmith, and Hb Cheverly) [92-97].

Severe anemia — In vitro and animal studies suggest that pulse oximetry readings may be affected by profoundly decreased hemoglobin concentration [27]. In vivo, low hemoglobin concentrations appear to cause falsely low readings when the SaO₂ is below 80 percent [9]. However, this effect is not clinically significant until the hemoglobin level is less than 5 g/dL [53,98].

Venous congestion — Venous congestion due to tricuspid valve incompetence or cardiomyopathy may yield falsely low SaO₂ readings due to the generation of venous pulsations. This results from the instrument detecting less oxygenated, pulsatile venous blood as part of the arterial sample, thereby underestimating the actual SaO₂ [9,99,100]. (See 'Calibration and calculation' above.)

Venous pulsations may occur when an adhesive probe is too tight around the finger resulting in falsely low readings. Similarly, pulsations can also occur when the probe is in a dependent position (eg, forehead probe in a patient in the Trendelenburg position) and rarely in patients with arteriovenous shunting.

When venous pulsations are suspected, loosening the probe, repositioning the probe or the patient, and/or drawing an arterial blood gas should be attempted.

Nail polish — The use of nail polish can potentially affect pulse oximeter readings if the polish absorbs light at 660 nm and/or 940 nm [4,101-104]. A small study of volunteers wearing black, green, and blue nail polish revealed a drop in SaO₂ of 3, 5, and 6 percent, respectively [17]. Red nail polish does not appear to have an effect on pulse oximetry readings. Newer devices appear to be less affected with the greatest reductions in SpO₂ found in those with black or brown polish not exceeding 2 percent [105].

The problem may be avoided by mounting the probe on the finger sideways, rather than in a dorsal-ventral orientation (eg, single measurements in an outpatient facility) or removing the nail polish (eg, patients in a critical care unit) [9]. An alternate site can also be used (eg, earlobe, forehead, or toe provided similar nail polish is not on the toe).

Artificial acrylic nails may also affect the accuracy of pulse oximetry readings, depending on the device used. This problem can be solved by mounting the probe sideways, using an alternative site, or removing one of the acrylic nails by soaking in acetone [106].

Vital dyes — Vital dyes, such as methylene blue (used to treat methemoglobinemia, or during endoscopic polypectomy), indocyanine green (used for measuring cardiac output, for ophthalmic angiography, or for measuring liver blood flow), fluorescein (ophthalmic angiography) and isosulfan blue (used intraoperatively to mark breast and melanoma tumors), can cause erroneously low pulse oximetry readings due to absorption of light at 660 nm or 940 nm [17,107-112]. Methylene blue has the greatest impact as it absorbs significantly at 670 nm. However, these effects tend to be transient and resolve rapidly as the dyes are diluted and metabolized [17,18].

Others — All the causes of an inadequate signal such as motion artifact, poor probe positioning, hypoperfusion, and hypothermia may result in low readings. (See 'Inadequate waveform' above.)

Fetal hemoglobin gives pulse oximetry readings clinically indistinguishable from those of adult hemoglobin such that pulse oximetry is as reliable in newborns as in adult SpO₂ [18].

CO-OXIMETRY — Abnormal hemoglobins or hemoglobin variants may interfere with pulse oximetry if their absorption properties are similar to those of oxyhemoglobin or deoxyhemoglobin. Conventional pulse oximeters which use two light emitting diodes can only detect these two substances and cannot detect abnormal forms of hemoglobins. Multi-wavelength co-oximeters use several, rather than two, wavelengths of light (eg, four to eight) to detect oxyhemoglobin, deoxyhemoglobin, carboxyhemoglobin, and methemoglobin. They require a sample of arterial whole blood and a specific laboratory request for its measurement [25,35]. (See 'Carboxyhemoglobin' above and 'Glycohemoglobin A1c' above and 'Methemoglobin' above and 'Sulfhemoglobin' above and 'Sickle hemoglobin' above and 'Inherited forms of abnormal hemoglobin' above.)

SUMMARY AND RECOMMENDATIONS

●Pulse oximetry measures peripheral arterial oxygen saturation (SpO₂) as a surrogate marker for tissue arterial oxygenation (SaO₂). Pulse oximetry uses spectrophotometry to determine the proportion of hemoglobin that is saturated with oxygen (ie, oxyhemoglobin) in peripheral arterial blood. A variety of probes are available, none of which have clear superiority over another. (See 'Principles and equipment' above.)

●Pulse oximetry is a rapid, noninvasive tool that can provide continuous assessment of oxygenation and is associated with few complications. However, it cannot detect hyperoxemia or arterial oxygen or carbon dioxide tension. (See 'Advantages and disadvantages' above and 'Complications' above.)

●Pulse oximetry is indicated in any setting where hypoxemia may occur. It provides accurate assessment of tissue oxygenation in most patients. However, clinicians should pay attention to trends on oximetry, and when treating patients with supplemental oxygen for hypoxemia, clinicians should target levels that are desirable for the specific etiology, while simultaneously avoiding oxygen toxicity. The clinician should be aware of the limitations and errors associated with pulse oximetry and have a low threshold to obtain arterial blood for analysis. (See 'Applications' above and 'Interpreting the results' above.)  

●Pulse oximetry is subject to artifactual and patient-related sources of error. The best defense against error is a high index of suspicion. If a saturation reading is in doubt, equipment error can be quickly ruled out by putting the probe on the healthcare worker’s own finger. Once assured that equipment is functioning, the clinician should investigate for other potential sources of error, most of which are readily identified and resolvable (table 1). (See 'Troubleshooting sources of error' above.)