Pulsoximetria în practica pediatrică

Introduction

Pulse oximetry is a noninvasive and painless method of measuring the level of oxygen saturation in the blood. It can detect even the smallest changes in the efficiency of transporting oxygen from the heart to the extremities, such as the hands and feet.

The principle of operation of the pulse oximeter is based on spectrophotometry and on Beer’s law, measuring changes in light absorption by two forms of hemoglobin: oxygenated and reduced. The device emits light that passes through the probe and reaches the light detector. If a finger is placed between the light source and the detector, the light will have to pass through the finger to reach the detector. Some of the light will be absorbed by the finger, and the part that is not absorbed reaches the light detector.

Pulse oximetry – a cornerstone in modern pediatric healthcare – swiftly and noninvasively offers clinicians invaluable insights into a patient’s oxygenation status which is considered the “fifth vital sign”^(1,2).

In the pediatric setting, where oxygen saturation (SpO₂) levels hold profound implications for respiratory function and overall well-being, pulse oximetry is a vital tool in the daily practice of healthcare providers. It plays a crucial role in assessing respiratory function and oxygenation status in children of all ages, from neonates to adolescents⁽³⁾.

This technology has revolutionized the management of pediatric patients by providing real-time data on oxygen levels, enabling timely intervention and improved patient outcomes. It is used in many pediatric conditions, such as monitoring respiratory status, assessment of oxygenation during procedures, screening for congenital heart defects, monitoring sleep-disordered breathing, and home monitoring for conditions that need more than usual care⁽⁴⁾.

Pulse oximetry operates on the spectrophotometry principle, which measures light absorption by oxygena­ted and deoxygenated hemoglobin molecules in the blood. The pulse oximeter emits two wavelengths of light, typically red and infrared, through a translucent part of the patient’s body, such as a finger, toe or earlobe⁽⁵⁾.

Oxygenated hemoglobin absorbs more infrared light, while deoxygenated hemoglobin absorbs more red light. By analyzing the ratio of absorbed light at these wavelengths, the pulse oximeter calculates the SpO₂ level, expressed as a percentage (%SpO₂)⁽⁶⁾.

Oximetry is based on the principle that oxyhemoglobin (O₂Hb) and deoxyhemoglobin (deoxyHb) have different light absorption spectra. According to Beer-Lambert’s law, the concentration of a solute in a solution can be measured by the intensity of light absorption through the solution. The amount of light absorbed depends, among other things, on the solute and the wavelength of the light used.

Pulse oximeters use two light-emitting diodes (LEDs) in their sensors, one emitting red light (660 nm wavelength) and the other infrared (wavelength 940 nm), and a photoreceptor (Figure 1). The photoreceptor measures the amount of light of each wavelength absorbed through the tissues. To determine the proportion of absorption that is due to pulsating arterial blood, the pulse oximeter takes hundreds of measurements per second and determines the times of maximum and minimum absorption for each wavelength. The difference of these two moments corresponds to pulsatile arterial blood. Each type of hemoglobin has its own absorption for red light and infrared light⁽⁸⁾ (Figure 2). The pulse oximeter determines the absorption of the two absorbed wavelengths, then establishes a ratio between the two:

Deoxyhemoglobin is characterized by greater absorption of red light (wavelength range 600-750 nm) compared to oxyhemoglobin, whereas oxyhemoglobin exhibits higher absorption in the infrared spectrum (850-1000 nm)⁽⁹⁾.

As light passes through human tissues, it is absorbed to varying degrees by bones, blood vessels, fluids, skin, venous blood and arterial blood, including various types of hemoglobin. The absorption of light changes depending on the amount of blood in the tissue layers and the relative amounts of oxygenated and deoxygenated hemoglobin change. The measurement of changes in light absorption allows for the estimation of heart rate and arterial oxygen saturation. To measure accurately, the oximeter must distinguish between background (or constant) absorption and pulsatile absorption, which is any change in absorption caused by the change in blood volume with each heartbeat. Background absorption can change when there is a change in the shape or position of the tissues through which the light passes, which can cause false readings⁽¹⁰⁾.

A conventional pulse oximeter measures the ratio of the absorption of two wavelengths of light, separates the changes it assumes to be the result of pulsatile activity and oxygenation, averages the readings over a short period of time, and then looks at the result of the absorption ratio in a table or calibration curve of the corresponding arterial saturations.

The relationship between pO₂ and SaO₂ can be represented by the oxygen dissociation curve, which represents oxygen saturation (y-axis) as a function of the partial pressure of oxygen (x-axis). The sigmoid or S-shape of the curve is due to the positive coope­rativity of hemoglobin. In the pulmonary capillaries, the partial pressure of oxygen is high, allowing more molecules of oxygen to bind hemoglobin, until reaching the maximum concentration. At this point, little additional binding occurs, and the curve flattens out, representing hemoglobin saturation. At the systemic capillaries, pO₂ is lower and can result in large amounts of oxygen released by hemoglobin for metabolically active cells, which is represented by a steeper slope of the dissociation curve.

The strength by which oxygen binds to hemoglobin is affected by several factors, and can be represented as a shift to the left or right in the oxygen dissociation curve. A rightward shift of the curve indicates that hemoglobin has a decreased affinity for oxygen, thus oxygen actively unloads. A shift to the left indicates increased hemoglobin affinity for oxygen and an increased reluctance to release oxygen. Several physiologic factors are responsible for shifting the curve left or right, such as pH, carbon dioxide (CO₂), temperature, and 2,3-disphosphoglycerate.

Perfusion index

Perfusion index (PI) determines the pulse strength at the sensor site. The normal PI ranges from 0.02% to 20%. If the perfusion index is at or below 0.4%, showing weak pulse strength, then the oximeter reading can be unreliable.

The PI is defined as the ratio of the pulsed (I_p) to the non-pulsed (I_np) component of the infrared reaching the photoreceptors of a pulse oximeter. It is a figure obtained using an algorithm corresponding to the formula: PI=(I_p/I_np)x100. Exponential corrections may be the result of local circulation phenomena. From this perspective, an increase in PI generally reflects vasodilation, whereas a decrease in PI generally reflects vasoconstriction or even a local disorder such as hypothermia, vasospasm, or microembolism at the measurement site. However, more global circulatory parameters such as blood volume, stroke volume and venous return also have an impact. Thus, PI is a complex indicator that reflects systolic-diastolic changes affecting the macro- and microcirculation⁽¹¹⁾.

Plethysmographic curve

Plethysmographic curve, developed from the pulsatile variations in the absorption of infrared light by oxyhemoglobin, makes it possible to attest to the arterial origin of the signal. It represents an immediate “quality control” regarding the interpretability of the displayed SpO₂. The presence of a regular sinusoid with a dicrotic wave cannot in fact be reproduced by pseudopulsatile waves linked to ambient light, such as, for example, scialytic or phototherapy lamps. The variations in light absorption that generate the curve are essentially systolic and proportional to the volume variations. They can be interpreted, under certain conditions, as a hemodynamic parameter.

The history and evolution of SpO₂ monitoring

During World War II, a young physiologist named Glenn Allen Millikan developed a lightweight optical device that, when placed on the earlobe, could provide a noninvasive, continuous estimate of SaO₂. He coined the term “oximeter” for this monitoring device which was to be used in aeronautical research to assess pilots’ oxygenation during high-altitude flying⁽¹²⁾.

The Millikan ear oximeter estimated saturation by illuminating the ear with two frequencies of light, thus using the earlobe as a cuvette containing the blood sample. Unfortunately, there are many absorbers in the earlobe other than HbO₂ and Hb (i.e., skin, cartilage and other tissues). The Millikan oximeter solved this problem by first calibrating the device on a bloodless ear. This was accomplished by inflating a cuff around the earlobe, thus compressing the earlobe and removing blood. This bloodless compressed ear was used as the zero reference point for the oximeter. Another problem in obtaining an estimate of arterial hemoglobin saturation was that the earlobe contained not only arterial but also capillary and venous blood. To obtain data that were primarily related to arterial blood, the ear oximeter sensor was heated to 44°C, producing a hyperemic earlobe composed of arterialized blood. After the war, this device was used clinically, and it was found to be able to accurately detect previously unrecognized cyanosis in patients⁽¹³⁾.

In 1947, Julius Comroe used an ear oximeter to conduct a classic study that demonstrated the unreliability of clinical signs in detecting cyanosis. In this study, volunteers breathed various fractions of inspired oxygen to produce varying degrees of hemoglobin saturation, while medical students and physicians attempted to clinically detect cyanosis. After 3673 observations, it was concluded that these medical personnel were unable to consistently detect cyanosis until it was severe (SaO₂ ~80%)⁽¹³⁾.

These early studies foreshadowed the widespread use of pulse oximetry in clinical medicine today. They illustrated that oxygen saturation could be measured easily and noninvasively, and that without these objective data, severe degrees of desaturation could go unnoticed by trained medical personnel. Why did it take nearly 40 years to move from these early studies to the widespread use of the technique today? The problem was one of practicality.

The first commercial device, although functional, was difficult to be used in clinical practice. The calibration technique was time-consuming, positioning the probe on the ear was difficult, and the heated sensor could actually cause burns to the earlobe⁽¹³⁾.

In the mid-1970s, Hewlett-Packard Corporation developed a multi-wavelength ear oximeter that required no in vivo calibration process, but still required earlobe heating. It used an eight-wavelength oximeter to estimate saturation, resulting in eight equations that allowed it to empirically self-calibrate. This device was also somewhat bulky and cost nearly $10,000 in 1975. It had improved accuracy and became the standard for noninvasive oximetry by the late 1970s.

In the mid-1970s, an engineer working for Nihon Kohden Corporation, named Takuo Aoyagi, revolutionized the field of noninvasive oximetry by developing what would become the pulse oximeter. This engineer was actually working on a project to estimate cardiac output noninvasively by injecting a dye into a peripheral vein. A dye dilution curve was obtained by measuring the absorbance as this dye circulated through the ear and was detected by an oximeter-like device. He had difficulty obtaining a smooth curve because the dye perfused the ear and observed small pulsations in the absorbing signals in the red and infrared light wavelengths. He noted that the amplitude of these pulsations changed if the subject desaturated and immediately realized that he had developed a method for estimating arterial hemoglobin saturation by analyzing the pulsatile component of the absorbance in red and infrared light. By using this pulsatile signal, he did not only overpass the need to calibrate the device on a bloodless ear, but he also obtained a signal that was related only to arterial blood.

The basic assumption of a pulse oximeter is that anything pulsating and absorbing red and infrared light between the light source and the light detector must be arterial blood. Oddly enough, the first device developed by Nihon Kohden once again had a bulky ear sensor and suffered from insufficient accuracy. At this point, an engineer in Boulder, Colorado, USA, named Scott Wilber, made two substantial improvements to the pulse oximeter that allowed it to develop into a clinically useful monitor. He used solid-state LEDs and photodetectors to produce a small, lightweight probe. Secondly, he incorporated a microprocessor into the monitor itself to allow more complex processing of the pulsatile data, which improved the accuracy of the devices.

New generation pulse oximeters are manufactured with improved algorithms that minimize erroneous motion-related data by filtering out body movements. Furthermore, because they use multiple wavelengths, these new pulse oximeters are capable of measuring the concentration of hemoglobin, carboxyhemoglobin and methemoglobin⁽¹⁴⁾.

The importance of monitoring peripheral oxygen saturation in children

The human eye has a poor capacity to detect hypoxia. Classically, in fact, its diagnosis is based on the appearance of cyanosis which only appears when the reduced hemoglobin (Hb) level reaches 50 g/L in the capillary blood, this level corresponding to an oxygen saturation (SaO₂) of approximately 75% under normal perfusion conditions⁽¹⁵⁾.

Simple clinical signs, including respiratory rate, nasal retractions or flaring, grunting, cyanosis, pallor and general appearance, are used to assess the cardiorespiratory status of infants and children. Previous researchers have found that, although these clinical signs are frequently present, their absence does not reliably exclude the possibility of serious cardiopulmonary disease or lower respiratory tract infections^(16,17).

The physician currently has no other simple method to detect hypoxemia in a patient, since blood gas measurement is not routinely used. Oximetry must therefore be integrated with other parameters estimating the severity of respiratory failure, and may be used in many situations. The assessment of the severity of dyspnea secondary to pneumonia, pulmonary embolism, pneumothorax or an asthma attack are the most common uses of oximetry. In this context, care must be taken when saturation is lower than 95%, because below this value, small variations in SpO₂ can mean significant changes in PaO₂. A saturation below 92% is an absolute severity criterion, justifying hospitalization.

Pulse oximetry has been used in pediatrics for almost 30 years to monitor respiratory function and adjust oxygen therapy. Technological advances in plethysmographic signal processing suggest a potential interest for hemodynamic assessment. The peripheral perfusion index (PI), which evaluates the pulsatile component of the signal, is proposed as a continuous and noninvasive indicator of distal circulation. Its clinical use has mainly been developed in neonatology, as a diagnostic aid or complementary to macrocirculatory exploration. In the perspective of heart disease screening, a threshold value below 0.70 is suggestive of critical obstruction on the left ventricular outflow tract. In premature newborns, several observations highlight its relevance for the diagnosis of systemic low flow or patent ductus arteriosus. The indices derived from respiratory variability of oximeter plethysmography have mainly been the subject of study in children in the operating room, with the aim of assessing their blood volume and predicting their response to vascular filling. However, the results are contradictory and do not currently allow, in this context, to rationalize the administration of fluids reliably by the use of these parameters. Further studies are necessary, particularly in pediatric intensive care, to demonstrate that pulse oximetry can be a hemodynamic tool with a validity comparable to that established in adult intensive care medicine⁽¹⁸⁾.

Pulse oximetry has lots of uses in the pediatric medical care, being an everyday assessment tool. Some of the clinical applications of pulse oximetry are found in Table 1.

Pulse oximetry during resuscitation provides real-time clinical information about heart rate and oxygen saturation and facilitates important decision-making related to interventions such as positive pressure ventilation, cardiac compressions and oxygen titration. Prior to pulse oximetry, the evaluation of color was the only way to assess oxygenation in newborns during delivery room stabilization, but this, too, was subjective and inaccurate. This was further compounded by the routine and often needless use of 100% oxygen during resuscitation. The dangers of hyperoxemia during resuscitation, even for a short period, have been well documented, and more than 3 minutes of exposure to 100% oxygen is associated with long-term effects of oxygen toxicity, including childhood cancers. This is thought to be related to DNA damage from reactive oxygen species⁽¹⁹⁾.

Limitations of pulse oximetry 

The limitations of pulse oximetry can generally be classified as safe or potentially dangerous. Safe limits refer to circumstances in which inaccuracy of the displayed SpO₂ can be suspected and its cause is recognizable. In this case, the observer is usually warned by the device (alarm) of the pitfall. A potentially dangerous limitation is considered to be any situation in which the inaccuracy is difficult to recognize; the displayed SpO₂ is erroneous, but the observer is not warned of the pitfall⁽⁹⁾.

Motion artifact represents the most common limitation of pulse oximetry⁽²⁰⁾. Since the normally pulsatile (arterial) component of light absorption accounts for no more than 5% of the total absorbed energy, any movement that alters the remaining fraction of absorption (especially when due to venous blood) will affect the signal-to-noise ratio and drive SpO₂ to lower than true values. Fortunately, motion artifacts can be recognized by motion alarms or distorted plethysmographic waveforms. However, rhythmic movements or vibrations with a frequency similar to heart rate (0.5 to 3.5 Hz) can be particularly bothersome⁽²¹⁾.

Adequate arterial pulsation at the measurement site is essential to distinguish the true signal from background noise. Low perfusion states, such as low cardiac output, shock, hypothermia, vasoconstriction, arterial occlusion or during pressure cuff inflation, may impair device function and result in lower SpO₂ readings or delayed recognition of acute hypoxemia^(22,23).

For infants with cold extremities, local rubbing or heating before device application may temporarily improve perfusion; however, for hypothermic patients, forehead probe monitoring is an alternative option. Newer-generation devices are equipped with signal extraction algorithms and may perform better in low-perfusion states. 

In theory, skin pigmentation has a constant level of absorption subtracted in the SpO₂ calculation and, therefore, should not influence the performance of the device. However, dark skin pigmentation has been implicated for altered SpO₂, particularly at SaO₂ values of 80%⁽²⁴⁾.

Pulse oximetry can detect irregular heart rhythms by evaluating the plethysmographic waveform. Currently available devices have signal extraction technologies capable of recognizing such events.

Electromagnetic energy from electrosurgical cautery devices and mobile phones can interfere with pulse oximeters and lead to erroneous SpO₂ readings, due to excessive heating of the sensor, with consequent thermal damage.

The displayed SpO₂ is the result of converting the absorption ratio to percent saturation, using specific calibration algorithms. These algorithms are derived by correlating the absorption ratio with arterial gas SaO₂ measurements in healthy young volunteers over a range of desaturation values. Because it is unethical to desaturate volunteers below 80% SaO₂ levels, lower SpO₂ values are derived by extrapolation and, therefore, are less accurate. In addition, because the subjects recruited for calibration are healthy young adults, the applicability of the calibration data to patients of extremes of age has been questioned.

Most conventional pulse oximeters exhibit a clinically significant delay between a sudden change in blood oxygenation and the corresponding change in displayed SpO₂ values. This delay depends on the complexity of the algorithms used, and can exceed 15 to 20 seconds. Although newer-generation devices have improved response times and the desaturation events may be detected earlier if the probe is placed more centrally (e.g., at the earlobe), pulse oximetry should not be used as a substitute for cardiorespiratory monitoring in critically ill patients.

Lower SpO₂ readings can occur when the probe is poorly positioned, particularly on the little fingers of neonates and infants. In this case, the emitted light may be projected tangentially to the detector, sometimes without crossing an arterial bed, phenomena that have been described respectively as “penumbra” and “optical shunt” effects. This pitfall can be avoided by positioning the emitter and detector exactly opposite each other and by using appropriate optics for neonates and infants.

Intense white or infrared light can interfere with pulse oximetry and cause falsely low SpO₂ readings. This phenomenon, known as flooding, is caused by the excessive increase in light energy literally flooding the photodetector and pushing the absorption rate ratio toward unity; this corresponds to an SpO₂ of 85%. Although newer generation devices can detect light interference, healthcare professionals, especially those handling newborns exposed to phototherapy, should be aware of this potential limitation. Ambient light interference can be avoided by simply covering the sensor with a non-transparent material.

Carboxyhemoglobinemia represents the most dangerous limitation of pulse oximetry, because in the presence of COHb, the method estimates arterial oxygenation. This effect is caused by the specific characteristics of COHb, which has a similar absorption of red light as oxyhemoglobin. Therefore, SpO₂ values should be verified by SaO₂ measurements using a co-oximetry method when the presence of COHb is suspected (e.g., carbon monoxide poisoning).

Methemoglobin (MetHb) absorbs approximately equal amounts of energy in the red and infrared spectra. In cases of significant methemoglobinemia (MetHb = 30%), the absorption rate ratio will tend towards unity (SpO₂ = 85%), thus underestimating high saturation values and overestimating severe hypoxemia. If the difference between SaO₂ and SpO₂ (the “SaO₂–SpO₂ gap”) exceeds 5%, the presence of abnormal hemoglobin molecules should be investigated by CO-oximetry. Pulse CO-oximeters, taking advantage of new technologies, have been shown to accurately measure both COHb and MetHb.

Fetal hemoglobin and hemoglobin S have similar light absorption characteristics to adult hemoglobin, and do not interfere with pulse oximetry. However, physicians should be aware that abnormal hemoglobin molecules affect oxygen dissociation curve, so the displayed SpO₂ value may not reliably reflect tissue oxygenation, especially for children with sickle cell disease.

Anemia does not appear to affect the accuracy of pulse oximetry, at least for hemoglobin levels of 5 g/dL and if cardiovascular function is preserved. Similarly, polycythemia does not appear to interfere with pulse oximetry.

In cases of significant tricuspid regurgitation and in states of hyperdynamic circulation, the pulsatile variation of venous blood can affect the signal-to-noise ratio and lead to erroneous SpO₂ readings.

Intravenous dyes like indocyanine green and indigo carmine can cause lower SpO₂ readings.

Conclusions

Peripheral oxygen saturation monitoring is an easy, noninvasive and rapid way to inform the physician about the patient’s cardiorespiratory status. It indicates the severity of respiratory impairment in many pathological conditions in children. Despite its limitations, pulse oximetry remains an indispensable tool for monitoring children in diverse clinical settings.

Autor corespondent: Bogdan A. Stana E-mail: bogdan.stana@gmail.com

CONFLICT OF INTEREST: none declared.

FINANCIAL SUPPORT: none declared.

This work is permanently accessible online free of charge and published under the CC-BY.