Pulse oximetry can be performed with disposable or reusable sensors, and hospitals and birthing centers must select the type of sensor that best meets the needs of their facility.1 The SACHDNC recommendations state that either disposable or reusable pulse-oximetry sensors are appropriate for CCHD screening.8 Reusable sensors have the advantage of reducing the overall cost of screening for CCHD. However, the reusable sensors must be cleaned between uses to minimize the risk of infection.
The Maryland State Advisory Council on Hereditary and Congenital Disorders estimates that the cost of reusable sensors can be amortized to approximately $1 per use compared with approximately $12 per disposable sensor.9 Partially reusable sensors, which require less cleaning than standard reusable sensors and cost less than fully disposable sensors, are another option.
For newborn patients, sensors with close coupling to the skin, such as those that are taped rather than clamped, show better performance for CCHD screening.1 Clinicians should keep in mind that sensor type can influence the performance of pulse oximetry, and therefore only recommended combinations of devices and sensors should be used.1
|Case Study: Maryland|
Based on recommendations from the Maryland State Advisory Council on Hereditary and Congenital Disorders, in 2011 the Maryland legislature signed into law a bill to facilitate the implementation of CCHD screening for all newborns in the state.9 Each year, Maryland has approximately 74,000 live births. By implementing a CCHD screening program, the state anticipates identifying 10 newborns with CCHD each year who would have otherwise been undetected by current prenatal and newborn examination protocols.
A key feature of the Maryland CCHD screening legislation is a quality-assurance initiative that includes ongoing surveillance and program evaluation. The Maryland Department of Health and Mental Hygiene will employ a half-time nurse to develop a dedicated surveillance system for the CCHD screening program. The surveillance initiative will track screening rates—including total screens, false-positive rates, and false-negative rates—as well as follow-up data on infants with positive pulse-oximetry screens.
Once launched, the ongoing surveillance component will consist of monitoring the electronic database to identify hospitals with unusual rates of failed or missed screens and providing technical assistance to those hospitals.
Newborn fingers and toes are often too small for proper attachment of pulse-oximeter sensors, requiring sensor placement on the sole of the foot or palm of the hand. (It should be noted, however, that the finger or toe may be suitable for probe placement, if in the judgment of the clinician the child is large enough.)
According to the current SACHDNC recommendations for CCHD screening, oxygen saturation readings should be obtained in the right hand and one foot.8 Moreover, the sensors should be used according to the specifications outlined by the manufacturer, with the palm, foot, great toe, or thumb serving as the optimal sites for placement.
In addition to the ideal placement of sensors in newborns, it is essential that the operator obtain pre-ductal and post-ductal sensor measurements to ensure an accurate oxygen-saturation reading. For instance, researchers have placed a sensor on one foot as well as the right hand to determine any difference between readings, with a difference of >3% considered abnormal.
The strategy of using two sensor readings may improve detection of left outflow tract obstructions, such as coarctation of the aorta or interrupted aortic arch, which may result in an arm saturation of 99% and a leg saturation of 95%.2 One large study identified two patients whose diagnosis of CCHD would have been missed if only the foot SpO2 reading had been recorded.2
Ongoing research will continue to evaluate the optimal approach to CCHD screening. Toward this goal, several recent studies explored different options for sensor placement.13,14 In one study, researchers evaluated the accuracy and precision of pulse oximetry at five different sensor locations (the finger, palm, toe, sole, and ear) in infants and children with known cyanotic heart disease.13
In this study, the sole was the most accurate sensor location. Regardless of sensor placement, however, the accuracy of pulse oximetry deteriorated at low saturation states (<90%) and tended to overestimate oxygen saturation.
Another study showed that pulse-oximetry sensors were less accurate when used on the sole of the foot or palm of the hand when oxygen saturation fell below 90% in newborns and children up to 2 years of age.15 Clinicians should keep in mind that this limitation can be mitigated if the sensors are used as instructed by the manufacturer.
Additional practice pearls
These guidelines offer additional practice pearls for optimal pulse-oximetry use, including the following:
- Adult pulse-oximetry clips should not be used when obtaining an SpO2 reading for an infant. An adult pulse-oximetry clip in a newborn will yield an inaccurate reading.10
- Blood flow is necessary to obtain an accurate SpO2 reading. Therefore, do not attempt to obtain a reading on the same extremity as an automatic blood-pressure cuff, which impedes normal blood flow.10
- Bilirubin lamps, surgical lights, and other bright or infrared lights can affect the accuracy of the pulse-oximetry reading. Therefore, avoid placing the infant in bright or infrared light while pulse oximetry is being performed.10
- Substances with dark pigmentation, such as dried blood and nail-polish dyes, can alter pulse-oximetry readings. Therefore, confirm that the skin is clean and dry before placing the sensors. Skin color and jaundice, however, will not affect the SpO2 reading.10
When using disposable pulse-oximetry sensors, use a new, clean sensor for every newborn. Likewise, when using reusable sensors, be sure to clean the sensor with the recommended disinfectant between each infant. In addition to increasing the risk of transmitting infection, dirty sensors can decrease the accuracy of the pulse-oximetry reading. In each case, the sensor should be secured to the sensor site with a disposable wrap.10
Understanding screening performance
Pulse oximetry is not 100% accurate, however, and some readings will result in false-positive or false-negative screens.1 Understanding the limitations of pulse-oximetry screening is important for interpreting screening results.
A small percentage of infants who have a positive pulse-oximetry screen will not have CCHD, and some may have other conditions that require intervention. In a recent meta-analysis of data on pulse-oximetry screening in 229,421 asymptomatic newborns, pulse oximetry was highly specific (99.9%) and moderately sensitive (76.5%) for the detection of CCHD.16
The false-positive rate was 0.14%, but fell to 0.05% when pulse oximetry was performed after 24 hours from birth, without any compromise in overall sensitivity. Although the goal of pulse oximetry was to detect CCHD, other life-threatening disorders of noncardiac origin were also identified via screening, including group B streptococcal pneumonia and pulmonary hypertension.
Current CCHD screening algorithms include steps to decrease the rate of false-positive readings. For instance, screening after 24 hours of life is a key step toward enhancing the performance of CCHD screening.1 Some evidence suggests that screening infants who are alert (rather than sleeping) will decrease the risk of false-positive results by reducing the likelihood of low oxygen saturations caused by hypoventilation in deep sleep.1 In addition, performing pulse oximetry around the time of the newborn hearing screening may improve the efficiency of CCHD screening.1
The SACHDNC screening protocol recommends taking a second SpO2 measurement — and if needed, a third reading — for infants with an initial SpO2 reading <95% in both extremities, or if there is >3% absolute difference in oxygenation between the right hand and foot.1 These second and third measurements are included to reduce the risk of a false-positive screen by confirming the presence of low oxygenation. Infants with oxygen saturation <90% in any screen do not need repeat pulse oximetry, however. These infants should be referred immediately for further evaluation.
Follow-up echocardiography and pediatric cardiology consultations are a major source of secondary costs related to CCHD screening.2 Therefore, there is a major cost incentive for hospitals and birthing centers to reduce the false-positive rates of pulse oximetry by implementing standardized protocols for CCHD screening.
Although current protocols include steps to enhance the efficiency of screening, they do not remove the possibility of a false-positive screen entirely. By comparison, though, the false-positive rate with a standard physical examination alone for CCHD is nearly 10 times higher than with the addition of pulse oximetry.17,18 Additional technological advances should improve the performance of screening tests in the future.
Pulse oximetry does not detect all types of CCHD, so it is possible for a newborn with a negative screening result to have CCHD or other congenital heart defects. Data from large studies provide some insight into the patterns of false-negative screens. In the large Swedish study of nearly 40,000 newborn screens, 5 out of 60 ductal-dependent lesions were not detected by pulse oximetry.17
In each case of a false-negative screen, the newborn had coarctation of the aorta, a type of CCHD that can present late and, therefore, be missed during the recommended screening window.4,17 A Norwegian study of 50,008 newborn screenings involved 49,684 infants with SpO2 readings ≥95%, including eight with false-negative screens.19 Of these, four were diagnosed with CCHD before discharge on the basis of clinical findings. Therefore, the overall false-negative rate was 4 of 49,684 screens, or 0.008%.
Missing a diagnosis of CCHD has significant clinical and financial costs. In the Swedish trial, newborns who were discharged with undiagnosed CCHD had a higher proportion of severe acidosis than those who were diagnosed in hospital, resulting in a longer stay in the intensive-care unit and a higher mortality rate during corrective surgery.17