Executive Summary Checklist

Congenital Heart Disease (CHD) is one of the most common types of birth defects. Critical Congenital Heart Disease (CCHD), including ductal-dependent lesions, represents 40% of death caused by CHD. CCHD is life-threatening and is typically identified in the first year of life. Early intervention in CCHD is imperative and remains an important clinical challenge. Historically, due to the lack of physical signs and difficulties in screening mild cyanosis in newborns, a third of babies were discharged unchecked. A fetal ultrasound can identify increased structural abnormalities and proportions; however, this detailed ultrasound is operator-dependent and potentially inconsistent. Pulse oximetry screening is a universally accepted test that increases overall detection of CCHD to over 90% and identifies babies with non-cardiac, hypoxemic conditions such as congenital pneumonia, early-onset sepsis, and pulmonary hypertension as well.
To address the failure to detect CCHD in newborns, we should implement the following actionable steps:
  • SpO2 <90% from any site; or
  • SpO2 <95% from the right hand or either foot
  • If initial SpO2 measurement is <95%, proceed with up to two additional SpO2 measurements.
  • If the second and third SpO2 measurements read >95% the screening is negative.
  • If the second and third SpO2 measurements are <95% the screening is positive.
  • >3% difference in SpO2 measurements between the right hand and either foot (repeat three times as described in the bullet above)

The Performance Gap

Congenital heart disease (CHD) is the most common birth defect, affecting approximately 8 in 1,000 live-born infants \cite{Reller_2008,Bernier_2010}. Nearly 40,000 infants are born with CHD per year in the US, and 1.35 million globally \cite{Hoffman_2002,22078432}. Critical congenital heart disease (CCHD), including ductal dependent lesions, affects between one-quarter and one-third of these infants \cite{Oster_2013,26086632,25963011}. CCHD represents about 40% of the deaths from congenital anomalies and the majority of the deaths due to CHD that occur in the first year of life.(Hoffman 2002) In 2012, before newborn screening programs were introduced in the United States, it was estimated that between 70-100 infants died each year from late-diagnosed CCHD \cite{Govindaswami_2012}. It is now believed that the number of deaths is closer to 120 per year \cite{28837548}. [MG these two sentences don't work well together; after screening, are we seeing increased or decreased death?]
Antenatal ultrasound and physician examination after birth improve detection and perinatal outcomes for certain forms of CCHD \cite{Tworetzky_2001,Bonnet_1999}. Evidence showed that prenatal detection increased every year (2006-2012); prenatal detection now occurs in 34% of patients \cite{Quartermain_2015}. The benefit of a CCHD diagnosis before birth allows for counseling and coordination of delivery at an experienced cardiac center.
The gap in patient safety is that more than 30 percent of CCHD deaths have been attributed to late or missed diagnosis \cite{Chang_2008}. It is estimated that 2,000 infants/year die or are undiagnosed in the US and some 300,000 infants/year die globally \cite{Salvi_2016}. The burden of undiagnosed cases in the developing world is significant, with fewer than half of CHD cases diagnosed in the first week of life \cite{Hoffman_2013}. The magnitude of the problem has been extensively documented \cite{Singh_2014,de_Wahl_Granelli_2014,Ewer_2014,Ewer_2014a,Ewer_2013,Ewer_2013a,GRANELLI_2007}.
Pulse oximetry noninvasively measures oxygen saturation (SpO2) and pulse rate.  In 2009, de-Wahl Granelli et al published a breakthrough cohort study in which 39,821 infants were screened for CCHD by identifying abnormal SpO2 measurements from Signal Extraction Technology (SET) pulse oximetry. SET's ability to measure through motion and low-perfusion is essential for accurate CCHD screening \cite{de_Wahl_Granelli_2009}. In a separate CCHD screening study of 20,055 asymptomatic newborns, Ewer et al, confirmed the importance of utilizing SET technology that can “produce accurate saturations that are stable in active neonates and in low perfusion states, making them suitable for use in the first few hours of a newborn baby’s life" \cite{22284744}. In 2014, Zhao et al reported similarly positive results from a prospective study using SET in more than 100,000 newborns in China \cite{24768155}.
The addition of pulse oximetry screening to antenatal ultrasound and physical examination may increase detection rates for CCHD to over 90%. Furthermore, the detection of non-critical CHDs and significant non-cardiac neonatal conditions, such as respiratory problems or early-onset sepsis, is an additional benefit. However, clinicians need to be aware that, although combining pulse oximetry screening with other screening methods will reduce this diagnostic gap, some babies will still be missed. The Journal of Pediatrics has published a study estimating the number of infants with critical congenital heart defects (critical CHDs) potentially detected or missed through universal screening for critical CHDs using pulse oximetry \cite{23266220}. CDC researchers estimated that about 1,755 infants with critical CHDs would be diagnosed late (meaning on or after the third day after birth). Of these, about half (875 infants) with a critical CHD would be detected through newborn screening using pulse oximetry, but an equal number (880 infants) might still be missed each year in the United States.
Most studies report that the lesions most often missed are those causing obstruction to aortic outflow (e.g. coarctation and interrupted arch), which may not necessarily be detected in antenatal ultrasound, physical examination, or by abnormal SpO2 values from pulse oximetry. However, an additional SET pulse oximetry measurement may increase detection of CCHD with obstructions to aortic outflow. This measurement is called perfusion index (PI), which is an assessment of strength of perfusion at the monitored site. In a 2007 study, Granelli showed that adding abnormal PI to pulse oximetry screening may increase sensitivity to identifying CCHD with an obstruction to the aortic outflow. The authors of this study also noted that adding PI to the screening criteria may also result in an increase in false positives. [MG reference?]
In 2011, the federal CCHD workgroup, with members selected by the US Health and Human Services Secretary's Advisory Committee on Heritable Disorders in Newborns and Children, the American Academy of Pediatrics, the American College of Cardiology Foundation, the Newborn Foundation, the March of Dimes, and the American Heart Association, developed a report: Strategies for Implementing Screening for Critical Congenital Heart Disease \cite{21987707}. After a thorough review, the workgroup relied upon a thorough body of evidence and independent published studies to recommend that “screening be performed with motion tolerant pulse oximeters that report functional oxygen saturation, have been validated in low-perfusion conditions, have been cleared by the FDA for use in newborns, and have a 2% root mean-square accuracy.”
Several domestic and international studies have shown parents are predominantly satisfied with pulse oximetry screening and those whose babies had a false positive result were no more anxious than those with true negative tests (Ewer 2012). Parents generally perceived it as an important and valuable test to detect ill babies. Additionally, all staff groups (healthcare assistants, midwives, nurses and doctors) were predominantly positive about the testing procedure and perceived the test as important.
Screening for CCHD not only reduces pain and suffering of infants and families but can also reduce costs associated with severe cardiovascular and other organ or neurological compromise upon delayed admission to a cardiac unit – and has been tied to significantly reduced mortality, fewer poor surgical outcomes, and lower incidence of prolonged ventilation and potential developmental issues \cite{23918890}.
Relative to the developing world, the prevalence of certain heart lesions varies significantly on the global map, as does the burden of hypoxemia-related conditions such as neonatal pneumonia, sepsis, necrotizing enterocolitis (NEC), and PPHN.(Hoffman 2013) Every year nearly 41% of all under-five child deaths are among newborn infants, babies in their first 28 days of life or the neonatal period \cite{world2012newborns}. Three-quarters of all newborn deaths occur in the first week of life, and 1/3 of these newborn deaths are from infection, such as pneumonia, tetanus, and sepsis.30 Each of these conditions are likely to manifest with below normal oxygen saturation. Some are preventable deaths in that when diagnosed in a timely fashion, a course of antibiotics and/or supplemental oxygen therapy can save a life or improve an outcome.

Considerations regarding algorithms for screening

A recent review describes the experience of CCHD screening in the United States in reference to optimizing the algorithm for screening,  educating all stakeholders and performing screening using the proper equipment \cite{27244826}. There are many factors to consider when determining the optimal screening algorithm, including the balance of sensitivity and specificity, resource utilization, cost, high altitude and timing of screening. For this reason, other screening protocols have been evaluated in the United States and in other countries \cite{27940777,27603536}For this reason, other screening protocols have been evaluated in the United States and in other countries. For example, infants at high altitude may have a lower oxygen saturation than those at sea level with potential implications at elevations over 6,800 feet. Therefore, to identify the optimal algorithm in particular settings, it may be necessary to modify the screening protocol described in this document, including the saturation cutoff points and the timing of screening.
A certain degree of controversy still remains, and debate continues regarding the most appropriate time to screen, the most effective screening pathway, what saturations are acceptable, which conditions are we trying to identify and screening outside the well-baby nursery.
When evaluating algorithms, it is important to consider sensitivity, specificity, and false-positive and false-negative rates. It is also vital that screening leads to timely diagnosis (ie, before presentation with acute collapse).
Be all this as it may, if SpO2 is < 90% in either limb the infant needs to be assessed immediately. If SpO2 is between 90-94% in one or both limbs and the infant does not look completely healthy, clinical assessment is mandatory without delays for repeated measurements. If an infant is completely healthy, the measurement should be repeated as described. Finally, there is no need to do an echocardiogram immediately, as many babies with positive screening do not have CCHD. [Mg this last sentence is confusing and should be removed,  a failed algorithm occurs in a two hour period; we don't want babies sent home without an echo because there is no need to do an echo immediately]
In summary, the lack of a systematic approach to prevent failure to rescue in CCHD significantly affects patient safety, quality, and cost of care. Universal newborn screening with pulse oximetry technology has been shown to increase the detection of CCHD by identifying potential abnormalities that are not apparent in prenatal or postnatal examinations.  Closing the performance gap with CCHD will require hospitals, healthcare systems and all members of the neonatal healthcare team (RN’s, RT’s and MD’s) to commit to action in the form of specific leadership, practice, and technology plans for all newborn infants.

Leadership Plan