- Email: [email protected]
Pulse Oximetry Screening for Critical Congenital Heart Disease in Planned Out-of-Hospital Births
Pulse Oximetry Screening for Critical Congenital Heart Disease in Planned Out-of-Hospital Births Jennifer J. Lhost, BS1, Elizabeth M. Goetz, MD, MPH2, Jody D. Belling, RN, MSN2, W. Marijke van Roojen, LM, CPM3, Gretchen Spicer, LM, CPM4, and John S. Hokanson, MD2 Objectives To describe the use of pulse oximetry screening (POS) for critical congenital heart disease (CCHD). Study design This observational study of Wisconsin out-of-hospital births was performed from January to November, 2013. Licensed midwives, Amish birth attendants, and public health nurses were trained in the use of pulse oximetry to detect CCHD, supplied with pulse oximeters, and reported screening results and clinical outcomes. Results Results of POS in 440 newborns were reviewed; 173/440 births were from Amish or Mennonite communities. Prenatal ultrasonography was performed in less than one-half of the pregnancies and in only 13% of Amish and Mennonite women. A total of 432 babies passed the screening, 5 babies were incorrectly assigned to have passed or failed, and 3 babies failed the screening. Two of the babies who failed the screening were treated for sepsis and the third had congenital heart disease. There was 1 false negative result (coarctation of the aorta and ventricular septal defect). Conclusions This study provides information on the use of POS for CCHD in out-of-hospital births and shows that POS can be successfully implemented outside the hospital setting. Although the failure rate in this small sample was higher than reported in studies of hospital births, those babies failing the screening had significant disease processes that were identified more rapidly because of the screening. (J Pediatr 2014;165:485-9).
C
ritical congenital heart disease (CCHD) affects 1-3/1000 babies born in the US every year.1-4 The majority of these infants are diagnosed by prenatal ultrasound or by clinical examination shortly after birth. However, a subset of infants with CCHD will be missed by both prenatal ultrasound and clinical examination and may present with significant cardiac morbidity or mortality in the neonatal period.5,6 Recent efforts to screen infants for CCHD with pulse oximetry screening (POS) have focused on identifying infants with clinically undetectable CCHD prior to discharge from the birth hospitalization.7-10 The US Secretary of Health and Human Services recommended universal newborn POS for CCHD in September 2011. The American Academy of Pediatrics endorsed these recommendations in December 201111 and in 2013 included a recommendation for POS for planned home births.12 All screening recommendations are based on evidence derived from babies born in the hospital setting, and little is currently known about POS in the out-of-hospital (OOH) birth population. According to the Centers for Disease Control and Prevention, the number of OOH births increased from 0.56%-0.72% of US deliveries between 2004 and 2009.13 Wisconsin had the fifth highest rate of OOH delivery in the nation with 1.66% of births occurring at home or in a birthing center.13 Midwifery standards of practice for OOH birth in Wisconsin are defined by the State Department of Safety and Professional Services. Licensed midwives are required by the rules and regulations guiding their practice to remain with the mother and baby after birth until the both mother and baby are stable and for no less than 2 hours. On average, midwives report staying 3-4 hours after birth. They are also required From the School of Medicine and Public Health, to reevaluate maternal and infant well-being within 36 hours of birth. The timing Department of Pediatrics, University of Wisconsin, Madison, WI; Wisconsin Guild of Midwives, Birthwise of this reevaluation is subject to medical indication, clinic schedules, weather, Health and Birth Center; and Wisconsin Guild of distance to the home, and the preferences of the family. Newborn blood Midwives, Local Delivery Midwifery Service, Iola, Wisconsin screening is typically performed at this visit. Hearing screening may also be perFunded by Health Resources and Services Administraformed at the same visit, or deferred to a subsequent appointment. tion (H46MC24057). The authors declare no conflicts of interest. Some families, particularly members of plain clothes communities, have unatPortions of the study were presented as a poster at the tended deliveries or use a birth attendant who is a member of their community. Midwest Pediatric Cardiology Society Scientific Session, Chicago, IL, September 20, 2013, at the Critical The training, experience, and newborn care practices of these nonlicensed birth Congenital Heart Disease Stakeholders Meeting:State 1
2
3
4
CCHD OOH POS SHINE
Critical congenital heart disease Out-of-hospital Pulse oximetry screening Screening Hearts in Newborns
Implementation, Association of Public Health Laboratories, at the Pediatric Congenital Heart Association Meeting, Silver Springs, MD, February 27-28, 2014, and at the Wisconsin Association for Perinatal Care’s Statewide Perinatal Conference, Wisconsin Dells, WI, April 27-29, 2014. 0022-3476/$ - see front matter. Copyright ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2014.05.011
485
THE JOURNAL OF PEDIATRICS
www.jpeds.com
attendants are highly variable. Previous data collected in Wisconsin suggest that the incidence of missed or delayed diagnosis of CCHD is significantly higher in the OOH birth population (1/2684) than among hospital born babies (1/28 350).6
Methods To further understand the role of POS for CCHD in the OOH population, the Wisconsin Screening Hearts in Newborns (SHINE) Project has included the OOH birth population in a 3-year statewide project designed to provide information on POS, assist in the implementation of POS, and evaluate the performance of POS in hospital and OOH settings. All hospitals and licensed midwives in the state of Wisconsin were invited to participate with ongoing enrollment beginning January 1, 2013. The SHINE Project is a collaborative effort between the University of Wisconsin, the Medical College of Wisconsin, the Wisconsin Guild of Midwives, the Wisconsin Department of Health Services, and the Wisconsin State Laboratory of Hygiene and is funded by a demonstration grant from the Health Resources and Services Administration (a branch of the US Department of Health Services). The functions of the Wisconsin SHINE project were reviewed by the University of Wisconsin Health Sciences Institutional Review Board and determined to be quality assurance measures and not human subject research. The assessment of POS in the OOH population presented in this article was also evaluated using the “Not Research Determination Decision Tool” prepared by the University of Wisconsin Health Sciences Institutional Review Board and was again determined to not to be human subjects research. A discussion of the advantages and disadvantages of prenatal and postnatal screening is a routine part of the midwives’ practice and patient level discussion of the SHINE project, and its data collection with each mother was the responsibility of each participating midwife. Licensed midwives belonging to the Wisconsin Guild of Midwives were trained in the use of POS to detect CCHD and the use of a handheld pulse oximeter approved for use in infants by the Food and Drug Administration14 (Masimo Rad5v; Masimo Corporation, Irvine, California). Participating midwives offered POS to their clients with screening and reporting performed on a voluntary basis. POS was done according to the protocol recommended by the US Secretary of Health and Human Services.15 The recommended time for screening was between 24 and 48 hours after birth, and oxygen saturation was measured in the right hand and either foot using a Masimo Rad5v (Masimo Corporation) handheld pulse oximeter and a reusable probe designed for use with the Masimo oximeter (Masimo Corporation). All midwives used a reusable probe but carried a disposable probe in case of failure of the reusable probe. Midwives continued to enroll throughout study period. Late in the course of the project, 2 Amish birth attendants and 2 public health nurses working with the plain clothes community 486
Vol. 165, No. 3 were also enrolled in the Wisconsin SHINE project and received the same training as the licensed midwives. Infants were considered to pass the screen if oxygen saturation was $95% in the hand or foot and there was #3% difference in saturation between the hand and the foot. Infants were considered to fail the screen of oxygen saturation was <90% in the hand or foot. An equivocal reading was $90% and <95% in hand and foot or $3% difference between hand and foot. In the case of an equivocal reading the screen was repeated in 1 hour. The screen could be repeated twice with the same criteria for pass, fail, or equivocal. An infant with 3 equivocal results was considered to have failed.15 The phrase “two sites/three strikes” was used to remind providers of the algorithm. Each midwife was given laminated copies of the screening protocol. The protocol was also available online on the SHINE project website (www. wisconsinshine.org). The presence of 1 of the 7 primary or 5 secondary target CCHD lesions was the endpoint of the POS algorithm. Infants who passed the POS required no further evaluation. A protocol was established for failed screening that included contacting a “hotline” that would respond to questions regarding the algorithm or data collection methods and would provide consultation and clinical support for any infant failing the screening. Access to an on-call pediatric cardiologist was available to the participating midwives at all times. The role of the on-call pediatric cardiologist was to provide clinical guidance and to facilitate the medical evaluation of babies when appropriate. This remote assistance also included contact with the medical facility receiving the baby to review the implications of a failed POS. Midwives were encouraged, although not required to investigate the resources available locally to assess those infants who failed their POS before incorporating it into their practices. The American Academy of Pediatrics guidelines recommend a “comprehensive evaluation for causes of hypoxemia” prior to performance of echocardiography to exclude CCHD.15 Options for referral of babies failing their POS depended on the location and clinical status of the infant and the resources available at the nearest medical facility. Neonatal echocardiography was available in only 37.5% of the Wisconsin hospitals responding to a 2011 survey on POS.16 Options for evaluation included referral to the nearest hospital that could evaluate for causes of hypoxia but could not perform echocardiography, referral to the nearest hospital that had neonatal echocardiography available, or partnering with a local physician to evaluate hypoxia in a clinic or urgent care facility. For infants referred for evaluation where no echocardiogram was available, infants could be transferred to higher level of care for echocardiogram based on the assessment of local providers with echocardiography recommended for those babies without an explanation for their hypoxia. Results of the POS (pass/fail/not screened) were recorded on the state newborn screening blood card and submitted to the Wisconsin State Lab of Hygiene. Midwives also completed a more detailed data set that included basic Lhost et al
ORIGINAL ARTICLES
September 2014 demographic information including a first-degree family history of CCHD, timing of screening, and screening results. Screening and demographic information results were then entered into a database at the State Lab of Hygiene and reviewed for deviation from the algorithm. As part of the Wisconsin SHINE project, the charts of all patients under 1 year of age admitted to the American Family Children’s Hospital (Madison) or the Children’s Hospital of Wisconsin (Milwaukee) with 1 of the 12 CCHD diagnoses are reviewed to determine the mechanism of diagnosis and if POS had been performed.
Results During the first 11 months of data collection, 29 licensed midwives, 2 Amish birth attendants, and 2 public health nurses serving the Amish community completed training in POS and reported POS results from their practices. A total of 1196 newborn screening cards were returned to the State Lab of Hygiene from OOH births. POS results were reported on 449/1196 OOH births in Wisconsin (37.5%) (Figure 1). This data set also includes 17 infants who were born in hospitals, but discharged home at <24 hours old and underwent POS at an OOH location. A complete data set was available on 440/449 newborns with POS results reported (98%). As the Amish birth attendants and public health nurses started gathering data much later in the project, only 15/440 babies with complete data sets were from Amish birth attendants or public health nurses.
The population of newborns born OOH with complete data set includes 173 (38.5%) babies born to mothers from the plain clothes community. Only 23 (13%) of these babies were known to have prenatal ultrasounds. Of the babies born outside of the plain clothes community, 74% of babies (196/265) had prenatal ultrasounds. Five families reported a history of CCHD in a first degree relatives, including Ellis-Van Creveld Syndrome, Ebstein anomaly, and aortic stenosis. The type of heart defect was not specified in 2 cases. Only 2 of these 5 babies had prenatal ultrasounds. The average maternal age was 30.6 years with a SD of 5.7 years and range of 18.5-50.2 years. The average birth weight was 3.6 kg with a SD of 0.5 kg. There were 217 girls and 223 boys. The average age at the time of POS was 40 21 hours, ranging from 2 hours to 7 weeks (Figure 2). Seven newborns did not have an age reported. Newborns were screened early (less than 24 hours old) in 20 cases (4.5%), with 14 (3.2%) at less than 22 hours old. Delayed screening between 48 and 96 hours was performed in 69 (16%) of babies. Late screening, defined as greater than 96 hours old, occurred in 8 newborns (1.8%). Among those with a complete dataset, 432 passed the screening, 3 failed the screening, and in 5, the screening algorithm was misinterpreted (Table). Four of the 437 infants were considered to have failed a completed screening algorithm (failure rate 0.9%). Of the 432 passing the screening, there has been 1 known false negative. This infant with right hand and foot saturations of 97% and 96%, respectively, was found to have a coarctation of the aorta and ventricular septal defect when
Figure 1. Results of POS. Number of newborn screening cards received and pulse oximetry results. Pulse Oximetry Screening for Critical Congenital Heart Disease in Planned Out-of-Hospital Births
487
THE JOURNAL OF PEDIATRICS
www.jpeds.com
Vol. 165, No. 3 represented 1.98% of Wisconsin births. During the reporting period, 25 hospitals participated in the Wisconsin SHINE Project. A total of 59 241 newborn screening cards were returned to the Wisconsin State Lab of Hygiene from hospital births. POS results were reported on 16 034 (27.1%) of these cards as part of the hospital birth component of the SHINE project. Of those reported, 23 infants hospital born infants failed their POS (0.14%).
Discussion
Figure 2. Age in hours at time of screening. Number of newborns and age in hours at screening.
evaluated for heart failure at 4 months of age. This infant went on to successful repair. The 3 babies who failed their POS had pathology other than CCHD. One newborn had an atrioventricular canal defect with common atrium and a mildly hypoplastic right ventricle, but as this child had a successful 2-ventricle repair, they would not be considered to have 1 of the 12 target CCHD lesions. One baby was treated for sepsis, and another was treated for sepsis with a possible inborn error of metabolism. The screening algorithm was misinterpreted in 5 babies of 440 babies with complete data sets (1.1%). One infant was inappropriately reported to have failed, 3 did not have equivocal results repeated, and measurements were repeated in 1 infant that should have been classified as an immediate screening failure (Table). In 8 of 440 newborns screened (1.8%), a repeat set of POS measurements were or should have been performed, a rate similar but slightly higher the 1.5% reported in a cohort of hospital born babies.17 A third set of POS measurements should have been performed in 1 of these 8 infants (patient 4, Table). During the reporting period, The Wisconsin State Laboratory of Hygiene received newborn screening cards on 60 437 babies (Figure 1). The 1196 OOH births
The OOH birth population represents a unique challenge to newborn POS. Families choose OOH birth for a wide range of religious, cultural, philosophical, or financial reasons. Mothers choosing OOH delivery may also desire greater privacy and increased decision-making authority. Midwives also have widely varied practice sizes and styles that may facilitate or impede screening. Anticipated barriers to screening in the OOH population included access to equipment, training of midwives to do POS, costs associated with screening, timing of screening 24-48 hours after delivery, costs to families associated with follow-up to a failed screen, and concerns about reporting results to a state database including parental fears of lost healthcare decision-making authority for their child. A strength of the Wisconsin SHINE Project was the early collaboration with the Wisconsin Guild of Midwives. Pediatric cardiologists and pediatricians involved in the Wisconsin SHINE Project attended Guild meetings and gave presentations about POS as the project was developed and involved midwives in the development of screening recommendations and protocols. The number of midwives electing to participate in this study, and the number of families agreeing to POS suggests an acceptance of POS in the OOH population and the reporting rate in the OOH population (36.8%) exceeded that in the hospital born population (26.3%) during the study period (Figure 1). Previous data collected in Wisconsin suggest that missed CCHD is more common in the OOH population than in infants born in the hospital. Possible explanations for this include lower use of prenatal screening technologies such as prenatal ultrasound, a shorter observation period following delivery, and a higher incidence of CCHD in the OOH birth population. Wisconsin has a large plain clothes
Table. POS results requiring additional scrutiny Patient 1 2 3 4 5 6 7 8 9
Initial right hand
Foot
96% 99% 95% 94% 89% 97% 86% 90% 68%
94% 95% 99% 94% 90% 96% 84% 90% 65%
Repeat right hand
Foot
94% 97%
94% 96%
88% 90% 81%
89% 88% 77%
Bedside interpretation
Correct interpretation
Clinical outcome
Fail Pass Pass Pass Pass Pass Fail Fail Fail
Pass Incomplete Incomplete Incomplete Fail Pass Fail Fail Fail
Normal Normal Normal Normal Normal Coarctation/VSD AV canal/common atrium Sepsis Sepsis
AV canal, atrioventricular canal; VSD, ventricular septal defect.
488
Lhost et al
ORIGINAL ARTICLES
September 2014 (Amish and Mennonite) population who often choose to deliver outside of the hospital. Among plain clothes populations currently living in Wisconsin, there are known heritable conditions such as Ellis von Creveld that occur at higher frequency and are associated with CCHD. Preliminary findings from the Wisconsin SHINE Project also suggest a higher failure rate in the OOH population than the population born in hospital. In the Swedish study on which the recommended algorithm was first evaluated, 0.23% of babies failed (18/38 429 with CCHD, 31/38 429 with other disease processes, and 38/38 429 healthy newborns).7 Our data suggests a slightly lower failure rate for hospital born babies (0.14%) but a substantially higher failure rate in the OOH birth community (0.9%). In the same Swedish study, the false negative rate was 0.03%7 compared with the false negative rate of 0.23% in our OOH cohort. The recommended screening algorithm was misinterpreted or not completed in five of 440 infants. One infant in the OOH population was reported to have failed their POS because of a misinterpretation of the screening algorithm. Three infants had equivocal results that should have been repeated. The fifth infant had initial POS measurements that should have resulted in immediate failure, although subsequent measurements were normal. Upon further investigation, it appears that the initial measurements may have been obtained incorrectly. Of the 440 infants with a complete data set, a second set of POS measurements was necessary in 4 infants and should have been performed in another 4 infants, suggesting that the need for multiple measurements in this population is small (1.8%). By the strict criteria of detecting the 12 target CCHD lesions, there were no true positive results. However, all 3 babies correctly identified as having failed their POS had disease processes that were identified more quickly because of early detection of hypoxia. The screening failure and false negative rates in this very early data must be interpreted with caution because of the small numbers of patients. Determining the most appropriate response to a failed screening in the OOH setting requires attention to the clinical status of the infant, respect for the medical care preferences of the mother, and an understanding of the local health care resources. Because of the wide variability in these factors, no standard recommendation can be made and all responses must be individualized. An ongoing collaborative relationship between midwives and physicians is critical for the success of these efforts. n Submitted for publication Jan 29, 2014; last revision received Mar 24, 2014; accepted May 1, 2014.
Reprint requests: John S. Hokanson, MD, Division of Pediatric Cardiology, Department of Pediatrics, University of Wisconsin, H6/516c, 600 Highland Ave, Madison, WI 53792. E-mail: [email protected]
References 1. Hoffman JI, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol 2002;39:1890-900. 2. Wren C, Reinhardt Z, Khawaja K. Twenty-year trends in diagnosis of life-threatening neonatal cardiovascular malformations. Arch Dis Child Fetal Neonatal Ed 2008;93:F33-5. 3. Pradat P, Francannet C, Harris JA, Robert E. The epidemiology of cardiovascular defects, part I: a study based on data from three large registries of congenital malformations. Pediatr Cardiol 2003;24:195-221. 4. Mellander M, Sunnegardh J. Failure to diagnose critical heart malformations in newborns before discharge—an increasing problem? Acta Paediatr 2006;95:407-13. 5. Chang RK, Gurvitz M, Rodriguez S. Missed diagnosis of critical congenital heart disease. Arch Pediatr Adolesc Med 2008;162:969-74. 6. Ng B, Hokanson J. Missed congenital heart disease in neonates. Congenit Heart Dis 2010;5:292-6. 7. de-Wahl Granelli A, Wennergren M, Sandberg K, Mellander M, Bejlum C, Inganas L, et al. Impact of pulse oximetry screening on the detection of duct dependent congenital heart disease: a Swedish prospective screening study in 39,821 newborns. BMJ 2009;338:a3037. 8. Meberg A, Andreassen A, Brunvand L, Markestad T, Moster D, Nietsch L, et al. Pulse oximetry screening as a complementary strategy to detect critical congenital heart defects. Acta Paediatr 2009;98:682-6. 9. Riede FT, Worner C, Dahnert I, Mockel A, Kostelka M, Schneider P. Effectiveness of neonatal pulse oximetry screening for detection of critical congenital heart disease in daily clinical routine-results from a prospective multicenter study. Eur J Pediatr 2010;169:975-81. 10. Ewer AK, Middleton LJ, Furmston AT, Bhoyar A, Daniels JP, Thangaratinam S, et al. Pulse oximetry screening for congenital heart defects in newborn infants (PulseOx): a test accuracy study. Lancet 2011; 378:785-94. 11. Mahle WT, Martin GR, Beekman RH III, Morrow WR, Rosenthal GL, Snyder CS, et al. Endorsement of health and human services recommendation for pulse oximetry screening for critical congenital heart disease. Pediatrics 2012;129:190-2. 12. Watterberg KL , on behalf of the Committee on F, Newborn. Policy statement on planned home birth: upholding the best interests of children and families. Pediatrics 2013;1-16, 1020. 13. MacDorman MF, Mathews TJ, Declercq E. Home births in the United States 1990-2009. NCHS Data Brief 2012. 14. Zuckerman B. 510(k) Summary: K120657. fda.gov 2012; Available at: http://www.accessdata.fda.gov/cdrh_docs/pdf12/k120657.pdf. Accessed March 18, 2014. 15. Kemper AR, Mahle WT, Martin GR, Cooley WC, Kumar P, Morrow WR, et al. Strategies for implementing screening for critical congenital heart disease. Pediatrics 2011;128:e1259-67. 16. Beissel DJ, Goetz EM, Hokanson JS. Pulse oximetry screening for congenital heart disease in Wisconsin. Congenit Heart Dis 2011;6:521-2. 17. Kochilas LK, Lohr JL, Bruhn E, Borman-Shoap E, Gams BL, Pylipow M, et al. Implementation of critical congenital heart disease screening in Minnesota. Pediatrics 2013;132:e587-94.
Pulse Oximetry Screening for Critical Congenital Heart Disease in Planned Out-of-Hospital Births
489
C
ritical congenital heart disease (CCHD) affects 1-3/1000 babies born in the US every year.1-4 The majority of these infants are diagnosed by prenatal ultrasound or by clinical examination shortly after birth. However, a subset of infants with CCHD will be missed by both prenatal ultrasound and clinical examination and may present with significant cardiac morbidity or mortality in the neonatal period.5,6 Recent efforts to screen infants for CCHD with pulse oximetry screening (POS) have focused on identifying infants with clinically undetectable CCHD prior to discharge from the birth hospitalization.7-10 The US Secretary of Health and Human Services recommended universal newborn POS for CCHD in September 2011. The American Academy of Pediatrics endorsed these recommendations in December 201111 and in 2013 included a recommendation for POS for planned home births.12 All screening recommendations are based on evidence derived from babies born in the hospital setting, and little is currently known about POS in the out-of-hospital (OOH) birth population. According to the Centers for Disease Control and Prevention, the number of OOH births increased from 0.56%-0.72% of US deliveries between 2004 and 2009.13 Wisconsin had the fifth highest rate of OOH delivery in the nation with 1.66% of births occurring at home or in a birthing center.13 Midwifery standards of practice for OOH birth in Wisconsin are defined by the State Department of Safety and Professional Services. Licensed midwives are required by the rules and regulations guiding their practice to remain with the mother and baby after birth until the both mother and baby are stable and for no less than 2 hours. On average, midwives report staying 3-4 hours after birth. They are also required From the School of Medicine and Public Health, to reevaluate maternal and infant well-being within 36 hours of birth. The timing Department of Pediatrics, University of Wisconsin, Madison, WI; Wisconsin Guild of Midwives, Birthwise of this reevaluation is subject to medical indication, clinic schedules, weather, Health and Birth Center; and Wisconsin Guild of distance to the home, and the preferences of the family. Newborn blood Midwives, Local Delivery Midwifery Service, Iola, Wisconsin screening is typically performed at this visit. Hearing screening may also be perFunded by Health Resources and Services Administraformed at the same visit, or deferred to a subsequent appointment. tion (H46MC24057). The authors declare no conflicts of interest. Some families, particularly members of plain clothes communities, have unatPortions of the study were presented as a poster at the tended deliveries or use a birth attendant who is a member of their community. Midwest Pediatric Cardiology Society Scientific Session, Chicago, IL, September 20, 2013, at the Critical The training, experience, and newborn care practices of these nonlicensed birth Congenital Heart Disease Stakeholders Meeting:State 1
2
3
4
CCHD OOH POS SHINE
Critical congenital heart disease Out-of-hospital Pulse oximetry screening Screening Hearts in Newborns
Implementation, Association of Public Health Laboratories, at the Pediatric Congenital Heart Association Meeting, Silver Springs, MD, February 27-28, 2014, and at the Wisconsin Association for Perinatal Care’s Statewide Perinatal Conference, Wisconsin Dells, WI, April 27-29, 2014. 0022-3476/$ - see front matter. Copyright ª 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2014.05.011
485
THE JOURNAL OF PEDIATRICS
www.jpeds.com
attendants are highly variable. Previous data collected in Wisconsin suggest that the incidence of missed or delayed diagnosis of CCHD is significantly higher in the OOH birth population (1/2684) than among hospital born babies (1/28 350).6
Methods To further understand the role of POS for CCHD in the OOH population, the Wisconsin Screening Hearts in Newborns (SHINE) Project has included the OOH birth population in a 3-year statewide project designed to provide information on POS, assist in the implementation of POS, and evaluate the performance of POS in hospital and OOH settings. All hospitals and licensed midwives in the state of Wisconsin were invited to participate with ongoing enrollment beginning January 1, 2013. The SHINE Project is a collaborative effort between the University of Wisconsin, the Medical College of Wisconsin, the Wisconsin Guild of Midwives, the Wisconsin Department of Health Services, and the Wisconsin State Laboratory of Hygiene and is funded by a demonstration grant from the Health Resources and Services Administration (a branch of the US Department of Health Services). The functions of the Wisconsin SHINE project were reviewed by the University of Wisconsin Health Sciences Institutional Review Board and determined to be quality assurance measures and not human subject research. The assessment of POS in the OOH population presented in this article was also evaluated using the “Not Research Determination Decision Tool” prepared by the University of Wisconsin Health Sciences Institutional Review Board and was again determined to not to be human subjects research. A discussion of the advantages and disadvantages of prenatal and postnatal screening is a routine part of the midwives’ practice and patient level discussion of the SHINE project, and its data collection with each mother was the responsibility of each participating midwife. Licensed midwives belonging to the Wisconsin Guild of Midwives were trained in the use of POS to detect CCHD and the use of a handheld pulse oximeter approved for use in infants by the Food and Drug Administration14 (Masimo Rad5v; Masimo Corporation, Irvine, California). Participating midwives offered POS to their clients with screening and reporting performed on a voluntary basis. POS was done according to the protocol recommended by the US Secretary of Health and Human Services.15 The recommended time for screening was between 24 and 48 hours after birth, and oxygen saturation was measured in the right hand and either foot using a Masimo Rad5v (Masimo Corporation) handheld pulse oximeter and a reusable probe designed for use with the Masimo oximeter (Masimo Corporation). All midwives used a reusable probe but carried a disposable probe in case of failure of the reusable probe. Midwives continued to enroll throughout study period. Late in the course of the project, 2 Amish birth attendants and 2 public health nurses working with the plain clothes community 486
Vol. 165, No. 3 were also enrolled in the Wisconsin SHINE project and received the same training as the licensed midwives. Infants were considered to pass the screen if oxygen saturation was $95% in the hand or foot and there was #3% difference in saturation between the hand and the foot. Infants were considered to fail the screen of oxygen saturation was <90% in the hand or foot. An equivocal reading was $90% and <95% in hand and foot or $3% difference between hand and foot. In the case of an equivocal reading the screen was repeated in 1 hour. The screen could be repeated twice with the same criteria for pass, fail, or equivocal. An infant with 3 equivocal results was considered to have failed.15 The phrase “two sites/three strikes” was used to remind providers of the algorithm. Each midwife was given laminated copies of the screening protocol. The protocol was also available online on the SHINE project website (www. wisconsinshine.org). The presence of 1 of the 7 primary or 5 secondary target CCHD lesions was the endpoint of the POS algorithm. Infants who passed the POS required no further evaluation. A protocol was established for failed screening that included contacting a “hotline” that would respond to questions regarding the algorithm or data collection methods and would provide consultation and clinical support for any infant failing the screening. Access to an on-call pediatric cardiologist was available to the participating midwives at all times. The role of the on-call pediatric cardiologist was to provide clinical guidance and to facilitate the medical evaluation of babies when appropriate. This remote assistance also included contact with the medical facility receiving the baby to review the implications of a failed POS. Midwives were encouraged, although not required to investigate the resources available locally to assess those infants who failed their POS before incorporating it into their practices. The American Academy of Pediatrics guidelines recommend a “comprehensive evaluation for causes of hypoxemia” prior to performance of echocardiography to exclude CCHD.15 Options for referral of babies failing their POS depended on the location and clinical status of the infant and the resources available at the nearest medical facility. Neonatal echocardiography was available in only 37.5% of the Wisconsin hospitals responding to a 2011 survey on POS.16 Options for evaluation included referral to the nearest hospital that could evaluate for causes of hypoxia but could not perform echocardiography, referral to the nearest hospital that had neonatal echocardiography available, or partnering with a local physician to evaluate hypoxia in a clinic or urgent care facility. For infants referred for evaluation where no echocardiogram was available, infants could be transferred to higher level of care for echocardiogram based on the assessment of local providers with echocardiography recommended for those babies without an explanation for their hypoxia. Results of the POS (pass/fail/not screened) were recorded on the state newborn screening blood card and submitted to the Wisconsin State Lab of Hygiene. Midwives also completed a more detailed data set that included basic Lhost et al
ORIGINAL ARTICLES
September 2014 demographic information including a first-degree family history of CCHD, timing of screening, and screening results. Screening and demographic information results were then entered into a database at the State Lab of Hygiene and reviewed for deviation from the algorithm. As part of the Wisconsin SHINE project, the charts of all patients under 1 year of age admitted to the American Family Children’s Hospital (Madison) or the Children’s Hospital of Wisconsin (Milwaukee) with 1 of the 12 CCHD diagnoses are reviewed to determine the mechanism of diagnosis and if POS had been performed.
Results During the first 11 months of data collection, 29 licensed midwives, 2 Amish birth attendants, and 2 public health nurses serving the Amish community completed training in POS and reported POS results from their practices. A total of 1196 newborn screening cards were returned to the State Lab of Hygiene from OOH births. POS results were reported on 449/1196 OOH births in Wisconsin (37.5%) (Figure 1). This data set also includes 17 infants who were born in hospitals, but discharged home at <24 hours old and underwent POS at an OOH location. A complete data set was available on 440/449 newborns with POS results reported (98%). As the Amish birth attendants and public health nurses started gathering data much later in the project, only 15/440 babies with complete data sets were from Amish birth attendants or public health nurses.
The population of newborns born OOH with complete data set includes 173 (38.5%) babies born to mothers from the plain clothes community. Only 23 (13%) of these babies were known to have prenatal ultrasounds. Of the babies born outside of the plain clothes community, 74% of babies (196/265) had prenatal ultrasounds. Five families reported a history of CCHD in a first degree relatives, including Ellis-Van Creveld Syndrome, Ebstein anomaly, and aortic stenosis. The type of heart defect was not specified in 2 cases. Only 2 of these 5 babies had prenatal ultrasounds. The average maternal age was 30.6 years with a SD of 5.7 years and range of 18.5-50.2 years. The average birth weight was 3.6 kg with a SD of 0.5 kg. There were 217 girls and 223 boys. The average age at the time of POS was 40 21 hours, ranging from 2 hours to 7 weeks (Figure 2). Seven newborns did not have an age reported. Newborns were screened early (less than 24 hours old) in 20 cases (4.5%), with 14 (3.2%) at less than 22 hours old. Delayed screening between 48 and 96 hours was performed in 69 (16%) of babies. Late screening, defined as greater than 96 hours old, occurred in 8 newborns (1.8%). Among those with a complete dataset, 432 passed the screening, 3 failed the screening, and in 5, the screening algorithm was misinterpreted (Table). Four of the 437 infants were considered to have failed a completed screening algorithm (failure rate 0.9%). Of the 432 passing the screening, there has been 1 known false negative. This infant with right hand and foot saturations of 97% and 96%, respectively, was found to have a coarctation of the aorta and ventricular septal defect when
Figure 1. Results of POS. Number of newborn screening cards received and pulse oximetry results. Pulse Oximetry Screening for Critical Congenital Heart Disease in Planned Out-of-Hospital Births
487
THE JOURNAL OF PEDIATRICS
www.jpeds.com
Vol. 165, No. 3 represented 1.98% of Wisconsin births. During the reporting period, 25 hospitals participated in the Wisconsin SHINE Project. A total of 59 241 newborn screening cards were returned to the Wisconsin State Lab of Hygiene from hospital births. POS results were reported on 16 034 (27.1%) of these cards as part of the hospital birth component of the SHINE project. Of those reported, 23 infants hospital born infants failed their POS (0.14%).
Discussion
Figure 2. Age in hours at time of screening. Number of newborns and age in hours at screening.
evaluated for heart failure at 4 months of age. This infant went on to successful repair. The 3 babies who failed their POS had pathology other than CCHD. One newborn had an atrioventricular canal defect with common atrium and a mildly hypoplastic right ventricle, but as this child had a successful 2-ventricle repair, they would not be considered to have 1 of the 12 target CCHD lesions. One baby was treated for sepsis, and another was treated for sepsis with a possible inborn error of metabolism. The screening algorithm was misinterpreted in 5 babies of 440 babies with complete data sets (1.1%). One infant was inappropriately reported to have failed, 3 did not have equivocal results repeated, and measurements were repeated in 1 infant that should have been classified as an immediate screening failure (Table). In 8 of 440 newborns screened (1.8%), a repeat set of POS measurements were or should have been performed, a rate similar but slightly higher the 1.5% reported in a cohort of hospital born babies.17 A third set of POS measurements should have been performed in 1 of these 8 infants (patient 4, Table). During the reporting period, The Wisconsin State Laboratory of Hygiene received newborn screening cards on 60 437 babies (Figure 1). The 1196 OOH births
The OOH birth population represents a unique challenge to newborn POS. Families choose OOH birth for a wide range of religious, cultural, philosophical, or financial reasons. Mothers choosing OOH delivery may also desire greater privacy and increased decision-making authority. Midwives also have widely varied practice sizes and styles that may facilitate or impede screening. Anticipated barriers to screening in the OOH population included access to equipment, training of midwives to do POS, costs associated with screening, timing of screening 24-48 hours after delivery, costs to families associated with follow-up to a failed screen, and concerns about reporting results to a state database including parental fears of lost healthcare decision-making authority for their child. A strength of the Wisconsin SHINE Project was the early collaboration with the Wisconsin Guild of Midwives. Pediatric cardiologists and pediatricians involved in the Wisconsin SHINE Project attended Guild meetings and gave presentations about POS as the project was developed and involved midwives in the development of screening recommendations and protocols. The number of midwives electing to participate in this study, and the number of families agreeing to POS suggests an acceptance of POS in the OOH population and the reporting rate in the OOH population (36.8%) exceeded that in the hospital born population (26.3%) during the study period (Figure 1). Previous data collected in Wisconsin suggest that missed CCHD is more common in the OOH population than in infants born in the hospital. Possible explanations for this include lower use of prenatal screening technologies such as prenatal ultrasound, a shorter observation period following delivery, and a higher incidence of CCHD in the OOH birth population. Wisconsin has a large plain clothes
Table. POS results requiring additional scrutiny Patient 1 2 3 4 5 6 7 8 9
Initial right hand
Foot
96% 99% 95% 94% 89% 97% 86% 90% 68%
94% 95% 99% 94% 90% 96% 84% 90% 65%
Repeat right hand
Foot
94% 97%
94% 96%
88% 90% 81%
89% 88% 77%
Bedside interpretation
Correct interpretation
Clinical outcome
Fail Pass Pass Pass Pass Pass Fail Fail Fail
Pass Incomplete Incomplete Incomplete Fail Pass Fail Fail Fail
Normal Normal Normal Normal Normal Coarctation/VSD AV canal/common atrium Sepsis Sepsis
AV canal, atrioventricular canal; VSD, ventricular septal defect.
488
Lhost et al
ORIGINAL ARTICLES
September 2014 (Amish and Mennonite) population who often choose to deliver outside of the hospital. Among plain clothes populations currently living in Wisconsin, there are known heritable conditions such as Ellis von Creveld that occur at higher frequency and are associated with CCHD. Preliminary findings from the Wisconsin SHINE Project also suggest a higher failure rate in the OOH population than the population born in hospital. In the Swedish study on which the recommended algorithm was first evaluated, 0.23% of babies failed (18/38 429 with CCHD, 31/38 429 with other disease processes, and 38/38 429 healthy newborns).7 Our data suggests a slightly lower failure rate for hospital born babies (0.14%) but a substantially higher failure rate in the OOH birth community (0.9%). In the same Swedish study, the false negative rate was 0.03%7 compared with the false negative rate of 0.23% in our OOH cohort. The recommended screening algorithm was misinterpreted or not completed in five of 440 infants. One infant in the OOH population was reported to have failed their POS because of a misinterpretation of the screening algorithm. Three infants had equivocal results that should have been repeated. The fifth infant had initial POS measurements that should have resulted in immediate failure, although subsequent measurements were normal. Upon further investigation, it appears that the initial measurements may have been obtained incorrectly. Of the 440 infants with a complete data set, a second set of POS measurements was necessary in 4 infants and should have been performed in another 4 infants, suggesting that the need for multiple measurements in this population is small (1.8%). By the strict criteria of detecting the 12 target CCHD lesions, there were no true positive results. However, all 3 babies correctly identified as having failed their POS had disease processes that were identified more quickly because of early detection of hypoxia. The screening failure and false negative rates in this very early data must be interpreted with caution because of the small numbers of patients. Determining the most appropriate response to a failed screening in the OOH setting requires attention to the clinical status of the infant, respect for the medical care preferences of the mother, and an understanding of the local health care resources. Because of the wide variability in these factors, no standard recommendation can be made and all responses must be individualized. An ongoing collaborative relationship between midwives and physicians is critical for the success of these efforts. n Submitted for publication Jan 29, 2014; last revision received Mar 24, 2014; accepted May 1, 2014.
Reprint requests: John S. Hokanson, MD, Division of Pediatric Cardiology, Department of Pediatrics, University of Wisconsin, H6/516c, 600 Highland Ave, Madison, WI 53792. E-mail: [email protected]
References 1. Hoffman JI, Kaplan S. The incidence of congenital heart disease. J Am Coll Cardiol 2002;39:1890-900. 2. Wren C, Reinhardt Z, Khawaja K. Twenty-year trends in diagnosis of life-threatening neonatal cardiovascular malformations. Arch Dis Child Fetal Neonatal Ed 2008;93:F33-5. 3. Pradat P, Francannet C, Harris JA, Robert E. The epidemiology of cardiovascular defects, part I: a study based on data from three large registries of congenital malformations. Pediatr Cardiol 2003;24:195-221. 4. Mellander M, Sunnegardh J. Failure to diagnose critical heart malformations in newborns before discharge—an increasing problem? Acta Paediatr 2006;95:407-13. 5. Chang RK, Gurvitz M, Rodriguez S. Missed diagnosis of critical congenital heart disease. Arch Pediatr Adolesc Med 2008;162:969-74. 6. Ng B, Hokanson J. Missed congenital heart disease in neonates. Congenit Heart Dis 2010;5:292-6. 7. de-Wahl Granelli A, Wennergren M, Sandberg K, Mellander M, Bejlum C, Inganas L, et al. Impact of pulse oximetry screening on the detection of duct dependent congenital heart disease: a Swedish prospective screening study in 39,821 newborns. BMJ 2009;338:a3037. 8. Meberg A, Andreassen A, Brunvand L, Markestad T, Moster D, Nietsch L, et al. Pulse oximetry screening as a complementary strategy to detect critical congenital heart defects. Acta Paediatr 2009;98:682-6. 9. Riede FT, Worner C, Dahnert I, Mockel A, Kostelka M, Schneider P. Effectiveness of neonatal pulse oximetry screening for detection of critical congenital heart disease in daily clinical routine-results from a prospective multicenter study. Eur J Pediatr 2010;169:975-81. 10. Ewer AK, Middleton LJ, Furmston AT, Bhoyar A, Daniels JP, Thangaratinam S, et al. Pulse oximetry screening for congenital heart defects in newborn infants (PulseOx): a test accuracy study. Lancet 2011; 378:785-94. 11. Mahle WT, Martin GR, Beekman RH III, Morrow WR, Rosenthal GL, Snyder CS, et al. Endorsement of health and human services recommendation for pulse oximetry screening for critical congenital heart disease. Pediatrics 2012;129:190-2. 12. Watterberg KL , on behalf of the Committee on F, Newborn. Policy statement on planned home birth: upholding the best interests of children and families. Pediatrics 2013;1-16, 1020. 13. MacDorman MF, Mathews TJ, Declercq E. Home births in the United States 1990-2009. NCHS Data Brief 2012. 14. Zuckerman B. 510(k) Summary: K120657. fda.gov 2012; Available at: http://www.accessdata.fda.gov/cdrh_docs/pdf12/k120657.pdf. Accessed March 18, 2014. 15. Kemper AR, Mahle WT, Martin GR, Cooley WC, Kumar P, Morrow WR, et al. Strategies for implementing screening for critical congenital heart disease. Pediatrics 2011;128:e1259-67. 16. Beissel DJ, Goetz EM, Hokanson JS. Pulse oximetry screening for congenital heart disease in Wisconsin. Congenit Heart Dis 2011;6:521-2. 17. Kochilas LK, Lohr JL, Bruhn E, Borman-Shoap E, Gams BL, Pylipow M, et al. Implementation of critical congenital heart disease screening in Minnesota. Pediatrics 2013;132:e587-94.
Pulse Oximetry Screening for Critical Congenital Heart Disease in Planned Out-of-Hospital Births
489