- Email: [email protected]
Prospective Surveillance of Extreme Neonatal Hyperbilirubinemia in Australia
Prospective Surveillance of Extreme Neonatal Hyperbilirubinemia in Australia Angela McGillivray, FRACP, MMedEd1,2, Jan Polverino, RN, RM, BSocSc1,2, Nadia Badawi, FRACP, PhD2,3,4, and Nick Evans, MRCPCH, DM1,2 Objectives To determine the incidence, causes, associated factors, and short-term outcomes of extreme neonatal hyperbilirubinemia in Australia in order to identify opportunities for prevention.
Study Design This was a prospective population-based surveillance study in collaboration with the Australian Pediatric Surveillance Unit between April 1, 2010, and March 31, 2013. Case definition was: infants >34 weeks gestation with a peak total serum bilirubin $450 mmol/L and or clinical evidence of bilirubin encephalopathy. Clinicians completed questionnaires detailing demographic and clinical data including: peak serum bilirubin, signs of bilirubin encephalopathy, etiology, associated pathology, management, and short-term outcomes. Results The questionnaire return rate was 95%, and 87 infants met the case definition. The Australian incidence of extreme neonatal hyperbilirubinemia is estimated to be 9.4/100 000 live births. Main etiologies were: idiopathic ABO blood group incompatibility, glucose-6-phosphate dehydrogenase deficiency, and Rhesus isoimmunization. There were no significant differences in short-term outcomes between inpatient and outpatient cases. Cases with a hemolytic etiology were significantly more likely to have extremely high levels of hyperbilirubinemia (P < .002). Conclusion The incidence of extreme neonatal hyperbilirubinemia in Australia is comparable with previous studies. Robust pre- and post-discharge assessment and management strategies of neonatal hyperbilirubinemia are essential to prevent neurodisability. Universal glucose-6-phosphate dehydrogenase screening and serial bilirubin monitoring may optimize preventative strategies. (J Pediatr 2015;-:---). See editorial, p
E
xtreme neonatal hyperbilirubinemia, although rare, can cause neonatal bilirubin encephalopathy and permanent impairment with choreoathetoid cerebral palsy and deafness. It is a condition that in most circumstances can be prevented. There have been recent international concerns regarding a possible reemergence of neonatal bilirubin encephalopathy. This has resulted in the publication of 6 surveillance studies, all in developed countries in the northern hemisphere.1-7 Notwithstanding methodological and case definition variation, voluntary clinician case-reporting in collaboration with local rare disease surveillance units has been the predominant approach. Extreme neonatal hyperbilirubinemia is variably defined in these studies by peak bilirubin levels in the range of 340-510 mmol/L. Overall, the incidence internationally is estimated between 2 and 53 cases per 100 000 live births.1-7 After a cluster of cases of bilirubin encephalopathy locally, it was recognized that there was a lack of incidence data in Australia on which to base a discussion about preventative strategies. Our aim was to determine the incidence of extreme neonatal hyperbilirubinemia in Australia, to ascertain causes and associated risk factors and thus identify opportunities for enhanced future preventative strategies. This will include a neurodevelopmental follow-up study to establish the important longer-term outcomes of this population. The surveillance stage is reported in this paper.
Methods This prospective surveillance study utilized voluntary clinician reporting of cases in collaboration with the Australian Pediatric Surveillance Unit (APSU) from April 1, 2010, to March 31, 2013. The APSU was established in 1993 to facilitate the reporting and surveillance of rare conditions and infectious disease outbreaks. Approximately 1400 clinicians (predominantly pediatricians) are contacted on a monthly basis, primarily by e-mail, and requested to make reports according to the From the Royal Prince Alfred Hospital, Camperdown; 1
2
The University of Sydney, Sydney; 3The Children’s Hospital at Westmead, Westmead; and 4The University of Notre Dame, Darlinghurst, New South Wales, Australia
APSU G6PD TCB TSB
Australian Pediatric Surveillance Unit Glucose-6-phosphate dehydrogenase Transcutaneous bilirubin Total serum bilirubin
The Cerebral Palsy Alliance provided funding for the Australian Pediatric Surveillance Unit costs (IRG2009 [to N.E.]). The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2015.08.048
1
THE JOURNAL OF PEDIATRICS
Vol. -, No. -
www.jpeds.com
contemporary conditions of interest. The study was approved by the Sydney Local Health District Human Research Ethics Committee. Cases were defined as: an infant of 34 weeks gestation or more with a peak total serum bilirubin (TSB) $450 mmol/L in the first month, or with clinical and/or magnetic resonance imaging consistent with bilirubin toxicity. On notification of a case, reporting clinicians received a de-identified questionnaire requesting demographic and relevant clinical data including sex, ethnicity, gestation, delivery mode, feeding methods, presentation and discharge timing, peak TSB levels, signs of bilirubin encephalopathy, etiology, associated pathology, and treatment (Appendix; available at www.jpeds.com). Subsequently, the research team contacted clinicians who had reported cases by e-mail, letter, and/or telephone as necessary to expedite and optimize questionnaire return. Data were held securely both digitally and in hardcopy at the primary research site. Accuracy of database information was ensured by a triple check of all questionnaire data entry by two members of the research team. The incidence calculation denominator was from the Australian Institute of Health and Welfare8 and the Australian Bureau of Statistics.9 The annual Australia’s Mothers and Babies report produced by the Australian Institute of Health and Welfare provides accurate, national, gestational week-specific live birth data; thus, the most current available data (up to December 2012) was used for incidence calculation. Thereafter, data from the Australian Bureau of Statistics was extrapolated to complete the epoch. Data were summarized by descriptive statistics. Comparisons between groups were performed using Wilcoxon rank sum test, Pearson c2, or Fisher exact test where appropriate. The analyses were done using IBM SPSS Statistics version 22 (IBM, Armonk, New York).
Results
Neither of the babies with seizures was reported as having tonal or feeding abnormalities. Based on the 5 cases presenting with opisthotonus, the incidence of bilirubin encephalopathy is 0.6/100 000 live births. The median peak bilirubin level was 483 mmol/L (range 370-983). There were 56 cases (64%) with peak bilirubin levels between 450 and 500 mmol/L, 19 (22%) between 500 and 600 mmol/L, and 11 (13%) above 600 mmol/L. The national distribution of reported cases reflects the geographical general population distribution of Australia, with the majority of cases in the most populous states of New South Wales and Victoria. Demographic and perinatal variable data of the 87 cases is summarized in Table I. Fifty cases were male (58%) with the sex of one case not reported. Clinician-reported ethnicity was provided in 72 reports. Of these, 69% of cases were reported as non-Caucasian. Eight cases were born by cesarean (9.3%) compared with a national average of 32%.10 With regards to primary mode of feeding, 67% were exclusively breastfed in the first 2 weeks compared with a national average that ranges between 61% and 90%.10 Table II summarizes the underlying conditions and treatments. The majority (48.6%) of cases were caused, as determined by the reporting clinicians, by idiopathic hyperbilirubinemia. Thirty-six percent of cases had an underlying hemolytic etiology, which included: Rhesus D isoimmunization, direct antiglobulin test positive ABO-incompatibility, glucose-6-phosphate dehydrogenase (G6PD) deficiency, hereditary spherocytosis, and oxidative hemolysis. Clinicians documented 32% of cases had associated clinical signs of dehydration based on their clinical impression. Thirty-six percent of cases had hypoalbuminemia (<30 g/L) and 16% of the overall group received albumin
Table I. Demographic and perinatal variables of 87 cases Variables*
Over the 3-year period, 159 notifications of extreme neonatal hyperbilirubinemia were made by Australian pediatric clinicians. Of these, 150 notifications yielded questionnaires; equating to a 95% return rate. Eighty-seven reported patients met the case definition: 86 neonates had peak bilirubin levels $450 mmol/L and one case had a peak bilirubin level of 370 mmol/L but presented with clinical signs consistent with bilirubin encephalopathy, which resolved following phototherapy. There were 28 reports that did not meet the inclusion criteria and 35 reports were duplicates. There were 893 693 estimated live births over 34 weeks gestation in Australia between April 1, 2010, and March 31, 2013.8,9 Thus, the current incidence of extreme neonatal hyperbilirubinemia in Australia is calculated to be 9.4/100 000 live births. Of the 87 cases, 34 infants had lethargy and poor feeding (79 responses), 3 had hypotonia, 1 in association with lethargy and poor feeding (72 responses), 5 had opisthotonus (72 responses), and 2 had convulsions (71 responses). 2
Sex n = 86 Male Ethnicity (clinician determined) n = 72 Asian Caucasian Middle Eastern Aboriginal Pacific Islander African Other Birth weight (g), mean (SD) n = 85 Gestational age (wk), mean (SD) n = 87 Mode of delivery in hospital n = 86 Normal vaginal Vaginal breech Instrumental Cesarean Feeding method n = 86 Breast Artificial Combination Readmitted n = 87
N (%) 50 (58.1) 38 (52.8) 20 (27.8) 5 (6.9) 3 (4.2) 3 (4.2) 2 (2.8) 1 (1.3) 3298 (498) 38 (1.5) 53 (61.7) 2 (2.3) 23 (26.7) 8 (9.3) 58 (67.4) 1 (1.2) 27 (31.4) 68 (78.2)
*Values are numbers (%) unless otherwise stated.
McGillivray et al
- 2015
ORIGINAL ARTICLES
Table II. Causes and treatment of severe hyperbilirubinemia N (%) Idiopathic Probable ABO incompatibility (DAT positive) Possible ABO incompatibility (DAT negative) G6PD deficiency Rh Other hematological (spherocytosis, Oxhem) Other (sepsis, cephalohematoma) Unknown Associated factors Dehydration n = 84 Low albumin (<30 g/L) n = 53 Culture positive infection n = 86 Other serious associated morbidity n = 84 Treatment Phototherapy n = 85 Albumin n = 81 Immunoglobulin n = 81 XT n = 84
37 (48.7) 10 (13.2) 7 (9.2) 7 (9.2) 3 (3.9) 7 (9.2) 5 (6.6) 11 27 (32.1) 19 (35.8) 7 (8.1) 4 (4.8) 84 (98.8) 13 (16.0) 7 (8.6) 17 (20.2)
DAT, direct antiglobulin test; Oxhem, oxidative hemolysis; Rh, Rhesus isoimmunization; XT, exchange transfusion.
treatment. Almost all infants (98.8%) received phototherapy treatment and 20% received at least one exchange transfusion. All cases were born in hospital. Extreme hyperbilirubinemia was identified in 78% of cases postdischarge, primarily by hospital- and community-based nurses. Table III presents the characteristics of cases either diagnosed as inpatients or outpatients of their birth hospital. Discharged infants were diagnosed significantly later at 5 days (range 2-16 days) compared with 3 days (range 0-9 days) for cases identified in hospital (P < .001). However, there was no significant difference in bilirubin levels between the cases identified in hospital compared with those diagnosed postdischarge: 496 vs 481 mmol/L (P = 049). All eighty-seven cases survived to discharge. Of the 68 whose discharge neurologic examination was reported, 3 were abnormal. The results of newborn hearing screening were provided for 76 cases, of which 4 required further audiological assessment. Magnetic resonance brain imaging was
carried out in only 11 of the 87 cases. Four patients had scans consistent with bilirubin toxicity, a further 3 were abnormal but not consistent with bilirubin toxicity, and the remaining 4 were normal. Table IV provides a summary of the 13 cases with known adverse short-term outcomes, their underlying etiology, treatment, and short-term outcomes. A neurodevelopmental follow-up study to the age of 3 years is underway to ascertain longer-term outcomes in the study cohort. Univariate analyses for known risk factors and peak bilirubin levels over 500 mmol/L were conducted and are summarized in Table V (available at www.jpeds.com). Infants with an underlying hemolytic cause for their extreme hyperbilirubinemia were significantly more likely to have bilirubin levels greater than 500 mmol/L compared with those with nonhemolytic hyperbilirubinemia (P < .002). There were no significant differences in the distribution of sex, clinician-reported ethnicity, gestational age, feeding mode, and readmission rates between cases with maximum peak bilirubin levels greater or equal to 500 mmol/L.
Discussion The incidence of extreme neonatal hyperbilirubinemia is estimated to be 9.4/100 000 live births and is comparable with previous surveillance studies using voluntary clinician report methodology.2-6 The study by Bjerre et al from Denmark likely provides the most accurate incidence estimate of 45/ 100 000 live births from their population-based linked national laboratory record analysis.5 The incidence of bilirubin encephalopathy of 0.6/100 000 in this study is comparable to that of the UK and Danish surveillance studies (0.9 and 0.4/100 000, respectively)2,5 in addition to the findings of a recent population-based study of the incidence of cerebral palsy consistent with kernicterus from Northern California (0.57/100 000).11 Although the estimated incidence of extreme hyperbilirubinemia in Australia is consistent with published surveillance studies from developed countries,
Table III. Characteristics of cases identified in readmitted cases compared with in-hospital cases Variables
Cases readmitted, n = 68
Cases identified in hospital, n = 19
P value
Max serum bilirubin (mmol/L) median (range) Gestation (wk) mean (SD) Birth weight (g) mean (SD) Age at diagnosis (d) median (range) postdischarge surveillance n = 61 Hospital-based midwives Community-based nurses General practitioner Other Referral source n = 64 Hospital-based midwives Community-based nurses General practitioner/pediatrician Other
481 (370-762) 38 (1.3) 3260 (500) 5 (2-16) N (%) 38 (62.3) 20 (32.8) 2 (3.3) 1 (1.6) N (%) 27 (42.1) 17 (26.6) 14 (21.9) 6 (9.4)
496 (454-983) 38 (1.9) 3460 (465) 3 (0-9)
.43 .89 .096 <.001
Prospective Surveillance of Extreme Neonatal Hyperbilirubinemia in Australia
3
THE JOURNAL OF PEDIATRICS
Vol. -, No. -
www.jpeds.com
Table IV. Characteristics of cases with adverse short-term outcomes. Demographic and clinical characteristics of all cases with either peak SBR >600 mmol/L or abnormal discharge neurology or abnormal MRI Sex
Ethnicity
Gestation (wk)
BW (g)
Feeding mode
Max SBR, mmol/L
1 2 3 4
M M M M
Asian – Asian Caucasian
40 38 39 40
3390 3380 3303 3684
Combination Breast Breast Breast
5 6 7 8 9 10 11 12 13
F M M M F M M M M
Caucasian Caucasian Caucasian African Asian Caucasian Caucasian – Aboriginal
40 35 35 38 36 37 37 40 38
2650 2470 3548 3736 3000 2965 2700 3500 2980
Breast – Breast Combination Breast Breast Breast Combination Breast
610 762 546 983 day 12 (conj 632) 630 729 525 612 628 647 669 740 760
Case
Cause
Readm
XT
G6PD G6PD IVH/SH/CH Rh
Day 3 Day 3 Day 5 No
N Y Y N
OxHae OxHae Rh ABO+ Idio ABO G6PD ABO+ Rh
Day 3 Day 12 No No Day 4 Day 4 Day 4 Day 3 No
Y Y N Y N N Y Y N
Associated morbidity Dehyd none Dehyd Hydrops HIE Low Plts none Multisyst none Dehyd none none – none
Neuro at D/C AbN
MRI AbN
Hearing result
Y N N –
Y Y Y Y
BRefer – – –
Y – Y N N N – N N
N Y Y N – – – – –
BRefer BRefer – BPass BPass BPass BPass – BPass
–, unknown; ABO+, DAT positive ABO incompatibility; BPass, bilateral pass result on newborn hearing screen; BRefer, bilateral refer result on newborn hearing screen; CH, large cephalohematoma; conj, conjugated bilirubin; Dehyd, clinically dehydrated; Ethnicity, clinician reported; F, female; HIE, hypoxic ischemic encephalopathy; Idio, idiopathic jaundice; IVH, intraventricular hemorrhage; M, male; MRI, magnetic resonance imaging; Multisyst, multisystem failure; N, no; Neuro at D/C AbN, neurological examination at discharge, abnormal; Plts, platelets; SH, subdural hemorrhage; Y, yes.
the global annual risk of neonatal hyperbilirubinemia-related adverse outcomes is estimated to be 24 million, with mortality rates at their highest in Sub-Saharan Africa and South Asia.12 Clinical associations with extreme hyperbilirubinemia in this population included male sex, non-Caucasian ethnicity (predominantly Asian), vaginal delivery, and (nonexclusive) breastfeeding. The causes and associations of extreme neonatal hyperbilirubinemia were largely in keeping with the findings of previous surveillance studies. Common associations with extreme neonatal hyperbilirubinemia described in the literature are male sex, non-Caucasian ethnicity, hemolytic etiology (most commonly blood-group incompatibility), breastfeeding, and vaginal delivery, although reporting of the wide range of factors influencing neonatal hyperbilirubinemia varies. Comparable short-term outcomes for inpatient and community-managed cases in addition to a lower than expected early exclusive breastfeeding rate were identified among reported cases in this study. It is possible that these observations reflect active early supplemental newborn feeding management in both types of postnatal care setting. Similarly, peak bilirubin levels of inpatient and community-managed cases were found to not differ significantly in the UK, Dutch, and Danish studies, and most cases from the community presented significantly later.2,5,7 The Canadian study, however, found discharged cases had higher bilirubin levels (488 vs 428 mmol/L) in addition to presenting later (5.2 vs 2.6 days).4 The predominance of idiopathic hyperbilirubinemia as the main etiology is likely reflected in the other surveillance studies reporting of cases of “unknown” etiology, ranging from 64% in the Canadian study to 25% in the report from The Netherlands.2,4,5,7 The significance of hemolytic hyperbilirubinemia as a risk factor for extreme hyperbilirubinemia is well 4
documented.2-7,13,14 This study underlines the importance of early identification of hemolytic etiologies and makes the case for screening, particularly for G6PD deficiency, either universal or targeted, in the Australian newborn population. Although routine screening for ABO incompatibility has not been shown conclusively to be an effective strategy in the prevention and management of extreme hyperbilirubinemia, the evidence for G6PD screening is more encouraging.15,16 The introduction of newborn G6PD screening in combination with parental and health professional education programs has been associated with a decreased incidence of extreme hyperbilirubinemia and bilirubin encephalopathy in Greece, Hong Kong, Philippines, Saudi Arabia, Singapore, and Taiwan.17-22 The World Health Organization Working Committee suggested screening be instituted in population groups with a male G6PD incidence of 3%-5% or more.23 Deficient G6PD alleles are distributed worldwide, with a conservative estimate of 400 million affected individuals.24 Highest prevalence populations are reported from Africa, Southern Europe, the Middle-East, South-East Asia, and Central and Southern Pacific Islands.24 On account of Australia’s relative geographical proximity to some of these regions in addition to recent immigration trends, the need for local newborn health professionals to be alert to this risk factor for neonatal bilirubin encephalopathy is a priority. Furthermore, consideration should be given to the introduction of newborn G6PD screening programs in migrant population dense health service areas or, indeed, national universal screening. Moreover, this study confirms that neonatal bilirubin encephalopathy remains a risk in the Australian newborn population and, thus, robust clinical vigilance strategies are essential. The majority (78%) of cases in this study were readmitted from home but were not found to be at increased risk of poorer outcome in terms of peak total bilirubin or short-term neurologic outcome. This is likely due to effective McGillivray et al
- 2015 postdischarge surveillance processes. It is essential that this is maintained and extended to all newborns in Australia. The most recent American Academy of Pediatrics guideline recommends systematic assessment and management of neonatal hyperbilirubinemia, which includes stratification of clinical risk factors in combination with universal predischarge bilirubin screening; either by measuring TSB or transcutaneous bilirubin (TCB).25 TCB monitoring provides a reliable estimate of TSB26 and can be utilized effectively, within the limitations of the devices, as part of a predischarge hyperbilirubinemia assessment tool, particularly when local nomogram data is available, to predict the risk of subsequent hyperbilirubinemia and planning for appropriate follow-up.25 However, all TCB devices must have quality assurance systems in place to ensure their appropriate use and functioning. This includes regular calibration and comparison with local laboratory TSB measurements and training for all users.25 TCB measurements are recommended only in infants who are >35 weeks, >24 hours old, and only pretreatment. TSB measurements should be used outside these parameters, and if treatment thresholds are met or surpassed.27 Additionally, the National Institute for Health and Care Excellence guideline from the UK prioritizes the provision of parental and carer neonatal jaundice information and identification of key risk factors by healthcare providers to optimize early identification of potentially pathological hyperbilirubinemia.27 The limitations of this study include the likely underestimate of incidence due to the voluntary nature of clinician reporting, albeit a monthly reminder sent to all clinicians to report cases. The etiological data is reliant on individual clinicians’ practice, particularly with regard to investigations for hyperbilirubinemia and reporting thereof, and therefore limits the generalizability of our findings. Our conclusions are also limited by the cohort size and incomplete data particularly, with regard to short-term outcomes such as discharge neurologic examination and newborn hearing screening data. The analysis of ethnicity as a risk factor within this cohort was limited by the nature of the clinician-reported data and the lack of an appropriate national comparator. The strengths of this study include its prospective data collection design and high questionnaire return rate. In conclusion, extreme neonatal hyperbilirubinemia remains a serious risk for Australian newborn babies. A national discussion engaging a community perspective is indicated to best define the preventative strategies most relevant to this population. Optimal, universal, systematic, and reliable neonatal hyperbilirubinemia management processes are an essential part of neonatal health services to enable prevention of potentially devastating neurologic consequences. These should include predischarge strategies such as those suggested in the American Academy of Pediatrics13,25 and National Institute for Health and Care Excellence27 guidelines to facilitate the early identification and treatment of hyperbilirubinemia. Combined universal serial TSB/TCB monitoring, early stratification of risk factors, and G6PD
ORIGINAL ARTICLES deficiency screening programs offer effective opportunities to prevent hyperbilirubinemia-associated neurological impairment. Maintenance of effective postdischarge newborn surveillance for all infants is imperative to ensure early discharge from hospital is not a risk for chronic bilirubin encephalopathy. n We would like to thank the APSU, all reporting clinicians, Dr Noel French, Dr Peter Gray, Associate Professor Ross Haslam, Associate Professor Alison Kent, Dr Andrew Watkins, Dr Ingrid Rieger, Dr Adrienne Gordon, Dr Crista Wocadlo, Dr Karen Walker, and Dr Peter Dargaville for their support of this study. Submitted for publication May 17, 2015; last revision received Jul 15, 2015; accepted Aug 24, 2015. Reprint requests: Angela McGillivray, FRACP, MMedEd, Royal Prince Alfred Hospital, Camperdown, NSW, Australia. E-mail: angela.mcgillivray@sswahs. nsw.gov.au
References 1. Bhutani VK, Rodriguez M , HCA Perinatal Safety Initiative Kernicterus Taskforce. Recent epidemiologic indices of extreme hyperbilirubinemia and kernicterus for a diverse multi-state newborn population in USA. PAS 2005;57:1255. 2. Manning D, Todd P, Maxwell M, Platt MJ. Prospective surveillance study of extreme hyperbilirubinemia in the newborn in the UK and Ireland. Arch Dis Child Fetal Neonatal Ed 2007;92:F342-6. 3. Ebbesen F, Andersson C, Verder H, Grytter C, Pedersen-Bjergaard L, Petersen JR, et al. Extreme hyperbilirubinemia in term and near-term infants in Denmark. Acta Paediatr 2005;94:59-64. 4. Sgro M, Campbell D, Shah V. Incidence and causes of extreme neonatal hyperbilirubinemia in Canada. Can Med Assoc J 2006;175:587-90. 5. Bjerre JV, Petersen JR, Ebbesen F. Surveillance of extreme hyperbilirubinemia in Denmark. A method to identify the newborn infants. Acta Paediatr 2008;97:1030-4. 6. Zoubir S, Arlettaz Mieth R, Berrut S, Roth-Kleiner M , for the Swiss Pediatric Surveillance Unit (SPSU). Incidence of extreme hyperbilirubinemia in Switzerland: a nationwide population-based prospective study. Arch Dis Child Fetal Neonatal Ed 2011;96:F310-1. 7. Gotink MJ, Benders MJ, Lavrijsen SW, Rodrigues Pereira R, Hulzebos CV, Dijk PH. Extreme neonatal hyperbilirubinemia in The Netherlands. Neonatology 2013;104:137-42. 8. Hilder L, Zhichao Z, Parker M, Jahan S, Chambers GM. Australia’s mothers and babies 2012. Perinatal statistics series no. 30. Cat. no. PER 69. Canberra: AIHW; 2014. 9. Australian Bureau of Statistics, 2015, Births, Australia, 2013. http://www. abs.gov.au/AUSSTATS/[email protected]/DetailsPage/3301.02013?Open Document. Accessed January 30, 2015. 10. Australian Institute of Health and Welfare. Australia’s health 2014. Australia’s health series no. 14. Cat. no. AUS 178. Canberra: AIHW; 2014. 11. Wu YW, Kuzniewicz MW, Wickresmasinghe AC, Walsh EM, Wi S, McCulloch CE, et al. Risk for cerebral palsy in infants with total serum bilirubin levels at or above the exchange transfusion threshold a population-based study. JAMA Pediatr 2015;169:239-46. 12. Bhutani VK, Zipursky A, Blencowe H, Khanna R, Sgro M, Ebbesen F, et al. Neonatal hyperbilirubinemia: incidence and impairment estimates for 2010 at regional and global levels. Pediatr Res 2013;74(Suppl 1): 86-100. 13. American Academy of Pediatrics, Subcommittee on Hyperbilirubinemia. Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics 2004;114:297-316. 14. Johnson L, Bhutani VK, Karp K, Sivieri EM, Shapiro SM. Clinical report from the pilot USA Kernicterus Registry (1992 to 2004). J Perinatol 2009; 29:S25-45.
Prospective Surveillance of Extreme Neonatal Hyperbilirubinemia in Australia
5
THE JOURNAL OF PEDIATRICS
15. Levine DH, Belton H, Meyer P. Newborn screening for ABO hemolytic disease. Clin Pediatr 1985;24:391-4. 16. Han P, Kiruba R, Ong R, Joseph R, Tan KL, Wong HB. Hemolytic disease due to ABO incompatibility: incidence and value of screening in an Asian population. Aust Paediatr J 1988;24:35-8. 17. Missiou-Tsagaraki S. Screening for glucose-6-phosphate dehydrogenase deficiency as a preventative measure: prevalence among 1 286 000 Greek newborn infants. J Pediatr 1991;119:293-9. 18. Padilla CD, Therell BL. Newborn screening in the Asia Pacific region. J Inherit Metab Dis 2007;30:490-506. 19. Padilla CD. Newborn screening in the Philippines. Southeast Asian J Trop Med Public Health 2003;34(Suppl 3):87-8. 20. Mallouh AA, Imseeh G, Abu-Osba YK, Hamdan JA. Screening for glucose-6-phosphate dehydrogenase deficiency can prevent extreme neonatal hyperbilirubinemia. Ann Trop Pediatr 1992;12:391-5. 21. Joseph R, Ho LY, Gomez JM, Rajdurai VS, Sivasankaran S, Yip YY. Mass newborn screening for glucose-6-phosphate dehydrogenase deficiency in
6
Vol. -, No. -
www.jpeds.com
22.
23. 24. 25.
26.
27.
Singapore. Southeast Asian J Trop Med Public Health 1999;30(Suppl 2): 70-1. Hsaio KJ, Chiang S, Wu KF. Experience of neonatal G6PD deficiency screening in Taiwan. In: Wilcken B, Webster D, eds. Neonatal screening in the nineties. Leura, NSW, Australia: The Kelvin Press; 1991. p. 217-8. WHO Working Group. Glucose-6-phosphate dehydrogenase deficiency. Bull World Health Organ 1989;67:601-11. Cappellini MD, Fiorelli G. Glucose-6-phosphate deficiency. Lancet 2008; 371:64-74. Maisels MJ, Bhutani VK, Bogen DB, Newman TB, Stark AR, Watchko JF. Hyperbilirubinemia in the newborn infant $35 weeks’ gestation: an update with clarifications. Pediatrics 2009;124:1193-8. Nagar G, Vandermeer B, Campbell A, Kumar M. Reliability of transcutaneous bilirubin devices in preterm infants: a systematic review. Pediatrics 2013;132:871-81. NICE. Neonatal Jaundice, https://www.nice.org.uk/guidance/cg98; 2010. Accessed July 14, 2015.
McGillivray et al
- 2015
ORIGINAL ARTICLES SEVERE NEONATAL HYPERBILIRUBINAEMIA OR EXCHANGE TRANSFUSION Australian Paediatric Surveillance Unit If you have any questions about this questionnaire please contact Dr Angela McGillivray on 0403 786298 or A/Prof Nick Evans on [email protected] Tel: +612 9515 8760 Fax: +612 9550 4375 if you wish to discuss this
REPORTING CLINICIAN: 1. APSU Dr Code/Name:
/___________________________________
2. Month/Year of Report:____/___
3. Date questionnaire completed:
PATIENT DETAILS: 4. First 2 letters of first name:
5. First 2 letters of surname:
/
/
/ Time of Birth: _____(24 hr clock) 7. Sex: M F 8. Postcode of family: Aboriginal Torres Strait Islander Both Aboriginal and Torres Strait Islander Caucasian Asian Pacific Islander Middle Eastern African Other __________ 10. Child’s skin colour: Fair Dark Oriental 11. Parents’ country of birth: Mother ________________________ Father ________________________ 6. Date of Birth: 9. Child’s Ethnicity:
/
If this patient is primarily cared for by another physician who you believe will report the case, please complete the questionnaire details above this line and return to APSU. Please keep the patient’s name and other details in your records. If no other report is received for this child we will contact you for information requested in the remainder of the questionnaire. The primary clinician caring for this child is: Name: Hospital: Instructions: Answer each question by ticking the appropriate box or writing your response in the space provided. DK= Don't Know, NA = Not applicable
PERINATAL INFORMATION 12. Gestation _________weeks DK 13. Birthweight __________grams DK 14. Antenatal management of Rhesus disease? Yes No DK If yes, details please: ___________________________________________________________________________ 15. Mode of delivery: Vaginal Vaginal breech Ventouse Forceps Caesarean 16. Where did the birth take place: Hospital Home 17. Apgar scores: 1 minute:__ 5 minutes:___ 10 minutes:___ 18. Arterial cord gas result if available:pH__Base deficit__ 19. Marked bruising eg. large cephalohaematoma: Yes No DK If yes, please give details ___________________________________________________________________________ 20. Mode of feeding in first 2 weeks: Breast only Formula only Combination Describe timings if mode of feeding has changed_________________________________________________________ PRESENTATION and DIAGNOSIS No DK If No Pls go to Q27 21. Was this infant re-admitted with jaundice after initially being discharged? Yes / / If yes, date of initial discharge: If initially discharged from hospital before 48 hours old, please give hours of age at discharge:____________ 22. What kind of postdischarge surveilla nce was there after initial discharge? None Hospital Based Midwifery Discharge Support GP Community Based Nursing or Midwifery Support Paediatrician Other (Specify):_____________________ / / Time: ___________(24 hr clock) 23. Time and date of re-admission Date 24. Source of referral for the re-admission: Hospital Based Midwifery Discharge Support GP Community Based Nursing or Midwifery Support Paediatrician Self referred Other (Specify):__________________________ DK 25. Weight on re-admission:________ (grams) Not weighed Yes No DK Plasma sodium _______mmol/L Not measured DK 26. Dehydrated on re-admission DIAGNOSIS 27. Date and time of diagnosis of severe hyperbilirubinaemia: 28. How was diagnosis confirmed?
/
Total serum bilirubin ≥ 450μmol/L
/
Time: ___________(24 hr clock)
Need for exchange transfusion
Retrospective diagnosis on basis of MRI changes? Retrospective diagnosis on basis of clinical kernicterus? 29. Clinical features at time of severe hyperbilirubinaemia: Lethargy and poor feeding: Yes No DK Hypotonia: Yes No DK Opisthotonus: Yes No DK Convulsions: Yes No DK Other (Specify)__________________________________________________________________________________ 30. Highest Bilirubin result recorded for this infant: ________ μmol/L 31. Total duration of elevated bilirubin ≥ 450 μmol/L: _______(hrs and minutes) 32. Serum bilirubin results PRE TREATMENT(μmol/L). Please include all results before any treatment was commenced. Please attach de-identified printout of all serum bilirubin results if available. DATE
TIME (24 hr clock)
TOTAL BILIRUBIN μmol/L
DATE
TIME (24 hr clock)
TOTAL BILIRUBIN μmol/L
Appendix. Australian Paediatric Surveillance Unit Neonatal Hyperbilirubinaemia Reporting Paediatrician Questionnaire. (Continues)
Prospective Surveillance of Extreme Neonatal Hyperbilirubinemia in Australia
6.e1
THE JOURNAL OF PEDIATRICS
Vol. -, No. -
www.jpeds.com
33. Serum bilirubin results POST TREATMENT(μmol/L). Please include all results after treatment commenced. Please attach de-identified printout of all serum bilirubin results if available. DATE
TIME (24 hr clock)
TOTAL BILIRUBIN μmol/L
DATE
TIME (24 hr clock)
TOTAL BILIRUBIN μmol/L
34. Associated dehydration with hyperbilirubinaemia? Yes No DK / / Time: ___________(24 hour clock) Plasma sodium _______mmol/L Not measured Date No DK 35. Did the infant have associated culture positive systemic infection? Yes If Yes, SITE: ___________________________ ORGANISM: ____________________________________________ / / Time: ___________(24 hour clock) Not measured DK 36. Lowest albumin level______ g/l Date 37. Lowest blood pH ________ Date / / Time: ___________(24 hour clock) Not measured DK 38. Did the infant have other serious morbidity? Yes No DK If Yes, Please specify (eg. hypoxic-ischaemic encephalopathy, hypoglycaemia, liver disease)_________________________________________________________ 39. Cause of hyperbilirubinaemia: Physiological ABO incompatibility: Probable (Coombs positive) Possible (Coombs negative) Rhesus isoimmunisation Glucose-6-phosphate dehydrogenase deficiency Other _______________________________________________________________________________________ MANAGEMENT 40. Phototherapy used? DATE
Yes
Start (24 hr clock)
No
DK
If Yes, please record times and dates:
End (24 hr clock)
DATE
Start (24 hr clock)
End (24 hr clock)
41. Albumin infusion? Yes No DK If yes, time and date commenced______________________ 42. Immunoglobulin infusion Yes No DK If yes, number of doses __________ 43. Exchange transfusion Yes No DK If yes, how many ________________ Time and dates _________________________________________________________________________________ 44. Was magnetic resonance imaging brain scan done? Yes No DK If Yes, Date / / and Result: Normal Increased signal on T2-weighted images in globus pallidus Abnormal but not consistent with bilirubin toxicity, (specify): ________________________________________ Other: (specify): __________________________________________________________________________ OUTCOME 45. Did the baby survive Yes No DK If Yes, date of discharge: / / No DK Serum bilirubin closest to discharge: _____ μmol/L Neurological status normal at discharge? Yes No DK If Yes, Result:________________________________________________ Hearing screen done? Yes If baby died, date of death: / / Was a post-mortem conducted Yes No DK No DK Or other associated pathology Yes No DK If Yes, did the post-mortem show kernicterus Yes If Yes, (specify): ___________________________________________________________________________________ FOLLOW UP TRACKING INFORMATION Please provide details of the physician from whom follow-up information can be obtained: Please print Name:_________________________ Phone No:______________ E-mail:________________ Do you have any other comments about this infant? ________________________________________________________________________________________________ ________________________________________________________________________________________________ THANK YOU FOR COMPLETING THIS QUESTIONNAIRE Please return this questionnaire in the addressed reply-paid envelope to: Dr Angela M Gillivray, Clinical Neonatology Fellow, Newborn Care, Royal Prince Alfred Hospital, Missenden Rd, Camperdown, NSW 2050, Australia.
Appendix. (Continued).
6.e2
McGillivray et al
- 2015
ORIGINAL ARTICLES
Table V. Univariate analysis of severe neonatal hyperbilirubinemia with peak levels over 500 mmol/L Risk factors for SBR >500 mmol/L Sex Male Female Ethnicity Asian Other Cause of hyperbilirubinemia Non-hemolytic Hemolytic* Gestational age Term Preterm Feeding mode Breast Formula Combination Readmission Inpatient Readmitted
n (%)
P value .422
18/50 (36) 10/36 (27.8) .397 14/34 (41) 12/38 (31.6) <.002 10/54 (18.5) 15/27 (55.6) .2 23/77 (29.9) 5/10 (50) .309 17/58 (29.3) 1/1 (100) 9/27 (33.3) .295 8/19 (42.1) 20/68 (29.4)
Oxhem, oxidative hemolysis; Rh, Rhesus isoimmunization; SBR, serum bilirubin. *Hemolytic causes: Rh, glucose-6-phosphate deficiency, hereditary spherocytosis, Oxhem, and ABO blood group incompatibility with DAT+.
Prospective Surveillance of Extreme Neonatal Hyperbilirubinemia in Australia
6.e3
Study Design This was a prospective population-based surveillance study in collaboration with the Australian Pediatric Surveillance Unit between April 1, 2010, and March 31, 2013. Case definition was: infants >34 weeks gestation with a peak total serum bilirubin $450 mmol/L and or clinical evidence of bilirubin encephalopathy. Clinicians completed questionnaires detailing demographic and clinical data including: peak serum bilirubin, signs of bilirubin encephalopathy, etiology, associated pathology, management, and short-term outcomes. Results The questionnaire return rate was 95%, and 87 infants met the case definition. The Australian incidence of extreme neonatal hyperbilirubinemia is estimated to be 9.4/100 000 live births. Main etiologies were: idiopathic ABO blood group incompatibility, glucose-6-phosphate dehydrogenase deficiency, and Rhesus isoimmunization. There were no significant differences in short-term outcomes between inpatient and outpatient cases. Cases with a hemolytic etiology were significantly more likely to have extremely high levels of hyperbilirubinemia (P < .002). Conclusion The incidence of extreme neonatal hyperbilirubinemia in Australia is comparable with previous studies. Robust pre- and post-discharge assessment and management strategies of neonatal hyperbilirubinemia are essential to prevent neurodisability. Universal glucose-6-phosphate dehydrogenase screening and serial bilirubin monitoring may optimize preventative strategies. (J Pediatr 2015;-:---). See editorial, p
E
xtreme neonatal hyperbilirubinemia, although rare, can cause neonatal bilirubin encephalopathy and permanent impairment with choreoathetoid cerebral palsy and deafness. It is a condition that in most circumstances can be prevented. There have been recent international concerns regarding a possible reemergence of neonatal bilirubin encephalopathy. This has resulted in the publication of 6 surveillance studies, all in developed countries in the northern hemisphere.1-7 Notwithstanding methodological and case definition variation, voluntary clinician case-reporting in collaboration with local rare disease surveillance units has been the predominant approach. Extreme neonatal hyperbilirubinemia is variably defined in these studies by peak bilirubin levels in the range of 340-510 mmol/L. Overall, the incidence internationally is estimated between 2 and 53 cases per 100 000 live births.1-7 After a cluster of cases of bilirubin encephalopathy locally, it was recognized that there was a lack of incidence data in Australia on which to base a discussion about preventative strategies. Our aim was to determine the incidence of extreme neonatal hyperbilirubinemia in Australia, to ascertain causes and associated risk factors and thus identify opportunities for enhanced future preventative strategies. This will include a neurodevelopmental follow-up study to establish the important longer-term outcomes of this population. The surveillance stage is reported in this paper.
Methods This prospective surveillance study utilized voluntary clinician reporting of cases in collaboration with the Australian Pediatric Surveillance Unit (APSU) from April 1, 2010, to March 31, 2013. The APSU was established in 1993 to facilitate the reporting and surveillance of rare conditions and infectious disease outbreaks. Approximately 1400 clinicians (predominantly pediatricians) are contacted on a monthly basis, primarily by e-mail, and requested to make reports according to the From the Royal Prince Alfred Hospital, Camperdown; 1
2
The University of Sydney, Sydney; 3The Children’s Hospital at Westmead, Westmead; and 4The University of Notre Dame, Darlinghurst, New South Wales, Australia
APSU G6PD TCB TSB
Australian Pediatric Surveillance Unit Glucose-6-phosphate dehydrogenase Transcutaneous bilirubin Total serum bilirubin
The Cerebral Palsy Alliance provided funding for the Australian Pediatric Surveillance Unit costs (IRG2009 [to N.E.]). The authors declare no conflicts of interest. 0022-3476/$ - see front matter. Copyright ª 2015 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jpeds.2015.08.048
1
THE JOURNAL OF PEDIATRICS
Vol. -, No. -
www.jpeds.com
contemporary conditions of interest. The study was approved by the Sydney Local Health District Human Research Ethics Committee. Cases were defined as: an infant of 34 weeks gestation or more with a peak total serum bilirubin (TSB) $450 mmol/L in the first month, or with clinical and/or magnetic resonance imaging consistent with bilirubin toxicity. On notification of a case, reporting clinicians received a de-identified questionnaire requesting demographic and relevant clinical data including sex, ethnicity, gestation, delivery mode, feeding methods, presentation and discharge timing, peak TSB levels, signs of bilirubin encephalopathy, etiology, associated pathology, and treatment (Appendix; available at www.jpeds.com). Subsequently, the research team contacted clinicians who had reported cases by e-mail, letter, and/or telephone as necessary to expedite and optimize questionnaire return. Data were held securely both digitally and in hardcopy at the primary research site. Accuracy of database information was ensured by a triple check of all questionnaire data entry by two members of the research team. The incidence calculation denominator was from the Australian Institute of Health and Welfare8 and the Australian Bureau of Statistics.9 The annual Australia’s Mothers and Babies report produced by the Australian Institute of Health and Welfare provides accurate, national, gestational week-specific live birth data; thus, the most current available data (up to December 2012) was used for incidence calculation. Thereafter, data from the Australian Bureau of Statistics was extrapolated to complete the epoch. Data were summarized by descriptive statistics. Comparisons between groups were performed using Wilcoxon rank sum test, Pearson c2, or Fisher exact test where appropriate. The analyses were done using IBM SPSS Statistics version 22 (IBM, Armonk, New York).
Results
Neither of the babies with seizures was reported as having tonal or feeding abnormalities. Based on the 5 cases presenting with opisthotonus, the incidence of bilirubin encephalopathy is 0.6/100 000 live births. The median peak bilirubin level was 483 mmol/L (range 370-983). There were 56 cases (64%) with peak bilirubin levels between 450 and 500 mmol/L, 19 (22%) between 500 and 600 mmol/L, and 11 (13%) above 600 mmol/L. The national distribution of reported cases reflects the geographical general population distribution of Australia, with the majority of cases in the most populous states of New South Wales and Victoria. Demographic and perinatal variable data of the 87 cases is summarized in Table I. Fifty cases were male (58%) with the sex of one case not reported. Clinician-reported ethnicity was provided in 72 reports. Of these, 69% of cases were reported as non-Caucasian. Eight cases were born by cesarean (9.3%) compared with a national average of 32%.10 With regards to primary mode of feeding, 67% were exclusively breastfed in the first 2 weeks compared with a national average that ranges between 61% and 90%.10 Table II summarizes the underlying conditions and treatments. The majority (48.6%) of cases were caused, as determined by the reporting clinicians, by idiopathic hyperbilirubinemia. Thirty-six percent of cases had an underlying hemolytic etiology, which included: Rhesus D isoimmunization, direct antiglobulin test positive ABO-incompatibility, glucose-6-phosphate dehydrogenase (G6PD) deficiency, hereditary spherocytosis, and oxidative hemolysis. Clinicians documented 32% of cases had associated clinical signs of dehydration based on their clinical impression. Thirty-six percent of cases had hypoalbuminemia (<30 g/L) and 16% of the overall group received albumin
Table I. Demographic and perinatal variables of 87 cases Variables*
Over the 3-year period, 159 notifications of extreme neonatal hyperbilirubinemia were made by Australian pediatric clinicians. Of these, 150 notifications yielded questionnaires; equating to a 95% return rate. Eighty-seven reported patients met the case definition: 86 neonates had peak bilirubin levels $450 mmol/L and one case had a peak bilirubin level of 370 mmol/L but presented with clinical signs consistent with bilirubin encephalopathy, which resolved following phototherapy. There were 28 reports that did not meet the inclusion criteria and 35 reports were duplicates. There were 893 693 estimated live births over 34 weeks gestation in Australia between April 1, 2010, and March 31, 2013.8,9 Thus, the current incidence of extreme neonatal hyperbilirubinemia in Australia is calculated to be 9.4/100 000 live births. Of the 87 cases, 34 infants had lethargy and poor feeding (79 responses), 3 had hypotonia, 1 in association with lethargy and poor feeding (72 responses), 5 had opisthotonus (72 responses), and 2 had convulsions (71 responses). 2
Sex n = 86 Male Ethnicity (clinician determined) n = 72 Asian Caucasian Middle Eastern Aboriginal Pacific Islander African Other Birth weight (g), mean (SD) n = 85 Gestational age (wk), mean (SD) n = 87 Mode of delivery in hospital n = 86 Normal vaginal Vaginal breech Instrumental Cesarean Feeding method n = 86 Breast Artificial Combination Readmitted n = 87
N (%) 50 (58.1) 38 (52.8) 20 (27.8) 5 (6.9) 3 (4.2) 3 (4.2) 2 (2.8) 1 (1.3) 3298 (498) 38 (1.5) 53 (61.7) 2 (2.3) 23 (26.7) 8 (9.3) 58 (67.4) 1 (1.2) 27 (31.4) 68 (78.2)
*Values are numbers (%) unless otherwise stated.
McGillivray et al
- 2015
ORIGINAL ARTICLES
Table II. Causes and treatment of severe hyperbilirubinemia N (%) Idiopathic Probable ABO incompatibility (DAT positive) Possible ABO incompatibility (DAT negative) G6PD deficiency Rh Other hematological (spherocytosis, Oxhem) Other (sepsis, cephalohematoma) Unknown Associated factors Dehydration n = 84 Low albumin (<30 g/L) n = 53 Culture positive infection n = 86 Other serious associated morbidity n = 84 Treatment Phototherapy n = 85 Albumin n = 81 Immunoglobulin n = 81 XT n = 84
37 (48.7) 10 (13.2) 7 (9.2) 7 (9.2) 3 (3.9) 7 (9.2) 5 (6.6) 11 27 (32.1) 19 (35.8) 7 (8.1) 4 (4.8) 84 (98.8) 13 (16.0) 7 (8.6) 17 (20.2)
DAT, direct antiglobulin test; Oxhem, oxidative hemolysis; Rh, Rhesus isoimmunization; XT, exchange transfusion.
treatment. Almost all infants (98.8%) received phototherapy treatment and 20% received at least one exchange transfusion. All cases were born in hospital. Extreme hyperbilirubinemia was identified in 78% of cases postdischarge, primarily by hospital- and community-based nurses. Table III presents the characteristics of cases either diagnosed as inpatients or outpatients of their birth hospital. Discharged infants were diagnosed significantly later at 5 days (range 2-16 days) compared with 3 days (range 0-9 days) for cases identified in hospital (P < .001). However, there was no significant difference in bilirubin levels between the cases identified in hospital compared with those diagnosed postdischarge: 496 vs 481 mmol/L (P = 049). All eighty-seven cases survived to discharge. Of the 68 whose discharge neurologic examination was reported, 3 were abnormal. The results of newborn hearing screening were provided for 76 cases, of which 4 required further audiological assessment. Magnetic resonance brain imaging was
carried out in only 11 of the 87 cases. Four patients had scans consistent with bilirubin toxicity, a further 3 were abnormal but not consistent with bilirubin toxicity, and the remaining 4 were normal. Table IV provides a summary of the 13 cases with known adverse short-term outcomes, their underlying etiology, treatment, and short-term outcomes. A neurodevelopmental follow-up study to the age of 3 years is underway to ascertain longer-term outcomes in the study cohort. Univariate analyses for known risk factors and peak bilirubin levels over 500 mmol/L were conducted and are summarized in Table V (available at www.jpeds.com). Infants with an underlying hemolytic cause for their extreme hyperbilirubinemia were significantly more likely to have bilirubin levels greater than 500 mmol/L compared with those with nonhemolytic hyperbilirubinemia (P < .002). There were no significant differences in the distribution of sex, clinician-reported ethnicity, gestational age, feeding mode, and readmission rates between cases with maximum peak bilirubin levels greater or equal to 500 mmol/L.
Discussion The incidence of extreme neonatal hyperbilirubinemia is estimated to be 9.4/100 000 live births and is comparable with previous surveillance studies using voluntary clinician report methodology.2-6 The study by Bjerre et al from Denmark likely provides the most accurate incidence estimate of 45/ 100 000 live births from their population-based linked national laboratory record analysis.5 The incidence of bilirubin encephalopathy of 0.6/100 000 in this study is comparable to that of the UK and Danish surveillance studies (0.9 and 0.4/100 000, respectively)2,5 in addition to the findings of a recent population-based study of the incidence of cerebral palsy consistent with kernicterus from Northern California (0.57/100 000).11 Although the estimated incidence of extreme hyperbilirubinemia in Australia is consistent with published surveillance studies from developed countries,
Table III. Characteristics of cases identified in readmitted cases compared with in-hospital cases Variables
Cases readmitted, n = 68
Cases identified in hospital, n = 19
P value
Max serum bilirubin (mmol/L) median (range) Gestation (wk) mean (SD) Birth weight (g) mean (SD) Age at diagnosis (d) median (range) postdischarge surveillance n = 61 Hospital-based midwives Community-based nurses General practitioner Other Referral source n = 64 Hospital-based midwives Community-based nurses General practitioner/pediatrician Other
481 (370-762) 38 (1.3) 3260 (500) 5 (2-16) N (%) 38 (62.3) 20 (32.8) 2 (3.3) 1 (1.6) N (%) 27 (42.1) 17 (26.6) 14 (21.9) 6 (9.4)
496 (454-983) 38 (1.9) 3460 (465) 3 (0-9)
.43 .89 .096 <.001
Prospective Surveillance of Extreme Neonatal Hyperbilirubinemia in Australia
3
THE JOURNAL OF PEDIATRICS
Vol. -, No. -
www.jpeds.com
Table IV. Characteristics of cases with adverse short-term outcomes. Demographic and clinical characteristics of all cases with either peak SBR >600 mmol/L or abnormal discharge neurology or abnormal MRI Sex
Ethnicity
Gestation (wk)
BW (g)
Feeding mode
Max SBR, mmol/L
1 2 3 4
M M M M
Asian – Asian Caucasian
40 38 39 40
3390 3380 3303 3684
Combination Breast Breast Breast
5 6 7 8 9 10 11 12 13
F M M M F M M M M
Caucasian Caucasian Caucasian African Asian Caucasian Caucasian – Aboriginal
40 35 35 38 36 37 37 40 38
2650 2470 3548 3736 3000 2965 2700 3500 2980
Breast – Breast Combination Breast Breast Breast Combination Breast
610 762 546 983 day 12 (conj 632) 630 729 525 612 628 647 669 740 760
Case
Cause
Readm
XT
G6PD G6PD IVH/SH/CH Rh
Day 3 Day 3 Day 5 No
N Y Y N
OxHae OxHae Rh ABO+ Idio ABO G6PD ABO+ Rh
Day 3 Day 12 No No Day 4 Day 4 Day 4 Day 3 No
Y Y N Y N N Y Y N
Associated morbidity Dehyd none Dehyd Hydrops HIE Low Plts none Multisyst none Dehyd none none – none
Neuro at D/C AbN
MRI AbN
Hearing result
Y N N –
Y Y Y Y
BRefer – – –
Y – Y N N N – N N
N Y Y N – – – – –
BRefer BRefer – BPass BPass BPass BPass – BPass
–, unknown; ABO+, DAT positive ABO incompatibility; BPass, bilateral pass result on newborn hearing screen; BRefer, bilateral refer result on newborn hearing screen; CH, large cephalohematoma; conj, conjugated bilirubin; Dehyd, clinically dehydrated; Ethnicity, clinician reported; F, female; HIE, hypoxic ischemic encephalopathy; Idio, idiopathic jaundice; IVH, intraventricular hemorrhage; M, male; MRI, magnetic resonance imaging; Multisyst, multisystem failure; N, no; Neuro at D/C AbN, neurological examination at discharge, abnormal; Plts, platelets; SH, subdural hemorrhage; Y, yes.
the global annual risk of neonatal hyperbilirubinemia-related adverse outcomes is estimated to be 24 million, with mortality rates at their highest in Sub-Saharan Africa and South Asia.12 Clinical associations with extreme hyperbilirubinemia in this population included male sex, non-Caucasian ethnicity (predominantly Asian), vaginal delivery, and (nonexclusive) breastfeeding. The causes and associations of extreme neonatal hyperbilirubinemia were largely in keeping with the findings of previous surveillance studies. Common associations with extreme neonatal hyperbilirubinemia described in the literature are male sex, non-Caucasian ethnicity, hemolytic etiology (most commonly blood-group incompatibility), breastfeeding, and vaginal delivery, although reporting of the wide range of factors influencing neonatal hyperbilirubinemia varies. Comparable short-term outcomes for inpatient and community-managed cases in addition to a lower than expected early exclusive breastfeeding rate were identified among reported cases in this study. It is possible that these observations reflect active early supplemental newborn feeding management in both types of postnatal care setting. Similarly, peak bilirubin levels of inpatient and community-managed cases were found to not differ significantly in the UK, Dutch, and Danish studies, and most cases from the community presented significantly later.2,5,7 The Canadian study, however, found discharged cases had higher bilirubin levels (488 vs 428 mmol/L) in addition to presenting later (5.2 vs 2.6 days).4 The predominance of idiopathic hyperbilirubinemia as the main etiology is likely reflected in the other surveillance studies reporting of cases of “unknown” etiology, ranging from 64% in the Canadian study to 25% in the report from The Netherlands.2,4,5,7 The significance of hemolytic hyperbilirubinemia as a risk factor for extreme hyperbilirubinemia is well 4
documented.2-7,13,14 This study underlines the importance of early identification of hemolytic etiologies and makes the case for screening, particularly for G6PD deficiency, either universal or targeted, in the Australian newborn population. Although routine screening for ABO incompatibility has not been shown conclusively to be an effective strategy in the prevention and management of extreme hyperbilirubinemia, the evidence for G6PD screening is more encouraging.15,16 The introduction of newborn G6PD screening in combination with parental and health professional education programs has been associated with a decreased incidence of extreme hyperbilirubinemia and bilirubin encephalopathy in Greece, Hong Kong, Philippines, Saudi Arabia, Singapore, and Taiwan.17-22 The World Health Organization Working Committee suggested screening be instituted in population groups with a male G6PD incidence of 3%-5% or more.23 Deficient G6PD alleles are distributed worldwide, with a conservative estimate of 400 million affected individuals.24 Highest prevalence populations are reported from Africa, Southern Europe, the Middle-East, South-East Asia, and Central and Southern Pacific Islands.24 On account of Australia’s relative geographical proximity to some of these regions in addition to recent immigration trends, the need for local newborn health professionals to be alert to this risk factor for neonatal bilirubin encephalopathy is a priority. Furthermore, consideration should be given to the introduction of newborn G6PD screening programs in migrant population dense health service areas or, indeed, national universal screening. Moreover, this study confirms that neonatal bilirubin encephalopathy remains a risk in the Australian newborn population and, thus, robust clinical vigilance strategies are essential. The majority (78%) of cases in this study were readmitted from home but were not found to be at increased risk of poorer outcome in terms of peak total bilirubin or short-term neurologic outcome. This is likely due to effective McGillivray et al
- 2015 postdischarge surveillance processes. It is essential that this is maintained and extended to all newborns in Australia. The most recent American Academy of Pediatrics guideline recommends systematic assessment and management of neonatal hyperbilirubinemia, which includes stratification of clinical risk factors in combination with universal predischarge bilirubin screening; either by measuring TSB or transcutaneous bilirubin (TCB).25 TCB monitoring provides a reliable estimate of TSB26 and can be utilized effectively, within the limitations of the devices, as part of a predischarge hyperbilirubinemia assessment tool, particularly when local nomogram data is available, to predict the risk of subsequent hyperbilirubinemia and planning for appropriate follow-up.25 However, all TCB devices must have quality assurance systems in place to ensure their appropriate use and functioning. This includes regular calibration and comparison with local laboratory TSB measurements and training for all users.25 TCB measurements are recommended only in infants who are >35 weeks, >24 hours old, and only pretreatment. TSB measurements should be used outside these parameters, and if treatment thresholds are met or surpassed.27 Additionally, the National Institute for Health and Care Excellence guideline from the UK prioritizes the provision of parental and carer neonatal jaundice information and identification of key risk factors by healthcare providers to optimize early identification of potentially pathological hyperbilirubinemia.27 The limitations of this study include the likely underestimate of incidence due to the voluntary nature of clinician reporting, albeit a monthly reminder sent to all clinicians to report cases. The etiological data is reliant on individual clinicians’ practice, particularly with regard to investigations for hyperbilirubinemia and reporting thereof, and therefore limits the generalizability of our findings. Our conclusions are also limited by the cohort size and incomplete data particularly, with regard to short-term outcomes such as discharge neurologic examination and newborn hearing screening data. The analysis of ethnicity as a risk factor within this cohort was limited by the nature of the clinician-reported data and the lack of an appropriate national comparator. The strengths of this study include its prospective data collection design and high questionnaire return rate. In conclusion, extreme neonatal hyperbilirubinemia remains a serious risk for Australian newborn babies. A national discussion engaging a community perspective is indicated to best define the preventative strategies most relevant to this population. Optimal, universal, systematic, and reliable neonatal hyperbilirubinemia management processes are an essential part of neonatal health services to enable prevention of potentially devastating neurologic consequences. These should include predischarge strategies such as those suggested in the American Academy of Pediatrics13,25 and National Institute for Health and Care Excellence27 guidelines to facilitate the early identification and treatment of hyperbilirubinemia. Combined universal serial TSB/TCB monitoring, early stratification of risk factors, and G6PD
ORIGINAL ARTICLES deficiency screening programs offer effective opportunities to prevent hyperbilirubinemia-associated neurological impairment. Maintenance of effective postdischarge newborn surveillance for all infants is imperative to ensure early discharge from hospital is not a risk for chronic bilirubin encephalopathy. n We would like to thank the APSU, all reporting clinicians, Dr Noel French, Dr Peter Gray, Associate Professor Ross Haslam, Associate Professor Alison Kent, Dr Andrew Watkins, Dr Ingrid Rieger, Dr Adrienne Gordon, Dr Crista Wocadlo, Dr Karen Walker, and Dr Peter Dargaville for their support of this study. Submitted for publication May 17, 2015; last revision received Jul 15, 2015; accepted Aug 24, 2015. Reprint requests: Angela McGillivray, FRACP, MMedEd, Royal Prince Alfred Hospital, Camperdown, NSW, Australia. E-mail: angela.mcgillivray@sswahs. nsw.gov.au
References 1. Bhutani VK, Rodriguez M , HCA Perinatal Safety Initiative Kernicterus Taskforce. Recent epidemiologic indices of extreme hyperbilirubinemia and kernicterus for a diverse multi-state newborn population in USA. PAS 2005;57:1255. 2. Manning D, Todd P, Maxwell M, Platt MJ. Prospective surveillance study of extreme hyperbilirubinemia in the newborn in the UK and Ireland. Arch Dis Child Fetal Neonatal Ed 2007;92:F342-6. 3. Ebbesen F, Andersson C, Verder H, Grytter C, Pedersen-Bjergaard L, Petersen JR, et al. Extreme hyperbilirubinemia in term and near-term infants in Denmark. Acta Paediatr 2005;94:59-64. 4. Sgro M, Campbell D, Shah V. Incidence and causes of extreme neonatal hyperbilirubinemia in Canada. Can Med Assoc J 2006;175:587-90. 5. Bjerre JV, Petersen JR, Ebbesen F. Surveillance of extreme hyperbilirubinemia in Denmark. A method to identify the newborn infants. Acta Paediatr 2008;97:1030-4. 6. Zoubir S, Arlettaz Mieth R, Berrut S, Roth-Kleiner M , for the Swiss Pediatric Surveillance Unit (SPSU). Incidence of extreme hyperbilirubinemia in Switzerland: a nationwide population-based prospective study. Arch Dis Child Fetal Neonatal Ed 2011;96:F310-1. 7. Gotink MJ, Benders MJ, Lavrijsen SW, Rodrigues Pereira R, Hulzebos CV, Dijk PH. Extreme neonatal hyperbilirubinemia in The Netherlands. Neonatology 2013;104:137-42. 8. Hilder L, Zhichao Z, Parker M, Jahan S, Chambers GM. Australia’s mothers and babies 2012. Perinatal statistics series no. 30. Cat. no. PER 69. Canberra: AIHW; 2014. 9. Australian Bureau of Statistics, 2015, Births, Australia, 2013. http://www. abs.gov.au/AUSSTATS/[email protected]/DetailsPage/3301.02013?Open Document. Accessed January 30, 2015. 10. Australian Institute of Health and Welfare. Australia’s health 2014. Australia’s health series no. 14. Cat. no. AUS 178. Canberra: AIHW; 2014. 11. Wu YW, Kuzniewicz MW, Wickresmasinghe AC, Walsh EM, Wi S, McCulloch CE, et al. Risk for cerebral palsy in infants with total serum bilirubin levels at or above the exchange transfusion threshold a population-based study. JAMA Pediatr 2015;169:239-46. 12. Bhutani VK, Zipursky A, Blencowe H, Khanna R, Sgro M, Ebbesen F, et al. Neonatal hyperbilirubinemia: incidence and impairment estimates for 2010 at regional and global levels. Pediatr Res 2013;74(Suppl 1): 86-100. 13. American Academy of Pediatrics, Subcommittee on Hyperbilirubinemia. Management of hyperbilirubinemia in the newborn infant 35 or more weeks of gestation. Pediatrics 2004;114:297-316. 14. Johnson L, Bhutani VK, Karp K, Sivieri EM, Shapiro SM. Clinical report from the pilot USA Kernicterus Registry (1992 to 2004). J Perinatol 2009; 29:S25-45.
Prospective Surveillance of Extreme Neonatal Hyperbilirubinemia in Australia
5
THE JOURNAL OF PEDIATRICS
15. Levine DH, Belton H, Meyer P. Newborn screening for ABO hemolytic disease. Clin Pediatr 1985;24:391-4. 16. Han P, Kiruba R, Ong R, Joseph R, Tan KL, Wong HB. Hemolytic disease due to ABO incompatibility: incidence and value of screening in an Asian population. Aust Paediatr J 1988;24:35-8. 17. Missiou-Tsagaraki S. Screening for glucose-6-phosphate dehydrogenase deficiency as a preventative measure: prevalence among 1 286 000 Greek newborn infants. J Pediatr 1991;119:293-9. 18. Padilla CD, Therell BL. Newborn screening in the Asia Pacific region. J Inherit Metab Dis 2007;30:490-506. 19. Padilla CD. Newborn screening in the Philippines. Southeast Asian J Trop Med Public Health 2003;34(Suppl 3):87-8. 20. Mallouh AA, Imseeh G, Abu-Osba YK, Hamdan JA. Screening for glucose-6-phosphate dehydrogenase deficiency can prevent extreme neonatal hyperbilirubinemia. Ann Trop Pediatr 1992;12:391-5. 21. Joseph R, Ho LY, Gomez JM, Rajdurai VS, Sivasankaran S, Yip YY. Mass newborn screening for glucose-6-phosphate dehydrogenase deficiency in
6
Vol. -, No. -
www.jpeds.com
22.
23. 24. 25.
26.
27.
Singapore. Southeast Asian J Trop Med Public Health 1999;30(Suppl 2): 70-1. Hsaio KJ, Chiang S, Wu KF. Experience of neonatal G6PD deficiency screening in Taiwan. In: Wilcken B, Webster D, eds. Neonatal screening in the nineties. Leura, NSW, Australia: The Kelvin Press; 1991. p. 217-8. WHO Working Group. Glucose-6-phosphate dehydrogenase deficiency. Bull World Health Organ 1989;67:601-11. Cappellini MD, Fiorelli G. Glucose-6-phosphate deficiency. Lancet 2008; 371:64-74. Maisels MJ, Bhutani VK, Bogen DB, Newman TB, Stark AR, Watchko JF. Hyperbilirubinemia in the newborn infant $35 weeks’ gestation: an update with clarifications. Pediatrics 2009;124:1193-8. Nagar G, Vandermeer B, Campbell A, Kumar M. Reliability of transcutaneous bilirubin devices in preterm infants: a systematic review. Pediatrics 2013;132:871-81. NICE. Neonatal Jaundice, https://www.nice.org.uk/guidance/cg98; 2010. Accessed July 14, 2015.
McGillivray et al
- 2015
ORIGINAL ARTICLES SEVERE NEONATAL HYPERBILIRUBINAEMIA OR EXCHANGE TRANSFUSION Australian Paediatric Surveillance Unit If you have any questions about this questionnaire please contact Dr Angela McGillivray on 0403 786298 or A/Prof Nick Evans on [email protected] Tel: +612 9515 8760 Fax: +612 9550 4375 if you wish to discuss this
REPORTING CLINICIAN: 1. APSU Dr Code/Name:
/___________________________________
2. Month/Year of Report:____/___
3. Date questionnaire completed:
PATIENT DETAILS: 4. First 2 letters of first name:
5. First 2 letters of surname:
/
/
/ Time of Birth: _____(24 hr clock) 7. Sex: M F 8. Postcode of family: Aboriginal Torres Strait Islander Both Aboriginal and Torres Strait Islander Caucasian Asian Pacific Islander Middle Eastern African Other __________ 10. Child’s skin colour: Fair Dark Oriental 11. Parents’ country of birth: Mother ________________________ Father ________________________ 6. Date of Birth: 9. Child’s Ethnicity:
/
If this patient is primarily cared for by another physician who you believe will report the case, please complete the questionnaire details above this line and return to APSU. Please keep the patient’s name and other details in your records. If no other report is received for this child we will contact you for information requested in the remainder of the questionnaire. The primary clinician caring for this child is: Name: Hospital: Instructions: Answer each question by ticking the appropriate box or writing your response in the space provided. DK= Don't Know, NA = Not applicable
PERINATAL INFORMATION 12. Gestation _________weeks DK 13. Birthweight __________grams DK 14. Antenatal management of Rhesus disease? Yes No DK If yes, details please: ___________________________________________________________________________ 15. Mode of delivery: Vaginal Vaginal breech Ventouse Forceps Caesarean 16. Where did the birth take place: Hospital Home 17. Apgar scores: 1 minute:__ 5 minutes:___ 10 minutes:___ 18. Arterial cord gas result if available:pH__Base deficit__ 19. Marked bruising eg. large cephalohaematoma: Yes No DK If yes, please give details ___________________________________________________________________________ 20. Mode of feeding in first 2 weeks: Breast only Formula only Combination Describe timings if mode of feeding has changed_________________________________________________________ PRESENTATION and DIAGNOSIS No DK If No Pls go to Q27 21. Was this infant re-admitted with jaundice after initially being discharged? Yes / / If yes, date of initial discharge: If initially discharged from hospital before 48 hours old, please give hours of age at discharge:____________ 22. What kind of postdischarge surveilla nce was there after initial discharge? None Hospital Based Midwifery Discharge Support GP Community Based Nursing or Midwifery Support Paediatrician Other (Specify):_____________________ / / Time: ___________(24 hr clock) 23. Time and date of re-admission Date 24. Source of referral for the re-admission: Hospital Based Midwifery Discharge Support GP Community Based Nursing or Midwifery Support Paediatrician Self referred Other (Specify):__________________________ DK 25. Weight on re-admission:________ (grams) Not weighed Yes No DK Plasma sodium _______mmol/L Not measured DK 26. Dehydrated on re-admission DIAGNOSIS 27. Date and time of diagnosis of severe hyperbilirubinaemia: 28. How was diagnosis confirmed?
/
Total serum bilirubin ≥ 450μmol/L
/
Time: ___________(24 hr clock)
Need for exchange transfusion
Retrospective diagnosis on basis of MRI changes? Retrospective diagnosis on basis of clinical kernicterus? 29. Clinical features at time of severe hyperbilirubinaemia: Lethargy and poor feeding: Yes No DK Hypotonia: Yes No DK Opisthotonus: Yes No DK Convulsions: Yes No DK Other (Specify)__________________________________________________________________________________ 30. Highest Bilirubin result recorded for this infant: ________ μmol/L 31. Total duration of elevated bilirubin ≥ 450 μmol/L: _______(hrs and minutes) 32. Serum bilirubin results PRE TREATMENT(μmol/L). Please include all results before any treatment was commenced. Please attach de-identified printout of all serum bilirubin results if available. DATE
TIME (24 hr clock)
TOTAL BILIRUBIN μmol/L
DATE
TIME (24 hr clock)
TOTAL BILIRUBIN μmol/L
Appendix. Australian Paediatric Surveillance Unit Neonatal Hyperbilirubinaemia Reporting Paediatrician Questionnaire. (Continues)
Prospective Surveillance of Extreme Neonatal Hyperbilirubinemia in Australia
6.e1
THE JOURNAL OF PEDIATRICS
Vol. -, No. -
www.jpeds.com
33. Serum bilirubin results POST TREATMENT(μmol/L). Please include all results after treatment commenced. Please attach de-identified printout of all serum bilirubin results if available. DATE
TIME (24 hr clock)
TOTAL BILIRUBIN μmol/L
DATE
TIME (24 hr clock)
TOTAL BILIRUBIN μmol/L
34. Associated dehydration with hyperbilirubinaemia? Yes No DK / / Time: ___________(24 hour clock) Plasma sodium _______mmol/L Not measured Date No DK 35. Did the infant have associated culture positive systemic infection? Yes If Yes, SITE: ___________________________ ORGANISM: ____________________________________________ / / Time: ___________(24 hour clock) Not measured DK 36. Lowest albumin level______ g/l Date 37. Lowest blood pH ________ Date / / Time: ___________(24 hour clock) Not measured DK 38. Did the infant have other serious morbidity? Yes No DK If Yes, Please specify (eg. hypoxic-ischaemic encephalopathy, hypoglycaemia, liver disease)_________________________________________________________ 39. Cause of hyperbilirubinaemia: Physiological ABO incompatibility: Probable (Coombs positive) Possible (Coombs negative) Rhesus isoimmunisation Glucose-6-phosphate dehydrogenase deficiency Other _______________________________________________________________________________________ MANAGEMENT 40. Phototherapy used? DATE
Yes
Start (24 hr clock)
No
DK
If Yes, please record times and dates:
End (24 hr clock)
DATE
Start (24 hr clock)
End (24 hr clock)
41. Albumin infusion? Yes No DK If yes, time and date commenced______________________ 42. Immunoglobulin infusion Yes No DK If yes, number of doses __________ 43. Exchange transfusion Yes No DK If yes, how many ________________ Time and dates _________________________________________________________________________________ 44. Was magnetic resonance imaging brain scan done? Yes No DK If Yes, Date / / and Result: Normal Increased signal on T2-weighted images in globus pallidus Abnormal but not consistent with bilirubin toxicity, (specify): ________________________________________ Other: (specify): __________________________________________________________________________ OUTCOME 45. Did the baby survive Yes No DK If Yes, date of discharge: / / No DK Serum bilirubin closest to discharge: _____ μmol/L Neurological status normal at discharge? Yes No DK If Yes, Result:________________________________________________ Hearing screen done? Yes If baby died, date of death: / / Was a post-mortem conducted Yes No DK No DK Or other associated pathology Yes No DK If Yes, did the post-mortem show kernicterus Yes If Yes, (specify): ___________________________________________________________________________________ FOLLOW UP TRACKING INFORMATION Please provide details of the physician from whom follow-up information can be obtained: Please print Name:_________________________ Phone No:______________ E-mail:________________ Do you have any other comments about this infant? ________________________________________________________________________________________________ ________________________________________________________________________________________________ THANK YOU FOR COMPLETING THIS QUESTIONNAIRE Please return this questionnaire in the addressed reply-paid envelope to: Dr Angela M Gillivray, Clinical Neonatology Fellow, Newborn Care, Royal Prince Alfred Hospital, Missenden Rd, Camperdown, NSW 2050, Australia.
Appendix. (Continued).
6.e2
McGillivray et al
- 2015
ORIGINAL ARTICLES
Table V. Univariate analysis of severe neonatal hyperbilirubinemia with peak levels over 500 mmol/L Risk factors for SBR >500 mmol/L Sex Male Female Ethnicity Asian Other Cause of hyperbilirubinemia Non-hemolytic Hemolytic* Gestational age Term Preterm Feeding mode Breast Formula Combination Readmission Inpatient Readmitted
n (%)
P value .422
18/50 (36) 10/36 (27.8) .397 14/34 (41) 12/38 (31.6) <.002 10/54 (18.5) 15/27 (55.6) .2 23/77 (29.9) 5/10 (50) .309 17/58 (29.3) 1/1 (100) 9/27 (33.3) .295 8/19 (42.1) 20/68 (29.4)
Oxhem, oxidative hemolysis; Rh, Rhesus isoimmunization; SBR, serum bilirubin. *Hemolytic causes: Rh, glucose-6-phosphate deficiency, hereditary spherocytosis, Oxhem, and ABO blood group incompatibility with DAT+.
Prospective Surveillance of Extreme Neonatal Hyperbilirubinemia in Australia
6.e3