Elsevier

Clinical Biochemistry

Volume 42, Issue 3, February 2009, Pages 176-179
Clinical Biochemistry

Clinical implication of the difference between transcutaneous bilirubinometry and total serum bilirubin for the classification of newborns at risk of hyperbilirubinemia

https://doi.org/10.1016/j.clinbiochem.2008.09.108Get rights and content

Abstract

Objectives

1) To determine the relationship between transcutaneous bilirubin measurements (TcB), performed using BiliCheck™ or Minolta Air-Shields JM-103®, and total serum bilirubin levels (TSB) and 2) to evaluate the predictive accuracy of TcB measurements for risk using the nomogram recommended by the Canadian Pediatric Society (CPS).

Methods

A total of 154 healthy term neonates from the newborn nursery at McMaster Children's Hospital meeting the inclusion criteria were enrolled. TcB measurements were performed within 30 min of obtaining the TSB measurement.

Results

Both devices showed a good correlation with the laboratory method (BiliCheck™-Vitros, R2 = 0.86; Minolta Air-Shields JM-103®-Vitros, R2 = 0.85), but underestimated the serum bilirubin. Applying the risk classification using the 40th, 75th, and 95th percentile of the Bhutani nomogram a 6%, 0%, and 1% false negative rate was found for BiliCheck™ and 62%, 74% and 81% for the Minolta Air-Shields JM-103® device. After correcting for the differences using either the bias or the 95% CI the false negative rate was reduced to zero in all cases.

Conclusion

TcB measurements cannot be directly applied to a TSB nomogram but must be adjusted for any observed biases in order to avoid misclassifying newborns at risk for hyperbilirubinemia.

Introduction

The healthcare policy mandating early discharge of healthy, full-term infants combined with a higher prevalence of breast-feeding newborns has contributed to an increased number of neonatal hospital readmissions for hyperbilirubinemia, within days of being discharged [1]. The Canadian Pediatric Society (CPS) released new screening guidelines for jaundice with a recommendation that all infants should have their bilirubin measured in the first 72 h of life. Infants with a level consistent with increased risk should be carefully monitored and followed within 24 to 48 h of discharge [2]. Bilirubin concentration can be measured by capillary or venous blood sampling or by using transcutaneous devices.

Early detection and prevention of severe hyperbilirubinemia is essential to avoid bilirubin encephalopathy. Therefore, it is mandated that when infants go home, especially prior to 24 h of age, a follow up visit should be scheduled with a healthcare professional or a home visit arranged with a primary community provider who can perform a bilirubin test, as well as evaluate other potential health problems. Thus the primary responsibility for detecting significant post discharge jaundice rests within the healthcare system (family practitioners, midwives, and community nurses). In these settings, transcutaneous bilirubinometry seems to be a reasonable alternative to laboratory-based bilirubin testing to screen newborns for hyperbilirubinemia.

In recent years there have been major improvements in the assessment of bilirubin levels using transcutaneous bilirubinometry (TcB). At present two transcutaneous bilirubinometers are reputed to address the confounding effects of melanin and hemoglobin on TcB analysis; the Respironics BiliCheck™ and the Minolta Air-Shields JM-103® [3], [4]. The objectives of this study were 1) to determine whether transcutaneous bilirubin measurements, performed using the BiliCheck™ or JM-103 meters correlate with total serum bilirubin (TSB) levels, measured in our laboratory (VITROS 950) and 2) to evaluate the predictive accuracy of the TcB measurements pertinent to the risk classification of infants with jaundice based on the nomogram recommended by the CPS [2], [5].

Section snippets

Selection and description of participants

We selected healthy term or near term neonates from the nursery at McMaster Children's Hospital that were born in the period of July–August 2003 (BiliCheck™-Vitros (B-V) study) and December 2003 to February 2004 (Air-Shields-Vitros (AS-V)) who met the inclusion criteria. This was not designed as a 3-way comparison, and the Minolta Air-Shields JM-103® bilirubinometer was available in Canada at the end of 2003. Inclusion criteria were: i) healthy neonates greater than 35 weeks gestational age and

Population

A total of 154 admitted newborns met the inclusion criteria; 94 for Minolta Air-Shields JM-103®-Vitros (AS-V) and 60 for BiliCheck™-Vitros (B-V) comparisons. TSB values ranged from 69–343 µmol/L for the B-V infant group and 18–259 µmol/L for the AS-V. Mean gestational age was 39.2 ± 1.3 weeks for both cohorts and mean age and weight at the time when measurements were taken were 43.0 ± 31.9 h and 3391.6 ± 487.7 g for B-V and 40.6 ± 15.6 h and 3343.6 ± 399 g for the AS-V cohort. Slightly more vaginal

Discussion

Our results show a good correlation between both TcB devices (BiliCheck™ and Minolta Air-Shields JM-103®) and the laboratory method, in a study cohort representative of the population in Hamilton, Ontario. However, in both cases the TcB levels were lower than the serum bilirubin. Applying the risk classification developed by Bhutani et al. [5] and recommended by the CPS [2] we observed a 6% FN rate with BiliCheck™ and a FN rate of 62% with the Minolta Air-Shields JM-103® device. After

Conclusions

In summary, it is incorrect to plot TcB levels on a TSB based nomogram. The biases for these methods revealed in our study, have clinical implications for the classification of newborns at risk for hyperbilirubinemia and correction factors must be applied to the TcB measurements when used on the Bhutani nomogram. The first line healthcare professional must be aware of these limitations and work with the laboratory for optimal bilirubin level interpretation.

Acknowledgments

We are grateful to the patients who participated in the study. We thank the nurses on Ward 4C of the McMaster Hospital who performed the TcB measurements and Mr. M. Donahoe for help with the statistical analysis. We thank Trudell Canada for the loan of the BiliCheck™ and disposables and Hill-Rom Canada for the loan of the Minolta Air-Shields JM-103®. This paper is dedicated to the late Dr Keith Kim who assisted in the protocol preparation and performed this study during his training.

References (17)

  • M. Sgro et al.

    Incidence and causes of severe neonatal hyperbilirubinemia in Canada

    CMAJ

    (2006)
  • Guidelines for detection, management, and prevention of hyperbilirubinemia in term and late preterm newborn infants (35 or more weeks' gestation)

    Paediatr. Child Health

    (2007)
  • S. Ip et al.

    An evidence-based review of important issues concerning neonatal hyperbilirubinemia

    Pediatrics

    (2004)
  • F.F. Rubaltelli et al.

    Transcutaneous bilirubin measurement: a multicenter evaluation of a new device

    Pediatrics

    (2001)
  • V.K. Bhutani et al.

    Predictive ability of a predischarge hour-specific serum bilirubin for subsequent significant hyperbilirubinemia in healthy term and near-term newborns

    Pediatrics

    (1999)
  • Ortho-Clinical Diagnostics I, © Ortho-Clinical Diagnostics I, 2003, 2004. INSTRUCTIONS FOR USE BuBc. VITROS Chemistry...
  • J.M. Bland et al.

    Statistical methods for assessing agreement between two methods of clinical measurement

    Lancet

    (1986)
  • R. Keren et al.

    Identifying newborns at risk of significant hyperbilirubinaemia: a comparison of two recommended approaches

    Arch. Dis. Child.

    (2005)
There are more references available in the full text version of this article.

Cited by (45)

  • Transcutaneous Bilirubin Levels and Risk of Significant Hyperbilirubinemia in Early-Term and Term Newborns

    2021, JOGNN - Journal of Obstetric, Gynecologic, and Neonatal Nursing
    Citation Excerpt :

    The sternum region was not exposed to direct sunlight, and environmental lighting was constant during the study period. TcB levels higher than 13 mg/dl were confirmed with a serum total bilirubin measurement by using a colorimetric method (diazotized sulfanilic acid reaction; Roche Diagnostics GmbH, Mannheim, Germany) because TcB measurements of greater than 12 to 13 mg/dl may not show a close correlation with serum bilirubin levels and may show a significant discrepancy with simultaneous serum bilirubin measurements (Rodríguez-Capote et al., 2009). The diagnosis of significant hyperbilirubinemia was made considering risk factors and gestational and postnatal age in the AAP guidelines (AAP, Subcommittee on Hyperbilirubinemia, 2004; Maisels et al., 2009, 2014); and newborns with significant hyperbilirubinemia, were excluded from the study, and phototherapy was begun.

  • Noninvasive methods for bilirubin measurements in newborns: A report

    2021, Seminars in Perinatology
    Citation Excerpt :

    which also reported good correlations, ranging from 0.70 to 0.95. The Bilicheck has been evaluated in a large number of studies,39,50,53–55,60–76 which reported showed good correlations, ranging from 0.75 to 0.96. Lasly, the use of Bilimed was evaluated in two studies, which showed correlation coefficients ranging from 0.45 to 0.75.54,60

  • Transcutaneous bilirubinometry

    2014, Seminars in Perinatology
    Citation Excerpt :

    Yasuda et al.35 assessed the accuracy of the JM-103™ and JM-102™; they found that the JM-103™ had improved correlation with TSB. Rodriguez-Capote et al.87 compared the BiliChek™ with the JM-103™, with TSB measured by Ortho VITROS 950; correlations between TcB and TSB were excellent with both devices. The mean differences (2 SD) between TcB and TSB were −5.2 (40.4 to −50.8) μmol/L [−0.3 (2.4 to −3 mg/dL)] and −38.3 (1.8 to −78.4) μmol/L [2.2 (0.11 to −4.6 mg/dL)] for the BiliChek™ and JM-103™, respectively.

View all citing articles on Scopus
View full text