|Year : 2018 | Volume
| Issue : 1 | Page : 36-42
Correlation between transcutaneous bilirubin estimation and total serum bilirubin estimation in neonatal hyperbilirubinemia
Avijit Mandal1, Rohit Bannerji1, Jaydeb Ray1, Monjori Mitra1, Syed Md Azad1, Surupa Basu2
1 Department of Paediatrics, Institute of Child Health, Kolkata, West Bengal, India
2 Department of Biochemistry, Institute of Child Health, Kolkata, West Bengal, India
|Date of Submission||09-Jun-2017|
|Date of Acceptance||12-Sep-2017|
|Date of Web Publication||20-Jun-2018|
Dr. Rohit Bannerji
Institute of Child Health, 11, Biresh Guha Street, Kolkata - 700017, West Bengal
Source of Support: None, Conflict of Interest: None
AIMS: To determine the efficacy of measuring the transcutaneous bilirubin as a screening tool for clinically significant hyperbilirubinaemia in Indian infants.
MATERIALS AND METHODS: A cohort of 100 neonates who have clinical jaundice, admitted in NICU of Institute of Child Health, Kolkata from March 2014 to Jan 2015, were included in the study. Both the total serum bilirubin and transcutaneous bilirubin were measured for the above group. Inclusion criteria was neonates who had clinical jaundice and required extimation of serum bilirubin. Exclusion criteria was neonates who would receive phototherapy and/or exchange transfusion.
STATISTICAL ANALYSIS: Software used was SPSS.
RESULTS: 59 male and 41 female neonates were enrolled. A strong agreement was found between Total serum bilirubin and Transcutaneous bilirubin values. Mean gestational age was 34.91 weeks with a mean birth weight of 2492.2 gms.Mean age at the time of measurement was 86.13 hours.
CONCLUSION: The above strong agreement between TSB and TCB was same in the term population as well as the whole population. The results of this study supports Transcutaneous bilirubin as an effective screening tool for estimation of neonatal hyperbilirubinaemia in Indian infants.
Keywords: Neonatal hyperbilirubinemia, total serum bilirubin, transcutaneous bilirubin
|How to cite this article:|
Mandal A, Bannerji R, Ray J, Mitra M, Azad SM, Basu S. Correlation between transcutaneous bilirubin estimation and total serum bilirubin estimation in neonatal hyperbilirubinemia. BLDE Univ J Health Sci 2018;3:36-42
|How to cite this URL:|
Mandal A, Bannerji R, Ray J, Mitra M, Azad SM, Basu S. Correlation between transcutaneous bilirubin estimation and total serum bilirubin estimation in neonatal hyperbilirubinemia. BLDE Univ J Health Sci [serial online] 2018 [cited 2019 May 23];3:36-42. Available from: http://www.bldeujournalhs.in/text.asp?2018/3/1/36/234648
Neonatal jaundice or hyperbilirubinemia is a fairly common problem encountered in the newborn nursery. Complications of hyperbilirubinemia, namely, acute bilirubin encephalopathy and/or kernicterus are rare in infants having bilirubin levels below the 95th percentile. Kernicterus, which was believed to have disappeared almost completely, is now of greater concern for neonatologists and pediatricians because the earlier discharge from the hospital of mothers and neonates prevents an adequate and effective monitoring of jaundice., The problem is that most infants are now being discharged before the age of 48 h whereas their bilirubin levels peak almost always between 3 and 7 days of age. Even though visual evaluation of neonates for jaundice is a time-honored method having certain advantages, the disadvantages are that the correlation with total serum bilirubin (TSB) is poor, even when it is performed by experienced nursery personnel. The revised American Academy of Pediatrics guidelines  which have been recently laid down for the prevention and management of hyperbilirubinemia emphasize the importance of an objective assessment of bilirubin status for all neonates before being discharged from the hospitals, and the use of a transcutaneous device is well considered to be an option for this determination. In addition, the guidelines also recommended that “if there is any doubt about the degree of jaundice, the TSB or transcutaneous bilirubin (TcB) should be measured.” As TSB estimation is painful, a convenient accurate and noninvasive TcB measurement would be of enormous benefit to physicians, infants, and families.
Previous studies of TcB have produced results which were conflicting with some investigators emphasizing the limitation of this technique  and others on the other extreme side of the spectrum reporting that TcB is reliable enough to completely replace laboratory measurement of TSB.,
Aims and objectives
- To evaluate bilirubin level by transcutaneous bilirubinometer and by standard biochemical method (modified diazo reaction)
- To correlate results of TSB by conventional method and TcB in term and preterm neonates
- To decrease the number of invasive painful procedure for estimation of serum bilirubin.
| Materials and Methods|| |
- Study area: This study was conducted in Neonatal Intensive Care Unit (NICU) in Institute of Child Health, Kolkata
- Study population: A cohort expected to be around 100 neonates who have clinical jaundice and require estimation of serum bilirubin, admitted to NICU of Institute Child Health, Kolkata, constituted the study population
- Study period: This study was conducted from March 2014 to January 2015
- Sample size: The study sample was 100 neonates
- Sample design:
- Inclusion criteria: Neonates who have clinical jaundice and require estimation of serum bilirubin
- Exclusion criteria: Neonates who will receive phototherapy and an exchange transfusion.
- Study design: This was a prospective observational study
- Data collection: Infant's medical records were reviewed for collecting the demographic data. Gestational age, birth weight, sex, and at the time of study were recorded
- Study tools: The study tools are described below.
Total serum bilirubin measurement
TSB estimation was done by modified diazo method on the ROCHE INTEGRA-e 411+ autoanalyzer (mg/dl). This is a routine procedure which is performed regularly in the Biochemistry Department of Institute of Child Health. The assays used by this machine were previously studied by comparing with the gold standard bilirubin measurement of high-performance liquid chromatography (HPLC), and they were found to correlate well. In addition, there was one study which compared HPLC, modified diazo reaction, and TcB measurement and found good correlation among all three methods.
Transcutaneous bilirubin measurement
For TcB measurement, we used the Jaundice Meter JM-103 device (manufactured by Draeger Medical Systems, Inc., Lubeck, Germany). All the measurements were performed with a single Jaundice Meter JM-103 device in accordance with the manufactures' recommendations. The device is housed in the form of a handheld assembly which has a fiber-optic probe that is placed strategically which allows for a convenient application to the forehead or sternum of the infant. Once the appropriate skin application is established, the light source is triggered and the measured values appear on display.
In this study, TcB measurements were done on sternum within 15 min of TSB measurements. Averaging five readings, computed automatically by the Jaundice Meter when the desired number of measurements is set, provides more precise TcB measurement than using a single measurement. The device does not require user calibration. However, the system includes a checker that measures the intensity of light from the device to ensure the light output is acceptable for proper use.
Risk level determination
We divided the bilirubin level into two groups, namely, “clinically significant jaundice” and “nonclinically significant jaundice” based on the bilirubin nomogram which was developed by Bhutani et al. The bilirubin nomogram plots the age in hours versus the TSB. As the level of bilirubin varies normally with the age, this nomogram thus helps to determine whether an infant is at risk for developing clinically significant hyperbilirubinemia based on the bilirubin levels at a particular hour of life. If a bilirubin value is found to be higher than the 95th percentile for age curve, the nomogram predicts that the infant is in the “high-risk zone” for developing clinically significant hyperbilirubinemia. Similarly, bilirubin values between the 75th and 95th percentile curves predict the infant is in the “high-intermediate risk zone.”
The TSB and the average TcB values which were obtained for each infant were plotted on this nomogram. Since both the TSB and TcB values for each individual infant were obtained simultaneously, the same exact hour of life was used to plot both values. For clinically significance, TSB values which were above the 95th percentile and TcB values which were above the 75th percentile were used by us.
All bilirubin samples were obtained during the ages of life which are included in the nomogram (0–144 h old).
| Results and Analysis|| |
In this study, 100 neonates were enrolled, out of 59 were male and 41 were female. Demographic data such as gestational age, birth weight, and postnatal age are listed in [Table 1]. The study group was divided into term group (55 neonates) and preterm group (45 neonates) as explained in [Table 2]. Majority of the study participants (55%) were at term at birth [Table 3].
|Table 3: Distribution of study population according to their gestational age|
Click here to view
- TSB values ranged from 4.92 to 20.82 mg/dl, with 51% of neonates below the 75th percentile for age, 19% with levels between 75th and 95th percentiles for age, and 30% with levels above the 95th for age [Table 4]
- 26% of the study participants were low birth weight, i.e below 2500gm [Table 5]
- Scatter plot of TSB versus TcB [Figure 1] shows a strong direct linear correlation. Pearson's correlation coefficient, r = 0.94 with hypothesis test of correlation showing P ≤ 0.001. This indicates that there is a strong linear association with the two variables in our sample and same holds for the underlying population. Simple linear regression is calculated as TSB = −0.9016 + 0.9760 × TcB. This equation can be used for prediction TSB from TcB with a high degree of accuracy
- Agreement between TSB and TcB values is assed qualitative by construction as Bland–Altman plot [Figure 2]. The bias line indicates that difference of 1.2 mg/dl between averages of the two variables with the TcB being higher than the TSB. Majority of the data points fall within ± 1.96 times of the standard deviation (SD) of the difference between TSB and TcB values. This corroborates that there is strong agreement present. This agreement is also assessed quantitatively by calculating interclass correlation coefficient. This comes to 0.892 with 95% confidence interval (CI) of 0.477–0.960
- Mean gestational age was 37.9 weeks; mean birth weight was 2919.8 g. Mean age at the time when measurements were taken was 87.92 h [Table 6]
- In preterm babies mean gestational age was 34.91 weeks; mean birth weight was 2492.2gm. Mean age at the time when measurements were taken was 86.13 h [Table 7]
- From [Figure 3], it was found that the same correlation was present between TcB and TSB in term population as it also was in whole population. Pearson's correlation coefficient r = 0.94, P ≤ 0.001 and r2 = 0.89. Simple linear regression equation y = −06740 + 0.9631 × TcB
- The bias lines in Bland–Altman plot [Figure 4] in term neonates indicate that the difference of 1.2 mg/dl between TSB and TcB with the TcB being higher than the TSB. Majority of the data points fall within ± 1.96 times of the SD of the difference between TSB and TcB values. This corroborates that there is strong agreement present. This agreement is also assessed quantitatively by calculating interclass correlation coefficient. This comes to 0.897 with 95% CI of 0.617–0.981.
|Table 4: Distribution of study participants according to their serum bilirubin percentile wise|
Click here to view
|Table 5: Distribution of study population according to their birth weight|
Click here to view
|Figure 1: Scatter plot of total serum bilirubin versus transcutaneous bilirubin-whole population (n = 100)|
Click here to view
|Figure 2: Bland–Altman plot of total serum bilirubin versus transcutaneous bilirubin-whole population (n = 100)|
Click here to view
|Figure 3: Scatter plot of total serum bilirubin versus transcutaneous bilirubin-term population (n = 55)|
Click here to view
|Figure 4: Bland–Altman plot of total serum bilirubin versus transcutaneous bilirubin-term population|
Click here to view
Mean gestational age was 34.91 weeks; mean birth weight was 2492.2 g. Mean age at the time when measurements were taken was 86.13 h.
From [Figure 5], it was found that the correlation between TcB and TSB was linear and statistically significant. Pearson's correlation coefficient r = 0.92, r2 = 0.85, P ≤ 0.001. Simple linear regression equation y = −1.05 + 0.9843 × TcB.
|Figure 5: Scatter plot of total serum bilirubin versus transcutaneous bilirubin-preterm population (n = 45)|
Click here to view
From [Figure 6], it was noted that the agreement was same as in whole population and term population. Majority of the data points fall within ± 1.96 times of the SD of the difference between TSB and TcB values. This corroborates that there is strong agreement present. This agreement is also assessed quantitatively by calculating interclass correlation coefficient. This comes to 0.871 with 95% CI of 0.452–0.952.
|Figure 6: Bland–Altman plot total serum bilirubin versus transcutaneous bilirubin-preterm populationfall|
Click here to view
- Sensitivity of the test is 100%, specificity is 72.8%, positive predictive value is 61.2%, and negative predictive value is 100%. None of the 51 neonates with TcB below the 75th percentile for age had clinically significant hyperbilirubinemia (TSB above the 95th percentile for the age).
- [Figure 7] shows the 3 zones of risk according to TSB vs age Showing that the high risk zone is above 95th centile.
| Discussion|| |
According to Rubaltelli et al., it is pertinent to remember that the TSB represents a measurement of the circulating bilirubin while TcB is a measurement of the tissue bilirubin.
As per our present study results in Indian population as in other ethnic groups, the TcB level correlated well with the laboratory measurement of TSB. In the present study, overall correlation of TSB to TcB was found to be linear and statistically significant (r = 0.94, P ≤ 0.001). Similar regression statistics were evident in subset population categorized by gestational age (term r = 0.94, P ≤ 0.001 and preterm r = 0.92, P ≤ 0.001). This is in agreement with the previous studies.
The present study demonstrated that using TcB level which was higher than the 75th percentile for age was found to be 100% sensitive in identifying infants who might be at risk for developing clinically significant hyperbilirubinemia. None of the neonates with TSB higher than the 95th percentile had a TcB measurement which was <75th percentile [Table 8]. This is in agreement with the study done by Bhutani et al.
|Table 8: Predictive indices using >95th percentile total serum bilirubin and >75th percentile transcutaneous bilirubin|
Click here to view
In one study, Engle el al. reported a similar correlation between TcB and TSB in preterm and term infants with a correlation coefficient r = 0.77 (n = 121; P < 0.001).
Maisels et al. reported that TcB provided good agreement with laboratory determination of TSB with correlation coefficient r ≥ 0.90 in an essentially Caucasian population. Our present study also showed good correlation between TcB and TSB values with correlation coefficient r = 0.94.
Similar to previous studies, a significant correlation was observed between TSB values. However, as pointed out by Bland and Altman, correlation coefficient could be a poor indicator of the agreement between two clinical tests, and a far better measure of agreement would be the plot of differences of values obtained by the two methods vis-a-vis the mean of the values. Analysis of our results using the Bland–Altman plot, the bias line indicates that difference of 1.2 mg/dl between average of the two variables with the TcB being higher than the TSB. Majority of the data points fall within ± 1.96 times of the SD of the difference between TSB and TcB values. This corroborates that there is strong agreement present.
We estimated serum bilirubin in the current study in our clinical laboratory whereas previous studies that had evaluated transcutaneous bilirubinometer used HPLC as gold standard. However, Bhutani et al. in their recent report showed that agreement between TcB values and TSB determined by a variety of laboratory methods was comparable to the agreement between TcB and HPLC. Our laboratory performs approximately 2000 total bilirubin estimations monthly, maintaining rigorous quality control standards for bilirubin testing. The results of the present study thus recommend the use of the TcB measurement as a first-line screening tool for hyperbilirubinemia. For infants having a TcB level below the 75th percentile for age, it could be safely assumed that the infant has a low risk of developing clinically significant hyperbilirubinemia eliminating the need for further testing. In our NICU, where every newborn is routinely required to have a TSB level checked before discharge, this would facilitate the reduction in the number of measurements and hence a potentially painful blood draw.
The present study was not without limitations. First, a small number of neonates were included in this study. Second, different operators obtained the measurements even though we used the same JM-103 device for measuring the TcB for all cases (possibilities of introducing error). Third, our neonates being mostly referred cases from the different hospitals had relatively high levels of jaundice (leading to possible selection bias).
The two methods (TSB and TcB) do not in fact measure the same parameter. Even though the laboratory method measures only that component of bilirubin that is circulating in the blood, the amount of bilirubin that has shifted from the serum into the tissues is quantified by TcB.
If this indicated the serum bilirubin levels that were also available to move into the brain tissue, which being our real concern, then it could offer additional information from which clinical management decisions could be made. These unanswered questions, along with the effects of exchange transfusion and phototherapy on TcB measurement, the effect of drugs as also the accuracy in very low birth weight neonates, and neonates <30 weeks' gestational age, all require further studies.
| Conclusion|| |
The results of this study support the use of the transcutaneous bilirubin meter as a screening tool for clinically significant hyperbilirubinemia in Indian infants. The advantage of the TcB meter was that it had an excellent sensitivity, thus making it extremely improbable that an infant with a significant level of hyperbilirubinemia would miss detection with TcB measurement. Even though the specificity of the TcB level was 72%, the test was successfully able to identify a sizeable group of infants who did not need invasive testing for the evaluation of hyperbilirubinemia. As per our study results, we can recommend that all normal infants be screened for hyperbilirubinemia initially by measuring their TcB levels. If the results indicate a TcB level above the 75th percentile for age, it could then warrant for TSB level estimations.
However, substituting the TcB for TSB as a universal screening tool should await the results of a well-conducted clinical trial for both efficacy as well as cost-effectiveness with the goals of reduction in the hospital readmission rates and the subsequent risks of bilirubin-related neurological sequelae. The clinical value of TcB measurement will evolve over the coming years, not only with continuing technological developments but also with our gradual understanding of the complex relationships existing between accumulation of bilirubin and levels of albumin-bound and unbound bilirubin in both intravascular and extravascular spaces.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Johnson LH, Bhutani VK, Brown AK. System-based approach to management of neonatal jaundice and prevention of kernicterus. J Pediatr 2002;140:396-403.
Maisels MJ, Newman TB. Kernicterus in otherwise healthy, breast-fed term newborns. Pediatrics 1995;96:730-3.
Bhutani VK, Johnson L, Sivieri EM. Predictive ability of a predischarge hour-specific serum bilirubin for subsequent significant hyperbilirubinemia in healthy term and near-term newborns. Pediatrics 1999;103:6-14.
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.
Wong CM, Van DIjk PJ, Liang IA. A comparison of transcutaneous bilirubinometry: SpectRx Bilicheck vs. Minolta Air Shields. Arch Dis Child Fetal Neonatal Ed 2002;86:F190-2.
Engle WD, Jackson GL, Stehel EK, Sendelbach DM, Manning MD. Evaluation of a transcutaneous jaundice meter following hospital discharge in term and near-term neonates. J Perinatol 2005;25:486-90.
Kazmierczak SC, Robertson AF, Briley KP, Kreamer B, Gourley GR. Transcutaneous measurement of bilirubin in newborns: Comparison with an automated Jendrassik-grof procedure and HPLC. Clin Chem 2004;50:433-5.
Gourley GR, Bhutani VK, Johnson L, Kreamer B, Kosorok MR, Dalin C, et al
. Measurement of serum bilirubin in newborn infants: Common clinical laboratory methods versus HPLC. Pediatr Res 1999;45:283A.
Bhutani VK, Gourley GR, Adler S, Kreamer B, Dalin C, Johnson LH, et al.
Noninvasive measurement of total serum bilirubin in a multiracial predischarge newborn population to assess the risk of severe hyperbilirubinemia. Pediatrics 2000;106:E17.
Rubaltelli FF, Gourley GR, Loskamp N, Modi N, Roth-Kleiner M, Sender A, et al.
Transcutaneous bilirubin measurement: A multicenter evaluation of a new device. Pediatrics 2001;107:1264-71.
Maisels MJ, Ostrea EM Jr., Touch S, Clune SE, Cepeda E, Kring E, et al.
Evaluation of a new transcutaneous bilirubinometer. Pediatrics 2004;113:1628-35.
Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1986;1:307-10.
[Figure 1], [Figure 2], [Figure 3], [Figure 4], [Figure 5], [Figure 6], [Figure 7]
[Table 1], [Table 2], [Table 3], [Table 4], [Table 5], [Table 6], [Table 7], [Table 8]