Hyperbilirubinemia, cord bilirubin, neonate


Hyperbilirubinemia or jaundice is one of the most common medical problems in newborn infants. Jaundice is observed during the first week of life in approximately 60% term and 80% of preterm infants (1).Due to social and family constraints the paediatricians are practicing early discharge from hospitals(2).This increases the risk of significant jaundice where subsequently intervention might be required The problems encountered due to hyperbilirubinemia are neurological manifestations as seen in kernicterus leading brain damage.This is also seen in a full term newborn with no apparent evidence of hemolysis(3). Therefore, it is difficult to predict which infants are at increased risk for significant and relatively late hyperbilirubinemia, and there is an obvious need to implement follow-up programs or to develop predictive guidelines that will enable the physicians to predict or to identify which of the early discharged new born will develop significant hyperbilirubinemia (2).Knowledge of infants at risk of developing jaundice allows simple bilirubin reducing methods to be implemented before jaundice becomes significant and could influence a decision regarding early discharge from Hospital.Predicting the high risk neonates for subsequent hyperbilirubinemia will also help in detecting infants at low risk for postnatal hyperbilirubinemia and minimize an unnecessary prolongation of Hospitalization or Hospital readmission for hyperbilirubinemia (4). We therefore, undertook this study to predict the value of cord bilirubin level beyond which a newborn might develop jaundice requiring intervention.

Materials & Methods

The study was conducted in Department of Pediatrics, Swami Rama Himalayan University, Dehradun, over a period of 12 months from 1stJanuary 2015 to December 31st 2015. All healthy term and near term neonates delivered consecutively in the department of Obstetrics and Gynaecology at HIMS were prospectively enrolled in the study.

Aim and Objective

Primarily study was done to evaluate the predictive value of cord bilirubin level for identifying term and near term neonates for subsequent hyperbilirubinemia requiring therapeutic intervention.

Inclusion criteria

  • Sequentially borne, any type of delivery, both gender.
  • Full Term and late preterm ( Gestational Age > 34 weeks and < 42 weeks )
  • APGAR >7 first minute and 10 at fifth minute.

Exclusion criteria

  • Significant illness (sepsis, Respiratory Distress Syndrome, asphyxia, IDM) that could aggravate hyperbilirubinemia.
  • Birth weight below 2000 gm.
  • Rh and ABO incompatibility

A detailed maternal and perinatal history and clinical findings were recorded on a prestructured format along with the lab investigations. Gestational age of enrolled infants was determined on the basis of the date of the last menstrual period or first trimester ultrasound (wherever available). It was confirmed by the expanded Ballard score done within 24 hours of life (5). Written and informed consent was taken from parents or guardian. Cord bilirubin was performed at birth, by a blinded observer.Repeat serum bilirubin levels were assessed at 72 hours of life.All neonates were regularly assessed by Paediatric residents and closely observed for the development of any clinically significant icterus. Neonates whowere discharged before the third day of life were asked to return on day3 for estimation of serum bilirubin at 72 hours of life.Neonates who did not come for follow up or in whom the 72 hours serum bilirubin could not be collected were excluded from the study.The decision to start or stop phototherapy was based on AAP guidelines (6).Serum bilirubin levels > 14 mg/dl at 72 hours was taken as significant hyperbilirubinemia.Bilirubin and its fractions were measured by spectrophotometric technique on an automated analyser Unicel DXC 800 manufactured by Beckman Coulter.The study was approved by institutional ethics committee.

Statistical Analysis

Result was analysed by using statistical software SPSS 22.Qualitative data were expressed in frequency and percentage, while quantitative data were expressed in mean ± SD. Paired-T test, ROC curve and logistic regression were used to analyse the pattern of bilirubin levels.


A total of 767 deliveries occurred during the study period of which 100 were enrolled in the study (figure 1). The baseline characterstic of the mothers and new born enrolled in the study is given in Table 1 and 2 respectively.


Figure 1: Flow chart showing neonates enrolled

Table 1: Baseline characteristics of mother enrolled

Maternal Characterstics n(%)
Mode of Delivery
LSCS 54 (61.3%)
Normal 34 (38.6%)
Blood Group
A +ve 21 (23.3%)
B+ ve 21 (23.3%)
AB+ve 9(10%)
O+ve 33(37.3%)
AB-ve 1(1.2%)
B-ve 3(3.02%)
MSL 14(16%)
Oxytocin 2(2.27%)
Neonatal jaundice in previous sibling 1(1.2%)
PROM 2 (2.27%)

Table 2: Baseline characteristics of new born enrolled: (* Mean +/- SD, # n (%)

Neonatal Characterstics
Gestational age (week)* 37.82+/- 1.09
Sex #
Female 31(35.2%)
Male 57(64.7%)
Blood Group#
A+ve 8(20%)
B+ve 11(27.5%)
AB+ve 2(5%)
O+ve 19(47.5%)
B-ve 3(3%)
Birth Trauma# 0(0%)
Weight (kg)* 2.77 +/- 0.39
Cord Blood Hb (g/dl)* 14.86+/-1.626

Table 3: relationship between cord bilirubin and serum bilirubin at 72 hours

Cord bilirubin Neonates with serum bilirubin >14mg/dl (n=16) Neonates with serum bilirubin 2-13mg/dl (n=72) Total N=88
>2.02 mg 14(40%) 21(60%) 35(40%)
<2.02 mg 2(3%) 51(97%) 53(60%)


Figure 2: ROC curve showing association of cord bilirubin levels and requirement of phototherapy at 72 hours of life

With ROC analysis , the mean ± cord bilirubin level was 1.923 ± 0.966 mg/dl. The mean ± SD of T. Bilirubin at 72 hrs was 10.56 ± 3.18 mg/dl. There were 16 new born who had serum bilirubin >14 mg/dl at 72 hours (Table 3 ). The cord bilirubin of > 2.02 had sensitivity and specificity of 87.5% and 70.8% respectively with positive predictive value of 0.39 and Negative predictive value of 0.965. The strength and association of cord bilirubin > 2.02 mg/dl and requirement of phototherapy at 72 hours was found to be significant with p value <0.001


Serum bilirubin levels are usually 1-3 mg/dl at the time of birth and rise at the rate of less than 5 mg/dl/day, peaking at 2-3 days in term neonates (7). Our aim was to quantify the relationship between cord blood bilirubin with peak serum bilirubin levels at 72 -84 hours of life. Cord blood estimation was chosen because it is a non- invasive and the result is available within few hours after birth.Thus, the babies at risk for developing hyperbilirubinemia can be detected at birth in a non-invasive way if the neonate leaves the hospital within the first few postnatal days. Also, the use of cord blood bilirubin values may help to predict infants with low risk for hyperbilirubinemia and minimise an unnecessary prolongation of hospitalization.(8, 9, 10). In a study by Taksande et al prevalence was 9.5% only because the cut-off for significant hyperbilirubinemia on 3rd day of life was taken as 17 mg% (11). Nagwa etal observed that there was a clear correlation between Umbilical cord serum bilirubin and subsequent development of hyperbilirubinemia and newborn with cord bilirubin >4 mg/dl were a group at risk of developing severe hyperbilirubinemia and were presented to have mean serum bilirubin levels higher than 16 mg/dl at 72 hours with peak 17 ± 4.3 mg/dl at 68 ±17.5 hours of postnatal age(12). Knudsen in his study demonstrated that jaundiced new born presented higher umbilical cord bilirubin levels than new born without clinical jaundice. In addition , the number of jaundiced new born undergoing phototherapy was significantly higher when these levels were higher than 2.0 mg/dl, in comparison with the number of jaundiced new born with no need for treatment and whose bilirubin levels were lower than equal to 2.0 mg/dl (13).Knupferetal, observed that serum bilirubin >1.74 mg/dl on the first day of life had 97% sensitivity of predicting a subsequent serum bilirubin levels > 16 mg/dl at 72 hours of life.At this critical serum bilirubin value, the negative predictive value was 99.8% (14). Rataj etal, reported that if cord bilirubin was < 1 mg/dl jaundice occurred in 2.4 % new born, whereas 89% of infants with cord bilirubin above 2.5 mg/dl became jaundiced (15).Taksande etal, in their study predicted that cord bilirubin level > 2 mg/dl had 89.5% sensitivity, and high (98.7%) negative predictive value and fairly low (38.6%) of positive predictive value(11). In our study,the cord bilirubin level > 2.02 mg/dl had the highest sensitivity (87.5%), and this critical bilirubin levels had a very high negative predictive value (96%) and fairly low (39%) positive predictive value. According to our findings, a critical cut off level of cord bilirubin was 2.02 mg/dl predicted 88% of new born who developed jaundice.However, the cord bilirubin <2.02 mg/dl did not completely exclude the development of significant hyperbilirubinemia. Only 3% of new born with cord bilirubin levels of <2.02 mg/dl developed jaundice. A 96% of negative predictive value suggests that the measurement of cord serum bilirubin can help in identifying those new born who are unlikely to require further evaluation and intervention.


Cord bilirubin level >2.02 mg/dl had sensitivity and specificity of 89.5% and 87.5% and can be taken as a cut off value above which, a neonate would develop hyperbilirubinemia and would require phototherapy