Factors associated with extreme

hyperbilirubinaemia in neonates at the

University Hospital of the West Indies
H. Trotman, C. Henny-Harry
Department of Child Health, University of the West Indies, Mona, St Andrew, Jamaica
Aim: To determine factors associated with extreme hyperbilirubinaemia in neonates at the University
Hospital of the West Indies.
Methods: A retrospective review of all neonates with hyperbilirubinaemia requiring medical intervention at
the University Hospital of the West Indies between 1 January 2006 and 30 June 2007 was performed.
Factors associated with extreme hyperbilirubinaemia were determined using multiple logistic regression
models.
Results: A total of 170 neonates fulfilled the inclusion criteria for the study and 15 (9%) of them had extreme
hyperbilirubinaemia. The majority (97, 57%) were term infants and 103 (61%) were male. Exclusively
breastfed neonates were more likely to have extreme hyperbilirubinaemia (OR 2.6, 95% CI 0.010.6).
Neonates whose mothers received oxytocin during labour (OR 2.7, 95% CI 0.020.3) and those who were
G6PD-deficient (OR 2.6, 95% CI 0.010.5) were more likely to have extreme hyperbilurubinaemia.
Conclusion: Exclusive breastfeeding, oxytocin use in the mother during labour and G6PD deficiency in the
infant were found to be factors associated with extreme hyperbilirubinaemia.
Keywords: Extreme hyperbilirubinaemia, Neonates

Introduction
Jaundice is the yellow discoloration of the skin and sclera
in newborns caused by accumulation of bilirubin.1
Neonatal jaundice affects 60% of full-term infants and
80% of preterm infants in the first 3 days after birth.
Although transient, the condition accounts for up to
75% of hospital re-admissions in the first week of life.2
A study done in Canada showed that neonates with
severe hyperbilirubinaemia, total serum bilirubin TSB
.425 mmol/l, were predominantly term, male infants
who had been exclusively breastfed. The most common
aetiology was ABO incompatibility, followed by glucose6-phosphate dehydrogenase deficiency (G6PD).3 Similarly, a study in the UK and Ireland found that neonates
with a TSB >510 mmol/L tended to be term, male infants
who were predominantly breastfed. A haemolytic cause
for the jaundice was identified in nearly 50%.4
Additionally, researchers in the US found that neonates
with a TSB >513 mmol/L were mostly term infants and
infants who were exclusively breastfed.5
A review of cases of extreme hyperbilirubinaemia
(TSB >428 mmol/L) in Turkey showed that 96% of
the neonates were exclusively breastfed and that in

66% the underlying cause was idiopathic and in 20%

it was owing to iso-immunization.6 Bolivian researchers documented that 98% of neonates with TSB
>428 mmol/L were exclusively breastfeeding and that
one-third of them had weight loss .12% of their
birthweight and were dehydrated on presentation.7
Neonatal jaundice was studied in Jamaican neonates in 1963; ABO/Rh incompatibility, prematurity
and maternal diabetes were the only significant
underlying factors reported to predispose to neonatal hyperbilirubinaemia.8 Subsequent studies in Jamaican neonates (1972) showed similar results while
identifying G6PD as a significant cause of jaundice in neonates with moderate-to-extreme neonatal
hyperbilirubinaemia.9
There have been few recent studies from the
English-speaking Caribbean on neonatal jaundice
and none on extreme neonatal hyperbilirubinaemia.
This study aimed to define the factors associated with
extreme neonatal hyperbiluribinaemia (TSB >428
mmol/L) at a tertiary-level centre in a resource-limited
setting.

Methods
Correspondence to: H Trotman, Department of Child Health, University of
the West Indies, Mona, St Andrew, Jamaica.
Email: helen.trotmanedwards@uwimona.edu.jm

W. S. Maney & Son Ltd 2012

DOI 10.1179/2046905512Y.0000000014

This was a retrospective, descriptive study looking at all neonates with a discharge diagnosis of

Paediatrics and International Child Health

2012

VOL .

32

NO .

97

Trotman and Henny-Harry

Extreme hyperbilirubinaemia

hyperbilirubinaemia between 1 January 2006 and 30

June 2007 at the University Hospital of the West
Indies (UHWI), Jamaica. Cases were identified from
the neonatal unit log books; however, on review of the
individual charts, any case that did not require medical
intervention was excluded. Patients were divided into
two groups, those who had extreme hyperbilirubinaemia (TSB >428 mmol/L) and those with TSB
,428 mmol/L. Demographic, clinical and laboratory
data were extracted from the individual charts using a
data extraction sheet. Usual laboratory investigations
undertaken on admission include a complete blood
count with differential and blood film, blood group
and rhesus factor, direct Coombs test (DCT), TSB
and direct serum bilirubin, urea and electrolytes if the
infant is clinically dehydrated, a sepsis screen if
indicated and G6PD estimation when available.
Dockets for five (3%) of the patients could not be
retrieved; however, baseline demographic data were
collected from the logbooks of the Labour Ward, the
Newborn Special Care Nursery and the chemical
pathology and haematology laboratories. The demographics of these five patients were no different from
those of the study population.

regression models. Statistical significance was taken as

P,0.05. Analyses were performed using the Statistical
Package for the Social Sciences SPSS version 12.

Ethical approval
The Ethics Committee of the University of the West
Indies/University Hospital of the West Indies,
Faculty of Medical Sciences granted approval for
this study to be conducted.

Results
During the study period, 170 neonates fulfilled the
inclusion criteria for the study; 139 (82%) were term
or near-term infants (gestational age >34 weeks).
Sixty-five (38%) patients were admitted from the
labour ward, 62 (37%) from the postnatal ward, 35
(20%) from home and 8 (5%) from other hospitals.
Fifteen (9%) neonates had extreme hyperbilirubinaemia (TSB >428 mmol/L). Maternal age ranged from
16 to 46 years with a mean (SD) of 29.4 (6.3) years.
Fifty-four per cent of the mothers were primiparous.
A significantly greater proportion of mothers of
infants with extreme hyperbilirubinaemia had labour augmented with oxytocin than mothers of
infants without extreme hyperbilirubinaemia (P,
0.05) (Table 1). The majority of mothers had blood
group O (66%). There were 20 (12%) rhesus-negative
mothers, but only four (3%) had a positive indirect
DCT. The infants of these four mothers did not
have extreme hyperbilirubinaemia.
The majority of infants were male 103 (61%). The
mean (SD) gestational age was 36.2 (3.1) weeks and
ranged from 24 to 41 weeks. One hundred and twelve
(66%) were delivered vaginally and 58 (34%) by
caesarean section. Birthweight ranged from 670 to
4830 g with a mean (SD) of 2696 (799) g. Term
infants accounted for 100% of infants with extreme
hyperbilirubinaemia while they accounted for only
53% of neonates without extreme hyperbilirubinaemia. Ninety-three per cent of neonates with extreme
hyperbilirubinaemia were exclusively breastfed compared with 49% of neonates who did not have
extreme hyperbilirubinaemia (P,0.01) (Table 2).
All three neonates with extreme hyperbilirubinaemia
who were dehydrated were exclusively breastfed while

Case definition
Any neonate admitted to the Newborn Special Care
Unit at the UHWI with a total serum bilirubin level
requiring investigation and medical intervention was
included. Medical intervention was defined as the
need for phototherapy, exchange transfusion or both.
Extreme hyperbilirubinaemia was defined as a TSB
level >428 mmol/L.

Exclusion criteria
Neonates who did not require medical intervention
and neonates who had a direct serum bilirubin level
.20% of the value of their total serum bilirubin level
were excluded.

Statistical analysis
Descriptive analyses were performed. Differences between the groups were determined using the x2 test for
categorical variables and Students t-test for continuous
variables. Factors associated with extreme hyperbilirubinaemia were determined using multiple logistic

Table 1 Maternal demographics of neonates with hyperbilirubinaemia at UHWI 20062007

TSB (%)
,428 mmol/L
n5155
Oxytocin augmentation
Prolonged rupture of membranes .18 hrs
Gestational diabetes
Pregnancy-induced hypertension
Mode of delivery
Vaginal
Caesarean section

16
16
8
35

(10)
(10)
(5)
(23)

100 (65)
55 (35)

* P,0.05

98

Paediatrics and International Child Health

2012

VOL .

32

NO .

TSB(%)
>428 mmol/L
n515
5 (33)*
0
0
4 (27)
12 (80)
3 (20)

Trotman and Henny-Harry

Table 2 Demographics of neonates

binaemia at UHWI 20062007

Male
Female
Preterm (,37 wks)
Term (>37 wks)
Cephalhaematoma
Excessive bruising
Exchange transfusion
Exclusive breastfeeding
Formula-fed/supplemented
Nil per orally
Dehydration on admission

with

hyperbiliru-

TSB (%)
,428 mmol/L
n5155

TSB (%)
>428 mmol/L
n515

93
62
73
82
3
10
6
75
35
44
16

10
5
0
15
1
0
3
14
1
0
3

(60)
(40)
(47)
(53)
(2)
(7)
(4)
(49)
(23)
(28)
(10)

(67)
(33)
(100)
(7)
(20)*
(93)*
(7)
(20)

* P,0.05

12/16 (75%) of dehydrated neonates who did not have

extreme hyperbilirubinaemia were also exclusively
breastfed. The mean (SD) percentage weight loss on
admission for dehydrated infants with extreme
hyperbilirubinaemia was greater than that for dehydrated infants without extreme hyperbilirubinaemia
[11.5% (2.3) vs 10.7% (4.6)] but this did not achieve
statistical significance. Forty-five per cent of the
dehydrated infants were admitted from the postnatal
ward.
There was no significant difference between the two
groups regarding the day of life on which the peak
TSB was recorded. However, the neonates with
extreme hyperbilirubinaemia presented to hospital
significantly later and had significantly higher mean
TSB values on admission and mean peak TSB values
than the neonates who did not have extreme
hyperbilirubinaemia (Table 3).
In 59 (35%) neonates the underlying cause of the
hyperbilirubinaemia was ABO incompatibility; 38
(64%) of these were DCT-positive. Only 44 neonates
(26%) were tested for G6PD deficiency; 8 (18%) of
these were G6PD-deficient, which represents 5% of
the total study population. G6PD-deficient neonates
accounted for 27% of those with extreme hyperbilirubinaemia compared with only 3% of those without
extreme hyperbilirubinaemia (P,0.001) (Table 4).
Seven (88%) of the eight G6PD-deficient neonates
were male; all four G6PD-deficient neonates with
extreme hyperbilirubinaemia were male. The mean

Extreme hyperbilirubinaemia

(SD) peak TSB for these five neonates was 473 (65)
mmol/L and all values fell within the top 5% of peak
TSB values for the study population. Of the 16
neonates for whom an aetiology for their hyperbilirubinaemia could not be found, only three (18%) were
tested for G6PD.
G6PD deficiency (4, 27%) and ABO incompatibility (6, 40%) together accounted for ten (67%) of
the cases of extreme hyperbilirubinaemia. Exclusive
breastfeeding, G6PD deficiency and augmentation of
labour with oxytocin were found to be significantly
associated with TSB levels >428 mmol/L (P,0.05).
One patient was discharged with a diagnosis of
bilirubin encephalopathy but was lost to follow-up;
this was a 32-week premature infant whose TSB was
317 mmol/L and who had received an exchange
transfusion. There were two deaths, but these were
related to extreme prematurity and not directly to
hyperbilirubinaemia; neither neonate had extreme
hyperbilirubinaemia. Sixty-two patients (37%) were
followed up at the outpatient clinic at UHWI, six of
whom were in the extreme hyperbilirubinaemia
group. Thirty (50%) had hearing tests done and they
were all normal, two were from the extreme
hyperbilirubinaemia group. Sixty-one (98%) had
normal development at the time of the study; one
patient who did not have extreme hyperbilirubinaemia had impaired motor development but this infant
also had a myelomeningocele.
Significant factors associated with extreme hyperbilirubinaemia were offered into a multiple logistic
regression model; neonates who were exclusively
breastfed were more likely to have extreme hyperbilirubinaemia than those who were formula-fed or
supplemented (OR 2.6, 95% CI 0.010.6). Mothers
who received oxytocin during labour were more likely
to have extreme hyperbilurubinaemia (OR 1.7, 95%
CI 0.040.8) and neonates who were G6PD-deficient
were more likely to have extreme hyperbilurubinaemia (OR 2.4, 95% CI 0.010.6) than those whose
mothers did not receive oxytocin and those who were
not G6PD-deficient.

Discussion
Similar to previous studies, we found that neonates
with extreme hyperbilirubinaemia tended to be

Table 3 Characteristics of neonates with hyperbilirubinaemia at UHWI 20062007

Age of admission, days

Length of stay on postnatal ward, days
Total serum bilirubin on admission, mmol/L
Peak total serum bilirubin, mmol/L
Day of life peak bilirubin recorded, days
* P,0.05,

TSB ,428 mmol/L

Mean (SD)
n5155

TSB >428 mmol/L

Mean (SD)
n515

1.5
2.2
305
328
4.6

4.3
2.4
434
470
5.6

(2)
(0.7)
(89)
(54)
(2.0)

(4)*
(0.6)
(98){
(53){
(4.0)

P,0.001

Paediatrics and International Child Health

2012

VOL .

32

NO .

99

Trotman and Henny-Harry

Extreme hyperbilirubinaemia

exclusively-breastfed, male, term infants.35 In keeping with other studies, we found that ABO incompatibility was the most common underlying aetiology.3,4
Unlike previous Jamaican studies, rhesus iso-immunization was not a major cause of hyperbilirubinaemia,
and this probably reflects the efficacy of Rhogam use
in our population.8,9
Jamaica followed the recommendations of the
Innocenti Declaration in support of breastfeeding
and has adopted the Baby-Friendly Hospital
Initiative. Our study demonstrated that exclusively
breastfed neonates were three times more likely to
develop extreme hyperbilirubinaemia than those who
were formula-fed or supplemented. This association
between jaundice and breastfeeding has been documented in other studies.6,1013
Nineteen infants were dehydrated, the majority of
whom were being exclusively breastfed. Dehydrated
neonates with extreme hyperbilirubinaemia had greater weight loss on admission. Weight loss has been cited
in other studies as a risk factor for hyperbilirubinaemia, as a function of decreased energy and fluid
intake.11,14 Almost half of the dehydrated infants were
admitted from the postnatal ward. This might indicate
that monitoring of breastfeeding on the postnatal
ward is not as rigorous as it should be. Therefore, the
recognition of infants with difficulty breastfeeding in
the first 48 hours of life and the implementation of
breastfeeding support and supplementation, if medically indicated, needs to be more energetically pursued
on the postnatal ward.
This study shows that G6PD-deficient neonates
were twice as likely to develop extreme hyperbilirubinaemia. Only 26% of the neonates were tested for
G6PD deficiency and only 18% of the neonates in the
idiopathic group were tested. It is quite possible that,
if all the neonates in the study population, in
particular the males, had been tested, there would
have been a greater incidence than the 5% recorded.
In Trinidad, an incidence of 21% for G6PD deficiency was seen in infants with hyperbilirubinaemia.15
In 1972 it was reported that the incidence of G6PD
deficiency in jaundiced infants at UHWI was 20.5%.9

In that study, G6PD deficiency was detected in 16

(69.6%) of 23 neonates with unexplained moderateto-extreme jaundice. Their findings suggested that
G6PD may be an important cause of neonatal
jaundice in our setting. It therefore seems prudent
that G6PD testing be undertaken in all infants with
hyperbilirubinaemia requiring intervention, particularly males.
Augmentation of labour with oxytocin was associated
with extreme hyperbilirubinaemia. This association
between oxytocin use in labour and neonatal jaundice
has also been found by some authors while others have
disputed any such relationship.1621 Some postulate that
the pathogenesis of neonatal hyperbilirubinaemia is
owing to the vasopressin-like action of oxytocin causing
osmotic swelling of erythrocytes, leading to decreased
deformability and, hence, more rapid destruction,
resulting in neonatal hyperbilirubinaemia.16 Others,
however, believe that there is transplacentally acquired
neonatal hyponatraemia secondary to maternal oxytocin
use or infusion of electrolyte-free fluids in the mother
which affects red blood cell membrane function,
resulting in haemolysis of the cells.17 Some authors have
documented a dose-response effect with the degree of
jaundice being more extreme with increasing doses of
oxytocin.18
Our study showed a low follow-up rate of infants
at high risk of the negative effects of bilirubin toxicity
on neurodevelopmental outcome and measures must
be instituted to ensure better monitoring of these
infants after discharge. It is also of some concern that
hearing was assessed in only 50% of the infants who
were followed up at the UHWI and in only two of the
six infants with extreme hyperbilirubinaemia; it is
well recognized that the auditory system is highly
sensitive to bilirubin toxicity.22 In resource-limited
settings where universal screening of hearing in
neonates is not a reality, a programme to screen at
least high-risk infants should be implemented.
The retrospective nature of this study limited
analysis to data that were recorded. The study was
also limited to investigations that were done, and, in
cases such as G6PD deficiency where only a small
number of patients were tested, a true incidence rate
of this aetiology could not be determined. Missing
dockets (five patients) were also a limitation; however, this did not affect the validity of the study
results as accurate information was obtained from
other sources and these patients demographics were
similar to those of the general study population.
Based on this studys findings, it is recommended
that all neonates requiring phototherapy for hyperbilirubinaemia be routinely tested for G6PD deficiency, especially if they are male. Infants who are
found to be deficient should be observed more closely
for the development of hyperbilirubinaemia. Neonates

Table 4 Major causes of hyperbilirubinaemia in neonates

at UHWI 20062007

Aetiology

TSB (%)
TSB (%)
,428 mmol/L >428 mmol/L
n5155
n515

ABO incompatibility
Prematurity
Idiopathic
G6PD deficiency
Rhesus incompatibility
Culture-positive sepsis
Minor blood group incompatibility

53
19
14
4
6
5
4

(34)
(12)
(9)
(3)
(4)
(3)
(3)

6
0
2
4
0
1
0

(40)
(13)
(27)*
(7)

* P,0.001

100

Paediatrics and International Child Health

2012

VOL .

32

NO .

Trotman and Henny-Harry

whose mothers received oxytocin during labour should

be closely monitored and followed up after discharge
for the development of hyperbilirubinaemia. Additionally, mothers should not be discharged from the
postnatal ward until breastfeeding has been adequately established, and women with difficulty breastfeeding should be identified on the postnatal ward and
offered continued breastfeeding support after discharge from hospital. Exclusive breastfeeding, oxytocin use in the mother and G6PD deficiency were
associated with extreme hyperbilirubinaemia.

References
1 Stoll B, Kliegman R. Jaundice and hyperbilirubinemia in the newborn.
In: Behrman RE, Kliegman, Jenson HB, eds. Nelson Textbook of
Pediatrics. Philadelphia, PA: WB Saunders, 2004; pp 5928.
2 Britton JR, Britton HL, Beebe SA. Early discharge of the term
newborn: a continued dilemma. Pediatrics. 1994;94:2915.
3 Sgro M, Campbell D, Shah V. Incidence and causes of extreme
neonatal hyperbilirubinemia in Canada. CMAJ. 2006;175:58790.
4 Manning D, Todd P, Maxwell M, Platt MJ. Prospective
surveillance study of extreme hyperbilirubinaemia in the
newborn in the UK and Ireland. Arch Dis Child Fetal
Neonatal Ed. 2007;92:F3426.
5 Newman TB, Liljestrand P, Escobar GJ. Infants with bilirubin
levels of 30 mg/dL or more in a large managed care
organization. Pediatrics. 2003;111:130311.
6 Tiker F, Hande G, Hasan K, Aylin T, Berkan G. Extreme
hyperbilirubinemia in newborn infants. Clin Pediatr. 2006;
45:25761.
7 Salas A, Mazzil E; Exchange transfusion in infants with
extreme hyperbilirubinemia: an experience from a developing
country. Acta Pdiatr. 2008;97:7548.

Extreme hyperbilirubinaemia

8 Miller CG, Milner PF, Sue SL. Neonatal jaundice in Jamaican

children. West Indian Med J. 1963;12:142.
9 Gibbs WN, Gray RH. Glucose-6-phosphate dehydrogenase
(G6PD) deficiency and neonatal jaundice in Jamaica. Br J
Haematol. 1979;43:26374.
10 Narang A, Gathwala G, Kumar P. Neonatal jaundice: an
analysis of 551 cases. Indian Pediatr. 1997;34:42932.
11 Maisels MJ, Clifford K, Antle CE. Jaundice in the healthy
newborn infant: a new approach to an old problem. Pediatrics.
1988;81:50511.
12 Stanley I, Chung M, Kulig J, OBrien R, Subcommittee on
Hyperbilirubinemia. An evidence-based review of important
issues concerning neonatal hyperbilirubinemia. Pediatrics.
2004;114:13053.
13 Newman TB, Xiong B, Gonzales VM, Escobar GJ. Prediction
and prevention of extreme neonatal hyperbilirubinemia in a
mature health maintenance organization. Arch Pediatr Adolesc
Med. 2000;154:11407.
14 Bertini G, Dani C, Tronchin M, Rubaltelli F. Is breastfeeding
really favoring early neonatal jaundice? Pediatrics.
2001;107:e41.
15 Dillon-Remy MT, Manning-Alleyne P, Brat D, Charles W.
Neonatal jaundice at Port-of-Spain General Hospital, Abstract.
West Indian Med J. 1987;36 (suppl): 28.
16 Buchan PC. Pathogenesis of neonatal hyperbilirubinaemia after
induction of labour with oxytocin. Br Med J. 1979;2:12557.
17 Singhi S, Singh M. Pathogenesis of oxytocin-induced neonatal
hyperbilirubinaemia. Arch Dis Child. 1979;54:4002.
18 Chalmers I, Campbell H, Turnbull AC. Use of oxytocin and
incidence of neonatal jaundice. Br Med J. 1975;2:11618.
19 Noble AD. Oxytocin and neonatal jaundice. Br Med J.
1978;1:239.
20 Friedman L, Lewis P, Harvey D. Oxytocin and neonatal
jaundice. Br Med J. 1977;6096:1223.
21 Singhi SC, Choo Kang E, Hall, JSE. Intrapartum infusion of
oxytocin and glucose water and neonatal jaundice. West Indian
Med J. 1988;33:803.
22 Shapiro SM, Nakamura H. Bilirubin and the auditory system. J
Perinatol. 2001;21 (Suppl 1):S525.

Paediatrics and International Child Health

2012

VOL .

32

NO .

101

Copyright of Paediatrics & International Child Health is the property of Maney Publishing and its content may
not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written
permission. However, users may print, download, or email articles for individual use.