Page 92 - 16Neonatal Jaundice_compressed

Early prediction of serious hyperbilirubinaemia

Recommendations – 4.1 Tests that should be used to predict significant
hyperbilirubinaemia
Measure and record the serum bilirubin level urgently (within 2 hours) in all babies with
suspected or obvious jaundice in the first 24 hours of life.
Continue to measure the serum bilirubin level every 6 hours for all babies with suspected or
obvious jaundice in the first 24 hours of life until the level is both:
• below the treatment threshold
• stable and/or falling.
Arrange a referral to ensure that an urgent medical review is conducted (as soon as possible
and within 6 hours) for all babies with suspected or obvious jaundice in the first 24 hours of
life to exclude pathological causes of jaundice.
Interpret bilirubin levels according to the baby’s postnatal age in hours and manage
hyperbilirubinaemia according to threshold table (Section 1.3) and treatment threshold graphs
(Section 1.6).
Do not measure bilirubin levels routinely in babies who are not visibly jaundiced.

Research recommendation
What is the comparative accuracy of the Minolta JM-103 and the BiliChek when compared
to serum bilirubin levels in all babies?
Why this is important
Evidence: The accuracy of transcutaneous bilirubinometers (Minolta JM-103 and BiliChek)
has been adequately demonstrated in term babies below treatment levels (bilirubin
< 250 micromol/litre). New research is needed to evaluate the accuracy of different
transcutaneous bilirubinometers in comparison to serum bilirubin levels in all babies.
Population: Babies in the first 28 days of life. Subgroups to include preterm babies, babies
with dark skin tones, babies with high levels of bilirubin and babies after phototherapy.
Exposure: Bilirubin levels taken from different transcutaneous bilirubinometers. Comparison:
Bilirubin levels assessed using serum (blood) tests. Outcome: Diagnostic accuracy (sensitivity,
specificity, positive predictive value, negative predictive value), parental anxiety, staff and
parental satisfaction with test and cost effectiveness. Time stamp: Sept 2009

4.2 Tests that do not predict hyperbilirubinaemia

Umbilical cord blood bilirubin

Description of included studies
26
30
31
29
Four studies of EL II conducted in various countries (Germany, India, Denmark and Spain )
have been included. The study population was made up of healthy term babies in three
30
studies, 26;29;31 while in the German study the population included healthy term babies who
were appropriate for gestational age, healthy term babies who were small for gestational age
30
and healthy preterm babies (gestational age > 34 weeks). Data from this study were extracted
and analysed separated for both appropriate for gestational age and small for gestational age. In
three studies cord blood bilirubin was measured within 2 hours of birth and the standard
reference test (laboratory serum bilirubin measurement) was carried out within 3–4 days, while
in the German study blood testing was done only in those babies who had a Minolta JM-102
transcutaneous bilirubin reading > 16 reflectance units. A meta-analysis was conducted with
data from three studies 26;30;31 that had defined hyperbilirubinaemia as serum bilirubin levels
≥ 290 micromol/litre. The threshold values of cord blood bilirubin in these studies were ≥ 30,
29
> 34 and ≥ 37 micromol/litre, respectively. In the Danish study, the ability of cord blood
bilirubin at levels ≥ 35 micromol/litre (best cut-off value derived from the ROC curve) to

63