Page 145 - 16Neonatal Jaundice_compressed

Neonatal jaundice

and in the other 159 EL 1+. Neither study reported on the randomisation method. One study 159
reported using sealed envelopes for allocation concealment.
The mean gestational ages were 30.7 ± 2.0 weeks and 33.6 ± 1.9 weeks, the mean ages at time
of entry to the study were 64.4 ± 15.2 hours and 68.1 ± 25.5 hours, the mean birthweights were
1192 ± 238 g and 1998 ± 541 g, and the mean baseline serum bilirubin levels were
180 ± 38 micromol/litre and 200 ± 16 micromol/litre. One study 158 reported gender and 58
participants (65.9%) were male.

Review findings
There were no reported cases of exchange transfusions or treatment failures in either group.
There were fewer cases of rebound jaundice in the conventional phototherapy group (eight
versus 12) but this difference was not statistically significant.

Phototherapy in both studies was terminated once a predefined serum bilirubin level was
reached so it was not possible to calculate the mean decrease in serum bilirubin. Babies in the
LED phototherapy had a statistically significantly shorter duration of phototherapy
(MD = −9.15 hours, 95% CI −3.53 to −14.77) but heterogeneity was high (I² = 90%).

Overall evidence summary for phototherapy in preterm/low-birthweight babies
The pooled results of meta-analysis indicate that phototherapy is effective in the treatment of
hyperbilirubinaemia in preterm and low-birthweight babies.
Babies who received early phototherapy had a statistically significant lower mean peak in serum
bilirubin level. Early phototherapy was also found to statistically significantly decrease the risk
of exchange transfusion and treatment failure when compared with no treatment. However, in
the study that contributed most to this analysis, the exchange transfusion thresholds were very
cautious and would not be used in current clinical practice in the UK.
Multiple phototherapy did not show any clinical difference on any outcome when compared
with conventional phototherapy.
There was no statistically significant difference in the number of exchange transfusions or
treatment failures in studies comparing fibreoptic with conventional phototherapy. Fibreoptic
phototherapy was, however, statistically significantly better than conventional therapy in terms
of duration of treatment.

LED phototherapy was shown to shorten statistically significantly the duration of treatment
compared with conventional phototherapy in preterm babies. Conversely, there was a trend
towards a greater decrease in serum bilirubin levels among the group treated with conventional
phototherapy but this was not statistically significant.

Overall GDG translation from evidence for phototherapy in preterm/low-
birthweight babies

All modes of phototherapy when used and maintained according to the manufacturer’s
instructions are safe and effective as first-line medical treatment of hyperbilirubinaemia in
preterm babies.
The evidence supporting the use of early phototherapy in preterm babies is limited by the
relatively low thresholds for exchange transfusion used in one study that contributed most to
the analysis and does not reflect current clinical practice in the UK. Early initiation of
phototherapy in preterm babies is effective in reducing the duration of phototherapy and
reducing peak bilirubin levels. The GDG is of the opinion that this evidence supports the
choice of relatively low threshold levels for starting phototherapy in preterm babies based on
the fomula given in Section 7.1.1.
GDG experience is that fibreoptic devices are more acceptable to parents and nursing staff for a
number of reasons, including less glare than from overhead lamps, the fact that parents can hold
and feed the baby, and that no eye protection is needed. However, fibreoptic phototherapy was
less effective than conventional phototherapy in term babies. Monitoring the effect of treatment
is essential because despite phototherapy some babies require further medical interventions.

114