Page 210 - 16Neonatal Jaundice_compressed

Neonatal jaundice

Which factors affect the relationship between neonatal hyperbilirubinaemia and kernicterus or other adverse outcomes (neurodevelopmental,
auditory)?

Bibliographic details Study type and Patient characteristics Methodology and interventions Results Reviewers Comments
Evidence level

Newman TB; Study Type: Cohort of all infants with 1) Relationship of clinical and Maternal and prenatal factors associated Unselected population but exclusion
Nested case- BW = 2000 g and GA = 36 weeks demographic factors associated with significant hyperbilirubinaemia (those criteria not defined
Year: 2000 control study born alive at 11 hospitals of a with hyperbilirubinaemia with P < 0.05 in bivariate analysis) Confounding variables controlled for
health maintenance organisation evaluated by bivariate analysis during multivariate analysis
Country: USA Evidence Level: II during a two year period and OR Maternal factors Test & Reference test described
(n = 51 387) adequately
9  Race, Reference test a standard test Blinding –
Cases: 2) Risk factors significant in the  maternal age, Not reported
Babies with maximum TSB levels univariate model entered into  family history of jaundice in a
multiple regression analysis to
newborn,
≥ 427 micromol/litre within the find independent predictors of  vacuum delivery
first 30 days after birth hyperbilirubinaemia – both by
n = 73 including and excluding early
Mean BW: Not reported Neonatal factors
Mean GA: Not reported jaundice cases  Male sex,

Gender: Males = 67.1% Early jaundice cases (n = 14)  lower GA,
Ethnicity: Not reported (only defined as babies with TSB  early jaundice,
maternal race specified)  cephalohaematoma,
exceeding recommended  bruising,
phototherapy threshold for age  breastfeeding at time of
Controls: during birth hospitalisation, discharge
Random sample of babies from the those given phototherapy during
birth hospitalisation,
cohort with maximum TSB levels Factors independently associated with
< 427 micromol/litre when jaundice noted at less than significant hyperbilirubinaemia from
n = 423 20 hours of age and TSB not multivariate regression analysis (OR with
measured within 6 hours of that
Mean BW: Not reported time. 95% CI)
Mean GA: Not reported
Gender: Males = 54.4% 3) Risk index developed by All cases (n = 73)
Ethnicity: Not reported (only
maternal race specified) assigning points equal to the OR
for risk factors that were Early jaundice: OR 7.3 (2.8–19)
significant in the logistic
GA (per wk): OR 0.6 (0.4–0.7)
For analyses examining the use of regression model with the Breastfeed only at discharge: OR 6.9 (2.7–
phototherapy only, additional
random sample of 30 babies with exclusion of early jaundice cases, 17.5)
maximum TSB levels of 342 to and predictive accuracy Asian race: OR 3.1 (1.5–6.3)
426 micromol/litre added to the compared by the c-statistic Bruising: OR 3.5 (1.7–7.4)
Cephalohaematoma: OR 3.2 (1.1–9.2)
(equal to area under ROC curve)
control group Maternal age ≥ 25 years: OR 2.6 (1.1–9.2)

Exclusion criteria: Reference standard:
Not defined Significant hyperbilirubinaemia Cases excluding early jaundice (n = 59)
defined as maximum TSB GA (per wk): OR 0.6 (0.4–0.7)
levels = 428 micromol/litre
within the first 30 days after Breastfeed only at discharge: 5.7 (2.1–

birth. 15.5)
Asian race: OR 3.5 (1.7–7.4)
178