Correlation between total serum bilirubin and clinico-laboratory parameters of babies admitted for neonatal jaundice in a resource-limited setting

Abstract

Background: Neonatal jaundice (NNJ) is one of the most frequent reasons newborn babies in low‐ to
middle‐income countries visit the hospital during the neonatal period. If not promptly managed, it could
lead to serious neurological sequela and mortality. The level and type of bilirubin in serum are an important
determinant of the management option in NNJ. Unfortunately, this simple but fundamental investigation
is not readily available in many hospitals in low‐ to middle‐income countries, and where available, needs
to be done serially, an expenditure most clients cannot afford.
Methodology: This study enrolled 83 newborns admitted and managed for NNJ at the Enugu State University
Teaching Hospital during an 18‐month period. We report correlation between total serum bilirubin (TSB)
and selected newborn clinical and laboratory parameters.
Results: A total of 83 newborns were admitted for NNJ during the study. The mean TSB (mTSB) was
307 ± 145.2 μmol/L. Preterm infants(T = 0.462, P = 0.500), neonates with birth weight <2.5 kg (T = 0.219,
P = 0.804), female neonate (T = 0.314, P = 0.578), and those that presented with skin level at the
lower body region (T = 28.52, P = 0.001) had a higher mTSB at presentation to the special care baby
unit. Similarly, mTSB at presentation was higher in neonates with PCV <40% (T = 0.005, P = 0.942),
negative malaria parasite film (T = 0.01, P = 0.933), and those that were Glucose‐6‐Phosphate
dehydrogenase deficient (T = 1.59, P = 0.221). Only skin level of jaundice at presentation was
significantly correlated to the TSB (R = 0.818 P = 0.001) while gestational age (R = 0.096, P = 0.500),
age at presentation (R = 0.197, P = 0.157), and birth weight (R = −0.107, P = 0.459) were not. Linear
regression analysis showed that for every regional rise in skin level of jaundice (X), the TSB (Y) in μmol/L
increases by a factor of 89.74 expressed in a regression equation as, Y = −10.66 + 89.74 X.
Conclusions: The skin level of jaundice can be used as a rough guide in the estimation of total bilirubin
in newborns with jaundice. This is especially useful in resource-limited setting where facilities for testing
bilirubin level are lacking.