Evaluating the ingram icterometer as a screening tool for significant neonatal hyperbilirubinemia at the Kenyatta National Hospital.

Abstract

Background: Neonatal Jaundice (NNJ) occurs in 30-60% term newborns and is significant (>221 umol/l) in 3.5-12% of these neonates. Kernicterus is the worst complication of NNJ and is associated with at least 70% morbidity and 10% mortality. In Kenya (Kilifi), severe NNJ accounts for about 22% admissions with an in-patient case fatality rate of 26%. At present all jaundiced newborns are screened by a serum bilirubin test. There is no noninvasive, sensitive, screening device in place to enable early detection of those neonates who may require intervention, hence the need to evaluate the icterometer. The icterometer, in studies conducted in Turkey, India and the USA, has shown a linear correlation with total serum bilirubin (TSB), with high sensitivity and specificity for detecting significant neonatal jaundice. Objectives: The main objective of this study was to determine the sensitivity and specificity of the Ingram icterometer for predicting the serum bilirubin levels in jaundiced term newborns, and the secondary objective was to determine the sensitivity and specificity of clinical assessment on the sole of the foot. Study design: Cross-sectional study. Study setting: Kenyatta National Hospital (KNH), paediatric filter clinic (PFC), new born unit (NBU), and the paediatric wards. Study population: Jaundiced term newborns: 2:37 weeks gestation or 2:2500 g birth weight. Sampling: Consecutive sampling of subjects who met the study inclusion criteria. Sample size: 143jaundiced neonates. Procedures: Transcutaneous bilirubin (TcB) measurements were done with the icterometer on neonates for whom the primary clinician had requested serum bilirubin. Only those neonates who had not had phototherapy or exchange transfusion were included. Two icterometer readings were done and the higher reading was taken. Serum bilirubin was determined in routine biochemistry laboratory (lab) at KNH. A follow up was done to document how many of the tested neonates had phototherapy and/or exchange transfusion, and what the eventual outcome was. Clinical assessment was done by blanching the sole of the foot and documenting presence or absence of jaundice.Results: A total of 143 jaundiced term neonates were recruited into the study. The mean gestational age was 39.3 weeks and the mean birth weight was 3100 g with the commonest comorbidity being neonatal sepsis. The sensitivity and specificity of the icterometer at index 3 and at a serum bilirubin cut-off of 221 mcmolll, was 99% and 55.3% respectively. At serum bilirubin cut-off of 257 mcmolll, the sensitivity and specificity of clinical assessment at the sole of the foot was 67% and 74.5% respectively. Poor outcome was associated with higher mean serum bilirubin levels. Conclusion: The icterometer at a cut-off index of 3 offers excellent sensitivity but only moderate specificity. The clinical assessment at the sole of the foot offers moderate sensitivity and specificity. Recommendations: The icterometer performs well to detect possibly serious jaundice and is recommended for routine screening in term jaundiced neonates. Although it lacks specificity, its performance compares favourably with clinical assessment which it might usefully replace