Papers

Jaundice in babies: implications for community screening for biliary atresia

BMJ 1995; 310 doi: https://doi.org/10.1136/bmj.310.6988.1172 (Published 06 May 1995) Cite this as: BMJ 1995;310:1172
  1. D A Kelly, consultant paediatric hepatologista,
  2. A Stanton, consultant community paediatricianb
  1. a Children's Hospital, Birmingham B16 8ET
  2. b Solihull Healthcare NHS Trust, Birmingham B37 7XR
  1. Correspondence to: Dr Stanton.
  • Accepted 18 January 1995

Although prompt referral of babies with biliary atresia ensures early surgery and improves outcome,1 late referral persists in the United Kingdom, leading to calls for systematic screening of babies with neonatal jaundice persisting beyond 14 days.2 This would entail reviewing the baby, establishing whether the stool is pigmented, testing urine for bilirubin, and measuring blood concentrations of conjugated and unconjugated bilirubin.2 3 To evaluate the feasibility of systematic screening, we ascertained the prevalence of (a) jaundice in babies aged 2, 4, and 6 weeks, and (b) dark urine or pale stools as reported by the parents or a health visitor. We also established how often stools or urine of jaundiced babies were noted by health visitors at their first visit to gauge the effect on workload.

Subjects, methods, and results

In Solihull Health District health visitors examine all babies between 12 and 14 days of age. We recorded jaundice at this primary visit for convenience of clinical practice. Over six months the health visitors noted name, date of birth, date of examination, method of feeding, and presence of clinical jaundice (defined as yellow skin pigmentation). If jaundice was present they noted the colour of the eyes, stools, and urine and whether the colour of stools and urine was based on their observation or a parental report. Further reports were submitted at 4 and 6 weeks if the babies were still jaundiced. The study excluded babies still in hospital but included those who had moved into the district in the first 14 days of life. Denominator data were taken from the national child health computer system.

Detection of jaundice by health visitors at primary visit and when babies were aged 4 and 6 weeks. Values are numbers of babies unless stated otherwise

View this table:

From 1 September 1992 to 25 February 1993, 1170 live babies were born to the residents of Solihull Health District or had moved into the district in their first 14 days of life. The table shows the prevalence of jaundice by age. At the primary visit 175 babies were thought to be jaundiced. Only 78 babies had yellow eyes. Stool was seen in seven cases, urine in 10, both in nine, and neither in 119; the report was incomplete in 30. Dark urine was reported by the parents in three cases, one baby also having pale stools; pale stool was reported by the parents in three and by a health visitor in one. All these babies were well and free of jaundice at 6 weeks.

Seven babies were still jaundiced at 6 weeks, six of whom were breast fed. At 1 year the bottle fed baby had mildly raised transaminase activities and one baby was found to be heterozygous for (alpha) antitrypsin deficiency, but none had clinically significant liver disease.

Comment

Although 175 babies were reported as being clinically jaundiced at the primary visit, only 78 had yellow eyes, suggesting that visual screening for hyperbilirubinaemia may yield many false positive results. The prevalence of persistent jaundice was higher than might have been predicted.4 This increase may be due to differences in the prevalence of breast feeding (58% of the mothers breast fed) or a genuine secular trend—others have noted an increase in the prevalence of neonatal jaundice.5 We found prolonged neonatal jaundice to be commoner in boys, as reported previously,5 which may be owing to a difference in maturity of hepatic bilirubin conjugation.

Defining clinical jaundice as yellow would lead to around 15% of babies requiring blood and urine sampling, representing a major impact on workload for health visitors, laboratories, and medical staff. Further investigation is needed to determine the cost-benefit characteristics of the suggested screening programme and other strategies.

We thank the health visitors of Solihull Healthcare NHS Trust for their help in collecting the data.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
View Abstract