Assessment of stool colour in community management of prolonged jaundice in infancy

Jaundice persisting beyond the first 2 wk of life is often regarded as an i ndication for investigation to exclude cholestatic liver disease. Most babi es with prolonged jaundice have breast milk-related jaundice, which is a be nign condition. Cholestatic liver disease is usually accompanied by pale st ools and yellow or orange urine. A community programme was established to a scertain the incidence of prolonged jaundice and determine whether abnormal stool and urine colour could be used to assist primary care staff in refer ral decisions. Data were collected on normal stool and urine colour and use d to devise a colour chart and information sheet for parents. Babies with p rolonged jaundice were identified and referred for investigation. In all, 3 661 babies were recruited into the study, of which 127 were jaundiced at 28 d of age. Of these, 125 were breastfed. The incidence of jaundice in breas tfed babies at 28 d was 9.2% (95% CI 7.8%-11.0%) Abnormal liver function te sts (LFTs) were common, but no baby had abnormal stool or urine colour and none was found to have liver disease. Jaundiced breastfed babies who are we ll are unlikely to have serious disease. Elevated LFTs are compatible with a diagnosis of breast milk-related jaundice. Prolonged jaundice in bottle-f ed babies, and persistent pallor of stools or yellow/orange urine, are rare and merit immediate referral. Parents and professionals can be advised to report pale stools without generating a large number of unnecessary referra ls. Further work is needed to determine whether a colour chart reduces the mean age of referral and treatment of infants with cholestatic liver diseas e.