Conflict of interest: None.
Yellow is pale: The complications and challenges of late diagnosis of extrahepatic biliary atresia
Article first published online: 8 OCT 2012
© 2012 The Authors. Journal of Paediatrics and Child Health © 2012 Paediatrics and Child Health Division (Royal Australasian College of Physicians)
Journal of Paediatrics and Child Health
Volume 49, Issue 2, pages 152–155, February 2013
How to Cite
Ho, S. S., Haller, W. and Catto-Smith, A. G. (2013), Yellow is pale: The complications and challenges of late diagnosis of extrahepatic biliary atresia. Journal of Paediatrics and Child Health, 49: 152–155. doi: 10.1111/j.1440-1754.2012.02501.x
Financial disclosure: None.
- Issue published online: 18 FEB 2013
- Article first published online: 8 OCT 2012
- Accepted for publication 11 October 2011.
- extrahepatic biliary atresia;
- intracerebral haemorrhage;
- vitamin K deficiency
Extrahepatic biliary atresia classically presents in the neonatal period with jaundice and pale stools. The lack of bile pigment in stool can be unrecognised, delaying diagnosis and surgical treatment. Vitamin K is given at birth to reduce the risk of haemorrhagic disease of the newborn, but this may be inadequate to prevent the development of coagulopathy secondary to fat soluble vitamin malabsorption. We present the case of a 3 month old infant who presented with an intracerebral haemorrhage and coagulopathy thought to be secondary to fat malabsorption resulting from delayed diagnosis of extrahepatic biliary atresia. This was despite the perinatal administration of intramuscular vitamin K. His parents did not recognise the stool pallor as being abnormal. This case illustrates the importance of educating parents on the significance of pale stools, and also the risk of coagulopathy in extrahepatic biliary atresia despite perinatal intramuscular vitamin K.
A 95-day-old term male infant was brought to a regional emergency department with 2-day history of worsening jaundice, vomiting and bruising. The child was the second of non-consanguineous Cambodian parents and had an unremarkable birth and neonatal history. Physical examination revealed a lethargic infant with scleral icterus and bruises on his left hand. There was firm hepatomegaly, but no splenomegaly or ascites.
Soon after arrival, his condition deteriorated with a bulging anterior fontanelle. He had a brief generalised tonic clonic seizure leading to intubation.
Laboratory investigations showed normal inflammatory markers and serum electrolytes. His liver function profile revealed a predominantly cholestatic picture with conjugated hyperbilirubinaemia (total bilirubin 121 µmol/L, conjugated bilirubin 71 µmol/L, normal range (NR) <7 µmol/L) and significant elevation of gamma-glutamyl transferase of 967 U/L (NR <40 U/L). There was a significant coagulopathy with international normalised ratio (INR) >10. This corrected completely with intravenous vitamin K and fresh frozen plasma prior to transfer to our institution.
A cranial CT confirmed a left frontal lobe haemorrhage with secondary hydrocephalus (Fig. 1), which was decompressed with an external ventricular drainage. His coagulation profile remained subsequently normal with regular vitamin K supplementation.
The health record of the child identified a standard dose of 1 mg of vitamin K (phytomenadione – vitamin K1) given intramuscularly on day 1 after birth. The child had been exclusively breastfed since and thrived well. Importantly, his parents described his stool colour as having been always yellow, including the specimens collected shortly before hepatoportoenterostomy. On examining the same stool, it (Fig. 2) was actually pale.
An abdominal ultrasound showed a contracted gall bladder. A subsequent liver biopsy was suggestive of large bile duct obstruction. The operative findings confirmed the diagnosis of extrahepatic biliary atresia (EHBA). Hepatoportoenterostomy was performed on day 114 of life, having been delayed because of the need for neurosurgery. His stool colour became pigmented on day 8 after surgery (Fig. 2). He was subsequently discharged home without any neurological sequelae. He continues to pass pigmented stools, though now has firm hepatosplenomegaly, persistently elevated transaminases and mild thrombocytopenia suggestive of cirrhosis and early portal hypertension.
- 1Late vitamin K deficiency bleeding (VKDB) from long-standing cholestasis in undiagnosed chronic liver disease such as extrahepatic biliary atresia (EHBA) occurs despite intramuscular vitamin K prophylaxis given at birth.
- 2Early detection of EHBA is necessary to avoid catastrophic late VKDB as well as deferring morbidity and mortality associated with chronic liver disease.
- 3Pale stool colour is an important early clinical sign of liver disease in the jaundiced neonate and infant. Standardisation of stool colour assessment with a colour card may help to reduce inter-individual variability. Stool colour cards may have a role as a universal screening test for early detection of neonatal cholestasis.
Jaundice is very common in breastfed neonate. It is crucial to differentiate the infant with benign breastmilk jaundice from those with conjugated hyperbilirubinaemia as neonatal cholestasis is often associated with liver disease. EHBA is an important cause of neonatal cholestasis and is characterised by progressive inflammation and obliteration of large bile ducts eventually leading to cirrhosis. Pale stool colour and jaundice are early clinical features, while failure to thrive, hepatosplenomegaly and ascites occur later in the course of disease.
The association of EHBA with late vitamin K deficiency bleeding (VKDB) is rarely seen in Australia since introduction of vitamin K prophylaxis at birth. VKDB occurs in long-standing cholestasis with malabsorption of vitamin K or in the setting of exclusive breastfeeding.1 Associated morbidity and mortality are high.1
A single intramuscular dose of vitamin K at birth has proven successful in reducing the overall incidence of late VKDB.1 Intramuscular administration of vitamin K appears to produce a depot of the drug with relatively long-term availability2 in contrary to oral or intravenous dosing regimes which appear to be ineffective in preventing catastrophic late VKDB.1,3 However, our case report underlines that even parenteral vitamin K prophylaxis does not efficiently prevent VKDB in the setting of long-standing undiagnosed chronic liver disease. Early identification of this high-risk group of infants is therefore crucial.
Apart from preventing the potential catastrophic consequences of VKDB, early detection and treatment of EHBA is critical for the medium- and long-term hepatic outcome of the patient with EHBA. Timely hepatoportoenterostomy (Kasai procedure), the accepted first-line treatment, delays the need for liver transplantation by slowing the progression of cirrhosis and development of end-stage liver disease.
The median age of hepatoportoenterostomy in most Western countries range between 55 and 65 days.4–6 Up to 20% of EHBA cases in recent series from Canada and USA have hepatoportoenterostomy beyond 90 days of life.5,6 Unpublished data from the Royal Children's Hospital, Melbourne, a main referral hospital for EHBA in Victoria, showed a median age at hepatoportoenterostomy of 85 days (range 4–124 days) for the last 6 years between 2004 and 2010 (Total cases = 27). Thirty-seven per cent of all hepatoportoenterostomies were performed after 90 days of life.
Different factors have been shown to contribute to late referral and diagnosis of EHBA. There is lack of awareness in both parents and health-care providers that jaundice beyond 2 weeks of age needs urgent investigations.6 Jaundice is often wrongly attributed to breastfeeding. Dark skin colour can make early detection of jaundice difficult.7 One of the decisive factors leading to a delay in diagnosis of the cholestatic infant is the misinterpretation of stool colour: as in our case, the pigmentation of pale stool is regularly overestimated,6 yellow is actually pale.
The difficulties of late diagnosis and its implications on treatment success have enhanced research into ways of early detection and screening. Biochemical screening tools have proven unsuccessful due to lack of specificity (e.g. measuring bile salts from Guthrie cards8); detecting consistent conjugated bilirubin results using dry blood spots is yet to be achieved.8
A promising screening tool seems to be a stool colour card defining normal and abnormal stool colours (Fig. 3) which has been successfully introduced in Taiwan9 and Argentina.4 The stool colour card is integrated into the child health booklet, allowing a more reliable and consistent interpretation of stool pigment determination through parents and primary health-care providers during the 1-month health check.9 The stool colour card as a screening tool was reported to have 97.1% sensitivity and 99.1% specificity in detecting EHBA before 60 days of age. Kasai procedures beyond 90 days were eliminated since its introduction in Taiwan.9
This unusual case exemplifies the ongoing challenge of late diagnosis of EHBA as it highlights the catastrophic complication of VKDB which is not prevented by intramuscular vitamin K in this high-risk group of infants. Timely diagnosis is therefore crucial, but misinterpretation of stool colour is a frequent cause for delay. Standardisation of infant stool colour interpretation by integrating a stool colour card into the child health book has the potential to overcome the community's and health professional's subjective interpretation of stool colour. It could thereby play a role as a universal screening tool for early detection of neonatal cholestasis.
We would like to thank Professor Mei-Hwei Chang, Department of Pediatrics, National Taiwan University Hospital, and the Bureau of Health Promotion, Department of Health, Taiwan for their kind permission of using their stool colour card.
- 6Biliary atresia: the timing needs a changin. Can. J. Public Health 2009; 100: 475–7., , , .
Multiple Choice Questions
- 1Which of the following methods of delivering vitamin K prophylaxis prevents late vitamin K deficiency bleeding in all children?
- a. Oral
- b. Intravenous
- c. Intramuscular
- d. Subcutaneous
- e. None
A single intramuscular dose of vitamin K at birth has proven successful in reducing the overall incidence of late VKDB.1 Intramuscular administration of vitamin K appears to produce a depot of the drug with relatively long-term availability2 in contrary to oral or intravenous dosing regimes which appear to be less effective in preventing catastrophic late VKDB. However, infants with long-standing cholestasis, particularly those who are breastfed, are at risk of vitamin K prophylaxis failure, even if vitamin K is given intramuscularly.
- 2Which of the following is the best early indicator of neonatal and infant cholestasis?
- a. Jaundice beyond 2 weeks of age
- b. Hepatomegaly
- c. Prolonged prothrombin time
- d. Conjugated serum bilirubin
- e. Pruritus
Prolonged jaundice does not allow differentiations between benign breastmilk jaundice and cholestatic liver disease, hence further investigations are required. Determination of the conjugated bilirubin fraction is indeed the diagnostic test of choice being considered abnormal if above 15% of total serum bilirubin.
Hepatomegaly and pruritus are important, yet unspecific features of chronic liver disease in infancy and often do not occur until later during the course of disease.
A prolonged prothrombin time is usually a late indicator of chronic liver disease, either reflecting synthetic hepatic failure as a consequence of advanced cirrhosis or vitamin K deficiency due to long-standing cholestasis.
- 3The following factors have been shown to contribute to late referral and diagnosis of EHBA except?
- a. Jaundice attributed to breastfeeding
- b. Overestimation of pale stool colour
- c. Dark skin colour
- d. Language barrier
- e. Splenomegaly
There is lack of awareness in both parents and health-care providers that jaundice beyond 2 weeks of age needs urgent investigations.6 Jaundice is often wrongly attributed to breastfeeding. Dark skin colour can make early detection of jaundice difficult.7 Language barriers may further hamper timely diagnosis, which is also noted in our case report.
Pale stool colour is an early and sensitive sign of infant liver disease, remains however often undetected due to overestimation6: yellow is actually pale. Standardisation of stool colour assessment in early infancy through the parents has been successfully implemented in several countries allowing earlier detection of infant liver disease such as EHBA.
Splenomegaly is as hepatomegaly reflective of progressed liver disease and advanced portal hypertension occurring late in the course of disease. However, it does not lead to a delay in referral and diagnosis of infant cholestasis.