 |
|
Brief Reports Indian Pediatrics 2001; 38: 893-898 |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
Evaluation of Cholestatic Jaundice in Young Infants |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
John Matthai Cholestatic jaundice occurs in young infants due to a variety of causes(1). Irres-pective of the etiology, its pathophysiological consequences on liver cell function are the same. Early recognition and treatment may prevent permanent liver damage at least in some of the patients. Much of the data avail-able on this subject is from western literature. This study was done to analyze the various causes that lead to cholestatic jaundice in infants in a developing country. Subjects and Methods The study was undertaken between November 1996 and October 1999. Infants below 6 months of age with cholestatic jaundice were the subjects. Babies with neonatal sepsis were excluded. A complete blood picture, total liver function test (LFT), gamma glutamyl transpeptidase (G.G.T.), lipid profile, ultrasound abdomen, TORCH screening and percutaneous liver biopsy (when prothrombin time normalized) were done. Liver biopsy was not done in 5 infants - 1 with hypothyroidism and 4 with metabolic liver disease and uncorrected grossly pro-longed prothrombin time. Radio-isotope scans were done in selected patients. Specific biochemical tests were done as required in individual patients for confirmation of diagnosis. The diagnosis of biliary atresia was made on histopathology and/or laparo-tomy(2). Infants whose percutaneous liver biopsy or biochemical profile and ultrasound was suggestive of biliary atresia, were subjected to laparotomy. Neonatal hepatitis was confirmed by histopathology and sero-logy, while bile duct paucity was diagnosed on histopathology. Byler’s disease was diagnosed in children in whom itching was a predominant symptom, the alkaline phospha-tase was high and the GGT was very low and the histopathology was supportive of the diagnosis(3). Metabolic liver disease was considered in infants who presented early with evidence of gross hepatocellular dys-function, without evidence of congenital infection, and had the appropriate metabolic tests positive. All infants were followed up until death or for at least 1 year. Results There were 36 infants with cholestatic jaundice (20 males and 16 females). The mean age at presentation was 87 days, (range 20-180 days). The mean reported age of onset of jaundice was 40 days (range 5-120 days). The etiology of cholestatic jaundice is given in Table I. In 2 infants - one with metabolic liver disease and one with congenital cyto-megalovirus hepatitis, skin and mucosal bleeds were the presenting symptoms. Table I - Causes of Cholestatic Jaundice
Five of the 7 infants with biliary atresia were operated (Kasai’s procedure). In 2 infants, percutaneous liver biopsy was carried out, but surgery was not insisted on, since they were 140 days and 150 days old and the liver biopsy showed extensive fibrosis. The mean birth weight of infants with biliary atresia was 3 kg and the mean age at presentation was 90 days, the mean age of onset of jaundice was however 33 days. In the 5 operated patients, the mean age at surgery was 90 days (range 55-120 days). None of the 5 infants achieved complete bile drainage after surgery and all had irrversible liver cell failure by 9 months of age. Three patients died before 1 year of age. The biochemical profile is given in Table II. The prothrombin time was prolonged at admission in 3 patients, but was corrected after vitamin K therapy. The echotexture and echogenicity of the liver was normal on ultrasound in all the 5 operated patients. Among the 14 with neonatal hepatitis, 6 were due to cytomegalovirus and 2 were positive for Herpes. The TORCH screen IgM was all negative in the other six patients. VDRL was negative in all fourteen infants. The mean birth weight was 2.l kg (range 1.3-2.8kg). The mean age at presentation was 89 days and the mean age of onset of jaundice was 27 days. Their biochemical profile is given in Table II. Among the 14 with congenital infection, 4 infants recovered completely and had normal liver function tests at 1 year of age. Their bilirubin level was below 7 mg/dl, SGPT below 100 units/L and prothrombin time was normal at presentation. Four infants died and 5 had irreversible liver cell failure by 8 months of age. One infant was lost to follow up. Of the 3 infants with Byler’s disease, 2 died before 18 months of age and 1 had near normal liver function at 2 years of age. All 3 were treated with phenobarbitone and ursodeoxycholic acid. The surviving child developed gall stones at 1½ years and required cholecystectomy. The biochemical profile of the 3 patients is given in Table II. The prothrombin time was normal in all the patients. Among the 5 infants with metabolic liver disease, 1 had low serum a1, antitrypsin levels, 1 had hereditary fructose intolerance, 2 had perinatal hemochromatosis (siblings) and 1 had tyrosinemia. All 5 patients with metabolic liver disease presented before 45 days of birth and had grossly prolonged prothrombin time. In one, the prothrombin time was corrected with vitamin K. All had SGPT below 100 U/L, disproportionately low for the billrubin level. Their course was relentlessly progressive with death before 5 months of age. In 4 patients, no firm diagnosis was possible, despite investigations. All of them recovered completely both clinically and biochemically by 6 months of age. All of them presented before 60 days of age, had normal birth weight and thrived well. Their biochemical characteristics are given in Table II. Their liver biopsy showed normally preserved liver architecture without bile plugs, mild increase in fibrous tissue and mild infiltration of the portal region with lympho-cytes. With phenobarbitone, the bilirubin levels returned to normal in 3 weeks, SGPT in 4-6 weeks but the ALP remained in the 600-800 U/L range even 6 months later. Six of 7 patients with biliary atresta had gall bladders measuring less than 2.5 cms on ultrasound scan after a 3 hour fast. In none of these 6 patients was the gall bladder size decreased after feeding. Among the remaining 29, 6 had gall bladder size below 2.5 cm. In 23 of these 29 patients, the gall bladder contracted after feeding. In the overall analysis, 10 (NH-4, idiopathic-4, hypothyroidism-1, Byler’s disease-1) out of a total of 36 had normal liver function, a minimum of 1 year after cholestasis. They are being followed up for assessment of any chronic liver disease result-ing from the scars that may have occured during the cholestasis in early infancy. Discussion Cholestasis in young infants has a varied etiology including congenital infections, malformations, metabolic disorders as well as inherited functional disorders of unknown etiology. Early diagnosis is important as the effects of cholestasis are profound and wide-spread. Secondary cirrhotic changes occur early and correctable conditions can also become untreatable if the diagnosis is delayed. Delayed presentation was the single most important factor, which resulted in the poor prognosis of cholestatic syndrome of infancy in developing countries(4). In our study, the mean age of presentation to hospital was 87 days, though jaundice was noticed on an average at 40 days of age. Yachha et al. have reported earlier onset of jaundice in their patients (5). This is in marked contrast to that of developed countries(6). Lack of awareness about the condition among the general practitioners and the public is the probable reason for the late referral. Unlike observations in developed coun-tries, congenital infection was the commonest cause of cholestatic jaundice in our study. This is similar to earlier observations(7). Cytomegalovirus was the commonest etio-logical agent in our study, as has also been reported by Chang et al.(8). Most of these infants had no other stigmata of a congenital infection, other than low birth weight. Differentiation of biliary atresia from other causes of cholestatic jaundice is of crucial importance(9). With liver histo-pathology as the gold standard, we evaluated the usefulness of the various parameters for this purpose. In the differentiation of biliary atresia from the other causes of cholestatic jaundice, only birth weight, serum cholesterol and serum GGT were found to be useful, using the independent samples test. In differentiating biliary atresia specifically from neonatal hepatitis, only serum cholesterol and serum GGT were found to be of discrimina-ting value. None of the other parameters were of practical value. Using the receiver operating characteristic curve, it was found that cholesterol above 275 mg/dl and serum GGT above 175 U/L favored a diagnosis of biliary atresia. The usefulness of GGT in cholestatic jaundice has already been reported(10). In our study, the alkaline phosphatase levels were relatively high even in patients with neonatal hepatitis. Whitington et al. have suggested that the cholestatic enzymes (ALP and GGT) in combination may be useful in diagnosis. Patients with biliary atresia have high levels of both ALP and GGT; those with Byler’s disease have high ALP and low GGT and those with neonatal hepatitis have values in between(11). As an extension of that principle, we calculated the usefulness of the ratio of ALP to GGT for differentiating biliary atresia from neonatal hepatitis. In our study, patients with biliary atresta had an ALP: GGT ratio of 3:1, those with neonatal hepatits had a ratio of 12:1, while infants with Byler’s disease had a ratio of 36:1. Using the receiver-operating characteristic curve it was calculated that a ratio of less than 7 favored a diagnosis of biliary atresia rather than neonatal hepatitis. The usefulness of ultrasound in cholestatic syndromes was also evaluated(12). Two U/S features - gall bladder size less than 2.5 cm and absence of contractility after feeding have been reported useful in the dignosis of biliary atresia. In our study, it was found that when only one criteria (gall bladder size <2.5 cm) was used, then the sensitivity of ultra-sound in diagnosis of biliary atresia was 86% while the specificity was only 79%. When both criteria were applied, the sensitivity remained the same, while the specificity went up to 86%. This is similar to observations published earlier(7,12). Since the number of patients in this study are small, its statistical significance cannot be assessed. Elevated serum cholesterol, serum GGT and ALP:GGT ratio, as well as the ultrasound findings, can be considered suggestive of biliary atresia. A cholestatic infant with these features should have a diagnostic laparotomy even if the percutaneous liver biopsy histo-pathology does not show the classical changes of biliary atresia. The poor outcome after Kasai operation is very striking. All the 5 patients had irreversible liver cell failure by 12 months of age. This may be due to the late stage at which surgery was performed. The histo-pathology of the liver was also striking - in addition to bile duct proliferation and bile plugs, which are characteristic of biliary atresia, there was marked lymphocytic infiltration and portal fibrosis in all the patients. Reo virus-3 has been implicated in the etiology of biliary atresia. It is possible that biliary atresia in our set up is an active infectious cholangiopathy unlike that reported in western literature. Bhave et al. have also reported very poor results after hepato-portoenterostomy(13). The outcome of patients treated by Kasai operation before 60 days of age is dramatically better than those undergoing surgery subsequently(14). Infants operated at later ages invariably require orthotopic liver transplantation. Surgery probably just delays the age of liver trans-plantation in such patients(15). Questions have therefore been raised on the necessity of a Kasai operation in those infants who present late, if a liver transplantation cannot be done later on. Considering the reliability of ultra-sound, GGT, cholesterol and ALP:GGT ratio in the diagnosis of biliary atresia, and the safety and specificity of needle biopsy, laparotomy and per-operative cholangiogram should be undertaken in selected cases and not as a routine in all cholestatic infants. Since the number of infants in this study is small, multicentric trials may be needed to confirm some of the results of this study. Acknowledgement We wish to acknowledge Dr. Leslie Smiles, Associate Professor of pathology, Dr. Anil Mathew, Assistant Professor of Bio-statistics. Dr. Priya and Dr. Jothy, Assistant Professors of Pediatrics and Mrs. S. Umayal, Secretary for their help. Contributors: JM designed and co-ordinated the study. He drafted the manuscript and will act as the guarantor. SP participated in data collection and also helped in drafting the paper. Funding: None. Competing interests: None stated.
|
