STUDY OF BLOOD GLUCOSE LEVELSIN CHILDREN WITH FULMINANTVIRAL HEPATITIS
Sunil Karande*, Saroj Parekh*
*Senior Lecturer in Paediatrics, Seth GS Medical College and KEM Hospital, Parel, Mumbai 400 012. **(Retd.) Honorary Professor of Pediatrics, Seth GS Medical College and Bai Jerbai Wadia Hospital for Children, Parel, Mumbai 400 012.
Aim : To assess the degree of disturbance in blood glucose (BG) homeostasis in children with fulminant viral hepatitis (FVH).
Methods : BG levels on admission and serially at periodic intervals on intravenous glucose infusions were recorded in 18 FVH cases and in 18 disease controls. Hypoglycaemia (BG < 45 mg/dl) was an urgent indication to change to a higher glucose concentration infusion.
Results : At the time of admission hypoglycaemia was recorded in 2 (11.1%) FVH cases and both died. Serial BG estimations on intravenous glucose infusions revealed 3 distinct patterns : (i) steady normoglycaemia (BG 61-180 mg/dl) was recorded in 7 (38.9%) FVH cases, despite which 3 (16.7%) died. Steady normoglycaemia was recorded in all 18 (100%) disease controls including 3 who died. Mean BG levels± SD in FVH cases Vs disease controls who maintained steady normoglycaemia were 102.7 ± 24.2 mg/dl Vs 100.4 ± 16.2 mg/dl (NS). (ii) normoglycaemia with intermittent low normal BG levels (BG 45-60 mg/dl) was recorded in 8 (44.4%) FVH cases and all 8 died. Mean BG level ± SD was 60.8 ± 13.1 mg/dl. In 3 of these 8 FVH cases intractable hypoglycaemia was recorded terminally. (iii) normoglycaemia with intermittent hyperglycaemia (BG > 180 mg/dl) was recorded in 3 (16.7%) FVH cases and all 3 died. Mean BG level ± SD was 266.2 ± 85.4 mg/dl. In 1 of these 3 FVH cases hyperglycaemia was recorded even terminally.
Conclusions : In children with FVH an inability to maintain normoglycaemia inspite of intravenous glucose infusions reflects a significant decrease in functioning hepatocellular mass. Intractable hypoglycaemia and intermittent hyperglycaemia are biochemical which presage a fatal outcome.
Fulminant hepatic failure is a clinical syndrome which results from massive necrosis of hepatocytes or sudden and severe impairment of liver function, and is characterised by rapid development of hepatic encephalopathy.[1,2] Fulminant viral hepatitis (FVH) is the commonest cause of fulminant hepatic failure. FVH occurs in 1-2% of patients with acute viral hepatitis and usually overwhelms the patient within 10 days. FVH may develop so rapidly that jaundice is inconspicuous and the disease may be confused with an acute meningoencephalitis.
Glucose is the major metabolic fuel in the human body and is the only nutrient that can be utilised by the brain to maintain its normal functions. It is well known that the liver plays a key role in maintaining a normal blood glucose level.[4,5] In patients with severe liver disease it may become difficult to maintain a narrow range of blood glucose concentration. Though the pathogenesis of FVH is not fully understood, the management is basically supportive i.e. to maintain homeostasis as satisfactory as possible and to support life until spontaneous regeneration of the liver can occur.[1,2] Along with other supportive measures, glucose solutions are administered intravenously to prevent or correct hypoglycaemia.[1,2] The aim of our study was to assess the degree of disturbance in blood glucose homeostasis in children with FVH.
Thirty-six acutely ill children were included in our study. In the study group were 18 children (age range 4 months to 11.5 years) with FVH (Table 1). Acute viral hepatitis was diagnosed on the basis of clinical prodromata viz. anorexia, nausea, vomiting, fever, tender hepatomegaly with or without icterus, and with raised serum transaminases. These 18 children were diagnosed as having FVH on the basis of rapidly developing hepatic encephalopathy, shrinking liver size and an abnormally prolonged prothrombin time which did not improve after parenteral vitamin K administration.[1,2] Serum HBsAg was negative in all the 18 FVH cases. FVH cases received recommended available treatment viz. intravenous glucose and electrolyte infusions, mannitol, cimetidine, antibiotics, oral neomycin, Lactisyn (Franco-Indian) and high bowel washes. Fresh frozen plasma and/or fresh blood transfusions were given whenever indicated. Postmortem liver biopsy examination was done in 11 FVH cases.
In the disease control group there were 18 acutely ill children (age range 3.5 months to 11 years) with various non-hepatic illnesses (Table 1). The minimum criterion for designating a disease control case as being acutely ill was a child with a poor general condition at the time of admission and requiring intravenous fluid therapy. Every disease control had normal liver function tests viz. normal serum transaminases, bilirubin, proteins and prothrombin time. Seven children had acute gastroenteritis with dehydration. Eight were acute respiratory cases (4 had pneumonia, 2 bronchial asthma, 1 bronchiolitis and 1 pulmonary miliary kochs). Three were central nervous system cases (2 had tuberculous meningoencephalitis and 1 bulbospinal poliomyelitis). Children in whom liver dysfunction could develop secondarily e.g. children with kwashiorkar, septicaemia, congestive heart failure were not included in the disease control group.
TABLE 1 Comparative data in fulminant viral hepatitis (FVH) cases and disease controls Characteristics FVH cases
(n = 18)
(n = 18)
< 2 years
Hypoglycaemia on admission
Duration of intravenous glucose infusions
Number of deaths
4.5 hrs - 14 days
7 hrs - 12 days
Figures in parenthesis are percentages.
*Malnourished as judged by Indian Academy of Pediatrics classification; difference not significant (Chi-Square test with Yates correction). +1 case each of pulmonary miliary kochs, tuberculous meningoencephalitis, bulbospinal poliomyelitis.
Blood glucose level on admission and serially at periodic 2 to 4 hourly intervals on intravenous glucose saline were recorded in all 36 cases. Venous blood (1 ml) was collected in a flouride bulb and blood glucose level was estimated by the ‘glucose oxidase’ method. Whenever hypoglycaemia was suspected viz. if the child had convulsions, worsening of the sensorium or gasping respiration, blood glucose range was urgently estimated by a Dextrostix (Bayer Diagnostics), using the bottle label colour chart for interpretation. If the Dextrostix suggested hypoglycaemia, a venous blood sample was collected for accurately documenting the blood glucose level.
Children with FVH were started on either 5% or 10% glucose saline infusions at 1.5 L/m/day.[7,8] Hypoglycaemia was an urgent indication to give the child a 10% glucose bolus infusion (2-4 ml/kg/dose) and to change to a higher glucose concentration infusion viz. from 5% to 7.5%, or from 10% to 12.5% and if required even higher upto 15 to 20%, with an aim to correct hypoglycaemia and maintain normoglycaemia. The children in the disease control group received recommended treatment and 5% glucose saline infusions in the recommended doses, as required by the individual child’s clinical condition.
Blood glucose levels recorded on admission and serially on intravenous glucose infusions were analysed. The normal fasting blood glucose level in children varies between 50 and 90 mg/dl. A blood glucose level below 45 mg/dl was termed as hypoglycaemia. Standard paediatric textbooks do not clearly state the normal range for blood glucose levels recorded while a child is receiving intravenous glucose infusions. While analysing blood glucose levels recorded on glucose infusions we have termed a blood glucose level below 45 mg/dl as hypoglycaemia, 45 to 60 normoglycaemia. Corresponding to the renal threshold for glucose, a blood glucose level above 180 mg/dl has been termed as hyperglycaemia.,
At the time of admission, hypoglycaemia was recorded in 2 (11.1%) FVH cases who eventually died within 12 hours of admission. In the remaining 16 (88.9%) FVH cases and in all 18 (100%) disease controls, normoglycaemia was recorded on admission.
Serial blood glucose estimations on intravenous glucose infusions revealed 3 distinct patterns (Tables 2,3).
(i) Seven (38.9%) FVH cases maintained steady normoglycaemia. Five FVH cases were on 5% and 2 were on 10% glucose saline drips. Inspite of maintaining steady normoglycaemia 3 FVH cases (who were on 5% glucose saline) died. All 18 (100%) disease controls maintained steady normoglycaemia on 5% glucose saline. Average blood glucose levels were likely to be the same in both FVH cases and disease controls who maintained steady normoglycaemia (Table 3).
(ii) In 8 (44.4%) FVH cases (2 were on 5% and 6 were on 10% glucose saline) normoglycaemia with intermittent low normal blood glucose levels were recorded. All the 8 FVH cases died. In 3 FVH cases (who were on 10% glucose saline) intractable hypoglycaemia was recorded terminally. Hypoglycaemia did not get corrected for long inspite of changing to a higher glucose concentration infusion, even upto 15 to 20%. These 3 FVH cases died within 0.5 to 4 hours of the hypoglycaemia occurring.
(iii) In 3 (16.7%) FVH cases (who were on 10% glucose saline) normoglycaemia with intermittent hyperglycaemia was recorded. All the 3 FVH cases died. In 1 child hyperglycaemia was recorded even terminally.
Table 2 Mortality rate(%) correlated with blood glucose(BG) pattern recorded on intravenous glucose infusions in fulminant viral hepatitis(FVH) cases and disease controls. BG pattern recorded FVH cases
No of cases No of deaths No of cases No of deaths Steady normoglycaemia 7(38.9) 3(16.7) 18(100) 3(16.7) Normoglycaemia with intermittent low normal BG* 8(44.4) 8(44.4) 0(0) - Normoglycaemia with intermittent hyperglycaemia 3(16.7) 3(16.7) 0(0) - Total 18(100) 14(77.8) 18(100) 3(16.7) Figures in parenthesis are percentages * 3 FVH cases also had intractable hypoglycaemia terminally. BG 45-60 mg/dl = low normal; 61-180 mg/dl = normalglycaemia; >180 mg/dl = hyperglycaemia.
Table 3 BG pattern recorded FVH cases
Significance Steady normoglycaemia 102.7±24.2
NS+ Normoglycaemia with intermittent low normal BG* 60.8± 13.1
- Normoglycaemia with intermittent hyperglycaemia 266.2± 85.4
- * 3 FVH cases also had intractable hypoglycaemia terminally. +NS = not significant; by unpaired 't' test. BG 45-60 mg/dl = low normal; 61-80 mg/dl = normoglycaemia; > 180 mg/dl = hyperglycaemia.
Our study highlights the difficulty in maintaining a narrow range of blood glucose concentration in children with FVH. Standard paediatric textbooks do not clearly state the optimum range of blood glucose levels in a child with FVH being treated with intravenous glucose infusions, as it is probably easier said than done.,,
The liver plays a key role in maintaining blood glucose homeostasis. When the blood glucose level rises after a meal or starts to fall during a period of fasting, it is the liver which functions as a buffer system to normalise the blood glucose level.[4,5] Glucose absorbed from the gut after a meal or glucose directly entering the blood stream as an intravenous infusion reaches the liver and enters the hepatocytes by an active carrier-mediated process. Immediately it is converted to glucose-6-phosphate (‘phosphorylation of glucose’) which is utilised by the liver for its various metabolic functions and for lipid synthesis (‘glycolysis’). Also, glucose-6-phosphate is converted to glycogen and stored in the liver (‘glycogenesis’).[4,5]
During a period of fasting as the blood glucose level starts to fall the glycogen in the liver is converted back to glucose-6-phosphate (‘glycogenolysis’), which is hydrolysed to form free glucose. The liver releases this free glucose back into the blood stream. Glycogen stores in the liver are sufficient to maintain normal blood glucose levels for upto 24 hours. Over prolonged periods of fasting, glucose-6-phosphate is also produced in the liver from lactate, pyruvate, alanine, and glycerol by reversing glycolysis (‘gluconeogenesis’).[4,5]
Thus a normally functioning liver can release large amounts of glucose back into the blood stream to prevent hypoglycaemia from occurring, and is also capable of handling a blood glucose load to bring a rising blood glucose level back within the normal range. In our study, all disease controls (who had a normally functioning liver) maintained steady normoglycaemia on intravenous glucose infusions (Tables 2,3). But in the FVH group only 7 (38.9%) maintained steady normoglycaemia, despite which 3 died. FVH is an illness in which mortality exceeds 60 to 70%.,, In our study mortality in the FVH group was 77.8% (Table 2). The only survivors in the FVH group were 4 (22.2%) children in whom steady normoglycaemia was maintained. It is believed that blood glucose homeostasis can be maintained by as few as 15 to 20% of normally functioning hepatocytes. Probably these 4 survivors had an adequate residual functioning hepatocellular mass and with supportive treatment could tide over the crisis till spontaneous regeneration of the liver could occur.
In 8 (44.4%) FVH cases normoglycaemia with intermittent low normal blood glucose levels was recorded on intravenous glucose infusions and all 8 died (Tables 2,3). Blood glucose level after falling to a low normal level (45 to 60 mg/dl) would subsequently come back within the normoglycaemic range (61 to 180 mg/dl) on the same glucose infusion being given at a uniform rate. Recording intermittent low normal blood glucose levels on intravenous glucose infusions indicates the failing ability of the damaged liver to release adequate glucose back into the blood stream and should serve as an indication to change to a higher glucose concentration infusion. Such a measure may help maintain steady normoglycaemia and probably improve chances for survival. Three of these 8 FVH cases eventually developed intractable hypoglycaemia and died within the next 0.5 to 4 hours.
Hypoglycaemia recorded in a child with FVH on intravenous glucose infusion is an urgent indication to give 10% glucose bolus infusion (2-4 ml/kg/dose) and to change to a higher glucose concentration infusion. The aim is to achieve and maintain normoglycaemia till spontaneous regeneration of the liver can occur. Hypoglycaemia in FVH indicates failure of the damaged liver to release adequate free glucose into the blood stream and is a poor prognostic indicator.,,, Though in our study all children with FVH in whom hypoglycaemia was recorded died (including the 2 in whom hypoglycaemia was recorded on admission), hypoglycaemia need not necessarily presage a fatal outcome. Psacharopoulos et al have reported 2 survivors out of 9 children with FVH in whom hypoglycaemia was recorded at some stage during the course of their illness. The exact mechanism which can explain intractable hypoglycaemia in FVH inspite of glucose infusions (even upto 15 to 20%) is not well understood.,
In 3 (16.7%) FVH cases, normoglycaemia with intermittent hyperglycaemia was recorded on intravenous glucose infusions and all 3 died (Tables 2,3). Blood glucose level after rising above 180 mg/dl (even upto 443 mg/dl in one instance) would subsequently come back within the normoglycaemic range (61 to 180 mg/dl) on the same 10% glucose infusion being given at a uniform rate. Recording intermittent hyperglycaemia on intravenous glucose infusions indicates the failing ability of the damaged liver to metabolise the glucose load reaching it. It is known that cirrhotic patients can become hyperglycaemic following an oral glucose load and in FVH metabolic derangements can be similar to those seen in chronic liver disease. When cirrhotic patients develop encephalopathy high carbohydrate feeding takes precedence over any impairment of glucose tolerance secondary to the liver disease. In our study, since hyperglycaemia was an intermittent phenomenon, changing to a lower glucose concentration infusion was not considered since such a measure could have possibly lead to hypoglycaemia.
In conclusion, our results suggest that in children with FVH the chances for survival lessen as the glucoregulatory role of the hepatocytes gets compromised. An inability to maintain blood glucose homeostasis inspite of adequate intravenous glucose infusions reflects a significant decrease in functioning hepatocellular mass. Intractable hypoglycaemia and intermittent hyperglycaemia are biochemical findings which presage a fatal outcome.
- Suchy FJ. Fulminant hepatic failure. In : Behrman RE, Vaughan VC III, Nelson WE, Eds. Nelson Textbook of Pediatrics. Philadelphia : WB Saunders Co. 1987; 844-5.
- Parekh S, Rajadhyaksha S, Agarwal B. Fulminant hepatic failure in childhood. Indian J Pediatr 1985; 52 : 587-92.
- Sherlock S, Dooley J. Diseases of the Liver and Biliary System. London : Blackwell Scientific publications. 1993; 260-92.
- Guyton AC. Textbook of Medical Physiology. Philadelphia : WB Saunders Co. 1991; 744-53.
- Leonard JV. Disorders of carbohydrate metabolism. In : McLaren DS, Burman D, Belton NR, Williams AF, Eds. Textbook of Paediatric Nutrition. Edinburgh : Churchill Livingstone. 1991; 127-34.
- Dietzler DN, Smith CH. Carbohydrates. In : Sonnenwirth AC, Jarett L, Eds. Gradwohl’s Clinical Laboratory Methods and Diagnosis. St Louis : CV Mosby Co. 1980; 210-49.
- Mowat AP. Liver Disorders in Childhood. London : Butterworths and Co. 1987; 126-35.
- Silverman A, Roy CC. Pediatric Clinical Gastroenterology. St Louis : CV Mosby Co. 1983; 655-74.
- Robson AM. Parenteral fluid therapy. In : Behrman RE, Vaughan VC III, Nelson WE, Eds. Nelson Textbook of Pediatrics. Philadelphia : WB Saunders Co. 1987; 191-207.
- Mabry CC. Reference ranges for laboratory tests (Table 29-2). In : Behrman RE, Vaughan VC III, Nelson WE, Eds. Nelson Textbook of Pediatrics. Philadelphia : WB Saunders Co. 1987; 1535-8.
- Schwartz R. Hypoglycemia. In : Behrman RE, Vaughan VC III, Nelson WE, Eds. Nelson Textbook of Paediatrics. Philadelphia : WB Saunders Co. 1987; 1264-73.
- Ogata ES. Carbohydrate metabolism in the fetus and neonate and altered neonatal glucoregulation. Pediatr Clin North Am 1986; 33 : 25-45.
- Poley JR. The liver. In : Anderson CM, Burke V, Gracey M, Eds. Paediatric Gastroenterology. Melbourne : Blackwell Scientific Publications. 1987; 529-658.
- Psacharopoulos HT, Mowat AP, Davies M, Portmann B, Silk DBA, Williams R. Fulminant hepatic failure in childhood : an analysis of 31 cases. Arch Dis Child 1980; 55 : 252-8.
- Aynsley-Green A, Soltesz G. Hypoglycemia in Infancy and Childhood. Current Reviews in Paediatrics 1. Edinburgh : Churchill Livingstone. 1985; 103-50.
- Samson RI, Trey C, Timme AH, Saunders SJ. Fulminating hepatitis with recurrent hypoglycemia and hemorrhage (case report). Gastroenterology 1967; 53 : 291-300.
- Podolsky DK, Isselbacher KJ. Derangements of hepatic metabolism. In : Isselbacher KJ, Braunwald E, Wilson JD, Martin JB, Fauci AS, Kasper DL, Eds. Harrison’s Principles of Internal Medicine. New York : McGraw - Hill Inc. 1994; 1448-53.