Ascites in childhood liver disease
http://www.100md.com
《美国医学杂志》
Department of Gastroenterology (Pediatric Gastroenterology), Sanjay Gandhi Postgraduate Institute of Medical Sciences, Lucknow 226014, India
Abstract
Ascites is a common clinical problem in children with liver disease. The peripheral arterial vasodilation hypothesis is mostly accepted as the pathophysiological basis of ascites. The most important complication is spontaneous ascitic fluid infection in the form of spontaneous bacterial peritonitis (SBP) and its variants. Aerobic gram-negative bacteria, primarily Escherichia coli, are the most common isolates. Diagnostic paracentesis is done in patients with ascites when diagnosed first time and at the beginning of each admission to hospital. Ascitic fluid is evaluated for cell count with differential, albumin level, total protein, and culture. Serum-ascites albumin gradient (SAAG) is the best single test for classifying ascites into portal hypertensive (SAAG >1.1 g/dL) and non-portal hypertensive (SAAG <1.1 g/dL) causes. In patients with tense ascites LVP should be performed. A neutrophil count of > 250 cells /mm[3] is highly suggestive of bacterial peritonitis. Intravenous cefotaxime is the empiric antibiotic of choice. Long-term administration of oral norfloxacin 5-7.5 mg/Kg once a day in cirrhotic patients with ascitic fluid protein content of <1g/dL or prior episode of SBP is recommended for prevention of SBP. Oral dual diuretic therapy of single morning dose of spironolactone along with furosemide in the ratio of 5:2 is recommended. While obtaining satisfactory diuretic response dual diuretic therapy can be changed over to monotherapy with spironolactone. Patients should be on sodium restricted diet. Management of ascites might ultimately require liver transplantation.
Keywords: Ascites; Child; Peritonitis; Cirrhosis; Portal; Hypertension; Spontaneous; Bacterial
Ascites is pathologic fluid accumulation within the abdominal cavity. In liver disease ascites represents a state of excess total-body sodium and water. Three theories of ascites formation have been proposed.[1] (A) The underfilling theory : primarily there is inappropriate sequestration of fluid within the splanchnic vascular bed as a consequence of portal hypertension (PHT) that produces decrease in effective circulating blood volume. This activates the plasma rennin, aldosterone, and sympathetic nervous system, resulting in renal sodium and water retention. (B) The overflow theory: primary abnormality is inappropriate renal retention of sodium and water in the absence of volume depletion. Basis of this theory is that patients with cirrhosis have intravascular hypervolemia rather than hypovolemia. (C) The peripheral arterial vasodilation hypothesis : The major factor of ascites formation is splanchnic vasodilation.[2] Cirrhosis causes increased hepatic resistance to portal flow that results in PHT and shunting of blood to the systemic circulation. Local production of vasodilators, mainly nitric oxide due to PHT results in splanchnic and peripheral arterial vasodilatation. This leads to decrease in effective arterial blood volume (EABV). Progressive deterioration of liver functions, PHT, splanchnic arterial vasodilatation and reduced plasma oncotic pressure due to low serum albumin all contribute to development of ascites. Changes in renal hemodynamics and tubular function favor the development of refractory ascites and hepatorenal syndrome.
Clinical features
In ascites there is history of increasing abdominal girth or inappropriate weight gain. Infants may have history of periorbital edema. Patients with ascites should be asked about risk factors that contribute to liver diseases. These are neonatal or infantile cholestatic symptoms, chronic viral hepatitis or jaundice, injections, transfusion of blood or blood products and family history of liver diseases. In adolescents history of intravenous drug use, sexual promiscuity and tattoos may give clue to liver disease. On physical examination presence of ascites is suggested by abdominal distension with or without everted umbilicus, bulging flanks, shifting dullness, fluid wave and Puddle sign. The physical examination should focus on the signs of PHT and liver disease that include jaundice, palmar erythema, spider angioma, and often enlarged liver in infants and children. The liver may be difficult to palpate if a large amount of ascites is present. Dilated abdominal collaterals and caput medusa may be seen in ascites due to liver disease while collaterals in flanks and at back are suggestive of inferior vena cava block. Elevated jugular venous pressure may suggest a cardiac origin of ascites and should be always looked for especially in unexplained ascites, where cause may be constrictive pericarditis. Patients with cardiac disease or nephrotic syndrome may have anasarca. Ascites occurs in 44.1% of children with cirrhosis.[3] It is also known to occur in subacute liver failure, acute viral hepatitis, fulminant hepatic failure (16.5-55%) [4] and non-cirrhotic portal hypertension due to extrahepatic portal venous obstruction (12.8%).[3] Imaging detects ascites at a subclinical level.
Complications
Mechanical, bacterial and metabolic complications occur in ascites.[5] Mechanical complications result in respiratory impairment, physical immobility, and compression of great vessels, hernias, promotion of gastroesophageal reflux, impaired gastric motility and obstructive sleep apnea syndrome. Disturbances of electrolyte, altered drug pharmacokinetics, hepatic encephalopathy and hepato-renal syndrome are known metabolic complications. The most important complication of ascites is spontaneous bacterial peritonitis (SBP).
Spontaneous ascitic fluid infection in liver disease
This is characterized by the spontaneous infection of ascitic fluid in the absence of any intra-abdominal source of infection. Various types of SBP are shown in table1. SBP has a prevalence of 10 to 30% among patients with ascites.[6],[7] Bacteremia is thought to precede the development of SBP, mainly as a result of bacterial translocation from the intestinal lumen and the reduction in the phagocytic activity of the macrophages in the liver. Gut motility alterations, along with bacterial overgrowth and changes in intestinal permeability, probably play a role in bacterial translocation. Aerobic gram-negative bacteria, primarily Escherichia More Details coli, are the most common isolates, but the frequency of episodes caused by gram-positive bacteria including methicillin-resistant Staphylococcus aureus has recently increased.[8],[9] These changes in bacteriological spectrum are due to increasing number of invasive procedures and hospitalization in intensive care units. The presence of at least 250 polymorphonuclear (PMN) cells/mm 3 of ascitic fluid is diagnostic of this condition. Patients with culture-negative neutrocytic ascites (CNNA) have similar signs, symptoms, and the same prognosis as SBP and should therefore be treated similarly with empiric antibiotics. If antibiotic therapy is delayed then 34.5 % of these patients become culture positive. Asymptomatic monomicrobial non-neutrocytic bacterascites (MNB) usually signifies colonization and does not require antibiotic therapy as 62% of these patients resolve the colonization without antibiotic therapy. However, symptomatic patients with clinical features suggestive of infection even with PMN count < 250 cells/mm 3 progress to SBP and therefore warrant empiric antibiotic treatment.[10] Polymicrobial bacterascites occurs when ascitic fluid contains multiple organisms and the PMN cell count is <250/ mm 3. The latter usually results from inadvertent puncture of the intestines during attempted paracentesis and occurs in about 1/1000 paracentesis. Risk factors contributing to intestinal puncture are those of paralytic ileus, presence of multiple surgical scars, and operator inexperience.
Ascitic fluid analysis
Diagnostic paracentesis should be done in patients with ascites when diagnosed first time and at the beginning of each admission to hospital. A repeat paracentesis should be done if symptoms and signs or laboratory abnormalities suggest peritoneal or systemic infection. Abdominal pain, tenderness, vomiting, diarrhea, ileus, fever, leucocytosis, acidosis, septic shock, hepatic encephalopathy or renal failure should direct the physician for diagnostic paracentesis. Since bleeding following diagnostic paracentesis is sufficiently uncommon the prophylactic use of FFP or platelets transfusions before the procedure is not recommended.[10]
Procedure of paracentesis : Caution guidelines followed are: (a) avoid scars from previous surgery; chances of entering a viscus are high due to localized bowel adhesions, (b) empty urinary bladder before the procedure to avoid perforation, (c) no paracentesis through cellulites affected area. Major steps of the procedure are: (1) anesthetize the puncture site under aspectic conditions, (2) patient position in semisupine, lateral decubitus or sitting, 3) puncture site midline 2 cm below the umbilicus; in neonates just lateral to rectus muscle in the right or left lower quadrants, a few centimeters above the inguinal ligament, (4) intravenous catheter 16-22 gauge attached to a syringe is inserted, (5) needle is inserted at a 45 degrees aiming cephalad while one hand pulls the skin caudally until entering peritoneal cavity; a "Z" tract is created when the skin is released and needle is removed, (6) continuous negative pressure is applied and once fluid appears in the syringe introducer is removed and the catheter is left in place, stopcock is attached and desired amount of fluid is obtained.[11]
Laboratory studies : Routinely ascitic fluid should be evaluated for cell count with differential, albumin level, total protein, and culture.[12] Serum albumin is simultaneously done to calculate serum-ascites albumin gradient (SAAG).[13] In case of new-onset ascites cytology is also done. Inspection of freshly tapped ascitic fluid is usually helpful. Mostly ascitic fluid is transparent and tinged yellow, however, in deeply jaundiced patients, the fluid can have deep yellow color and a detergent like appearance due to presence of bilirubin. Patients with cirrhosis can have slightly opalescent to mildly cloudy fluid because of peritoneal lymphatic tear or rupture (triglyceride concentration of 100-400 mg/dL).
I. Cell count: Ascitic fluid cell count is the single most important laboratory parameter in patients of cirrhosis. Normal ascitic fluid contains <500 leukocytes/mm 3sub and <250 polymorphonuclear leukocytes/mm 3sub . Inflammatory condition of any etiology results in elevated white blood cell count. A neutrophil count of > 250 cells/mm 3sub is highly suggestive of bacterial peritonitis. A small amount (0.05-0.1 ml) of ascitic fluid is required for cell count and the fluid should be collected in an EDTA tube to prevent cell aggregation and consequent spurious counts. The mean total cell count of ascitic fluid can triple to values of 1,000 cells/mm 3sub with rapid diuresis and contraction of the ascitic pool. However, polymorphonuclear leukocytes count remains stable during diuresis, probably as a result of the shorter half life of these cells.
II. Culture: Overall culture positivity is 40-60% by routine methods, however, bedside inoculation of 10 mL of ascitic fluid directly in blood culture bottles increases positivity by 90%.
III. Gram stain: In SBP the median concentration of bacteria is one organism/mL. Approximately 10,000 bacteria/ mL are required for detection by Gram's stain. Thus in SBP Gram's stain is only 7% and 10% positive in un-centrifuged and centrifuged samples respectively. Therefore, routine Gram's stain and centrifugation of ascitic fluid is not recommended. Gram stain is of immense value when gut perforation is suspected irrespective of cirrhotic state. In the setting of secondary bacterial peritonitis the characteristic laboratory analysis of the ascitic fluid will show PMN cells to be greater than 250 cells/mm 3sub (usually in thousands) and multiple organisms on Gram's stain and culture and at least 2 of the following three criteria: lactate dehydrogenase more than the upper limit of normal for serum, glucose <50mg/dL and total protein >1g/dL. The above criterion is 100% sensitive but has a low specificity of 45%.
IV. Serum-ascites albumin gradient : In the past, ascitic fluid has been classified on the basis of protein content either as exudative 3 2.5 g/dL or transudative <2.5 g/dL with an accuracy of approximately 56% for detecting exudative causes. Drawbacks of this classification are (a) about 20% of cirrhotic patients are incorrectly labeled exudative type (b) two thirds of cirrhosis patients while on diuretics develop exudative ascites, (c) other disorders like malignant ascites and tuberculous peritonitis may be labeled having transudative ascites while being actually exudative and (d) cardiac ascites as exudative when actually transudative. This classification has now been replaced by serum-ascites albumin gradient (SAAG) that is calculated by subtracting the ascitic fluid albumin value from the serum albumin value, and it correlates directly with portal pressure. The SAAG is the best single test for classifying ascites into portal hypertensive (SAAG >1.1 g/dL) and non-portal hypertensive (SAAG <1.1 g/dL) causes. Preferably the serum and ascitic fluid specimens should be obtained simultaneously. The accuracy of the SAAG results is approximately 97% in classifying portal hypertensive and non-portal hypertensive ascites. The terms high-albumin gradient (SAAG >1.1 g/dL) and low-albumin gradient (SAAG <1.1 g/dL) have replaced the terms transudative and exudative in the description of ascites.[13] A high gradient is associated with diffuse parenchymal liver disease and occlusive portal and hepatic venous disease, liver metastasis and hypothyroidism. Certain limitations of using SAAG are: (a) the gradient may be falsely low if the patient with cirrhosis has a serum albumin level <1.1 g/dL, and also in disease state of hypergammaglobulinemia (>5 g/dL); b) errors may occur if the albumin assay is inaccurate in the low range, the samples are not withdrawn at relatively same time, and if the patient is in shock; (c) a falsely high value of SAAG may occur in chylous ascites as lipid fractions tend to interfere with laboratory determination of albumin.[12]
Management
Algorithm of treating children having ascites due to liver disease is given in the figure.
Large volume paracentesis (LVP) is defined as removal of 50 mL or more of ascitic fluid /kg of dry body weight.[14], [15] In patients with tense ascites LVP should be performed. Studies have shown that a mean volume of 118 ± 56 mL/kg can be safely removed at one time.[14] LVP should be done under cover of 0.5-1g of albumin/Kg of dry weight either as 5% or 20% infusion slowly over a period of 2 hours beginning at the time of catheter insertion.[14] Alternatively albumin infusion required is calculated as 8 g/L of ascitic fluid removed.[10]
Treatment of SBP
Treatment should be started empirically if SBP is suspected clinically, regardless of the availability of laboratory results. In community-acquired SBP and in patients not on SBP prophylaxis, Escherichia coli siella pneumoniae are seen in up to 60% of isolates.[6],[16] About 25% are Gram-positive cocci, mostly streptococcal species, usually pneumococci. Anerobes are rarely seen. Intravenous cefotaxime (or a similar third generation cephalosporin) is the empiric antibiotic of choice as it covers 95% of the flora and has been shown to cure SBP episodes in 85% of patients.[6],[10] Patients who develop SBP while on norfloxacin prophylaxis are more likely to have infections caused by Gram-positive cocci or quinolone-resistant Gram-negative bacilli.[8],[9] Cefotaxime is effective even in these latter cases. The weight adjusted pediatric dosage would be 50 mg/kg/dose to a maximum of 2 g/dose every 8 hrly. Optimal duration of treatment is 5 days. Once culture results are available, antibiotic modifications may be necessary, but aminoglycosides should still be avoided because of the risk of renal failure. Alternatively oral ofloxacin in hospitalized patients without septic shock, encephalopathy (grade II or higher), azotemia (serum creatinine greater than 3 mg/dL), gastrointestinal bleed, or ileus has been shown to be effective in SBP treatment.[17] Monomicrobial non-neutrocytic bacterascites (MNB) a variant of SBP table1 is managed differently. At admission if patients have symptoms of SBP despite PMN count of < 250 cells /mm 3sub should be treated with antibiotics. In case of no symptoms and upon availability of first positive culture a second paracentesis should be done. Antibiotics should then be administered if second paracentesis shows PMN count of > 250 cells /mm 3sub or if the repeat culture is still positive. [6],[10]
Prevention of SBP : Risk factors for development of SBP are ascitic fluid protein concentration <1 g/dL, variceal hemorrhage and prior episode of SBP. Long-term administration of oral norfloxacin 5-7.5 mg/Kg once a day in cirrhotic patients with ascitic fluid protein content of <1g/dL or prior episode of SBP is recommended for prevention of SBP. Patients having cirrhosis and upper gastrointestinal hemorrhage should be given short-term (7 days) norfloxacin 5-7.5 mg/Kg/dose orally twice a day, however, during active bleeding intravenous ofloxacin should be prescribed.[10]
Diuretic therapy
Oral diuretic therapy consists of single morning dose of spironolactone (0.3-3 mg/kg) along with furosemide (0.5-2 mg/kg) in the ratio of 5:2 that facilitates maintenance of normokalemia.[10],[14],[15] If weight loss and natriuresis are inadequate the doses of both spironolactone and furosemide should be increased simultaneously, maintaining the above ratio.[10] Reasons for starting patient on dual therapy is to have early mobilization of fluid with furosemide as spironolactone takes several days for a therapeutic response. Dual diuretic therapy can be changed over to monotherapy with spironolactone alone while obtaining satisfactory diuretic response. Spironolactone is best absorbed if administered with food, the diuretic effect is observed in 48 hours and has a half-life of up to 5 days. Therefore, the dose of spironolactone should be adjusted at intervals of 3-5 days. Spironolactone therapy may be complicated by hyperkalemia and tender gynecomastia. Furosemide may be temporarily withheld if hypokalemia occurs. When edema foot is present high dose of diuretics is safe as peripheral edema buffers any intravascular volume reduction. Diuretic-induced complications are those of hepatic encephalopathy, increase in serum creatinine, hyponatremia, hypokalemia or hyperkalemia. These complications should be monitored and corrected along with stoppage of diuretic therapy.
Along with diuretic therapy patients should be on sodium restricted diet. One gram of table salt contains 17 mEq of sodium. One gram of sodium approximates to 44 mEq of sodium. Restriction of sodium in diet is limited to 1 to 2mEq/kg/day for infants and children, and 1 to 2 g/day (44 to 88 mEq of sodium/day) in adolescents. Sodium content of commonly used dietary items as guide to therapy in Indian children is shown in table2. Patients with refractory ascites Figure1 not responding to diuretics may need alternative therapy in the form of LVP or liver transplantation.
Acknowledgement
We greatly appreciate help of Ms. Deepa Negi Assistant Dietician Department of Pediatric Gastroenterology for her inputs in manuscript preparation.
References
1. Dudley FJ. Pathophysiology of ascites formation. Gastroenterol. Clin North Am 1992; 21: 215-235.
2. Gines P, Cardenas A, Arroyo V, Rodes J. Management of cirrhosis and ascites. N Engl J Med 2004; 350 : 1646-1654.
3. Peter L, Dadhich SK, Yachha SK. Clinical and laboratory differentiation of cirrhosis and extrahepatic portal venous obstruction in children. J Gastroenterol Hepatol 2003; 18 : 185-189.
4. Poddar U, Thapa B R, Prasad A, Sharma AK, Singh K. Natural history and risk factors in fulminant hepatic failure. Archives of Disease in Childhood 2002; 87 : 54-56.
5. Kuntz E, Kuntz HT. Oedoma and ascites Hepatology, Principles and Practice . Kuntz E and Kuntz HT (Eds) Springer-Verlag, Berlin, Heidelberg 2002; 256-290.
6. Rimola A, Garcia-Tsao G, Navasa M et al. Diagnosis, treatment and prophylaxis of spontaneous bacterial peritonitis: a consensus document. J Hepatol 2000; 32 : 142-153.
7. Larcher VF, Manolaki N, Vegnent A et al. Spontaneous bacterial peritonitis with chronic liver disease: clinical features and etiologic factors. J Pediatr 1985; 106 : 907-912.
8. Fernαndez J, Navasa M, Gómez J, et al. Bacterial infections in cirrhosis: epidemiological changes with invasive procedures and norfloxacin prophylaxis. Hepatology 2002; 35 : 140-148.
9. Cholongitas E, Papatheodoridis GV, Lahanas A et al. Increasing frequency of Gram-Positive bacteria in spontaneous bacterial peritonitis. Liver Int 2005; 25 (1): 57-61.
10. Runyon BA. Management of adult patients with ascites due to cirrhosis. Hepatology 2004; 39 : 3 : 841-856.
11. Grady M. Procedures The Harriet Lane Handbook Gunn VL and Nechyba C (Eds) Mosby, Elsevier Science, New-Delhi 2002; 69.
12. McHutchison JG. Differential diagnosis of ascites. Semin Liver Dis 1997;17(3) : 191-202.
13. Runyon BA, Montano AA, Akriviadis EA et al. The serum-ascites albumin gradient in the differential diagnosis of ascites is superior to the exudate/ transudate concept. Ann Intern Med 1992 ; 117 : 215-220.
14. Kramer RE, Sokol RJ, Yerushalmi B et al. Large-volume paracentesis in the management of ascites in children. J Pediatr Gastroenterol Nutr 2001; 33 : 245-249.
15. Arikan C, Ozgenc F, Akman SA, Yagci RV, Tokat Y, Aydogdu S. Large-volume paracentesis and liver transplantation. J Pediatr Gastroenterol Nutr 2001; 33 : 245-924.
16. Vieira SMG, Matte U, Kieling CO et al. Infected and non-infected ascites in pediatric patients. J Pediatr Gastroenterol Nutr 2005; 40: 289-294.
17. Navasa M, Follo A, Llovet JM et al. Randomized, comparative study of oral ofloxacin versus intravenous cefotaxime in spontaneous bacterial peritonitis. Gastroenterology 1996; 111 : 1011-1017.(Yachha Surender Kumar, Kh)
Abstract
Ascites is a common clinical problem in children with liver disease. The peripheral arterial vasodilation hypothesis is mostly accepted as the pathophysiological basis of ascites. The most important complication is spontaneous ascitic fluid infection in the form of spontaneous bacterial peritonitis (SBP) and its variants. Aerobic gram-negative bacteria, primarily Escherichia coli, are the most common isolates. Diagnostic paracentesis is done in patients with ascites when diagnosed first time and at the beginning of each admission to hospital. Ascitic fluid is evaluated for cell count with differential, albumin level, total protein, and culture. Serum-ascites albumin gradient (SAAG) is the best single test for classifying ascites into portal hypertensive (SAAG >1.1 g/dL) and non-portal hypertensive (SAAG <1.1 g/dL) causes. In patients with tense ascites LVP should be performed. A neutrophil count of > 250 cells /mm[3] is highly suggestive of bacterial peritonitis. Intravenous cefotaxime is the empiric antibiotic of choice. Long-term administration of oral norfloxacin 5-7.5 mg/Kg once a day in cirrhotic patients with ascitic fluid protein content of <1g/dL or prior episode of SBP is recommended for prevention of SBP. Oral dual diuretic therapy of single morning dose of spironolactone along with furosemide in the ratio of 5:2 is recommended. While obtaining satisfactory diuretic response dual diuretic therapy can be changed over to monotherapy with spironolactone. Patients should be on sodium restricted diet. Management of ascites might ultimately require liver transplantation.
Keywords: Ascites; Child; Peritonitis; Cirrhosis; Portal; Hypertension; Spontaneous; Bacterial
Ascites is pathologic fluid accumulation within the abdominal cavity. In liver disease ascites represents a state of excess total-body sodium and water. Three theories of ascites formation have been proposed.[1] (A) The underfilling theory : primarily there is inappropriate sequestration of fluid within the splanchnic vascular bed as a consequence of portal hypertension (PHT) that produces decrease in effective circulating blood volume. This activates the plasma rennin, aldosterone, and sympathetic nervous system, resulting in renal sodium and water retention. (B) The overflow theory: primary abnormality is inappropriate renal retention of sodium and water in the absence of volume depletion. Basis of this theory is that patients with cirrhosis have intravascular hypervolemia rather than hypovolemia. (C) The peripheral arterial vasodilation hypothesis : The major factor of ascites formation is splanchnic vasodilation.[2] Cirrhosis causes increased hepatic resistance to portal flow that results in PHT and shunting of blood to the systemic circulation. Local production of vasodilators, mainly nitric oxide due to PHT results in splanchnic and peripheral arterial vasodilatation. This leads to decrease in effective arterial blood volume (EABV). Progressive deterioration of liver functions, PHT, splanchnic arterial vasodilatation and reduced plasma oncotic pressure due to low serum albumin all contribute to development of ascites. Changes in renal hemodynamics and tubular function favor the development of refractory ascites and hepatorenal syndrome.
Clinical features
In ascites there is history of increasing abdominal girth or inappropriate weight gain. Infants may have history of periorbital edema. Patients with ascites should be asked about risk factors that contribute to liver diseases. These are neonatal or infantile cholestatic symptoms, chronic viral hepatitis or jaundice, injections, transfusion of blood or blood products and family history of liver diseases. In adolescents history of intravenous drug use, sexual promiscuity and tattoos may give clue to liver disease. On physical examination presence of ascites is suggested by abdominal distension with or without everted umbilicus, bulging flanks, shifting dullness, fluid wave and Puddle sign. The physical examination should focus on the signs of PHT and liver disease that include jaundice, palmar erythema, spider angioma, and often enlarged liver in infants and children. The liver may be difficult to palpate if a large amount of ascites is present. Dilated abdominal collaterals and caput medusa may be seen in ascites due to liver disease while collaterals in flanks and at back are suggestive of inferior vena cava block. Elevated jugular venous pressure may suggest a cardiac origin of ascites and should be always looked for especially in unexplained ascites, where cause may be constrictive pericarditis. Patients with cardiac disease or nephrotic syndrome may have anasarca. Ascites occurs in 44.1% of children with cirrhosis.[3] It is also known to occur in subacute liver failure, acute viral hepatitis, fulminant hepatic failure (16.5-55%) [4] and non-cirrhotic portal hypertension due to extrahepatic portal venous obstruction (12.8%).[3] Imaging detects ascites at a subclinical level.
Complications
Mechanical, bacterial and metabolic complications occur in ascites.[5] Mechanical complications result in respiratory impairment, physical immobility, and compression of great vessels, hernias, promotion of gastroesophageal reflux, impaired gastric motility and obstructive sleep apnea syndrome. Disturbances of electrolyte, altered drug pharmacokinetics, hepatic encephalopathy and hepato-renal syndrome are known metabolic complications. The most important complication of ascites is spontaneous bacterial peritonitis (SBP).
Spontaneous ascitic fluid infection in liver disease
This is characterized by the spontaneous infection of ascitic fluid in the absence of any intra-abdominal source of infection. Various types of SBP are shown in table1. SBP has a prevalence of 10 to 30% among patients with ascites.[6],[7] Bacteremia is thought to precede the development of SBP, mainly as a result of bacterial translocation from the intestinal lumen and the reduction in the phagocytic activity of the macrophages in the liver. Gut motility alterations, along with bacterial overgrowth and changes in intestinal permeability, probably play a role in bacterial translocation. Aerobic gram-negative bacteria, primarily Escherichia More Details coli, are the most common isolates, but the frequency of episodes caused by gram-positive bacteria including methicillin-resistant Staphylococcus aureus has recently increased.[8],[9] These changes in bacteriological spectrum are due to increasing number of invasive procedures and hospitalization in intensive care units. The presence of at least 250 polymorphonuclear (PMN) cells/mm 3 of ascitic fluid is diagnostic of this condition. Patients with culture-negative neutrocytic ascites (CNNA) have similar signs, symptoms, and the same prognosis as SBP and should therefore be treated similarly with empiric antibiotics. If antibiotic therapy is delayed then 34.5 % of these patients become culture positive. Asymptomatic monomicrobial non-neutrocytic bacterascites (MNB) usually signifies colonization and does not require antibiotic therapy as 62% of these patients resolve the colonization without antibiotic therapy. However, symptomatic patients with clinical features suggestive of infection even with PMN count < 250 cells/mm 3 progress to SBP and therefore warrant empiric antibiotic treatment.[10] Polymicrobial bacterascites occurs when ascitic fluid contains multiple organisms and the PMN cell count is <250/ mm 3. The latter usually results from inadvertent puncture of the intestines during attempted paracentesis and occurs in about 1/1000 paracentesis. Risk factors contributing to intestinal puncture are those of paralytic ileus, presence of multiple surgical scars, and operator inexperience.
Ascitic fluid analysis
Diagnostic paracentesis should be done in patients with ascites when diagnosed first time and at the beginning of each admission to hospital. A repeat paracentesis should be done if symptoms and signs or laboratory abnormalities suggest peritoneal or systemic infection. Abdominal pain, tenderness, vomiting, diarrhea, ileus, fever, leucocytosis, acidosis, septic shock, hepatic encephalopathy or renal failure should direct the physician for diagnostic paracentesis. Since bleeding following diagnostic paracentesis is sufficiently uncommon the prophylactic use of FFP or platelets transfusions before the procedure is not recommended.[10]
Procedure of paracentesis : Caution guidelines followed are: (a) avoid scars from previous surgery; chances of entering a viscus are high due to localized bowel adhesions, (b) empty urinary bladder before the procedure to avoid perforation, (c) no paracentesis through cellulites affected area. Major steps of the procedure are: (1) anesthetize the puncture site under aspectic conditions, (2) patient position in semisupine, lateral decubitus or sitting, 3) puncture site midline 2 cm below the umbilicus; in neonates just lateral to rectus muscle in the right or left lower quadrants, a few centimeters above the inguinal ligament, (4) intravenous catheter 16-22 gauge attached to a syringe is inserted, (5) needle is inserted at a 45 degrees aiming cephalad while one hand pulls the skin caudally until entering peritoneal cavity; a "Z" tract is created when the skin is released and needle is removed, (6) continuous negative pressure is applied and once fluid appears in the syringe introducer is removed and the catheter is left in place, stopcock is attached and desired amount of fluid is obtained.[11]
Laboratory studies : Routinely ascitic fluid should be evaluated for cell count with differential, albumin level, total protein, and culture.[12] Serum albumin is simultaneously done to calculate serum-ascites albumin gradient (SAAG).[13] In case of new-onset ascites cytology is also done. Inspection of freshly tapped ascitic fluid is usually helpful. Mostly ascitic fluid is transparent and tinged yellow, however, in deeply jaundiced patients, the fluid can have deep yellow color and a detergent like appearance due to presence of bilirubin. Patients with cirrhosis can have slightly opalescent to mildly cloudy fluid because of peritoneal lymphatic tear or rupture (triglyceride concentration of 100-400 mg/dL).
I. Cell count: Ascitic fluid cell count is the single most important laboratory parameter in patients of cirrhosis. Normal ascitic fluid contains <500 leukocytes/mm 3sub and <250 polymorphonuclear leukocytes/mm 3sub . Inflammatory condition of any etiology results in elevated white blood cell count. A neutrophil count of > 250 cells/mm 3sub is highly suggestive of bacterial peritonitis. A small amount (0.05-0.1 ml) of ascitic fluid is required for cell count and the fluid should be collected in an EDTA tube to prevent cell aggregation and consequent spurious counts. The mean total cell count of ascitic fluid can triple to values of 1,000 cells/mm 3sub with rapid diuresis and contraction of the ascitic pool. However, polymorphonuclear leukocytes count remains stable during diuresis, probably as a result of the shorter half life of these cells.
II. Culture: Overall culture positivity is 40-60% by routine methods, however, bedside inoculation of 10 mL of ascitic fluid directly in blood culture bottles increases positivity by 90%.
III. Gram stain: In SBP the median concentration of bacteria is one organism/mL. Approximately 10,000 bacteria/ mL are required for detection by Gram's stain. Thus in SBP Gram's stain is only 7% and 10% positive in un-centrifuged and centrifuged samples respectively. Therefore, routine Gram's stain and centrifugation of ascitic fluid is not recommended. Gram stain is of immense value when gut perforation is suspected irrespective of cirrhotic state. In the setting of secondary bacterial peritonitis the characteristic laboratory analysis of the ascitic fluid will show PMN cells to be greater than 250 cells/mm 3sub (usually in thousands) and multiple organisms on Gram's stain and culture and at least 2 of the following three criteria: lactate dehydrogenase more than the upper limit of normal for serum, glucose <50mg/dL and total protein >1g/dL. The above criterion is 100% sensitive but has a low specificity of 45%.
IV. Serum-ascites albumin gradient : In the past, ascitic fluid has been classified on the basis of protein content either as exudative 3 2.5 g/dL or transudative <2.5 g/dL with an accuracy of approximately 56% for detecting exudative causes. Drawbacks of this classification are (a) about 20% of cirrhotic patients are incorrectly labeled exudative type (b) two thirds of cirrhosis patients while on diuretics develop exudative ascites, (c) other disorders like malignant ascites and tuberculous peritonitis may be labeled having transudative ascites while being actually exudative and (d) cardiac ascites as exudative when actually transudative. This classification has now been replaced by serum-ascites albumin gradient (SAAG) that is calculated by subtracting the ascitic fluid albumin value from the serum albumin value, and it correlates directly with portal pressure. The SAAG is the best single test for classifying ascites into portal hypertensive (SAAG >1.1 g/dL) and non-portal hypertensive (SAAG <1.1 g/dL) causes. Preferably the serum and ascitic fluid specimens should be obtained simultaneously. The accuracy of the SAAG results is approximately 97% in classifying portal hypertensive and non-portal hypertensive ascites. The terms high-albumin gradient (SAAG >1.1 g/dL) and low-albumin gradient (SAAG <1.1 g/dL) have replaced the terms transudative and exudative in the description of ascites.[13] A high gradient is associated with diffuse parenchymal liver disease and occlusive portal and hepatic venous disease, liver metastasis and hypothyroidism. Certain limitations of using SAAG are: (a) the gradient may be falsely low if the patient with cirrhosis has a serum albumin level <1.1 g/dL, and also in disease state of hypergammaglobulinemia (>5 g/dL); b) errors may occur if the albumin assay is inaccurate in the low range, the samples are not withdrawn at relatively same time, and if the patient is in shock; (c) a falsely high value of SAAG may occur in chylous ascites as lipid fractions tend to interfere with laboratory determination of albumin.[12]
Management
Algorithm of treating children having ascites due to liver disease is given in the figure.
Large volume paracentesis (LVP) is defined as removal of 50 mL or more of ascitic fluid /kg of dry body weight.[14], [15] In patients with tense ascites LVP should be performed. Studies have shown that a mean volume of 118 ± 56 mL/kg can be safely removed at one time.[14] LVP should be done under cover of 0.5-1g of albumin/Kg of dry weight either as 5% or 20% infusion slowly over a period of 2 hours beginning at the time of catheter insertion.[14] Alternatively albumin infusion required is calculated as 8 g/L of ascitic fluid removed.[10]
Treatment of SBP
Treatment should be started empirically if SBP is suspected clinically, regardless of the availability of laboratory results. In community-acquired SBP and in patients not on SBP prophylaxis, Escherichia coli siella pneumoniae are seen in up to 60% of isolates.[6],[16] About 25% are Gram-positive cocci, mostly streptococcal species, usually pneumococci. Anerobes are rarely seen. Intravenous cefotaxime (or a similar third generation cephalosporin) is the empiric antibiotic of choice as it covers 95% of the flora and has been shown to cure SBP episodes in 85% of patients.[6],[10] Patients who develop SBP while on norfloxacin prophylaxis are more likely to have infections caused by Gram-positive cocci or quinolone-resistant Gram-negative bacilli.[8],[9] Cefotaxime is effective even in these latter cases. The weight adjusted pediatric dosage would be 50 mg/kg/dose to a maximum of 2 g/dose every 8 hrly. Optimal duration of treatment is 5 days. Once culture results are available, antibiotic modifications may be necessary, but aminoglycosides should still be avoided because of the risk of renal failure. Alternatively oral ofloxacin in hospitalized patients without septic shock, encephalopathy (grade II or higher), azotemia (serum creatinine greater than 3 mg/dL), gastrointestinal bleed, or ileus has been shown to be effective in SBP treatment.[17] Monomicrobial non-neutrocytic bacterascites (MNB) a variant of SBP table1 is managed differently. At admission if patients have symptoms of SBP despite PMN count of < 250 cells /mm 3sub should be treated with antibiotics. In case of no symptoms and upon availability of first positive culture a second paracentesis should be done. Antibiotics should then be administered if second paracentesis shows PMN count of > 250 cells /mm 3sub or if the repeat culture is still positive. [6],[10]
Prevention of SBP : Risk factors for development of SBP are ascitic fluid protein concentration <1 g/dL, variceal hemorrhage and prior episode of SBP. Long-term administration of oral norfloxacin 5-7.5 mg/Kg once a day in cirrhotic patients with ascitic fluid protein content of <1g/dL or prior episode of SBP is recommended for prevention of SBP. Patients having cirrhosis and upper gastrointestinal hemorrhage should be given short-term (7 days) norfloxacin 5-7.5 mg/Kg/dose orally twice a day, however, during active bleeding intravenous ofloxacin should be prescribed.[10]
Diuretic therapy
Oral diuretic therapy consists of single morning dose of spironolactone (0.3-3 mg/kg) along with furosemide (0.5-2 mg/kg) in the ratio of 5:2 that facilitates maintenance of normokalemia.[10],[14],[15] If weight loss and natriuresis are inadequate the doses of both spironolactone and furosemide should be increased simultaneously, maintaining the above ratio.[10] Reasons for starting patient on dual therapy is to have early mobilization of fluid with furosemide as spironolactone takes several days for a therapeutic response. Dual diuretic therapy can be changed over to monotherapy with spironolactone alone while obtaining satisfactory diuretic response. Spironolactone is best absorbed if administered with food, the diuretic effect is observed in 48 hours and has a half-life of up to 5 days. Therefore, the dose of spironolactone should be adjusted at intervals of 3-5 days. Spironolactone therapy may be complicated by hyperkalemia and tender gynecomastia. Furosemide may be temporarily withheld if hypokalemia occurs. When edema foot is present high dose of diuretics is safe as peripheral edema buffers any intravascular volume reduction. Diuretic-induced complications are those of hepatic encephalopathy, increase in serum creatinine, hyponatremia, hypokalemia or hyperkalemia. These complications should be monitored and corrected along with stoppage of diuretic therapy.
Along with diuretic therapy patients should be on sodium restricted diet. One gram of table salt contains 17 mEq of sodium. One gram of sodium approximates to 44 mEq of sodium. Restriction of sodium in diet is limited to 1 to 2mEq/kg/day for infants and children, and 1 to 2 g/day (44 to 88 mEq of sodium/day) in adolescents. Sodium content of commonly used dietary items as guide to therapy in Indian children is shown in table2. Patients with refractory ascites Figure1 not responding to diuretics may need alternative therapy in the form of LVP or liver transplantation.
Acknowledgement
We greatly appreciate help of Ms. Deepa Negi Assistant Dietician Department of Pediatric Gastroenterology for her inputs in manuscript preparation.
References
1. Dudley FJ. Pathophysiology of ascites formation. Gastroenterol. Clin North Am 1992; 21: 215-235.
2. Gines P, Cardenas A, Arroyo V, Rodes J. Management of cirrhosis and ascites. N Engl J Med 2004; 350 : 1646-1654.
3. Peter L, Dadhich SK, Yachha SK. Clinical and laboratory differentiation of cirrhosis and extrahepatic portal venous obstruction in children. J Gastroenterol Hepatol 2003; 18 : 185-189.
4. Poddar U, Thapa B R, Prasad A, Sharma AK, Singh K. Natural history and risk factors in fulminant hepatic failure. Archives of Disease in Childhood 2002; 87 : 54-56.
5. Kuntz E, Kuntz HT. Oedoma and ascites Hepatology, Principles and Practice . Kuntz E and Kuntz HT (Eds) Springer-Verlag, Berlin, Heidelberg 2002; 256-290.
6. Rimola A, Garcia-Tsao G, Navasa M et al. Diagnosis, treatment and prophylaxis of spontaneous bacterial peritonitis: a consensus document. J Hepatol 2000; 32 : 142-153.
7. Larcher VF, Manolaki N, Vegnent A et al. Spontaneous bacterial peritonitis with chronic liver disease: clinical features and etiologic factors. J Pediatr 1985; 106 : 907-912.
8. Fernαndez J, Navasa M, Gómez J, et al. Bacterial infections in cirrhosis: epidemiological changes with invasive procedures and norfloxacin prophylaxis. Hepatology 2002; 35 : 140-148.
9. Cholongitas E, Papatheodoridis GV, Lahanas A et al. Increasing frequency of Gram-Positive bacteria in spontaneous bacterial peritonitis. Liver Int 2005; 25 (1): 57-61.
10. Runyon BA. Management of adult patients with ascites due to cirrhosis. Hepatology 2004; 39 : 3 : 841-856.
11. Grady M. Procedures The Harriet Lane Handbook Gunn VL and Nechyba C (Eds) Mosby, Elsevier Science, New-Delhi 2002; 69.
12. McHutchison JG. Differential diagnosis of ascites. Semin Liver Dis 1997;17(3) : 191-202.
13. Runyon BA, Montano AA, Akriviadis EA et al. The serum-ascites albumin gradient in the differential diagnosis of ascites is superior to the exudate/ transudate concept. Ann Intern Med 1992 ; 117 : 215-220.
14. Kramer RE, Sokol RJ, Yerushalmi B et al. Large-volume paracentesis in the management of ascites in children. J Pediatr Gastroenterol Nutr 2001; 33 : 245-249.
15. Arikan C, Ozgenc F, Akman SA, Yagci RV, Tokat Y, Aydogdu S. Large-volume paracentesis and liver transplantation. J Pediatr Gastroenterol Nutr 2001; 33 : 245-924.
16. Vieira SMG, Matte U, Kieling CO et al. Infected and non-infected ascites in pediatric patients. J Pediatr Gastroenterol Nutr 2005; 40: 289-294.
17. Navasa M, Follo A, Llovet JM et al. Randomized, comparative study of oral ofloxacin versus intravenous cefotaxime in spontaneous bacterial peritonitis. Gastroenterology 1996; 111 : 1011-1017.(Yachha Surender Kumar, Kh)