- Email: [email protected]
Hemodialysis-related portal-systemic encephalopathy
CASE REPORT
Hemodialysis-Related Portal-Systemic Encephalopathy Yoshifumi Ubara, MD, Jyunichi Hoshino, MD, Tetsuo Tagami, Naoki Sawa, MD, Hideyuki Katori, MD, Fumi Takemoto, MD, Masamichi Matsuda, MD, Shigeko Hara, MD, and Kenmei Takaichi, MD ● The authors encountered a case of portal-systemic venous shunt newly diagnosed after initiation of hemodialysis. A 68-year-old Japanese woman began hemodialysis because of symptoms of uremia including loss of appetite and pulmonary edema. Loss of consciousness occurred suddenly after her ninth session of hemodialysis. No hepatic functional abnormality was found other than hyperammonemia (314 g/dL [184 mol/L]). Loss of consciousness subsequently occurred often after hemodialysis. Color Doppler ultrasonography and magnetic resonance angiography depicted a large shunt between the left gastric vein and left renal vein resulting in portal flow entering the systemic circulation via the renal vein. Because the shunt was large, ligation of it was performed surgically. Results of histologic examination of a liver biopsy specimen obtained intraoperatively were normal. The patient became well postoperatively. This patient’s encephalopathy appeared to be caused by the flow of ammonia-rich portal venous blood into the systemic circulation via the large shunt owing to a decrease in intravenous pressure after rapid hemodialysis. Portal-systemic shunt encephalopathy should be recognized as a “new” neuropsychiatric disorder characteristic of patients undergoing hemodialysis. Am J Kidney Dis 44:E38-E42. © 2004 by the National Kidney Foundation, Inc. INDEX WORDS: Portal-systemic encephalopathy; hemodialysis; neuropsychiatric disorder; color Doppler ultrasonography; magnetic resonance angiography; hyperammonemia.
H
EPATIC ENCEPHALOPATHY usually is observed in patients with severe hepatocellular dysfunction. However, another type of hepatic encephalopathy may occur occasionally in patients without disorder of the hepatic parenchyma, ie, when portal venous blood flows into the inferior vena cava via a large shunt; this etiology has been called portal-systemic encephalopathy. Such encephalopathy occurring in the absence of intrinsic liver disease can be an important cause of morbidity.1-4 Recently in Japan, it has come to be recognized as one of the causes of disturbance of consciousness in patients on dialysis.5 We report a case of portal-systemic shunt encephalopathy that became apparent after initiation of dialysis and note that dynamic changes of blood flow during hemodialysis therapy may be closely related to this condition.
From the Nephrology Center and Department of Surgery, Toranomon Hospital, Tokyo, Japan. Received March 25, 2004; accepted in revised form May 14, 2004. Address reprint requests to Yoshifumi Ubara, Nephrology Center, Toranomon Hospital Kajigaya, 1-3-1, Kajigaya, Takatsu-kum Kawasaki-shi, Kanagawa-ken, 213-0015, Japan.E-mail: [email protected] © 2004 by the National Kidney Foundation, Inc. 0272-6386/04/4403-0025$30.00/0 doi:10.1053/j.ajkd.2004.05.035 e38
CASE REPORT A 68-year-old Japanese woman was admitted to our institution on October 16, 2000, because of severe edema in the lower extremities and loss of appetite. She exhibited proteinuria during her third pregnancy, at 30 years of age, in 1963, and proteinuria persisted. Although she was suspected to have chronic glomerulonephritis, renal biopsy was not performed. Thereafter, renal function gradually declined. The patient was admitted for further evaluation of her renal failure. On admission, she was 146 cm tall and weighed 44 kg. Her blood pressure was 154/100 mm Hg. Her abdomen was distended, and there was severe edema in her lower extremities. No abnormal neuropsychiatric findings were noted. Laboratory findings were as follows: erythrocyte count was 1.92 ⫻ 106/L, hemoglobin concentration was 6.2 g/dL (62 g/L), hematocrit level was 19.1%, leukocyte count was 7,100/ L, and thrombocyte count was 24.4 ⫻ 104/L. Total protein concentration was 6.6 g/dL (66 g/L), albumin level was 3.1 g/dL (31 g/L), urea nitrogen value was 50 mg/dL (17.9 mmol/ L), creatinine level was 5.1 mg/dL (451 mol/L), and uric acid was 4.9 mg/dL (291 mol/L). Sodium level was 146 mEq/L (mmol/L), potassium was 6.2 mEq/L (mmol/L), chloride was 121 mEq/L (mmol/L), calcium level was 6.0 mg/dL (1.50 mmol/L), phosphate was 4.3 mg/dL (1.39 mmol/L), and total cholesterol was 186 mg/dL (4.81 mmol/L). Total bilirubin was 0.3 mg/dL (5 mol/L), aspartate aminotransferase was 19 IU/L, alanine aminotransferase was level was 9 IU/L, lactate dehydrogenase level was 219 IU/L, and an indocyanine green test (R15) was 8%. C-reactive protein was 0.3 mg/dL. Serum immunoglobulin G concentration was 1,168 mg/dL (11.68 g/L), immunoglobulin A was 358 mg/dL (3.58 g/L), and immunoglobulin M was 27 mg/dL (0.27 g/L). The CH50 was 41 U. On viral examination, serum was negative for hepatitis C virus antibody (enzyme immunoassay) and for hepatitis B surface antigen.
American Journal of Kidney Diseases, Vol 44, No 3 (September), 2004: E38-E42
HEMODIALYSIS-RELATED PORTAL-SYSTEMIC ENCEPHALOPATHY
e39
Fig 1. Magnetic resonance angiography shows a large shunt (arrow) between the left gastric vein and left renal vein.
The urine pH was 5.5, negative for glucose, 1⫹ for protein, and 2⫹ for occult blood. The urine sediment contained 6 to 10 cells per high-power field. The urine was negative for BenceJones protein. Twenty-four–hour urine contained 1.25 g of protein and 4,740 g of beta 2 microglobulin and creatinine clearance was 5.8 mL/min (0.10 mL/s). Computed tomography and ultrasonography studies showed contracted kidneys bilaterally and normal liver architecture. A chest radiograph showed bilateral pleural effusion. Her cardiothoracic ratio was 71%. Dietary therapy including restriction of salt, protein, and water was performed immediately after admission, and diuretics and antihypertensive agents were administered. Forty-three days after her initial evaluation, hemodialysis was started 3 times weekly, because renal function had deteriorated, with a urea nitrogen of 128 mg/dL (45.7 mmol/L) and creatinine of 14.1 mg/dL (1,246 mol/L), and gastrointestinal symptoms such as loss of appetite, nausea, and vomiting began to occur. After initiation of hemodialysis, her gastrointestinal symptoms disappeared, and her appetite improved. Her dry weight began to increase owing to hypotension in the latter half of hemodialysis. Suddenly, after ninth treatment, she lost consciousness, although blood chemistry examination found no abnormality, with sodium of 142 mEq/L (mmol/L), potassium of 3.5 mEq/L (mmol/L), chloride of 104 mEq/L (mmol/L), calcium of 8.0 mg/dL
(2.00 mmol/L), phosphate of 2.4 mg/dL (0.77 mmol/L), total bilirubin of 0.3 mg/dL (5 mol/L), aspartate aminotransferase of 18 IU/L, alanine aminotransferase of 10 IU/L, lactate dehydrogenase of 214 IU/L, and glucose of 84 mg/dL (4.7 mmol/L). Because computed tomography of the brain was normal, her ammonia concentration was measured, and hyperammonemia was found, at 314 g/dL (184 mol/L; normal range, ⬍50 g/dL [⬍29 mol/L]). Twelve hours after initiation of fluid replacement at 40 mL/h, her consciousness became clear with decrease to 72 g/dL (42 mol/L) in her ammonia level. Loss of consciousness subsequently recurred often after hemodialysis. When the amount of fluid removed during a period of hemodialysis procedure became large, and her dry weight became set strictly so that blood pressure fell down in the latter half of hemodialysis, hyperammonemia-associated encephalopathy was apparent. Because ultrasonographic restudy found that hepatic parenchymal architecture was normal, blood analysis of portal vein was performed by using color Doppler. Color Doppler ultrasonography depicted a large shunt between the left gastric vein and left renal vein, with blood flow in the portal vein pouring into the inferior vena cava through left renal vein via a shunt from the left gastric vein. This portalsystemic shunt was confirmed by magnetic resonance angiography (Fig 1). Because embolization therapy using microcoils could not be performed because the shunt was large,
e40
UBARA ET AL
Fig 2. A schema of the findings in Fig 1. Red arrows show the flow of portal vein via the shunt. (A) left gastric vein, (B) splenic vein, (C) shunt, (D) left renal vein, (E) superior mesenteric vein, (F) inferior vena cava, (G) portal vein.
Fig 3. Ligation of the left gastric vein (black arrow) was performed surgically. Red thick arrows show the flow of blood in the portal vein. Red thin arrows show the flow of blood in the renal vein.
HEMODIALYSIS-RELATED PORTAL-SYSTEMIC ENCEPHALOPATHY
e41
Fig 4. Liver biopsy at the time of surgery shows normal histology. (Hematoxylin and eosin, original magnification ⴛ20.)
the ligation of the shunt was performed surgically (Figs 2 and 3). Histologic examination of a liver biopsy specimen obtained intraoperatively showed normal findings (Fig 4). The patient became well postoperatively and has experienced no episode of hepatic encephalopathy, with ammonia levels of 30 to 50 g/dL (18 to 29 mol/L), over the 3-year period since surgery.
DISCUSSION
Neuropsychiatric disorders in patients with renal failure have been called uremic encephalopathy. It usually improves immediately after dialysis has been initiated and has rarely been observed recently because dialysis therapy has been initiated earlier than in the past. Uremic toxins, including nitrogenous waste and parathyroid hormone as well as acidosis, have been suspected to be causative factors.6,7 Encephalopathy owing to aluminum intoxication had been found to occur in patients on long-term dialysis therapy. After replacement of aluminum phosphate–binding preparations by calcium derivatives and use of a reverse-osmotic apparatus for
processing water for dialysate, exposure to aluminum has been minimized.8 This encephalopathy has been observed rarely. Drugs administered to uremic patents may also cause encephalopathy if prescribed excessively. In Japan, attention recently has been paid to encephalopathy owing to portal systemic shunt formation, which became manifest after use of hemodialysis was started. This shunt may exist congenitally, may be formed spontaneously, or may develop after external abdominal injury or surgery.5,9 It usually remains nonsymptomatic until the end of life. However, with the addition of factors such as hemodynamic treatment of hemodialysis, shunt formation may become symptomatic. In anuric patients undergoing hemodialysis, a large fluid volume may be removed rapidly in 3 to 4 hours. Total body volume, including venous blood volume, will transiently decrease. When intravenous pressure decreases after hemodialysis, portal venous blood will flow readily into the inferior
e42
UBARA ET AL
vena cava via a large extrahepatic shunt. Once ammonia-rich portal venous blood flows into the systemic circulation, substances such as ammonia may cause central nervous system toxicity, resulting in hepatic encephalopathy. When the amount of fluid removed during a period of hemodialysis procedure becomes large, hyperammonemia tends to occur. Ligation of the shunt will improve encephalopathy.10,11 If histologic examination does not show abnormalities such as cirrhosis or fibrosis of the liver, the etiology of this encephalopathy may be hemodynamically mediated and may thus be termed portal-systemic encephalopathy. This does not usually occur in patients with peritoneal dialysis, because their body fluid can be removed very gradually over time. As the number of dialysis patients increase, the condition needs to be recognized as a potential cause of neuropsychiatric problems.12-14 We postulate the following speculation concerning the mechanism of portal-systemic shunt encephalopathy, based on the analysis of this case. Perhaps this encephalopathy may be caused by the flow of ammonia-rich portal venous blood into the systemic circulation via a large shunt due to a decrease in intravenous pressure after rapid hemodialysis. This encephalopathy should be recognized as a “new” neuropsychiatric disorder in patients undergoing hemodialysis (and unlikely to occur in those undergoing peritoneal dialysis). REFERENCES 1. Riordan SM, Williams R: Treatment of hepatic encephalopathy. N Engl J Med 337:473-479, 1997 2. Krieger D, Krieger S, Gass P, Theilmann L, Lichtnecker H: Manganese and chronic hepatic encephalopathy. Lancet 346:270-274, 1995
3. Jalan R, Seery JP, Taylor-Robinson SD: Review article: Pathogenesis and treatment of chronic hepatic encephalopathy. Ali Pharmaco Ther 10:681-697, 1996 4. Basile AS, Jones EA: Ammonia and GABA-ergic neurotransmission: Interrelated factors in the pathogenesis of hepatic encephalopathy. Hepatology 25:1303-1305, 1997 5. Shimono J, Azuma K, Hashiguchi M, Fujishima M: Recurring encephalopathy abolished by gastrorenal shunt ligation in a diabetic hemodialysis patient. Am J Gastroenterol 93:270-272, 1998 6. Fraser CL, Arieff AL: Brain abnormalities and peripheral neuropathy, in Massry SG, Glassock RJ (eds): Textbook of Nephrology (ed 4). Philadelphia, PA, Lippincott Williams & Wilkins, 2001, pp 1282-1287 7. Kovalik EC: Neurological manifestations, in Greenberg A (ed): Primer on Kidney Diseases (ed 3). San Diego, CA, National Kidney Foundation, 2001, pp 451-454 8. Ubara Y, Fushimi T, Tagami T, et al: Histomorphometric features of bone in patients with primary and secondary hypoparathyroidism. Kidney Int 63:1809-1816, 2003 9. Kerlan RK, Sollenberger RD, Palubinskas AJ, Raskin NH, Callen PW, Ehrenfeld WK: Portal-systemic encephalopathy due to a congenital portocaval shunt. Am J Radiol 139:1013-1015, 1982 10. Hanna SH, Smith RS, Henderson LM, Millikan WJ, Warren WD: Reversal of hepatic encephalopathy after occlusion of total portasystemic shunts. Am J Surg 142:285-289, 1981 11. Potts JR, Henderson M, Millikan WJ, Sones P, Warren WD: Restoration of portal venous perfusion and reversal of encephalopathy by ballon occlusion of portal systemic shunt. Gastroenterology 87:208-121, 1984 12. Onishi T, Nishitani M, Kanda M, Tsuji M, Shiotsu T, Nakamura S: A case report of portal-systemic encephalopathy with normal portal venous pressure and no cirrhosis of the liver in a hemodialysis patient. J Jpn Soc Dia Ther 29:327-330, 1996 13. Yokoo T, Kubo H, Ishikawa K, et al: Repetitive loss of consciousness during hemodialysis due to portal-systemic shunt in a ESRD patient with liver cirrhosis. J Jpn Soc Dia Ther 33:1115-1119, 2000 14. Yokoyama K, Ogura Y, Kawabata M, et al: Hyperammonemia in a patient with short bowel syndrome and chronic renal failure. Nephron 72:694-695, 1996
Hemodialysis-Related Portal-Systemic Encephalopathy Yoshifumi Ubara, MD, Jyunichi Hoshino, MD, Tetsuo Tagami, Naoki Sawa, MD, Hideyuki Katori, MD, Fumi Takemoto, MD, Masamichi Matsuda, MD, Shigeko Hara, MD, and Kenmei Takaichi, MD ● The authors encountered a case of portal-systemic venous shunt newly diagnosed after initiation of hemodialysis. A 68-year-old Japanese woman began hemodialysis because of symptoms of uremia including loss of appetite and pulmonary edema. Loss of consciousness occurred suddenly after her ninth session of hemodialysis. No hepatic functional abnormality was found other than hyperammonemia (314 g/dL [184 mol/L]). Loss of consciousness subsequently occurred often after hemodialysis. Color Doppler ultrasonography and magnetic resonance angiography depicted a large shunt between the left gastric vein and left renal vein resulting in portal flow entering the systemic circulation via the renal vein. Because the shunt was large, ligation of it was performed surgically. Results of histologic examination of a liver biopsy specimen obtained intraoperatively were normal. The patient became well postoperatively. This patient’s encephalopathy appeared to be caused by the flow of ammonia-rich portal venous blood into the systemic circulation via the large shunt owing to a decrease in intravenous pressure after rapid hemodialysis. Portal-systemic shunt encephalopathy should be recognized as a “new” neuropsychiatric disorder characteristic of patients undergoing hemodialysis. Am J Kidney Dis 44:E38-E42. © 2004 by the National Kidney Foundation, Inc. INDEX WORDS: Portal-systemic encephalopathy; hemodialysis; neuropsychiatric disorder; color Doppler ultrasonography; magnetic resonance angiography; hyperammonemia.
H
EPATIC ENCEPHALOPATHY usually is observed in patients with severe hepatocellular dysfunction. However, another type of hepatic encephalopathy may occur occasionally in patients without disorder of the hepatic parenchyma, ie, when portal venous blood flows into the inferior vena cava via a large shunt; this etiology has been called portal-systemic encephalopathy. Such encephalopathy occurring in the absence of intrinsic liver disease can be an important cause of morbidity.1-4 Recently in Japan, it has come to be recognized as one of the causes of disturbance of consciousness in patients on dialysis.5 We report a case of portal-systemic shunt encephalopathy that became apparent after initiation of dialysis and note that dynamic changes of blood flow during hemodialysis therapy may be closely related to this condition.
From the Nephrology Center and Department of Surgery, Toranomon Hospital, Tokyo, Japan. Received March 25, 2004; accepted in revised form May 14, 2004. Address reprint requests to Yoshifumi Ubara, Nephrology Center, Toranomon Hospital Kajigaya, 1-3-1, Kajigaya, Takatsu-kum Kawasaki-shi, Kanagawa-ken, 213-0015, Japan.E-mail: [email protected] © 2004 by the National Kidney Foundation, Inc. 0272-6386/04/4403-0025$30.00/0 doi:10.1053/j.ajkd.2004.05.035 e38
CASE REPORT A 68-year-old Japanese woman was admitted to our institution on October 16, 2000, because of severe edema in the lower extremities and loss of appetite. She exhibited proteinuria during her third pregnancy, at 30 years of age, in 1963, and proteinuria persisted. Although she was suspected to have chronic glomerulonephritis, renal biopsy was not performed. Thereafter, renal function gradually declined. The patient was admitted for further evaluation of her renal failure. On admission, she was 146 cm tall and weighed 44 kg. Her blood pressure was 154/100 mm Hg. Her abdomen was distended, and there was severe edema in her lower extremities. No abnormal neuropsychiatric findings were noted. Laboratory findings were as follows: erythrocyte count was 1.92 ⫻ 106/L, hemoglobin concentration was 6.2 g/dL (62 g/L), hematocrit level was 19.1%, leukocyte count was 7,100/ L, and thrombocyte count was 24.4 ⫻ 104/L. Total protein concentration was 6.6 g/dL (66 g/L), albumin level was 3.1 g/dL (31 g/L), urea nitrogen value was 50 mg/dL (17.9 mmol/ L), creatinine level was 5.1 mg/dL (451 mol/L), and uric acid was 4.9 mg/dL (291 mol/L). Sodium level was 146 mEq/L (mmol/L), potassium was 6.2 mEq/L (mmol/L), chloride was 121 mEq/L (mmol/L), calcium level was 6.0 mg/dL (1.50 mmol/L), phosphate was 4.3 mg/dL (1.39 mmol/L), and total cholesterol was 186 mg/dL (4.81 mmol/L). Total bilirubin was 0.3 mg/dL (5 mol/L), aspartate aminotransferase was 19 IU/L, alanine aminotransferase was level was 9 IU/L, lactate dehydrogenase level was 219 IU/L, and an indocyanine green test (R15) was 8%. C-reactive protein was 0.3 mg/dL. Serum immunoglobulin G concentration was 1,168 mg/dL (11.68 g/L), immunoglobulin A was 358 mg/dL (3.58 g/L), and immunoglobulin M was 27 mg/dL (0.27 g/L). The CH50 was 41 U. On viral examination, serum was negative for hepatitis C virus antibody (enzyme immunoassay) and for hepatitis B surface antigen.
American Journal of Kidney Diseases, Vol 44, No 3 (September), 2004: E38-E42
HEMODIALYSIS-RELATED PORTAL-SYSTEMIC ENCEPHALOPATHY
e39
Fig 1. Magnetic resonance angiography shows a large shunt (arrow) between the left gastric vein and left renal vein.
The urine pH was 5.5, negative for glucose, 1⫹ for protein, and 2⫹ for occult blood. The urine sediment contained 6 to 10 cells per high-power field. The urine was negative for BenceJones protein. Twenty-four–hour urine contained 1.25 g of protein and 4,740 g of beta 2 microglobulin and creatinine clearance was 5.8 mL/min (0.10 mL/s). Computed tomography and ultrasonography studies showed contracted kidneys bilaterally and normal liver architecture. A chest radiograph showed bilateral pleural effusion. Her cardiothoracic ratio was 71%. Dietary therapy including restriction of salt, protein, and water was performed immediately after admission, and diuretics and antihypertensive agents were administered. Forty-three days after her initial evaluation, hemodialysis was started 3 times weekly, because renal function had deteriorated, with a urea nitrogen of 128 mg/dL (45.7 mmol/L) and creatinine of 14.1 mg/dL (1,246 mol/L), and gastrointestinal symptoms such as loss of appetite, nausea, and vomiting began to occur. After initiation of hemodialysis, her gastrointestinal symptoms disappeared, and her appetite improved. Her dry weight began to increase owing to hypotension in the latter half of hemodialysis. Suddenly, after ninth treatment, she lost consciousness, although blood chemistry examination found no abnormality, with sodium of 142 mEq/L (mmol/L), potassium of 3.5 mEq/L (mmol/L), chloride of 104 mEq/L (mmol/L), calcium of 8.0 mg/dL
(2.00 mmol/L), phosphate of 2.4 mg/dL (0.77 mmol/L), total bilirubin of 0.3 mg/dL (5 mol/L), aspartate aminotransferase of 18 IU/L, alanine aminotransferase of 10 IU/L, lactate dehydrogenase of 214 IU/L, and glucose of 84 mg/dL (4.7 mmol/L). Because computed tomography of the brain was normal, her ammonia concentration was measured, and hyperammonemia was found, at 314 g/dL (184 mol/L; normal range, ⬍50 g/dL [⬍29 mol/L]). Twelve hours after initiation of fluid replacement at 40 mL/h, her consciousness became clear with decrease to 72 g/dL (42 mol/L) in her ammonia level. Loss of consciousness subsequently recurred often after hemodialysis. When the amount of fluid removed during a period of hemodialysis procedure became large, and her dry weight became set strictly so that blood pressure fell down in the latter half of hemodialysis, hyperammonemia-associated encephalopathy was apparent. Because ultrasonographic restudy found that hepatic parenchymal architecture was normal, blood analysis of portal vein was performed by using color Doppler. Color Doppler ultrasonography depicted a large shunt between the left gastric vein and left renal vein, with blood flow in the portal vein pouring into the inferior vena cava through left renal vein via a shunt from the left gastric vein. This portalsystemic shunt was confirmed by magnetic resonance angiography (Fig 1). Because embolization therapy using microcoils could not be performed because the shunt was large,
e40
UBARA ET AL
Fig 2. A schema of the findings in Fig 1. Red arrows show the flow of portal vein via the shunt. (A) left gastric vein, (B) splenic vein, (C) shunt, (D) left renal vein, (E) superior mesenteric vein, (F) inferior vena cava, (G) portal vein.
Fig 3. Ligation of the left gastric vein (black arrow) was performed surgically. Red thick arrows show the flow of blood in the portal vein. Red thin arrows show the flow of blood in the renal vein.
HEMODIALYSIS-RELATED PORTAL-SYSTEMIC ENCEPHALOPATHY
e41
Fig 4. Liver biopsy at the time of surgery shows normal histology. (Hematoxylin and eosin, original magnification ⴛ20.)
the ligation of the shunt was performed surgically (Figs 2 and 3). Histologic examination of a liver biopsy specimen obtained intraoperatively showed normal findings (Fig 4). The patient became well postoperatively and has experienced no episode of hepatic encephalopathy, with ammonia levels of 30 to 50 g/dL (18 to 29 mol/L), over the 3-year period since surgery.
DISCUSSION
Neuropsychiatric disorders in patients with renal failure have been called uremic encephalopathy. It usually improves immediately after dialysis has been initiated and has rarely been observed recently because dialysis therapy has been initiated earlier than in the past. Uremic toxins, including nitrogenous waste and parathyroid hormone as well as acidosis, have been suspected to be causative factors.6,7 Encephalopathy owing to aluminum intoxication had been found to occur in patients on long-term dialysis therapy. After replacement of aluminum phosphate–binding preparations by calcium derivatives and use of a reverse-osmotic apparatus for
processing water for dialysate, exposure to aluminum has been minimized.8 This encephalopathy has been observed rarely. Drugs administered to uremic patents may also cause encephalopathy if prescribed excessively. In Japan, attention recently has been paid to encephalopathy owing to portal systemic shunt formation, which became manifest after use of hemodialysis was started. This shunt may exist congenitally, may be formed spontaneously, or may develop after external abdominal injury or surgery.5,9 It usually remains nonsymptomatic until the end of life. However, with the addition of factors such as hemodynamic treatment of hemodialysis, shunt formation may become symptomatic. In anuric patients undergoing hemodialysis, a large fluid volume may be removed rapidly in 3 to 4 hours. Total body volume, including venous blood volume, will transiently decrease. When intravenous pressure decreases after hemodialysis, portal venous blood will flow readily into the inferior
e42
UBARA ET AL
vena cava via a large extrahepatic shunt. Once ammonia-rich portal venous blood flows into the systemic circulation, substances such as ammonia may cause central nervous system toxicity, resulting in hepatic encephalopathy. When the amount of fluid removed during a period of hemodialysis procedure becomes large, hyperammonemia tends to occur. Ligation of the shunt will improve encephalopathy.10,11 If histologic examination does not show abnormalities such as cirrhosis or fibrosis of the liver, the etiology of this encephalopathy may be hemodynamically mediated and may thus be termed portal-systemic encephalopathy. This does not usually occur in patients with peritoneal dialysis, because their body fluid can be removed very gradually over time. As the number of dialysis patients increase, the condition needs to be recognized as a potential cause of neuropsychiatric problems.12-14 We postulate the following speculation concerning the mechanism of portal-systemic shunt encephalopathy, based on the analysis of this case. Perhaps this encephalopathy may be caused by the flow of ammonia-rich portal venous blood into the systemic circulation via a large shunt due to a decrease in intravenous pressure after rapid hemodialysis. This encephalopathy should be recognized as a “new” neuropsychiatric disorder in patients undergoing hemodialysis (and unlikely to occur in those undergoing peritoneal dialysis). REFERENCES 1. Riordan SM, Williams R: Treatment of hepatic encephalopathy. N Engl J Med 337:473-479, 1997 2. Krieger D, Krieger S, Gass P, Theilmann L, Lichtnecker H: Manganese and chronic hepatic encephalopathy. Lancet 346:270-274, 1995
3. Jalan R, Seery JP, Taylor-Robinson SD: Review article: Pathogenesis and treatment of chronic hepatic encephalopathy. Ali Pharmaco Ther 10:681-697, 1996 4. Basile AS, Jones EA: Ammonia and GABA-ergic neurotransmission: Interrelated factors in the pathogenesis of hepatic encephalopathy. Hepatology 25:1303-1305, 1997 5. Shimono J, Azuma K, Hashiguchi M, Fujishima M: Recurring encephalopathy abolished by gastrorenal shunt ligation in a diabetic hemodialysis patient. Am J Gastroenterol 93:270-272, 1998 6. Fraser CL, Arieff AL: Brain abnormalities and peripheral neuropathy, in Massry SG, Glassock RJ (eds): Textbook of Nephrology (ed 4). Philadelphia, PA, Lippincott Williams & Wilkins, 2001, pp 1282-1287 7. Kovalik EC: Neurological manifestations, in Greenberg A (ed): Primer on Kidney Diseases (ed 3). San Diego, CA, National Kidney Foundation, 2001, pp 451-454 8. Ubara Y, Fushimi T, Tagami T, et al: Histomorphometric features of bone in patients with primary and secondary hypoparathyroidism. Kidney Int 63:1809-1816, 2003 9. Kerlan RK, Sollenberger RD, Palubinskas AJ, Raskin NH, Callen PW, Ehrenfeld WK: Portal-systemic encephalopathy due to a congenital portocaval shunt. Am J Radiol 139:1013-1015, 1982 10. Hanna SH, Smith RS, Henderson LM, Millikan WJ, Warren WD: Reversal of hepatic encephalopathy after occlusion of total portasystemic shunts. Am J Surg 142:285-289, 1981 11. Potts JR, Henderson M, Millikan WJ, Sones P, Warren WD: Restoration of portal venous perfusion and reversal of encephalopathy by ballon occlusion of portal systemic shunt. Gastroenterology 87:208-121, 1984 12. Onishi T, Nishitani M, Kanda M, Tsuji M, Shiotsu T, Nakamura S: A case report of portal-systemic encephalopathy with normal portal venous pressure and no cirrhosis of the liver in a hemodialysis patient. J Jpn Soc Dia Ther 29:327-330, 1996 13. Yokoo T, Kubo H, Ishikawa K, et al: Repetitive loss of consciousness during hemodialysis due to portal-systemic shunt in a ESRD patient with liver cirrhosis. J Jpn Soc Dia Ther 33:1115-1119, 2000 14. Yokoyama K, Ogura Y, Kawabata M, et al: Hyperammonemia in a patient with short bowel syndrome and chronic renal failure. Nephron 72:694-695, 1996