Case Report:Hepatorenal Syndrome: Resolution of Ascites by ContinuousRenal Replacement Therapy in an Alcoholic Coinfected withHepatitis B, C, and Human Immunodeficiency Viruses

Paul C. Hansard,1 Ricardo A. Manning,1 M.A. Haseeb,1,2,3 and Martin J. Salwen3

1Departments of Medicine, 2Microbiology & Immunology, and 3Pathology, Downstate Medical Center,State University of New York, and Kings County Hospital Center, Brooklyn, New York

Abstract. A 39-yr-old male with hepatorenal syndrome type 1 and refractory ascites was treated with continuous renal replacement therapy (CRRT) resulting in clinical improvement. He was positive forantibodies to hepatitis B, C, and human immunodeficiency viruses, and had a history of chronic alcoholand iv drug abuse. The patient had 4 hospital admissions during a 12-wk period. He first presented with advanced liver disease including pedal edema and a serum ammonia level of 56 µmol/L (reference range: 11 - 35 µmol/L). In subsequent admissions, he had asterixis, nausea, vomiting, jaundice, and worsening pedal edema. On his 4th admission, there was lethargy, tense ascites, decreased urinary output, bilateral edema of the lower extremities and scrotum, serum creatinine of 6.2 mg/dl (reference range: 0.6 - 1.5 mg/dl), and weight gain of 16 kg during the prior 8 wk. During the first 3 hospitalizations, he was treated withlactulose with slight improvement. On the 4th admission, he was started on low-dose dopamine (3 µg/kg/min) and 25% salt-poor albumin without clinical improvement. A pulmonary artery catheter was placedand hemofiltration by CRRT was performed for 5 days, with removal of 26.7 L of fluid and a net reductionof 11 kg of body weight. Serum creatinine decreased to 4.2 mg/dl during CRRT and was 2.2 mg/dl at hospital discharge 2 weeks later. His PaO2 improved from 66 to 78 mmHg and his systemic vascular resistance increased from 571 to 799 dyne•sec/cm5. CRRT was effective in relieving severe fluid retentionand producing marked clinical improvement. We suggest that CRRT should be considered for thetreatment of refractory ascites including that caused by hepatorenal syndrome.

Keywords: continuous renal replacement therapy, hemofiltration, hepatorenal syndrome; ascites

Introduction

Hepatorenal syndrome (HRS) has long beenknown to occur in individuals with advanced liver disease [1]. Its pathophysiologic mechanisms are poorly understood. Renal dysfunction or failureoccurs in patients with severe liver disease, but thekidneys demonstrate no intrinsic pathologicalteration. Renal dysfunction results from decreased renal perfusion, due in part to splanchnic arterial vasodilation caused by portal hypertension andincreased nitric oxide synthesis [2]. The complexity

of HRS is evident from the diagnostic criteria of the International Ascites Club that are based on theassessment of multiple measures of liver and kidney functions [2-4]. Management approaches for overtHRS, specifically type 1, short of liver trans-plantation, have not resulted in marked clinical improvement or increased survival. Mortality in cases of rapidly progressive HRS type 1 approaches100% within 10 wk after onset [5].

We report successful treatment of a patientwith HRS type 1. Clinical improvement ofrefractory ascites was accomplished by hemofiltra-tion with continuous renal replacement therapy(CRRT). Biochemical and physiological data arepresented to document the patient’s substantial improvement following completion of CRRT.

Address correspondence to Paul C. Hansard, M.D., KingsCounty Hospital Center, 451 Clarkson Ave., Brooklyn, NY11203, USA; tel 718 245 3820; fax 718 245 3764; e-mail phansard@pol.net.

0091-7370/06/0100-0096. $1.25. © 2006 by the Association of Clinical Scientists, Inc.

Available online at www.annclinlabsci.org

Annals of Clinical & Laboratory Science, vol. 36, no. 1, 200696

The CRRT procedure, as either continuous veno-venous hemofiltration (CVVH) or continuousarterio-venous hemofiltration (CAVH), is used toremove excess fluid. Blood is circulated through a hemofilter, yielding an ultrafiltrate with the com-position of plasma water, containing low molecular weight non-protein-bound solutes. The cellularcomponents and high molecular weight substancesin the blood are returned to the patient [6,7].

Case Report

The patient was a 39-yr-old male who was infectedwith hepatitis B, C, and human immunodeficiencyviruses. He had a history of chronic alcohol and iv drug abuse, but was negative for HIV-associated opportunistic infections.

First admission. The patient presented with lethargy, worsening resting tremors, and altered mental status. During 2 days before admission, he had several episodes of vomiting following meals.He was receiving therapy with zidovudine (AZT), lamivudine (3TC), indinavir, methadone, and thiamine. Physical examination findings were ecchymoses of the lower extremities and pedal edema, limiting mobility. Laboratory data on admission (Table 1) included an elevated serumammonia level of 56 µmol/L (reference range: 11 -35 µmol/L). Computer tomography (CT) of the head with contrast showed no abnormalities; Gramstain and culture of cerebrospinal fluid (CSF) werenegative. Anti-retroviral medications were temp-orarily discontinued and the patient was givenlactulose. There was gradual resolution of the lethargy and resting tremors. The patient left the hospital against medical advice on the third day.

Second admission. Approximately 2 wk later, the patient was readmitted due to hepatic encephal-opathy with asterixis, nausea, and vomiting.Physical findings were similar to those of the prior admission except that pedal edema was absent. Anabdominal sonogram showed hepatosplenomegalywith homogenous architecture. Blood and CSFcultures for bacteria and fungi were negative. The patient again improved in response to lactulose and was discharged after 2 days.

Third admission. Four weeks later, the patient presented with bipedal edema, jaundice, and

passage of bright red blood per rectum. Laboratorydata were similar to those of prior admissions(Table 1). He was treated with lactulose, vitaminK, and was transfused with 4 units each of packedred blood cells and fresh frozen plasma.

Fourth Admission. Eight wk after first hospitaladmission and 18 days after his last discharge, thepatient presented with a 10-day history of lethargy,progressive abdominal distention and tenderness, and decreased urinary output. He stated that for aweek he had been too weak to walk and that he had been anuric for the prior 24 hr. He denied recent alcohol consumption and his toxicology screen test identified only methadone, which was a prescribedmedication. He weighed 77 kg (baseline weight: 61 kg). Physical examination was significant foricterus, pallor, and spider nevi. There were bilateralpulmonary crepitations, bronchial breathing withegophony, and increased tactile fremitus. Hisabdomen was protuberant, consistent with ascites,but neither hepatomegaly nor splenomegaly was evident on palpation. He had testicular atrophy and bilateral lower extremity edema extending to the level of the thigh and inclusive of the scrotum.

Bilateral pleural effusions and interstitialinfiltrates consistent with pulmonary edema wereseen on chest X-ray. Abdominal X-ray showed nofree air or obstruction. Diagnostic thoracentesis (RBC, 4050/mm3; WBC, 6/mm3; glucose, 114mg/dl; LDH, 39 U/L; protein, 1.0 g/dl; pH 7.45) and paracentesis (RBC, 2090/mm3; WBC, 30/mm3; glucose, 84 mg/dl; protein, 0.5 g/dl) wereperformed and both fluids were determined to betransudates. His serum creatinine level was 6.2 mg/dl (reference range: 0.6 - 1.5 mg/dl). He was anemic and had prolonged plasma prothrombin time (PT) and partial thromboplastin time (PTT)(Table 1). Urine osmolality was 324 mOsm/kg;urine electrolytes were: Na, <10 mmol/L; K, 37.2mmol/L; and Cl, 15 mmol/L. Bacterial and fungal cultures of pleural fluid, ascitic fluid, blood, andurine were negative.

Based on these findings and his co-morbidities,the diagnoses of pre-renal azotemia or hepatorenalsyndrome were considered. The therapeutic goalwas to improve renal perfusion through hydrationwith non-invasive monitoring of vital signs. The patient was advised of the severity of his condition

Continuous renal replacement therapy in hepatorenal syndrome 97

Table 1. Laboratory data for serum or plasma samples and hemodynamic values of the patient for each of the 4 consecutive hospitalizations. Data for the first 3 admissions represent those recorded at admission. The fourth admission data represent thoserecorded before, during, and after initiation of CRRT. See text for abbreviations. Reference ranges are in parentheses.

Admission 1 2 3 4Days in hospital 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 26Day in Medical ICU 1 2 3 4 5 6 7 8 9 10 11 12 13 14Day CRRT 1 2 3 4 5

Diagnostic parameter

Na (135-145 mmol/L) 134 131 129 127 132 129 131 136 138 139 139 135 137 134 134 135 134 132 129

Cl (101-111 mmol/L) 111 107 108 104 109 105 109 106 108 113 112 114 114 110 104 111 109 109 107

Creatinine (0.6-1.5 mg/dl) 0.7 1.5 1.3 6.2 7.0 6.9 7.1 6.8 6.5 6.1 5.6 5.5 4.5 4.5 4.4 4.2 4.2 4.3 2.2

Urea nitrogen (0-22 mg/dl) 11 19 15 44 53 55 63 66 65 68 63 60 58 61 64 64 69 66 52

Protein, total (6.0-8.5 g/dl) 10.8 11.1 10.7 9.1 8.4 8.7 7.1 8.4 8.1 8.3 8 8.1 8.6 8.5 8.9 8.8 8.3 7.7 9.1

Albumin (3.3-6.1 g/dl) 1.4 1.7 1.8 1.5 1.6 1.6 1.6 1.7 2 2.2 2.1 2 2.5 2.6 2.7 2.5 2.6 2.5 2.5

AST (0-48 U/L) 179 153 178 107 100 98 92 99 94 85 102 92 91 89 88 97 80 74 120

ALT (10-40 U/L) 53 68 52 54 46 43 41 43 38 40 42 44 44 40 45 44 38 36 56

Bilirubin, total (0-1.2 mg/dl) 3.9 4.3 5.4 7 7.5 7.7 6.1 6.2 6.2 6.7 6.2 5.8 7 7.2 8.3 7 6.5 6.1 11.4

PT (11.5-13.5 sec) 26.4 22 26 24.1 24 23.6 40 39 40 38 41 40 41 24.4 29.1 26.6 28.6 24 25.2

PTT (24-33.6 sec) 65 53 62 53.7 51.9 50.7 49.5 54 63.9 69.9 >100 59.5

Ammonia (11-35 µmol/L) 56 59 26 40

GGT (8-54 U/L) 28 11 11 14 16 12 8 15

Cardiac index (2.5-4 L/min/m2) 5.6 5.1 4.9 4.4 4.6 4.5 4.9 5.2 4.4 4.7 4.4

Wedge pressure (8-12 mmHg) 11 7 8 11 9 7 6 5 5 5 6

Systemic Vascular Resistance (800-1200 dyne x sec/cm5) 571 771 710 841 745 767 657 632 791 726 799

Fig. 1. Patient’s daily net fluid balance (NFB) during days inthe medical ICU. Arrow = start of CRRT. NFB = total intake (oral + iv fluids) - total output (urine + ultrafiltrate).

Fig. 2. Patient’s pCO2 and PaO2 levels. Arrow = start ofCRRT. Bars represent pCO2 and the line represents PaO2.

Annals of Clinical & Laboratory Science, vol. 36, no. 1, 200698

and his ominous prognosis without intervention. He was advised of the risks and potential compli-cations as well as possible benefits of CRRT and invasive monitoring. The patient’s written consentfor CRRT was obtained and he was transferred to the medical intensive care unit (MICU).

A pulmonary artery catheter was placed via a median basilic approach to monitor hemodynamicresponse to therapy. The patient was started onlow-dose dopamine (3 µg/kg/min) and salt-poor albumin (25% albumin/50 ml). His urinary output increased without change in serum creatinine and he had weight reduction of only 0.6 kg during 7 days. It was determined that drug therapy alone was ineffective in achieving the desired volumereduction. A double-lumen femoral vein catheterwas placed and low-volume CRRT was begunusing CVVH with a blood flow of 125 ml/min, removing 300 - 400 ml of ultrafiltrate/hr. A Minntech RenaFlo II (HF-400) hemofilter(Minntech, Mineapolis, MN) was primed (using 1 L of 0.9% NaCl containing 5000 units of sodiumheparin) until the air was replaced with the solution.Routine anticoagulation of the hemofilter wasaccomplished by heparinization with 25,000 unitsof sodium heparin in 250 ml of 0.9% NaCladministered at approximately 400 units heparin/hr [6,7]. CRRT was continued for 5 days with the removal of a total of 26.7 L of ultrafiltrate. Therewas concurrent hemodynamic and biochemicalimprovement (Table 1, Fig. 1), as well as improved

oxygenation with decreased ventilation (Fig. 2).Serum creatinine decreased to 4.2 mg/dl during CRRT and was 2.2 mg/dl at hospital discharge 2wk later. PaO2 improved from 66 to 78 mmHg andsystemic vascular resistance increased from 571 to 799 dyne• sec/cm5.

The patient was discharged from the MICU onthe 15th day of hospitalization and from the hospital on the 26th day. There had been a net weight-loss of 11 kg and the patient’s body weightwas 61 kg on discharge (Fig. 3).

Discussion

Management and treatment of HRS with or without ascites has been a challenge to medical practitioners ever since its description [2,5]. InHRS with anasarca the prognosis is guarded andmost diuretic-refractory patients succumb. Treat-ment modalities have been limited to the use ofdiuretics, large volume paracentesis, and porto-systemic shunting, all with poor outcomes. Theonly effective treatment resulting in renal recovery for this condition is liver transplantation. Althoughrestoration of renal function appears to be the keyto reversing the HRS, conventional hemodialysis,its various modifications, and even kidney trans-plantation have not resulted in survival beyond a few days [2-5].

Our patient met the International Ascites Club’s Diagnostic Criteria of HRS type I [2,4]. Hehad a low systemic vascular resistance (SVR) and a high cardiac index (CI), consistent with an under-lying infection, portal hypertension, or vasodilation.All cultures were negative and no pharmacologicagent had been administered. There was no evidence of GI hemorrhage on the 4th admission, nor wasthe patient on diuretic therapy. Infusion of low-dose dopamine (3 µg/Kg/min) and albumin (25%) resulted in an increase in SVR (571 to 745 dyne•sec/cm5) and a decrease in CI (5.6 to 4.6 L/min/m2).There were improvements in SVR and CI, but there was no significant change in net fluid balance andCRRT was therefore initiated. CRRT resulted in reduction of the ascites and pulmonary interstitial edema, while the patient remained hemodynam-ically stable (Table 1); the serum creatinine level decreased and net fluid balance improved (Fig. 1).

Fig. 3. Patient’s body wt and serum creatinine levels (mg/dl)before and after sart of CRRT (indicated by arrow). Barsrepresent body wt and the line represents creatinine levels.

Continuous renal replacement therapy in hepatorenal syndrome 99

Since ultrafiltration is known to remove low molecular weight molecules from blood, thereduction in serum creatinine was expected.However, the continued creatinine reduction after CRRT had been stopped was surprising andsuggests that other molecules contributing to the pathogenesis of HRS may have been removed.Antidiuretic hormone, norepinephrine, nitric oxide, and neuropepitde Y have all been implicatedas possible potentiators of fluid retention and have molecular sizes that would permit their removal byCRRT. There is no known circulating toxinassociated with HRS. The temporal relationshipbetween circulating levels of these substances andthe clinical condition of such patients warrants investigation.

In our patient, CRRT resulted in resolution ofascites and edema, and improvement of renalfunction and hemodynamics. Despite significantcoagulopathy, the CRRT procedure was performedsafely and without incident. CRRT successfully achieved clinical improvement in this patient and should be considered in the therapy for refractory ascites including that resulting from HRS.

References

1. Hecker R, Sherlock S. Electrolyte and circulatorychanges in terminal liver failure. Lancet 1956;268:1121-1125.

2. Arroyo V, Guevara M, Ginès P. Hepatorenal syndrome in cirrhosis: pathogenesis and treatment. Gastro-enterology 2002;122:1658-1676.

3. Gines P, Guevara M, Arroyo V, Rodes J. Hepatorenalsyndrome. Lancet 2003;362:1819-1827.

4. Arroyo V, Gines P, Gerbes AL, Dudley FJ, Gentilini P, Laffi G, Reynolds TB, Ring-Larsen H, Scholmerich J. Definition and diagnostic criteria of refractory ascites and hepatorenal syndrome in cirrhosis. Hepatology1996;23:164-176.

5. Wong F, Blendis L. New challenge of hepatorenalsyndrome: prevention and treatment. Hepatology2001;34:1242-1251.

6. Hansard PC, Haseeb MA. Continuous renal replacementtherapy: an underutilized resource. Pulmonary Per-spectives 2001;18:4-6.

7. Hansard PC, Haseeb MA, Manning RA, Salwen MJ.Recovery of bacteria by continuous renal replacementtherapy in septic shock and by ultrafiltration from an in vitro model of bacteremia. Crit Care Med 2004;32:932-937.

Annals of Clinical & Laboratory Science, vol. 36, no. 1, 2006100