Hemodialysis alternative with ascites ultrafiltration for an end-stage renal failure patient associated with tense ascites secondary to decompensated liver cirrhosis

Hemodialysis Alternative With Ascites Ultrafiltration for an End-Stage Renal Failure Patient Associated With Tense Ascites Secondary to Decompensated Liver Cirrhosis Jyh-Chang Hwang, MD, Jen-An Chert, MD, and Hon-Yin Fung, MD • Patients with end-stage renal disease combined with tense ascites caused by decompensated liver cirrhosis are sometimes encountered in a hemodialysis center. A big problem for the management of these patients is the tendency of hypotension during ultrafiltration. Subsequent fluid accumulation, especially in the abdominal cavity, causes breathing difficulty and abdominal discomfort. We present a new technique, ascites ultrafiltration, to solve this problem. Using the same equipment as for ordinary hemodialysis, and incurring the same cost, we removed directly approximately 8 L of ascites fluid during each nearly 4-hour session. No hemodynamic instability was noted. We proved this technique to be an effective and safe alternative method for this group of patients. © 1996 by the National Kidney Foundation, Inc. INDEX WORDS: Tense ascites; ultrafiltration; liver cirrhosis; renal failure.

A

SCITES, a common sequel of advanced liver cirrhosis, causes abdominal distension and shortness of breath. The traditional treatment is based on bedrest, a low-sodium diet, and the administration of aldosterone antagonists and loop diuretics.1 The use of therapeutic paracentesis and peritoneovenous shunt is restricted to patients who are refractory to the conservative therapies; however, the long-term results are unsatisfactory. 2 Although Bernardi et al have presented a similar technique to extract approximately 8 L of ascites over a longer period, 3 this therapeutic strategy has been rarely mentioned in the literature, 4'5 especially in a chronic uremic patient. We present a patient with end-stage renal failure on maintenance hemodialysis who, using the same apparatus, received intermittent "ascites ultrafiltration" for the refractory ascites secondary to the decompensated liver cirrhosis. We also attempt to demonstrate the effectiveness and safety of this procedure in the management of patients with two organ failures. CASE REPORT A 53-year-old male patient with end stage renal disease has received regular hemodialysis three times a week at our dialysis center since March 1993. He was confirmed as having B-type hepatitis approximately 2 months before the initiation of renal replacement therapy. The patient was found to be hepatitis B surface antigen-positive and anti-hepatitis C virusnegative during entrance into the regular hemodialysis program. One episode of esophageal variceal bleeding occurred in January 1994. Splenomegaly with liver cirrhosis and ascites was diagnosed using abdominal sonography and computed tomography. Monthly routine biochemistry checkups showed serum albumin levels of approximately 2.2 to 2.5 g/dL throughout the whole course, while the serum alkaline phosphatase level was approximately 150 to 250 1U/L, y-glutamyl transferase (r-

GT) was 425 IU/L, glutamate oxaloacetic transaminase (GOT) was 45 to 70 IU/L, and glutamate pyruvate transaminase (GPT) 27 to 68 IU/L. Progressive massive ascites formation gave rise to abdominal discomfort and difficulty in breathing. Regular hemodialysis continued throughout this period. The ultrafiltration rate was maintained at approximately 0.3 to 0.5 L/hr, and the total volume of body fluid removal was only approximately 1.5 to 2.5 L in each dialysis session (Fig 1). The blood pressure was approximately 110/80 to 70/40 mm Hg or even lower during the intradialytic period (Fig 2), especially about 3 hours after the beginning of each hemodialysis. Because of the tendency toward hypotension, we failed to increase the ultrafiltration rate to remove excessive fluid from this patient even after albumin infusion. He subsequently gained weight and experienced progressive abdominal distension, so we tried "ascites ultrafiltration" intermittently beginning in November 1994 to relieve the intraabdominal pressure. We used the same artificial kidneys (Baxter CA170; Deerfield, IL) in both ordinary hemodialysis and ascites ultrafiltration. In this cellulose acetate hollow fiber dialyzer, the membrane has a surface area of 1.7 m 2 and an ultrafiltration rate coefficient of 11 mL/min/mm Hg. The heparin loading dosage during ascites ultrafiltration was 2,000 U and maintenance dosage was 1,000 U/hr, which was approximately twice the dosage used in hemodialysis for this patient. We inserted a femoral puncture catheter (Kendall; MedWest, Salt Lake City, UT) as an outlet route through the dependent portion of the lower abdomen at either side. We also used a gauge #16 intravenous catheter as an inlet at the counterlateral side of the lower abdomen. The tubing system (Kawasumi Laboratories; Tokyo, Japan) and dialysis machine

From the Division of Nephrology, Chi-Mei Foundation Hospital, Tainan, Taiwan. Received May 16, 1996; accepted in revised form July 17, 1996. Address reprint requests to Jyh-Chang Hwang, MD, No 901, Chung Hwa Rd, Yung Kang City, Tainan, Taiwan, 71010, ROC. © 1996 by the National Kidney Foundation, Inc. 0272 -6386/96/2806-001553.00/0

American Journal of Kidney Diseases, Vol 28, No 6 (December), 1996: pp 899-903

899

900

HWANG, CHEN, AND FUNG

8

6 5 UFR

4

(L) 3 2

Fig 1. Comparison of total ultrafiltration rate between ascites ultrafiltration and hemodialysis. Day 0: ascites ultrafiltration; days - 2 , 2, and 4: hemodialysis.

t 0 Day-2

Day0

Day2

Day4

Treatment Modalities

(Toray 201, single-patient machine [Tokyo, Japan]) were the same as used during hemodialysis. The flow rate for pumping ascites was approximately 250 mL/min. In March 1995, this patient received hemiorrhaphy for the progression of the umbilical hernia due to intractable ascites. By June 1995, he had been completed 11 sessions of ascites ultrafiltrafon. The frequency of ascites ultrafiltration was increased gradually from once a month initially to once every 2 weeks, then to about once every 10 days in the final stage (Table 1). The indication for ascites dialysis was the patient's complaint of abdominal distension and difficulty in breathing. The ultrafiltration rate was approximately 2 to 3 L/hr and the total ultrafiltration was approximately 5 to 8.85 kg for each session (Table 1). The total duration taken for the process was about 3 hours 18 minutes (2 hours 45 minutes to 3 hours 45 minutes), and the indication for termination of this process was the development of negative suction over the outer end. The hemodynamic status was satisfactory (Fig 2). Preultrafiltration systolic pressure was 116.83 _+ 15.70 mm

Hg, while postultrafiltration it was 112.83 _+ 17.87 mm Hg (P = 0.83). The patient's heart rate was similar between the preultrafiltration and postultrafiltration periods (89.83 _+ 15.08 beats/min to 85.83 _+ 4.49 beats/rain; P = 0.48), and no hypotension developed during nearly 4 hours of manipulation. The abdominal girdles were signifcantly reduced, especially near the area of the umbilicus. The abdominal circumference at the umbilicus level changed from 92.42 _+ 1.56 cm to 80.30 _+ 3.71 cm (P < 0.001) after ultrafiltration. No peritonitis episode developed during the treatment course. Ordinary hemodialysis was carried out between the ascites ultrafiltration sessions to maintain an adequate solute clearance. Finally, this patient died from an episode of intractable esophageal variceal bleeding in July 1995.

DISCUSSION Several hypotheses have been proposed to explain the mechanisms

NS

of developing

N$

120

lOO

•

~aN

~ ~° Fig 2. Comparison of blood pressure changes between ascites ultrafiltration and hemodialysis before and after treatment. Pre, preultrafiltration; post, postultrafiltration; Sys, systolic pressure; Dias, diastolic pressure. *P < 0.001.

40

20

0 PreSys

PostSys

PreDias Treatment Modalities

PostDias

a s c i t e s in

ASCITES ULTRAFILTRATION IN LIVER CIRRHOSIS Table 1. Frequency, Treatment Duration, and Total Ultrafiltration Rate During Ascites Ultrafiltration

Date

Ultrafiltration (kg)

Total Duration (h)

Nov 8, 1994 Dec 13, 1994 Jan 10, 1995 Feb 7, 1995 Feb 21, 1995 March 4, 1995 March 11, 1995 April 6, 1995 April 18, 1995 April 27, 1995 May 16, 1995

8.30 7.45 7.03 7.49 7.46 5.00 6.90 6.44 8.41 8.85 8.10

3:55 3:30 4:00 3:10 3:30 3:30 3:00 2:45 3:55 3:05 3:10

liver cirrhosis, such as the underfilling hypothesis, 6 the overflow theory] and the peripheral arteriolar vasodilation hypothesis. 8 The evidence for and against these hypotheses still requires further confirmation. However, the common pathway leading to the retention of sodium and water is "inadequate intravascular volume," although this condition is really a "hypervolemic" state. With the activation of the sympathetic nervous and renin angiotensin aldosterone systems, 9 the kidneys fail to autoregulate the disarrangement of body fluid status. Diuretics were not effective for maintaining the fluid homeostasis in chronic uremic patients. Ultratiltration during hemodialysis was a main route for water excretion in this patient. Unfortunately, due to low oncotic pressure and insufficient intravascular volume, the tendency of hypotension during the dialysis process caused ultrafiltration to be inadequate, and fluid retention eventually developed. Marked ascites causes abdominal distention, abdominal pain, anorexia, gastroesophageal reflux, ventilation restriction, pulmonary atelectasia, pneumonia, and umbilical or inguinal hernia. 9 The patient presented here often complained of intermittent abdominal distension, breathlessness, and progressive umbilical hernia when ascites accumulated; intermittent relief of abdominal pressure was therefore indicated. The aim of the treatment was to mobilize the intraabdominal fluid and to relieve the pressure. The conservative treatment modalities, including bed rest, low-sodium diet, and plasma expanders, such as albumin and fresh frozen plasma infu-

901

sion, 1° could not restrain the progression of ascites formation. In fact, approximately 5% to 10% of ascites are refractory to conservative medical treatments. 11 For the same cost as regular hemodialysis and without further requirement of other equipment, we tried "ascites ultrafiltration" for this patient. The major advantage of ascites ultrafiltration was effectiveness. It could remove almost all the fluid in the abdomen without interfering with the patient's hemodynamic status. In our experience, the rate of fluid removal could be maximized to 3 L/h, which was the upper limit of most dialysis machines, and no episode of hypotension was noted after more than 15 L was removed in one session. It was not necessary to supply albumin during ascites ultrafiltration to stabilize the blood pressure. Massive ascites will cause V/Q mismatch and lead to hypoxemia.la The aim of treatment for decompensated liver cirrhosis was to relieve the intra-abdominal pressure by removing the water but not all the components in the ascites. Fluid removal by ascites ultrafiltration, which retains the proteins in the peritoneal cavity by the semipermeable membrane, seems to be more logical than abdominal paracentesis, which would lose all the proteins in the ascites during water removal. Runyon et al13,14 have demonstrated that opsonic activity of ascitic fluid in cirrhosis is directly correlated with the concentration of defensive substances, such as immunoglobulins, complement, and fibronectin, and with the concentration of total protein in ascites. Simple tapping would deplete all these substances in the peritoneal cavity. Installing the access for ascites ultrafiltration took only about 10 minutes, which is similar to that of the puncture procedure in ordinary hemodialysis. Ascites ultrafiltration is a simple and effective method. Besides the "ascites ultrafiltration," there were four other acceptable methods to effectively remove the intra-abdominal fluid in this uremic patient: therapeutic paracentesis with or without albumin infusion, continuous ambulatory peritoneal dialysis, (3) slow ultrafiltration with prolonged duration, and peritoneovenous shunt. Why did we prefer the ascites ultrafiltration technique for this patient? Although weekly abdominal tapping could have been performed easily by any internist, a large amount of albumin would have been lost. In our patient, the preultrafiltra-

902

HWANG, CHEN, A N D FUNG

Table 2. The Biochemistry Changes After Ascites Ultra filtration Items Serum BUN (mg/dL) Cr (mg/dL) AIb (g/dL) Ascites BUN (mg/dL) Cr (mg/dL) AIb (g/dL)

Pre-Ascites Ultrafiltration

Post-Ascites Ultrafi[tration

Probability Value

70.83 ± 12.01 9.90 _+ 1.37 2.35 _+ 0.19

61.33 ± 9.18 8.93 _+ 1.07 2.25 ± 0.14

<0.01 <0.01 0.11

68.50 ± 10.37 9.83 ± 1.36 0.60 ± 0.13

22.50 ± 5.50 3.17 ± 0.82 1.55 ± 0.50

<0.01 <0.01 <0.01

NOTE. Data are expressed as mean values _+ SD. Abbreviations: BUN, blood urea nitrogen; Cr, creatinine; AIb, albumin.

tion concentration of albumin in the ascites was approximately 0.60 g/dL (Table 2), and more than 40 g of albumin would have been drained out weekly if 7 L of peritoneal fluid was tapped. One characteristic of liver cirrhosis is impaired hepatic synthesis function and the consequent hypoalbuminemia. 15 Progressive hypoalbuminemia would eventually develop under frequent paracentesis without adequate albumin replacement and nutrition supply. The cost of four vials of albumin infusion is slightly more than that of one session of ascites ultrafiltration. Therefore, the ascites ultrafiltration would not be more expensive than the total cost of abdominal paracentesis. The molecular weight of albumin is approximately 69,000 d. It is too large to be filtrated through the membrane of the hollow fiber, and the albumin will be reserved almost completely in the ascites after ascites ultrafiltration. It even could be concentrated to approximately three times the preultrafiltrafion concentration (Table 2). In the same way, the albumin would be completely lost through effluent dialysate if this patient were changed to continuous ambulatory peritoneal dialysis to treat the uremia. Chronic indwelling of the Tenckhoff catheter would also carry a greater risk of peritonitis, especially in this immunocompromised patient. Because of the tendency for hypotension (Fig 2), the maximal ultrafiltration rate could not be elevated by more than 0.5 L/hr during hemodialysis, and the total ultrafiltration rate in one session of hemodialysis was only approximately 2 kg (Fig 1). Although we could prolong the duration of hemodialysis, it was still difficult to increase the amount of removed water to more than

3 L within 6 hours. The slow ultrafiltration was time consuming and not effective. We also could increase the frequency of hemodialysis, such as to a frequency of four or five sessions per week. Although this would be helpful in removing the excess fluid, the treatment cost would be much higher. The ascites ultrafiltration was much more effective in removing fluid than hemodialysis (Fig 1). Several complications limited the use of the peritoneovenous shunt, in particular, coagulopathy and bleeding, 16 so we did not favor a LeVeen shunt for this patient. Because of limited experience and little description in the literature, 17'1s we do not know exactly the side effects and disadvantages of ascites ultrafiltrafion. Some patients complained of weakness and drowsiness on the day after the procedure, especially those patients with more than 10 L of ascites fluid removed. This might be due to redistribution of fluid and its composition between the intravascular compartment and the peritoneal cavity. Although the procedure was simple, the basic required equipment being a hemodialysis machine, an extracorporeal conduit, and a hollow fiber, only a hospital with a hemodialysis center could perform this procedure. The cost of this therapy in Taiwan would be $148 ($US) (the cost of hemodialysis). Although it represents a routine cost for the uremic patient, it is not a very cheap treatment modality. No peritonitis developed in this patient, but bacterial infection was still a potential risk for any procedure with catheter insertion into the peritoneal cavity. In summary, we found that intermittent ascites ultrafiltration was the most effective and simple albumin-reserving method for maintaining the fluid homeostasis in chronic uremic patients complicated with tense ascites secondary to decompensated liver cirrhosis. This technique also would not add any cost to the treatment of such cases, and is superior for managing the body fluid homeostasis of ascites patients compared with therapeutic paracentesis with plasma expander infusion, continuous ambulatory peritoneal dialysis, or prolonging the hemodialysis period or frequency. It was also suitable for those patients with tense ascites without chronic uremia. REFERENCES 1. Scholmerich J: Strategies in the treatment of ascites. Hepatogastroenterology 38:365-370, 1991

ASCITES ULTRAFILTRATION IN LIVER CIRRHOSIS 2. Gentilini P, La Villa G, Romanelli RG, Foschi M, Laffi G: Pathogenesis and treatment of ascites in hepatic cirrhosis. Cardiology 84:68-79, 1994 (suppl 2) 3. Bernardi M, Rimondi A, Gasbarrini A, Trevisani F, Caraceni P, Legnani C, Palareti G, Gasbarrini G: Ascites apheresis, concentration and reinfusion for the treatment of massive or refractory ascites in cirrhosis. J Hepatol 20:289295, 1994 4. Cadranel JF, Gargot D, Grippon P, Lnnel F, Bernard B, Valla D, Opolon P: Spontaneous dialytic ultrafiltration of the concentrate versus large volume paracentesis in cirrhotic patients with intractable ascites: A randomized study. Int J Artif Organs 15:432-435, 1992 5. Bruno S, Borzio M, Romagnoni M, Battezzati PM, Rossi S, Chiesa A, Podda M: Comparison of spontaneous ascites filtration and reinfusion with total paracentesis with intravenous albumin infusion in cirrhotic patients with tense ascites. BMJ 304:1655-1658, 1992 6. Witte MH, Witte CL, Dumont AE: Progress in liver disease: Physiological factors involved in the causation of cirrhotic ascites. Gastroenterology 61:742-750, 1971 7. Denison EK, Lieberman FL, Renolds TB: 9-c~-Flnorohydrocortisone induced ascites in alcoholic liver disease. Gastroenterology 61:497-503, 1971 8. Schrier RW, Arroyo V, Bernardi M, Epstein M, Henriksen JH, Rodes J: Peripheral arterial vasodilation hypothesis: A proposal for the initiation of renal sodium and water retention in cirrhosis. Hepatology 8:1151-1157, 1988 9. Gentilini P, La Villa G, Romanelli RG, Foschi M, Laffi G: Pathogenesis and treatment of ascites in hepatic cirrhosis. Cardiology 84:68-79, 1994 (suppl 2)

903 10. Forns X, Gines A, Gines P, Arroyo V: Management of ascites and renal failure in cirrhosis. Semin Liver Dis 14:83-96, 1994 11. Arroyo V, Epstein M, Gallus G, Gentilini P, RingLarsen H, Salerno F: Refractory ascites in cirrhosis: Mechanism and treatment. Gastroenterol Int 2:195-207, 1989 12. Ruff F, Hughes JMB, Stanley N, McCarthy D, Greene R, Aronoff A, Clayton L, Milic-Emili J: Regional lung function in patients with hepatic cirrhosis. J Clin Invest 50:24032413, 1971 13. Runyon BA, Morrissey RL, Hoers JC: Opsonie activity of human ascitic fluid. A potentially important protective mechanism against spontaneous bacterial peritonitis. Hepatology 5:634-637, 1985 14. Runyon BA: Patients with deficient ascitic fluid opsonic activity are predisposed to spontaneous bacterial peritonitis. Hepatology 8:632-635, 1988 15. Sherlock S, Dooley J (eds): Hepatic Cirrhosis, in SherIock S, Dooley J (eds): Diseases of the Liver and Biliary System (ed 9). Oxford, UK, Blackwell Scientific, 1992, pp 357-369 16. Volk BA, Scholmerich H, Wilms K, Hasler E, Kottgen W, G: Treatment of refractory ascites by retransfusion and peritoneovenous shunting. Dig Surg 2:93-96, 1985 17. Lai KN, Leung JWC, Loke J, Panesm" NS, Swaminathan R, Callance-Owen J: Ultrafiltration by hemofilter, a new therapeutic measure in intractable ascites. Int J Artif Organs 10:109-114, 1987 18. Rossaro L, Graziotto A, Bonato S, Plebani M, van Thiel DH, Burlina A, Naccarato R, Salvagnini M: Concentrated ascitic fluid reinfusion after cascade filtration in tense ascites. Dig Dis Sci 38:903-908, 1993