Chapter 22: Transjugular Intrahepatic Portosystemic Shunts for Ascites and Hepatic Hydrothorax: Results of Clinical Studies The transjugular intrahepatic portosystemic shunt (TIPS) procedure is a useful treatment option in the management of patients with ascites caused by liver disease. Indeed, at tertiary referral centers with active liver transplantation programs, refractory ascites may be the most common indication for TIPS.1 Yet ascites remains a controversial indication for TIPS, and this procedure is less commonly performed outside of tertiary referral centers or in patients who are not candidates for liver transplantation. This chapter reviews the current approach to TIPS for ascites with an emphasis on the evidence to support the use of TIPS in patients with refractory ascites. Ascites by definition is the accumulation of more than 25 cc of fluid in the peritoneal cavity. Ascites is characterized by amount according to grade as follows: grade 1 is minimal fluid detectable by ultrasonography only, grade 2 is moderate fluid usually detectable on physical examination and symptomatic, and grade 3 is severe or tense ascites.2 Although there are many causes of ascites, in the United States, more than 85% of patients with ascites have an underlying cause related to liver dysfunction. Because the etiology of liver disease in the context of ascites is usually cirrhosis, we refer in this chapter chiefly to “cirrhotic ascites.” But it is well to keep in mind that, uncommonly, ascites may be the sequelae of noncirrhotic liver disease such as Budd-Chiari syndrome. Cirrhotic ascites occurs in patients with portal hypertension (PHT). It is one of the four clinical manifestations of PHT, which include variceal bleeding, ascites, hypersplenism, and encephalopathy. However, the onset of ascites requires more than just elevation of portal pressure. Ascites develops in patients with PHT when arterial vasodilation results in a hyperdynamic circulation stimulating renal vasoconstriction and sodium retention. Therapies aimed at lowering portal pressure have long been known to ameliorate the clinical sequelae of variceal bleeding and ascites. For variceal bleeding, Garcia-Tsao et al3 have established a clear threshold value below which variceal bleeding does not occur (hepatic venous pressure gradient [HVPG] <12 mm Hg). When the portal pressure is lowered below this threshold, bleeding stops. However, the relationship of portal pressure to the occurrence of ascites is not so clear-cut. Although cirrhotic ascites is dependent on PHT, the exact threshold for its occurrence is not well established. Authors have suggested thresholds ranging from 8 to 12 mm Hg.4,5 More important, the development of ascites is also due to two additional important factors, sodium balance and the hyperdynamic circulation.2,6 This means that therapeutic interventions that lower portal pressure may not completely resolve ascites when the other two causes are not ameliorated. Thus it is important to note from a pathophysiologic perspective that the effect of TIPS on patients with ascites is much less predictable than the effect of TIPS on patients with variceal bleeding. Moreover, although the effect of TIPS on variceal bleeding is immediate and bleeding usually stops within days, the effect of TIPS on ascites may take months to 1 year to show full effect. Because this is a fairly long time interval, the patient’s behavior during this interval may influence results. In particular, noncompliance with a sodium-restricted diet or lack of abstinence from alcohol may influence the outcome of TIPS in follow-up. In the early 1990s, numerous clinical trials were conducted to evaluate the use of TIPS for patients with variceal bleeding. In many of these trials, it was observed that some patients who also had ascites responded well to TIPS. But the response was not predictable. By the mid-1990s, numerous investigators began to more carefully study the effect of TIPS in patients with ascites who did not have variceal bleeding. In general, investigators postulated that TIPS would be as effective as a surgical side-to-side portocaval at treating ascites but hopefully with less upfront operative morbidity and mortality. Before discussing TIPS outcomes in greater detail, it is worthwhile to review general approaches to the treatment of ascites and some definitions. Readers can find a more comprehensive coverage of current management strategies in the American Association for the Study of Liver Diseases (AASLD) Guidelines just updated in 20137,8 and the European Association for the Study of the Liver (EASL) Guidelines.9 After a diagnosis of cirrhotic ascites is secured by analysis of the ascitic fluid, dietary salt restriction (<2000 mg/day of sodium) is the initial treatment. For patients with grade 2 or 3 ascites, if sodium restriction alone fails to control ascites, then diuretics are initiated. Spironolactone, an aldosterone antagonist, is initiated at a dosage of 100 mg/day and can be increased to a dosage of 400 mg/day. Loop diuretics such as furosemide are added up to a dosage of 160 mg/day as necessary.6 Patients on diuretic therapy for ascites are monitored carefully for electrolyte abnormalities and renal failure. Complications of diuretics include hyponatremia (serum sodium <120 mmol/L), hypo- or hyperkalemia (K <3 mmol/L or >6 mmol/L), and renal failure (increase in creatinine by 100% to a value >2 mg/dL). Approximately 5% to 10% of patients with ascites cannot be adequately managed with diuretics and salt restriction.2 Patients whose ascites does not respond to maximal diuretic therapy or that recurs rapidly are deemed to have “refractory ascites,” and such patients are candidates for second-line therapy. As well, patients who have a clinically significant complication from diuretics such as renal impairment or hyponatremia are also candidates for second-line therapy. Second-line treatment options include large-volume paracentesis (LVP), TIPS, and peritoneovenous shunts. Peritoneovenous shunts have largely been abandoned because of their poor patency and high morbidity. This procedure is now only performed when the other second-line therapies are not possible and transplantation is not feasible.6 Large-volume paracentesis is defined as removal of more than 5 L of ascites.2,7 In this procedure, all abdominal ascites is usually removed (i.e., total paracentesis), and albumin is infused intravenously (8 g albumin per liter of fluid removed) to minimize the circulatory effects of the procedure. LVP may be performed every 2 weeks. If LVP is needed more often than every 2 weeks, patients are probably not adhering to sodium restriction. TIPS has been performed widely throughout the world since the early 1990s, and the technique is well established. Briefly, a needle is advanced from the hepatic vein into the portal vein (PV), and a channel is created to decompress the PV. The channel is lined by a metal stent to maintain patency. The desired portal pressure gradient after shunt formation is 8 to 12 mm Hg. However, the precise post-TIPS pressure gradient cannot usually be accurately controlled prospectively by the operator. It is important to recognize that a significant modification in TIPS technique occurred in the early 2000s. Most of the clinical trials that are summarized in this chapter were carried out before that time using “bare-metal” TIPS. But in the early 2000s, the polytetrafluoroethylene (PTFE)-covered stent was introduced for TIPS, and this innovation has resulted in significant improvements in patency. Another potential difference between bare-metal TIPS and covered TIPS is that with covered stent TIPS, the stent extends all the way into the inferior vena cava, and most operators measure postprocedure gradient from the PV to the right atrium. The endpoints of interest in evaluating patients treated by TIPS for refractory ascites are: • The amount of ascites: This result is most often treated as a categorical variable with three categories, complete resolution, improvement, or no improvement. • Survival: This endpoint is calculated using actuarial analysis either up to time of transplantation (transplant-free survival) or including transplantation (overall survival). • Complications: Encephalopathy, variceal bleeding, spontaneous bacterial peritonitis (SBP), renal failure, liver failure • Quality of life and cost Because patients treated for refractory ascites generally have more severe liver disease than those treated for variceal bleeding alone, survival is a particularly important endpoint. In general, the development of ascites in patients with cirrhosis confers a poor prognosis with a 1-year survival rate of 85% dropping to 50% at 2 years.6 The development of refractory ascites confers an even worse prognosis. The most significant complications from TIPS performed to treat refractory ascites in most published trials are encephalopathy and liver failure. Encephalopathy is the most frequent serious complication. Liver failure is the most serious complication and one that can precipitate the need for liver transplantation.
Introduction
Pathophysiology and Pressures
Treatment Options
First-Line Therapy
Second-Line Therapy
Results of Clinical Studies: TIPS for Refractory Ascites
Outcome Variables
Indications • Refractory ascites • Diuretic refractory ascites • Complications of LVP • Hepatic hydrothorax |
Contraindications • Absolute • Uncontrolled encephalopathy • Congestive heart failure (ejection fraction <50%) • Severe pulmonary artery hypertension • Unrelieved biliary obstruction |
• Relative • Age older than 70 years • Severe liver dysfunction (bilirubin > 5 mg/dL, CP > 12, MELD score >18) • Significant encephalopathy • INR > 2 |
CP: Child-Pugh score; INR: international normalized ratio; LVP: large-volume paracentesis; MELD: model for end-stage liver disease.
Indications and Contraindications
The indications and contraindications for TIPS in the setting of refractory ascites are shown in 
Table 22.1.
Prospective Cohort Trials: TIPS for Ascites
During the first decade of TIPS research in the 1990s, a number of cohort series focused on the use of TIPS to control ascites. The results of nine of these studies10–18 are summarized in 
Table 22.2. Note that in these uncontrolled studies, improvement in ascites varied widely (47% to 97%) as did the severity of liver disease in the patient populations studied with the percentage of patients with severe liver disease (Child’s class C) ranging from 41% to 75%.
Careful study of the effects of TIPS on patients with ascites has shown that urinary sodium excretion and serum creatinine improve steadily within 4 weeks of the procedure and can normalize in 6 month to 1 year of the procedure.1 As well, the circulatory dysfunction that accompanies PHT is ameliorated by TIPS, thus improving ascites. Detailed studies evaluating nutritional status after successful TIPS have shown positive improvement in nitrogen balance, dry weight, and total body fat.19–21 These beneficial effects can lead to improved quality of life.
Table 22.2 Transjugular Intrahepatic Portosystemic Shunt for Ascites: Early Cohort Studies10–18
Studies (n) | 9 |
Patients (n) | 295 |
Mean follow-up time | 6–17 months |
Technical success rate (%) | 93–100 |
Ascites improvement rate (%) | 47–96 |
Child’s C disease (%) | 14–90 |
1-year survival rate (%) | 41–75 |
