An understanding of the mechanisms that lead to ascites formation lends to understanding the treatment strategies and their shortcomings. The fundamental goal of ascites management is to induce a negative sodium balance. This is achieved through a combination of restriction of sodium intake and diuretics . Factors to consider in treatment strategies include grade or severity of ascites and initial presentation (Fig. 13.2).

A key consideration early in management is an analysis of the ascites itself. It is meant to confirm cirrhosis as a cause of the ascites and to exclude complicating conditions such as infections etc. Typically, cirrhosis-related ascites has a low total protein and albumin especially with respect to circulating albumin levels. Thus, the serum to ascites albumin gradient is usually more than 1.1 in those with cirrhosis . Cirrhosis-associated ascites typically has a low white blood cells (WBC) count. When the neutrophil count exceeds 250/mm³, spontaneous bacterial peritonitis is considered to be present. When in doubt regarding the presence of spontaneous bacterial infection, ascites fluid should be injected into blood culture bottles for bacterial culture. Hemorrhagic ascites should raise concern for malignant or tuberculous ascites. When the ascites appears milky, a chylous ascites is usually present and confirmed by a high triglyceride level (level > 200 mg/dL). In those who consume a large amount of alcohol, pancreatic ascites can be diagnosed by the presence of a high protein and amylase levels (usually 1000 IU/l) in ascites. Ascites cytology may be useful sometimes in diagnosis of peritoneal carcinomatosis; however, a negative study does not exclude the condition and may require additional imaging and direct visualization of the peritoneum with biopsies to confirm the diagnosis.

The International Ascites Club has proposed a treatment approached based on the quantitative severity (Table 13.1). In patients with grade 1 ascites, those with ascites detectable by ultrasound alone, no treatment is recommended. There are no data available as to the management or natural history with or without intervention. Grade 2 or moderate ascites is more readily encountered and the diagnosis made clinically. These patients have impairment in renal sodium excretion; sodium excretion is low comparative to sodium intake. Grade 3 ascites is defined as large or gross ascites with marked abdominal distension best addressed with sequential large volume paracentesis (LVP) followed by sodium restriction and diuretics.

Dietary sodium should be restricted to 2000 mg/day (88 mmol/day). Compliance with such dietary restrictions becomes a barrier to therapeutic efficacy. Salt substitutes containing high potassium content should be used cautiously particularly when used with potassium-sparring diuretics as this can result in hyperkalemia. A negative sodium balanced is successful in 10–20 % of cirrhotic patients particularly those presenting with their first episode of ascites [12]. There is no evidence to suggest a benefit in sodium restriction in cirrhotic patients who have never developed ascites.

Diuretics block sodium reabsorption along the nephron leading to natriuresis and passive water excretion. Most commonly used is a combination of spironolactone (an aldosterone antagonist) and furosemide (loop diuretic) at doses of 100 mg and 40 mg/day, respectively. The site of action on the nephron is important in understanding how these diuretics compliment each other. Spironolactone works in the distal tubule by blocking aldosterone resulting in a decrease in sodium reabsorption. Loop diuretics work more proximally to prevent sodium reabsorption. Mechanistically, loop diuretics alone are not therapeutically efficacious, as the sodium not reabsorbed in the loop of Henle would later be reabsorbed distally in the setting of a hyperaldosterone state [13]. The evidence for the use of spironolactone as monotherapy in patients with a first episode of ascites is extrapolated from studies comparing monotherapy with sequential use of furosemide to nonresponders versus dual therapy both with a stepwise increase in doses. Therefore, in patients with new ascites, spironolactone alone can be started at 100 mg/day and increased in a stepwise fashion every 7 days (100 mg steps) to a maximum of 400 mg/day [14, 15]. To prevent electrolyte derangement and acute kidney injury from diuretics, the goals of therapy should be weight loss of 0.5 kg/day in patients without peripheral edema and 1 kg/day in patients with peripheral edema [16]. Furosemide can be added at a dose of 40 mg/day in patients who are not responding to monotherapy and gradually increased to 160 mg/day. In patients with recurrent ascites, combination therapy with spironolactone and furosemide beginning with 100 and 40 mg, respectively, should be the strategy of choice with a stepwise simultaneous increase in doses maintaining the same ratio of dosages.

Side effects of spironolactone include hyperkalemia and decreased libido, impotence, and gynecomastia in men and menstrual irregularity in women, as a result of its antiandrogenic activity. Amiloride is an alternative in patients with tender gynecomastia, but was shown to be more expensive and less efficacious that spironolactone [17]. Tamoxifen has been reported to be effective in managing the symptoms of gynecomastia [18]. Clonidine, a central alpha-2 agonist, has sympatholytic activity in patients with cirrhosis. Simultaneous use of clonidine and spironolactone has been shown in studies to increase natriuresis and body weight loss more efficiently [19].

Ascites becomes refractory to diuretics and salt restriction in 10 % of cases. Refractory ascites (RA) has been defined as ascites that cannot be mobilized or the early recurrence of which cannot be prevented by medical therapy (Table 13.1) [20]. The prognosis associated with RA is poor, with about a 50 % 1-year survival rate [21]. As a consequence, patients with RA should be considered for liver transplantation. Current available treatments include LVP with albumin infusions, peritoneal shunts, or liver transplantation (Fig. 13.3).

Paracentesis is the first-line treatment of RA. It offers the advantage of quickly relieving tense ascites safer than high-dose diuretics and found to shorten duration of hospitalization, though survival is similar to those of diuretic therapy [22]. The frequency and volume of LVP is a reflection of the patient’s sodium intake. In general, a patient adherent to sodium restriction of 88 mmol/day will accumulate less than 4 L of ascites per week. Repeated LVP is relatively safe, despite a cirrhotic patient’s bleeding diathesis. The incidence of significant peritoneal bleeding complications during paracentesis has been reported 0.5–1 %, despite cirrhotic patients having coagulopathies and thrombocytopenia [23, 24]. In patients with renal failure (e.g., hepatorenal syndrome, HRS type 2), the risk of bleeding may be higher due to dysfunctional circulating platelets and may require an extended post paracentesis observation.

The most common complication of LVP is paracentesis-induced circulatory dysfunction (PICD) caused by effective hypovolemia and accompanying marked activation of the renin–angiotensin axis. PICD can result in worsening vasodilation, hyponatremia , and renal impairment in 20 % of cases [25]. More importantly, PICD may persist for months and is linked with subsequent adverse clinical events such as an increased rate of recurrent ascites , the development of HRS, and reduce survival [26]. The incidence of PICD correlates with the volume of ascites removed during paracentesis. Incidence of PICD is only 7 %, with little clinical consequence with a paracentesis less than 6 L [27]. However, the use of albumin given intravenously at a dose of 6–8 g/L of ascites removed can decrease the incidence of PICD when performing a paracentesis greater than 5 L. The frequency is approximately 75 % when LVP is performed without the administration of plasma expanders [17]. Albumin’s superiority over synthetic volume expanders (e.g., polygeline) was seen a double-blind, randomized pilot study showing a decrease in liver-related complications [28]. Beyond its role as a volume expander, albumin is thought to work on the endothelial dysfunction and circulatory disturbances associated with cirrhosis [26]. A meta-analysis showed that albumin is the most effective agent in the prevention of hemodynamic and clinical effects associated with PICD [29].

Not having addressed the underlying pathophysiology, patients with RA will undergo repeated LVP without any other intervention. The persistent ascites increases their risk for developing spontaneous bacterial peritonitis and HRS. Furthermore, there is an indirect impact on worsening nutrition: Ascites accumulation is associated with decreased caloric intake coupled with protein losses with repeated paracentesis.

The transjugular intrahepatic portosystemic shunt (TIPS) is an established procedure that has proven benefit in the treatment of patients with RA. TIPS reduces the portosystemic pressure gradient, one of the pathogenetic mechanisms of ascites formation, by functioning as a side-to-side portocaval shunt. Within 4 weeks after TIPS, urinary sodium excretion and serum creatinine improve significantly and can normalize within 6–12 months. This is associated with an increase in serum sodium concentration, urinary volume, and glomerular filtration rate together with a normalization of plasma renin activity, aldosterone, and noradrenaline concentrations during 4–6 months of follow-up [30, 31]. Patients should follow a sodium-restricted diet immediate post TIPS period and may require the use of diuretics to facilitate ascites clearance. Complete resolution of ascites is seen in two thirds and partial response in the other third within a 6-month follow-up period. At 12 months post TIPS, approximately 80 % of patients will completely clear their ascites [32]. Unrecognized cirrhotic cardiomyopathy is an identified risk factor for lack of ascites clearance after TIPS [33]. Cirrhotic cardiomyopathy is characterized by blunted contractile responsiveness to stress, and/or altered diastolic relaxation with electrophysiological abnormalities in the absence of other known cardiac disease and in the setting of cirrhosis [34].

Five randomized controlled trials have compared LVP versus TIPS as a treatment for ascites [35–39]. All have showed that TIPS is much more effective than LVP in controlling ascites, though at the expense of more episodes of hepatic encephalopathy (Table 13.2). Two separate meta-analyses showed a survival advantage with TIPS in carefully selected patients when compared to LVP (Table 13.3). Additionally, TIPS has shown to improve renal function, nutritional status, and improvement in quality of life [31, 40, 41]. The advent of polytetrafluoroethylene (PTFE)-covered stents has also improved long-term shunt patency.