Ascites and Hyponatremia

Introduction

Ascites is the pathologic accumulation of fluid in the peritoneal cavity, and its most common cause is cirrhosis. That fluid accumulates in the abdominal cavity has been known since ancient times, and it was Hippocrates who recognized that ascites (from the Greek askos , meaning a leather bag used to carry wine, water, or oil) derived from a diseased liver carried a grim prognosis. The development of ascites is one of the complications of cirrhosis that marks the transition from compensated to decompensated cirrhosis. Other complications that mark this transition are variceal hemorrhage, hepatic encephalopathy, and jaundice; however, ascites is the most common first decompensating event in patients with cirrhosis, and the one that carries the highest mortality.

Epidemiology

Ascites is present in 20% to 60% of patients with cirrhosis at the time of diagnosis, depending on the referral pattern. The cumulative probability of developing ascites ranges from 35% to 50% within 5 years in patients with compensated cirrhosis of all causes ( Fig. 15-1 ).

Pathogenesis

In cirrhosis, leakage of ascitic fluid into the peritoneal space occurs as a result of sinusoidal hypertension that in turn results from hepatic venous outflow block secondary to regenerative nodules and fibrosis. The other essential factor in the pathogenesis of cirrhotic ascites is plasma volume expansion, through sodium and water retention, that allows the replenishment of the intravascular volume and maintains the formation of ascites ( Fig. 15-2 ).

Clinical Features

The most frequent symptoms associated with ascites are increased abdominal girth (described as tightness of the belt or garments around the waist ) and recent weight gain. Ascites induces abdominal distention but this sign in itself has a poor specificity because it is present in other conditions, such as obesity, gas, tumors, and pregnancy. Small to moderate amounts of ascites can be identified by bulging flanks, flank dullness, and shifting dullness. The last two are the most sensitive physical signs in the diagnosis of ascites. Ascites can be classified in three grades: (1) mild ascites only detectable by ultrasound examination; (2) moderate ascites manifested by moderate symmetric distension of the abdomen; and (3) large or gross ascites with marked abdominal distension.

Diagnosis

Abdominal ultrasonography is the most cost-effective and least invasive method to confirm or exclude the presence of ascites and is therefore considered the gold standard in the diagnosis of ascites. It can detect amounts as small as 100 mL and even as small as 1 mL to 2 mL when the Morison pouch and the pelvic cul-de-sac are scanned. Abdominal ultrasonography is also useful in determining the best site to perform a diagnostic or therapeutic paracentesis, particularly in patients with a small degree of ascites or in those with loculated ascites. Additionally, findings on ultrasonography (nodular surface, splenomegaly) can indicate the presence of cirrhosis, and Doppler ultrasonography is the most useful initial test to investigate the presence of hepatic vein obstruction, an important and frequently overlooked cause of ascites. Therefore in patients with new-onset ascites, abdominal ultrasonography should always be performed and should include Doppler examination of the hepatic veins.

Differential Diagnosis

Although cirrhosis is the cause of ascites in more than 75% of patients, other causes of ascites such as peritoneal malignancy (12%), cardiac failure (5%), and peritoneal tuberculosis (2%) should be considered in the differential diagnosis of ascites.

A diagnostic paracentesis should be the first test performed in the diagnostic workup of a patient with ascites. It is a safe procedure with a very low incidence of serious complications, mostly transfusion-requiring hematomas that occur at a rate of 0.2% to 0.9%. Renal dysfunction appears to be more predictive of bleeding complications than clotting abnormalities, and therefore coagulopathy is not a contraindication to perform a diagnostic paracentesis. The preferred site for needle insertion is located in the left lower quadrant. Care should always be taken to avoid abdominal wall collaterals and to avoid the area of the inferior hypogastric artery, which lies midway between the anterior superior iliac spine and the pubic tubercle.

Uncomplicated ascitic fluid is transparent, straw colored to slightly yellow. The presence of blood in a nontraumatic tap (in which blood does not clot) may indicate the presence of malignant ascites. Milky fluid is indicative of chylous ascites, and although cirrhosis is the most common cause of nonsurgical chylous ascites, it represents only 0.5% to 1% of cases of cirrhotic ascites.

Ascites total protein content and serum-ascites albumin gradient (SAAG) are two inexpensive measurements that, taken together, are most useful in determining the cause of ascites and therefore in guiding the workup of patients with ascites. A high (>2.5 g/dL) ascites total protein content occurs in peritoneal processes (malignancy, tuberculosis) because of leakage of high-protein mesenteric lymph from obliterated lymphatics and/or from an inflamed peritoneal surface. A high ascites total protein content also occurs in cases of postsinusoidal or posthepatic sinusoidal hypertension when sinusoids are normal and protein-rich lymph leaks into the peritoneal cavity. In hepatic cirrhosis an abnormally low protein content of liver lymph has been demonstrated as a result of deposition of fibrous tissue in the sinusoids (capillarization of the sinusoid) that renders the sinusoid less leaky to macromolecules. On the other hand, the SAAG, which involves measurement of the albumin concentration of serum and ascitic fluid specimens and subtraction of the ascitic fluid value from the serum value, has been shown to correlate with hepatic sinusoidal pressure. A SAAG cutoff value of 1.1 g/dL has been shown to be the best to distinguish patients in whom ascites is secondary to liver disease and those with malignant ascites. Interestingly, this cutoff corresponds to a portal pressure gradient of 11 mm Hg to 12 mm Hg, the threshold pressure necessary for the development of ascites in cirrhotic patients ( Fig. 15-3 ). Therefore the SAAG and the ascites total protein content can distinguish among the three main causes of ascites—(1) cirrhosis, (2) peritoneal disease (malignancy or tuberculosis), and (3) posthepatic or postsinusoidal causes of ascites (e.g., heart failure, veno-occlusive disease)—and guide the further workup of the patient with ascites ( Fig. 15-4 ). In cirrhosis the SAAG is high and the ascites total protein content is low, in ascites secondary to peritoneal causes the SAAG is low and the ascites total protein content is high, and in posthepatic or postsinusoidal causes of ascites the SAAG is high and the ascites total protein content is high ( Fig. 15-5 ). A study performed in patients with new-onset ascites showed that serum B-type natriuretic peptide has a greater diagnostic accuracy than SAAG/ascites total protein content in the diagnosis of cardiac ascites, with levels above 364 pg/mL ruling in cardiac ascites and levels below 182 pg/mL ruling it out.

In patients with mixed ascites (e.g., cirrhosis with superimposed peritoneal malignancy) the SAAG is high and the ascites total protein content is low (i.e., the findings of ascites due to cirrhosis predominate). Of note, massive hepatic metastasis can lead to the development of ascites but because the mechanism of ascites formation is sinusoidal hypertension, these cases of malignant ascites will have a high SAAG.

The definitive test to determine whether ascites is the result of sinusoidal hypertension is measurement of hepatic sinusoidal pressure. The HVPG, obtained by hepatic vein catheterization and by subtraction of the free hepatic vein pressure (FHVP) from the wedged hepatic vein pressure (WHVP), is a measure of sinusoidal pressure. In cases of cirrhotic ascites the HVPG will be 10 mm Hg to 12 mm Hg or greater. In cases of cardiac ascites, both the WHVP and the FHVP will be elevated (reflecting elevated systemic pressures), and therefore the HVPG will be normal. In cases of peritoneal ascites (i.e., malignancy or tuberculosis), all hepatic venous pressure measurements (WHVP, FHVP, and HVPG) will be normal unless the patient has coexisting cirrhosis or heart failure. When performed properly, HVPG measurements are reproducible and safe. Additionally, hepatic vein catheterization for measurement of hepatic vein pressures allows the performance, in the same procedure, of a transjugular liver biopsy that will further help define the cause of ascites.

Associated Conditions

Hyponatremia develops in approximately 30% of cirrhotic patients with ascites and has been arbitrarily defined as a serum sodium concentration of 130 mEq/L or less ; however, even sodium levels below 135 mEq/L, the lower limit of normal, have been found to be of significant prognostic value. In the setting of cirrhosis, patients can have two types of hyponatremia: hypovolemic hyponatremia and hypervolemic hyponatremia.

Hypovolemic hyponatremia is a consequence of fluid losses either from the kidneys (most commonly from iatrogenic overdiuresis) or from gastrointestinal tract losses (i.e., diarrhea). Patients will typically have signs of dehydration and findings of prerenal azotemia due to the contraction of total plasma volume and will generally not have ascites or edema. In contrast, most patients with cirrhosis and ascites present with hypervolemic dilutional hyponatremia, characterized by increased urinary tubular water reabsorption due to the nonosmotic release of antidiuretic hormone. As described already, this in turn results from arterial vasodilatation, the main pathogenic mechanism in ascites formation. In a prospective inception cohort of patients with new-onset ascites, patients first developed hyponatremia, followed by the development of refractory ascites that in turn preceded the development of hepatorenal syndrome. Each of these stages was associated with progressively lower values of mean arterial pressure indicative of worsening vasodilatation, and with progressively higher Child-Pugh and Model for End-Stage Liver Disease (MELD) scores indicative of worsening liver function. Dilutional hyponatremia in cirrhosis represents an intermediate stage in the progression from ascites to hepatorenal syndrome. Therefore patients with dilutional hyponatremia usually have ascites and/or edema, may have concurrent kidney injury, and have a higher mortality, independent of the MELD score.

Although dilutional hyponatremia in cirrhosis is usually asymptomatic, it contributes to hepatic encephalopathy, is associated with impaired health-related quality of life, and increases the risk of developing central pontine myelinolysis after transplant (rapid correction early postoperative period).

Umbilical hernias develop in approximately 20% of cirrhotic patients with ascites (a rate significantly greater than the 3% in patients without ascites) and in up to 70% of patients with long-standing recurrent tense ascites. The main risks of these hernias are rupture and incarceration, a complication that has been observed mostly in patients in whom ascites resolves rapidly after paracentesis, peritoneovenous shunt (PVS), or transjugular intrahepatic portosystemic shunt (TIPS).

Hepatic hydrothorax develops in approximately 5% to 10% of patients with cirrhosis, most probably as the result of the transdiaphragmatic movement of fluid from the peritoneum to the pleural space through diaphragmatic defects. It usually develops in patients with ascites; however, hepatic hydrothorax may develop in patients without detectable ascites. Although large amounts of ascitic fluid can accumulate in the peritoneal cavity before resulting in significant patient discomfort, the accumulation of smaller amounts of fluid (1-2 L) in the pleural space results in severe shortness of breath and hypoxemia. Pleural effusion is right-sided in 85% of cases, left-sided in 13% of cases, and bilateral in 2% of cases. The diagnosis of hepatic hydrothorax is established by radionuclide scanning of the chest after the intraperitoneal injection of ^99m Tc-labeled sulfur colloid or macroaggregated serum albumin. In hepatic hydrothorax, presence of radiotracer in the pleural space is demonstrated generally within 2 hours of intraperitoneal injection.

Disease Complication: Spontaneous Bacterial Peritonitis

Spontaneous bacterial peritonitis (SBP) is an infection of ascites that occurs in the absence of a contiguous source of infection (e.g., intestinal perforation, intraabdominal abscess) and in the absence of an intraabdominal inflammatory focus (e.g., abscess, acute pancreatitis, cholecystitis). Spontaneous bacteremia and spontaneous bacterial empyema are other spontaneous infections in cirrhosis. The main mechanism for these spontaneous infections is translocation of bacteria from the gut lumen to mesenteric lymph nodes, to the systemic circulation (bacteremia), and then to existing fluids (ascites and/or hydrothorax). Therefore the organisms most commonly isolated in SBP are gram-negative organisms. However, because of the widespread use of antibiotic prophylaxis and inappropriate antibiotic use in patients with cirrhosis, a greater incidence of SBP due to antibiotic-resistant organisms has been reported in the literature.