Karen L. Krok

Ascites

Ascites is easily diagnosed when the abdomen is distended with a large amount of fluid by assessing for shifting dullness or flank dullness; patients need at least 1500 mL of peritoneal fluid to be detected reliably by physical examination. In less obvious cases, abdominal ultrasonography can detect as little as 100 mL of free intraperitoneal fluid and should be used in patients in whom the physical examination is unreliable.

Perform abdominal paracentesis in all ICU patients with new-onset ascites or a change in clinical status (i.e., worsening encephalopathy, increasing creatinine, increasing white blood cell count or hypo- or hyperthermia). Send ascitic fluid for cell count with differential, albumin, total protein, and bacterial culture. Prophylactic transfusions of fresh frozen plasma or platelets before paracentesis are not data supported. In fact, bleeding complications occur in fewer than 1 in 1000 paracenteses. In a study of 1100 large-volume paracenteses, there were no bleeding complications despite (1) no prophylactic transfusions, (2) platelet counts as low as 19,000 platelets/mm³ (54% < 50,000 platelets/mm³), and (3) INRs as high as 8.7 (75% > 1.5 and 26.5% > 2.0). Coagulopathy should preclude paracentesis only when there is disseminated intravascular coagulation or clinically evident hyperfibrinolysis (three-dimensional ecchymoses/hematoma).

Ascites is best characterized by calculating the serum-ascites albumin gradient (SAAG) from samples obtained on the same day. A SAAG > 1.1 g/dL establishes the presence of portal hypertension and abolishes the need to check an albumin with each subsequent paracentesis.

Multiple factors may precipitate ascites (Box 27.1). Medical management (including diuretics and sodium restriction) is effective in more than 90% of patients. In the ICU, avoid excessive sodium administration in intravenous fluids. For example, the volume of normal saline intended to keep an intravenous catheter open (10 mL/h for 24 hours) contains 850 mg of sodium (because 1000 mL of 0.9% NaCl contains 154 mmol of sodium and 154 mmol of chloride, or 3542 mg of sodium since 1 mmol of sodium = 23 mg). Thus, 240 mL of 0.9% NaCl contains 850 mg of sodium (because 3542 mg of sodium/1000 mL × 240 mL = 850 mg), nearly half of the 2000 mg daily sodium restriction applied to patients with ESLD.

Spironolactone is the predominant diuretic used in ESLD. Because of a prolonged half-life of the drug and its metabolites, spironolactone may be administered once a day (generally initiated at a dose of 50 mg once per day). Furosemide at a dose of 20 mg once per day may be added to increase diuresis. While monitoring serum electrolytes and renal function, diuretics may be adjusted at 3- to 4-day intervals to a maximum of spironolactone 400 mg/day and furosemide 160 mg/day. If painful gynecomastia or other side effects occur from spironolactone therapy, amiloride, starting at 5 mg/day to a maximum of 40 mg/day, may be substituted. The goal of treatment should be weight loss of 0.3 to 0.5 kg/day in patients without peripheral edema and 0.8 to 1.0 kg/day in patients with peripheral edema. In patients on diuretics who do not achieve the desired weight loss, determine a urinary sodium; patients with urine sodium > 90 mEq/day (i.e., urine sodium greater than theoretic—and prescribed—sodium intake) are not compliant with sodium restriction. A low-sodium diet is extremely challenging to maintain and may require a nutritional consult to be successful. Unless a patient is having respiratory compromise, do not give diuretics intravenously to patients with ESLD, as this may cause a rapid fluid shift and precipitate hepatorenal syndrome.

Refractory ascites cannot be mobilized or recurs early despite sodium restriction and diuretic treatment. Alternatively, refractory ascites may reflect the development of diuretic-induced complications that preclude the use of an effective diuretic dose. Refractory ascites occurs in only 10% of patients with ascites. Treatment strategies include repeated therapeutic paracentesis plus intravenous albumin or the use of a transjugular intrahepatic portosystemic shunt (TIPS). Large-volume paracentesis (4 to 8 L) may provide symptomatic relief in patients with ascites. If greater than 4 liters of fluid are removed, albumin should be administered intravenously at a dose of 8 g of albumin per liter of fluid removed—that is, if 5 L of fluid are removed, then 40 g of albumin should be given (5 L × 8 g/L removed = 40 g).

A TIPS is a nonsurgical method of portal decompression that consists of inserting an intrahepatic stent between one hepatic vein and the portal vein using a transjugular approach. Reduction in portal pressure is accompanied by a resolution of ascites in most patients. TIPS may be associated with several side effects and complications and is usually contraindicated in elderly patients aged > 70 to 75 years. These include hepatic encephalopathy, obstruction of shunt, congestive heart failure, hemolytic anemia, and impairment in liver function.

Hepatic Hydrothorax

Hepatic hydrothorax is a large pleural effusion (> 500 mL) in a patient with cirrhosis and no coexisting cardiopulmonary diseases. The prevalence in cirrhotic patients is 5% to 10%, and 85% of cases are right-sided. The most likely etiology is the free passage of ascites from the peritoneal cavity through defects located in the tendinous portion of the diaphragm. When hepatic hydrothorax is suspected, perform a diagnostic thoracentesis and send the fluid for cell count, Gram stain and culture, protein, lactate dehydrogenase (LDH), albumin, and bilirubin. In uncomplicated hepatic hydrothorax, the cell count is < 500 cells/mm³ and the total protein < 2.5 g/dL.

All patients should be started on a low-sodium diet and diuretics, as described for the treatment of ascites. A refractory hepatic hydrothorax persists despite fluid and sodium restriction and use of maximal tolerated doses of diuretics and warrants evaluation for TIPS placement. In general, avoid insertion of a long-term pleural catheter or chest tube as a potential nidus for infections and excessive fluid losses.

Spontaneous Bacterial Peritonitis (SBP)

Gram-negative bacteria, particularly Escherichia coli, are responsible for 80% of cases of SBP. Streptococcus viridans, Staphylococcus aureus, or Enterococcus fecalis will be isolated in the remaining 20% of cases. The diagnosis of SBP is made when the ascitic fluid has > 250/mm³ neutrophils. A positive culture is not needed for the diagnosis of SBP. A “clinical” diagnosis of infected ascitic fluid without a paracentesis is not adequate.

An intravenous third-generation cephalosporin (cefotaxime or ceftriaxone) for 5 to 7 days is the treatment of choice for SBP. Renal insufficiency may occur in up to one third of patients with SBP, related to impaired circulatory function with activation of the vasoconstrictor systems. Attempting to prevent this complication, administer albumin intravenously at a dose of 1.5 g/kg at the diagnosis of the infection and 1.0 g/kg 48 hours later.

Patients with SBP have a very poor 1-year survival of only 30% to 50%. Therefore, evaluate all patients for liver transplantation after recovery from their first episode of SBP. Treat patients indefinitely with oral SBP prophylaxis, with norfloxacin 400 mg daily, ciprofloxacin 250 mg daily, or double-strength trimethoprim-sulfamethoxazole at least five times a week.

Two conditions associated with an increased risk of SBP warrant primary prophylaxis of SBP. First, in patients with gastrointestinal hemorrhage, multiple studies have shown that a short-term (7-day) administration of oral norfloxacin 400 mg twice a day or intravenous ceftriaxone 1 gm/day reduces the incidence of SBP, bacteremia, and rebleeding. Second, patients with a serum creatinine > 1.2 mg/dL, ascitic fluid protein levels < 15 g/L, a Child-Pugh score > 9, or dilutional hyponatremia (serum sodium < 130 mEq/L) should receive prophylaxis. In a randomized, placebo-controlled trial, primary prophylaxis with oral norfloxacin (400 mg/day) reduced the 1-year probability of developing SBP (7% versus 61%) and hepatorenal syndrome (28% versus 41%) and improved the 3-month survival (94% versus 62%) and the 1-year survival (60% versus 48%) compared with placebo.