Chapter 8: Paracentesis and the LeVeen and Denver Shunts Large-volume paracentesis (LVP) and peritoneovenous shunts (PVS) are two of four commonly used methods of treating ascites that is refractory to optimal medical management (the other two, transjugular intrahepatic portosystemic shunt [TIPS] and liver transplant [LT], are discussed in other chapters of this book). All four methods compete for the same patient population, that is, patients with end-stage portal hypertension (PHT). It is not a fair competition. LT is the only one of the four that will significantly extend patient survival. Active alcohol or substance abuse, HIV or other systemic infections, life-limiting medical conditions (advanced cardiac, pulmonary, neurologic, or neoplastic conditions), uncontrolled psychiatric disorders, inability to comply with pre- and posttransplant regimens, and advanced age may eliminate a patient from consideration for LT. Patients with refractory ascites who are not candidates for LT or are waiting for a liver to become available are candidates for LVP and PVS placement. One of the most serious signs of decompensated liver disease is refractory ascites, which has a prevalence of 5% to 10% in patients with advanced PHT and a survival rate as low as 50% at 12 to 24 months after diagnosis.1 Ascites is considered refractory when it cannot be relieved by dietary sodium restriction and diuretic treatment. Sodium restriction to 2 g/day (88 mmol/day) and diuretics are the mainstay of therapy for cirrhotic ascites. Together these measures achieve successful diuresis in more than 90% of patients. About 10% to 15% of patients have sufficient spontaneous natriuresis to be managed by sodium restriction without the need for diuretics. The remainder requires oral diuretics; those recommended by the American Association for the Study of Liver Diseases (AASLD) are a combination of 100 mg of spironolactone and 40 mg of furosemide administered as a single morning dose.2 If necessary, these doses can be raised until a maximum of 400 mg and 160 mg, respectively, are reached. One or more additional diuretics may be added in low doses to treat inadequate diuresis. Potassium disturbances or painful gynecomastia may necessitate a reduction in the dose of spironolactone and the substitution of amiloride. Failure to control ascites by these measures may be due to an insufficient response to sodium restriction and diuretic therapy, termed diuretic-resistant ascites, or the occurrence of diuretic-induced complications such as hyperkalemia, hyponatremia, renal insufficiency, or hepatic encephalopathy, termed diuretic-intractable ascites. Contraindications to diuretic use include hepatic encephalopathy, a serum sodium level lower than 120 mmol/L, and renal insufficiency with a serum creatinine level above 2 mg/dL. Diuretics should be discontinued and other therapeutic options considered if any of these adverse events occurs.3 The main factor contributing to the development of ascites in a patient with cirrhosis is PHT, which results from increased intrahepatic resistance to blood flow and is exacerbated by splanchnic vasodilatation as a result of local production of vasodilators. A commonly accepted theory of ascites formation in PHT is that it develops secondary to the continuous retention of sodium. Splanchnic vasodilation caused by PHT causes arterial system underfilling (hypovolemia); as a result, vasopressin activity increases and causes fluid retention. The body is eventually unable to maintain homeostasis, and fluid starts leaking into the peritoneal cavity. Sodium and water retention develop as the sympathetic nervous system and renin–angiotensin–aldosterone system are activated to compensate for the arterial hypovolemia. This theory is believed to explain the development of hepatorenal syndrome (HRS) and dilutional hyponatremia seen in patients with refractory ascites. The splanchnic circulation remains patent in PHT as a result of its own production of vasodilators such as prostaglandin E, prostacyclin, and nitric oxide, causing low vascular resistance, increased cardiac output, and decreased blood pressure. As a result of elevated hepatic sinusoidal pressure, there is an increase in venous blood flow and therefore increased lymph formation, which exceeds the flow rate that can be transported by the thoracic duct into the central venous system, resulting in fluid leaks into the peritoneal cavity (i.e., ascites). Paracentesis was the only treatment available for ascites from the time of Hippocrates until the advent of dietary sodium restriction and oral diuretics in the 20th century. For the past 2000 years, a rather large, hollow-bore metallic cannula was used access the ascitic collection; this has been replaced over the past 60 years by much smaller needles and catheters and a more stringent attention to aseptic technique.1 Despite our progress during the past century, LVP is not a benign procedure; complications, including hyponatremia, renal failure, severe infection, gastrointestinal and abdominal wall bleeding, and paracentesis-induced circulatory dysfunction (PICD), occur in 20% to 50% of patients. In an era when moderate obesity is the norm and morbid obesity is not unusual, it is not always easy to diagnose even a large peritoneal fluid collection on physical examination. The classic signs of bulging flank and distended abdomen, shifting dullness to percussion, a fluid wave, auscultatory percussion, and the “puddle sign” (periumbilical dullness to percussion after the patient has been on his or her hands and knees for several minutes) can be masked by the patient’s girth. Thankfully, ultrasonography, which can detect as little as 100 mL of fluid, has become the “gold standard” and is readily available to any radiologist. Having determined that the patient has ascites, does he or she need paracentesis for relief of symptoms? Paracentesis can often give temporary relief of chest pain, dyspnea, abdominal pain, and anorexia. The AASLD practice guideline suggests a single LVP of 4 to 6 L followed by dietary sodium restriction and diuretic therapy for tense ascites; for patients with refractory ascites, the AASLD practice guideline recommends serial LVP sessions as needed. These guidelines consider albumin infusion as optional for LVP of more than 5 L but do not recommend it for paracentesis of lesser volume.2 Patients with tense ascites are frequently treated by the removal of 7 to 10 L of ascitic fluid. Removal of such large volumes has been determined to cause PICD, a disorder characterized by marked activation of the renin–angiotensin axis secondary to the further increase of an already established arteriolar vasodilatation, which is a frequent and potentially harmful complication of paracentesis of greater than 6 L. Although PICD is clinically silent, it has been associated with a rapid recurrence of tense ascites and shorter survival times. The rate of fluid extraction, mechanical modifications (caused by abdominal decompression), and release of vasodilator molecules (e.g., nitric oxide) from the vascular endothelium are postulated to play a major role in development of PICD. The main feature of PICD is a marked activation of the renin–angiotensin and sympathetic nervous system without changes in plasma volume, heart rate, or hematocrit. Therefore, the rationale for using plasma expanders such as albumin after paracentesis is to maintain the circulatory status and to prevent the subsequent activation of vasoconstrictor systems. In a prospectively randomized study by Sola-Vera et al,4 the incidence of PICD was significantly higher (P = 0.03) in the saline group versus the albumin group receiving total paracentesis; however, no significant differences were found when less than 6 L of ascitic fluid was evacuated (6.7% vs. 5.6% in the saline and albumin groups, respectively). Complications other than PCID were almost twice as frequent in the saline group in this study. Albumin was better than saline in the prevention of PICD. The incidence of PICD in patients receiving saline in Sola-Vera et al’s study is similar to that reported in patients treated with dextran 70 (34%) or polygeline (38%); therefore, these are inferior substitutes for albumin.4 Disseminated intravascular coagulation (DIC) and an acute abdomen requiring surgery are the only absolute contraindications to paracentesis. Significant bleeding was below 0.3% in 5337 patients with coagulopathy or thrombocytopenia in two retrospective studies. In both studies, an increased bleeding rate was associated with significant renal failure.5,6 Prophylactic transfusion with fresh frozen plasma did not reduce the rate of bleeding in one of the studies.5 Ultrasound guidance is mandatory if the patient in question has had an LT or abdominal surgery, which increase the likelihood of bowel adhesions and a lead to high probability of collateral veins on the abdominal wall that must be avoided. The patient should be asked to empty his or her bladder before the procedure. Prophylactic antibiotics should be given to cover the most common organisms (Escherichia coli, Klebsiella pneumoniae, and pneumococci) pending culture of the causative agent if the polymorphonuclear count is above 250 per cubic millimeter. Overviews of the indications, contraindications, preprocedure testing, and patient management related to LVP are summarized in As is frequently the case in interventional radiology (IR) procedures, the materials used are not as important as how they are used. Paracentesis and thoracentesis kits are sold by many companies; some interventionalists prefer to use the centesis trocar supplied in these kits; others prefer a multi-sidehole catheter introduced by the Seldinger technique for LVP. The procedure can be performed with little or no sedation. Informed consent and aseptic technique are mandatory. Examination of the abdomen by ultrasonography immediately before paracentesis is recommended but not mandatory for all cases. The standard of care demands that preventive measures be followed in determining the needle-entry site for paracentesis. 1. When the midline approach is taken, the needle should not be entered cephalad to the umbilicus because the recanalized umbilical vein commonly underlies the abdominal wall in this location. Table 8.1 Large-Volume Paracentesis
Introduction
Large-Volume Paracentesis
Table 8.1.
Procedure
Diagnostic indications | Fluid evaluation to determine the etiology of new-onset ascites, suspected spontaneous or secondary bacterial peritonitis, detection of cancer cells |
Therapeutic indications | Chest pain, respiratory compromise, anorexia, abdominal pain or pressure (including abdominal compartment syndrome) secondary to ascites |
Absolute contraindications | Hyperfibrinolysis, disseminated intravascular coagulopathy, acute abdomen that requires surgery |
Relative contraindications | Abdominal wall cellulitis, intraabdominal adhesions, distended bowel or bladder, pregnancy |
Procedural complications | Hyponatremia, hyperkalemia, renal failure, severe infection, GI bleeding, intraperitoneal bleeding, SBP, PCID, shortened survival time |
Preprocedure Laboratory Tests16 | |
INR | Routinely recommended for patients with liver disease |
aPTT | Routinely recommended for patients receiving IV UFH |
Platelet count | Not routinely recommended |
Hematocrit | Not routinely recommended |
Preprocedure Patient Management16 | |
INR > 2.0 | Threshold for treatment (i.e., FFP, vitamin K) |
PTT and Hct | No consensus |
Platelet count < 50,000/μL | Transfusion recommended |
Clopidogrel | Withhold for 5 days before procedure |
Aspirin | Do not withhold |
LMWH | Withhold one dose before procedure |
aPTT: activated partial thromboplastin time; FFP: fresh-frozen plasma; GI: gastrointestinal; Hct: hematocrit; INR: international normalized ratio; IV: intravenous; LMWH: low-molecular-weight heparin; PICD: paracentesis-induced circulatory dysfunction; SBP: spontaneous bacterial peritonitis; UFH: unfractionated heparin.
