WeiHou, MD, ArunJ. Sanyal, MBBS, MD*
Ascites is a common complication of cirrhosis. The development of ascites marks the
onset ofworsened prognosis and increased mortality inpatients with cirrhosis. Ascites
also causes considerable morbidity in the affected individual by producing abdominal
distension, respiratory distress, formation of hernias especially around the umbilicus,
worsening nutritional status, and increased susceptibility to infections. All of these
contribute to repeated hospitalizations in such patients and markedly impair their
quality of life. Appropriate management of ascites is thus an important pillar in the
care of a patient with cirrhosis. The current concepts about the pathophysiology, diag-
nosis, and management of ascites are reviewed in the following sections.
The main factor contributing to the development of ascites in a patient with cirrhosis is
the portal hypertension which results from increased intrahepatic resistance to blood
flow and is compounded by splanchnic vasodilatation as a result of local production of
vasodilators (Fig. 1).1–8Cirrhosis occurs as a consequence of chronic liver injury–
induced distortion of hepatic architecture and fibrosis. Increased resistance to portal
blood flow as a result of cirrhosis and vascular tone because of increased production
of vasoconstrictors, such as angiotensin, endothelin, cysteinyl-leukotrienes, and
thromboxane leads to gradual formation of portal hypertension, collateral vein circu-
lation, and shunting of blood to the systemic circulation. Splanchnic vasodilatation
develops as persistent portal hypertension results in local overproduction of vasodila-
tors such as nitric oxide (NO), calcitonin gene–related peptide, substance P, carbon
monoxide, and endogenous cannabinoids. Among these vasodilators, NO is a potent
This work is original and not under consideration elsewhere for publication. It was supported
in part by a grant from the NIH K24 DK 02755-09 to Dr. Sanyal.
Division of Gastroenterology, Hepatology and Nutrition, Department of Internal Medicine,
Virginia Commonwealth University School of Medicine, MCV Box 980341, Richmond, VA
* Corresponding author.
E-mail address: firstname.lastname@example.org (A.J. Sanyal).
? Cirrhosis ? Ascites ? Portal hypertension ? Refractory ascites
?Transjugular intrahepatic portosystemic shunts
? Spironolactone ? Furosemide
?Spontaneous bacterial peritonitis
Med Clin N Am 93 (2009) 801–817
0025-7125/09/$ – see front matter ª 2009 Elsevier Inc. All rights reserved.
and predominant vasodilator. Endothelial stretching and bacterial translocation are
responsible for the local overproduction of vasodilators and other cytokines.9,10
Recent data suggest that bacteria translocate to mesenteric lymph nodes in cirrhosis,
and consequently stimulation of cytokine production plays an important role in the
process of arterial vasodilatation.11–13
Splanchnic arteriolar vasodilation and consequent pooling of blood in the
splanchnic circulation causes a decrease in effective arterial blood volume and arterial
pressure. In response to this change in effective arterial blood volume and arterial
pressure, baroreceptor-mediated activation of the sympathetic nervous system
(SNS), renin-angiotensin-aldosterone system (RAAS), and antidiuretic hormone
(ADH) cause avid renal water and sodium retention in order to restore homeostasis.
Cirrhosis is also associated with increased sinusoidal pressure and decreased plasma
oncotic pressure. These combine to increase hepatic lymph formation. When the
capacity of the hepatic lymphatics to return hepatic lymph to the circulation is
exceeded, the excess lymph (composed of an ultrafiltrate of plasma containing the
retained sodium and water) spills into the peritoneal cavity, producing ascites.
The cardiac output and plasma volume increase in the early stages of liver cirrhosis,
contributes to a further fall in effective arterial blood volume and arterial pressure. This
causes marked activation of systemic vasoconstrictive mechanisms that particularly
affect the kidneys and decrease glomerular filtration and renal plasma flow. In its most
severe form, this leads to progressive renal failure, that is, hepatorenal syndrome.
The goals of the diagnostic assessment of a patient with ascites are to establish the
presence of ascites, determine its severity, determine its cause, and detect the
Fig. 1. The pathophysiology of cirrhosis and ascites is shown. Cirrhosis is associated with
splanchnic arterial vasodilation. This leads to a decrease in effective circulating volume
and a hyperdynamic circulation. The decrease in effective circulating volume causes activa-
tion of renal sodium and water retentive pathways (eg, RAAS, renal SNS, and ADH). The
resulting sodium and water retention leads to ascites as a result of spillage of excess sodium
and water from hepatic lymph into the peritoneal cavity. As the disease progresses,
a progressive decrease in effective circulating volume develops, causing severe renal vaso-
constriction and a decrease in glomerular filtration rate. The onset of cirrhotic cardiomyop-
athy accentuates this problem and tips the patient into hepatorenal syndrome. The
accompanying circulatory disturbance leads to organ failure and death. Sepsis is frequently
associated with this process.
Hou & Sanyal
presence of complications of ascites, which include spontaneous bacterial peritonitis
and renal failure. A good clinical assessment provides invaluable information about
A correct diagnosis of the cause of ascites is the essential first step to its successful
treatment. Cirrhosis accounts for approximately 85% of ascites in the United States,
whereas nonhepatic diseases cause most of the remaining cases (Box 1).14A history
of risk factors for liver diseases such as viral hepatitis, alcohol abuse, metabolic
syndrome, familial liver diseases, autoimmune disease, and so on should be obtained.
Cancer is the second most common cause of ascites. A history of cancer should lead
to exploring the possibility that ascites could be caused by peritoneal carcinomatosis.
Heart failure is the third common cause of ascites and a history of heart failure could
hint at a cardiogenic etiology as a potential cause of ascites. A history of tuberculosis,
kidney disease on dialysis, pancreatic disease, and so on are also relevant, and ques-
tions about these less common diseases should also be asked.
The physical examination should focus on stigmata of cirrhosis and signs suggest-
ing the presence of ascites. Stigmata of cirrhosis such as spider angioma and palmar
erythema may coexist with ascites. A full and bulging abdomen should lead to the
evaluation of ascites. Flank dullness on percussion is usually characteristic of ascites.
A positive shifting dullness usually suggests the presence of more than 1500 mL of
1. Portal hypertension
Presinusoidal causes, eg, portal vein thrombosis (usually ascites is mild if present at all)
Sinusoidal causes, eg, cirrhosis, vitamin A toxicity
Postsinusoidal causes, eg, venoocclusive disease, Budd-Chiari syndrome, constrictive
pericarditis, congestive heart failure
2. Neoplastic causes
3. Inflammatory causes
Infectious causes, eg, tuberculosis, Whipple disease
Chemical causes, eg, talc peritonitis
Immunologic disorders, eg, systemic lupus erythematosus, vasculitis
Allergic causes, eg, eosinophilic gastroenteritis
4. Miscellaneous causes
Ovarian hyperstimulation syndrome
Thoracic duct obstruction
Ascites: Diagnosis and Management
fluid.15An obese abdomen can masquerade as ascites, and an abdominal ultrasono-
gram may be required to establish the presence of ascites in an obese patient. An
abdominal ultrasonogram is usually performed in patients with ascites not only to
assess the presence of ascites and a mass, but also to investigate the hepatic echo-
genicity and vasculature.
Assessment of the Severity of Ascites
The International Ascites Club classifies ascites according to severity, complication,
and response to diuretic treatment. Ascites can be classified into grade 1 (mild), grade
2 (moderate), and grade 3 (large) according to severity; into uncomplicated according
to absence of complication; and into diuretic-resistant and diuretic-intractable
according to the response to diuretic treatment (Table 1).5,16,17
Laboratory Studies Including Ascites Fluid Analysis
History and physical examination are important first steps in establishing the diagnosis
of new-onset ascites, which should be further confirmed by an abdominal paracente-
sis and ultrasonography (Box 2). The presence of cirrhosis can be further assessed by
measuring tests of liver function such as the serum bilirubin, albumin, and international
normalized ratio (INR). These are often abnormal in patients with cirrhosis, although it
is possible to have ascites as a result of cirrhosis in the presence of near normal values
of these parameters. It is important to check renal function (serum creatinine) to estab-
lish a baseline and to determine if functional renal insufficiency is present. It is also
worth remembering that a serum creatinine of 1.5 mg/dL, which is often considered
to be near normal, represents considerable decrease in glomerular filtration in patients
with cirrhosis who have decreased muscle mass. The presence of an underlying hepa-
tocellular cancer should be sought with imaging studies and an alpha fetoprotein test.
Endoscopy is sometimes performed to look for varices as further corroborative
evidence for the presence of portal hypertension in cases where the diagnosis is
not clear-cut. Similarly, a liver biopsy is performed in selected patients with ascites
and liver disease of unknown etiology. Abdominal ultrasonography, CT, or MRI is
used to image the liver to screen for hepatocellular carcinoma, portal vein thrombosis,
and hepatic vein thrombosis.
Appropriate ascitic fluid analysis is probably the most efficient and effective method
of diagnosing the cause of ascites.15,18The left lower quadrant of the abdomen, 2
finger breadths cephalad and 2 finger breadths medial to the anterior superior iliac
crest, is the best location for paracentesis because it has thinner abdominal wall
and larger pool of fluid accumulation.19The prothrombin time is often prolonged
(approximately 71%) in patients with cirrhosis; however, the risk for bleeding is less
than 1% after paracentesis in these patients even without any interventions to correct
the coagulopathy.20The possibility for more serious complications such as hemoper-
itoneum and bowel perforation is remote, and they occur in less than 0.1% of
patients.21Coagulopathy should preclude paracentesis only when there is clinical
evidence of fibrinolysis or disseminated intravascular coagulation.20
In light of the presence of spontaneous bacterial peritonitis (SBP) in approximately
15% of hospitalized patients with cirrhosis and ascites, all such patients should be
screened for SBP at the time of admission to the hospital.22–25Ascitic fluid should
be analyzed in patients with new-onset ascites.2Ascitic fluid analysis to detect SBP
is required in all patients with any evidence of clinical deterioration such as fever,
abdominal pain, gastrointestinal bleeding, hepatic encephalopathy, hypotension, or
renal failure.26The SAAG has been proven superior to the total-protein-based
Hou & Sanyal
Classification ofascites according to severity, complication, and response to diuretic treatment
Ascites is only
Ascites that is not
Grade 2 (Moderate)
Grade 3 (Large)
Clinically marked or
preclude the use
of an effective
Ascites: Diagnosis and Management
exudate/transudate analysis in categorizing ascites in prospective studies and has
replaced the total-protein-based exudate/transudate concept in clinical practice.14,27
Calculation of SAAG is performed by measuring same-day albumin concentrations
of serum and ascitic fluid and then subtracting the ascitic fluid albumin value from the
serum albumin value. A SAAG value greater than or equal to 1.1 g/dL (11 g/L) predicts
ascites as a result of portal hypertension with approximately 97% accuracy.14A SAAG
greater than or equal to 1.1 g/dL may also be present in medical conditions such as
congestive heart failure, Budd-Chiari syndrome, and portal hypertension—a second
cause for ascites formation.14,17A SAAG less than 1.1 g/dL occurs commonly in peri-
toneal carcinomatosis, peritoneal tuberculosis, pancreatitis, serositis, and nephrotic
When there is no evidence of perforation of an intra-abdominal viscus or inflamma-
tion of intra-abdominal organs, an ascitic fluid neutrophil count greater than or equal to
250 cells/ mm3confirms the diagnosis of SBP.15,16,28The cell count is the most helpful
parameter in diagnosing SBP. A urine dipstick is a quick test to detect neutrophil in
ascites and may provide an early suspicion of SBP at bedside.29,30Gram stain of
ascitic fluid is not necessary because it is rarely helpful.31When culture of ascitic fluid
Evaluation ofa patient with cirrhosis andascites
Complete blood count, platelets
Evaluation of liver disease
Serum bilirubin, AST, ALT, alkaline phosphatase, serum albumin
Prothrombin time, INR
Tests to determine cause of liver disease, eg, hepatitis C antibody
Upper abdominal imaging by ultrasonogram/CT scana
Evaluation of renal functions
Serum creatinine and electrolytes
Urinalysis including microscopic examination
24-h urinary sodium and/or proteinb
Ascitic fluid analysis
Bacterial culture with bedside inoculation into blood culture bottle
Total protein and albumin
Glucose, lactate dehydrogenase, amylase, triglycerides, and cytology, if indicated according
to clinical situationc
aTo look for hepatocellular cancer or portal vein thrombosis.
bUsually done if urinalysis indicates proteinurea or if noncompliance with sodium restriction is
cUsually glucose, albumin, and protein are the only routinely performed tests initially. The
other tests are done if a diagnostic dilemma is present.
Hou & Sanyal
is performed, the fluid should be inoculated in blood culture bottles as opposed to
sterile culture vials. In sterile vials, growth of an organism is noted in about 50% of
cases whereas culture of ascitic fluid in blood culture bottles increases the probability
of identification of an organism to about 80%.32,33
Cell count and differential, albumin, total protein concentration, and SAAG are
tested in the initial screening of ascitic fluid if the ascites is believed to be likely uncom-
plicated on clinical grounds. Further testing is needed if the results of these tests are
abnormal. Cell count and differential are usually adequate in patients receiving serial
outpatient therapeutic paracentesis.15Lactate dehydrogenase, total protein, and
glucose may help differentiate spontaneous from secondary bacterial peritonitis.34
Ascitic fluid carcinoembryonic antigen (CEA) and alkaline phosphatase levels are
useful for the differentiation of primary from secondary bacterial peritonitis with intes-
tinal perforation. An ascitic fluid CEA greater than 5 ng/mL or ascitic fluid alkaline
phosphatase values greater than 240 IU/L suggest gut perforation into ascitic fluid.35
A high amylase level in ascitic fluid is diagnostic of pancreatic ascites. Ascitic fluid
amylase should be determined whenever there is an increased clinical suspicion for
pancreatic disease.36–38It is usually accompanied by increased total protein levels
and decreased SAAG.
Bloody ascites with red blood cell count greater than 50,000 cells/mm3occurs in
approximate 2% of patients with liver cirrhosis.39Hepatocellular carcinoma is the
cause of bloody ascites in about 30% of patients with cirrhosis. However, the cause
of bloody ascites cannot be found in about 50% of the cases. In patients with decom-
pensated cirrhosis, cancer antigen 125 (CA 125) is usually elevated in both blood and
ascites in proportion to the degree of ascites and does not necessarily indicate carci-
noma as the underlying cause of ascites.40–42
Ascitic fluid cytology is an expensive test and has a low yield if it is not used in selec-
tive patients. One study showed that only 7% ascitic fluid cytologies are positive.43In
peritoneal carcinomatosis, if adequate concentrated ascitic samples are sent and pro-
cessed promptly, it is highly positive. The sensitivity of cytology in detecting peritoneal
carcinomatosis is 96.7% if 3 samples are sent.43Breast, colon, gastric, or pancreatic
carcinoma is usually the underlying cause of peritoneal carcinomatosis. Mycobacteria
in ascitic fluid are difficult to detect. Smears for mycobacteria usually do not work, with
sensitivity approaching 0%; however, the sensitivity of ascitic fluid culture for myco-
bacteria is approximately 50%.44Patients with high pretest probability for tuberculous
peritonitis should be tested for mycobacteria on the first ascitic fluid sample.45Lapa-
roscopy with peritoneal biopsy and mycobacterial culture of tubercles are the most
rapid and accurate methods of diagnosing tuberculous peritonitis.
The fundamental goal of ascites management is to induce a negative sodium balance
(Box 3). Activation of sympathetic nervous system, RAAS, and ADH play an important
role in the pathophysiology of ascites formation. Physiologically, an upright position
activates these systems that theoretically worsen sodium and fluid retention and
decrease response to diuretics in patients with cirrhosis and ascites.46,47Moderate
physical activity may induce more profound effects on these systems.48,49These find-
ings support the traditional practice to recommend bed rest as part of the treatment
for ascites. However, bed rest is impractical, especially in patients with mild and
moderate ascites. Another concern is that bed rest may further decondition the
patient, weaken physical strength, and induce muscle atrophy in patients with
Ascites: Diagnosis and Management
cirrhosis and ascites. There are no controlled trials to support the theory of bed rest.
Therefore, bed rest is not generally recommended for the management of patients
with uncomplicated ascites.17
Dietary sodium restriction and oral diuretics are mainstays of treatment for patients
with cirrhosis and ascites. Stringent dietary sodium restriction mobilizes ascites in
patients with portal hypertension. Negative sodium balance leads to fluid and weight
loss in patients with cirrhosis and ascites. It is usually sodium restriction, not water
restriction, that induces weight loss, because water follows sodium passively.50Die-
tary sodium should be restricted to 2000 mg/d (88 mmol/d). Nonurinary excretion of
sodium is about 10 mmol/d. One of the goals of treatment is to achieve negative
sodium balance. In order to achieve negative sodium balance, urinary sodium excre-
tion should be greater than 78 mmol/d. If weight loss is not as expected, measurement
of urinary sodium excretion may be helpful.18Collection of 24-hour urine for sodium
measurement is cumbersome. Random urine ‘‘spot’’ sodium concentration is
a more quick and convenient method to determine urinary excretion with 97% accu-
racy.51Sodium restriction is effective in reducing the dose of diuretics, providing faster
solution for ascites, and a shorter hospital stay.52,53However, dietary sodium restric-
tion is successful in achieving a negative sodium balance in only about 10% to 15% of
patients with cirrhosis and ascites.54Compliance with sodium restriction is a practical
issue in daily management of patients with cirrhosis and ascites because most
patients are reluctant to go along with sodium restriction alone. There are no
controlled trials regarding fluid restriction. Most experts believe fluid restriction has
no role in the management of patients with uncomplicated ascites.15,17Water restric-
tion is recommended only when the serum sodium decreases to values below 130
1. General measures:
a. Sodium restriction
b. Maintain caloric intake at goal
c. Protein intake (1 g/kg/d unless patient is severely encephalopathic or catabolic)
d. Immunization: pneumococcal vaccine, influenza vaccine
2. Treatment of underlying liver disease:
a. Alcohol abstinence
b. Hepatitis B: antiviral therapy for E antigen positive subjects
c. Hemochromatosis: phlebotomy
d. Wilson disease: chelation therapy
3. Standard medical treatment (diuretics 1 paracentesis):
a. Distal tubule acting diuretics, eg, spironolactone
b. Loop-acting diuretics, eg, furosemide
c. Large volume paracentesis (>5 L)
4. Transjugular intrahepatic portasystemic shunts (TIPS)
5. Peritoneovenous shunts
6. Liver transplantation
Hou & Sanyal
Consideration should be given to treatment of the underlying cause of cirrhosis,
particularly for alcohol-related and hepatitis B–related cirrhosis. Abstinence can result
in dramatic improvement in liver function and even resolve ascites in the course of
a few months in patients with alcoholic hepatitis and cirrhosis. The benefits of absti-
nence are seen in patients with alcoholic liver disease and cirrhosis of varying
A nutritionist can play an important role in providing education regarding salt and
caloric intake. Protein-calorie malnutrition and weight loss are common among
patients with cirrhosis and ascites. Such patients often complain of dyspeptic symp-
toms such as early satiety, nausea, and postprandial fullness. A study has reported
significant reduction in median postprandial gastric volume and gastric accommoda-
tion; and paracentesis improves fasting gastric volume, tolerance to ingestion of
maximum volume, and caloric intake.57Decreased oral intake and absorption of nutri-
ents, increased energy expenditure, and altered fuel metabolism with a starvation
pattern of metabolism are the underlying mechanisms for malnutrition in patients
with cirrhosis and ascites.58–60Nutritional therapy with improved nutritional status
may reduce the occurrence of infection and perioperative morbidity. Although there
are no studies that show the value of nutritional status correction to improve ascites
management and hard outcomes such as mortality, absence of evidence does not
indicate the absence of an effect. One can imagine that good nutritional support
may be essential to make the patient a good candidate for liver transplant. Supple-
mental enteral nutrition isneeded in patients with severe malnutrition and may improve
liver function and hepatic encephalopathy.58–61
Specific therapies for ascites
Diuretics have been the mainstay of treatment for patients with cirrhosis and ascites
since they first became available. Patients with grade 1 ascites have mild ascites
that can only be detected by ultrasound. These patients should have a sodium-
restricted diet but do not require diuretics. Patients with grade 2 or higher severity
require diuretics to reduce edema and ascites. Activation of RAAS plays an important
role in the development of ascites. Spironolactone is the diuretic of choice because it
is an aldosterone antagonist that acts on the distal tubules in the kidney to increase
sodium excretion and conserve potassium. As a single agent, spironolactone has
been shown to be more efficacious than furosemide in a randomized clinical trial.62
However, it is used as a single agent mostly in patients with minimal fluid overload.63
Furosemide is a loop diuretic that causes marked natruresis and diuresis. It is less
efficacious than spironolactone as a single agent in the treatment of patients with
cirrhosis and ascites.62It is usually used as an adjunct to spironolactone treatment.
It should be used with caution because it can cause hyponatremia, hypokalemia,
and prerenal renal failure. Because of its good oral availability and intravenous admin-
istration–induced acute reduction in renal glomerular filtration rate, furosemide is
usually used asan oral agent.64,65Simultaneous use ofspironolactone and furosemide
increases the natriuretic effect and prevents hypokalemia.15Furosemide 40 mg and
spironolactone 100 mg daily are usually started as an initial dose in patients with
moderate to severe ascites. It usually takes 3 to 5 days for the diuretics to show their
maximal effects.66The doses of spironolactone and furosemide can be increased
simultaneously in a stepwise manner until the maximal doses of 400 mg spironolac-
tone and 160 mg furosemide every day are reached if desired weight loss and natri-
uresis are not attained.15,18
Dietary sodium restriction should always be implemented together with the use of
diuretics. The desired rate of daily weight loss depends upon the severity of edema.
Ascites: Diagnosis and Management
In patients with severe edema, diuretics can be given to patients with cirrhosis and
ascites without limitation of daily weight loss. Once ascites has resolved, daily
maximal weight loss of 0.5 kg is probably a reasonable goal.67This approach of
dual diuretics regimen in combination with dietary sodium restriction has been used
successfully in achieving improvement of ascites to an acceptable level in more
than 90% of patients in a large, multicenter, randomized, controlled clinical trial.68
Efforts should be made to avoid overdiuresis that can lead to decreased intravascular
volume, prerenal kidney impairment, hepatic encephalopathy, and hyponatremia.69
Serum creatinine greater than 2.0 mg/dL, or serum sodium less than 120 mmol/L indi-
cate that diuretics should be discontinued and alternative treatment considered.15The
hepatic encephalopathy associated with volume contraction is best treated with
albumin infusions. In patients with cirrhosis who have ascites, side effects of spirono-
lactone include hyperkalemia and decreased libido, impotence, and gynecomastia in
men and menstrual irregularity in women, as a result of its antiandrogenic activity.
Patients with organic renal disease may not tolerate spironolactone because of hyper-
kalemia, a serious complication that frequently limits its use in such a situation. Ami-
loride may be considered as an alternative in patients with tender gynecomastia, but it
was shown to be more expensive and less effective than spironolactone in a random-
ized controlled clinical study.65Tamoxifen at a dose of 20 mg twice daily has been re-
ported to be effective in managing the gynecomastia.63
Clonidine is a centrally acting alpha-2 agonist that has sympatholytic activity in
patients with cirrhosis.70–72Simultaneous administration of clonidine and spironolac-
tone has been shown in studies to increase natriuresis and body weight loss more effi-
ciently, to induce an earlier diuretic response, and fewer complications such as
hyperkalemia and renal impairment.73–75Diuretic use is one of the underlying causes
of muscle cramps. In patients with cirrhosis and ascites receiving diuretic therapy,
muscle cramps may require a reduction in diuretic dosage. Quinidine sulfate at
a dose of 400 mg daily or intravenous albumin at 25 g/wk reduces the frequency
and severity of diuretic-induced muscle cramps in patients with cirrhosis and
Diuretics alone may be inadequate in managing patients with large or refractory
ascites. Therapeutic paracentesis does not correct the underlying pathophysiological
process that results in ascites formation in patients with cirrhosis but usually relieves
symptoms caused by abdominal tension. In prospective studies the safety of a single
5 L or less paracentesis without postprocedure colloid infusion for intravascular
volume expansion has been shown.78,79Large volume paracentesis (LVP) can be per-
formed safely with the administration of intravenous albumin infusion (8 g/L of ascites
removed).80LVP with intravenous albumin infusion rapidly removed ascites, was more
effective in maintaining an ascites-free state, and was associated with fewer compli-
cations and shorter hospital stay when compared with diuretic therapy.69,81Total par-
acentesis followed by postparacentesis volume expansion with albumin is as safe as
serial LVP alone.78,80
A complication of LVP is the development of postparacentesis circulatory dysfunc-
tion.82This is characterized by worsening vasodilation, hyponatremia, activation of
sodium-retentive hormones and sometimes azotemia. It is a marker for poorer
outcomes and can be partially prevented by the use of albumin given intravenously
at a dose of 6 to 8 g per liter of ascites removed.83Although albumin was formerly
thought to be a simple volume expander, it is now thought that albumin may have
Hou & Sanyal
important effects on the endothelial dysfunction and circulatory disturbances associ-
ated with cirrhosis.84
Transjugular intrahepatic portosystemic shunts
These procedures decompress the portal vein by providing a low-resistance channel
between the intrahepatic portion of the portal vein and the hepatic vein. They increase
venous return to the heart and improve the effective circulating volume. This leads to
improved renal perfusion and a decrease in renal tubular sodium reabsorption,
thereby causing a natriuresis. Transjugular intrahepatic portosystemic shunts (TIPS)
effectively removes ascites and maintains an ascites-free state. However, these bene-
fits are offset by an increased incidence of hepatic encephalopathy.85,86The incidence
of hospitalization and overall survival are not impacted by TIPS. The outcomes after
TIPS depend on the model end-stage liver disease (MELD) score.87The ideal candi-
date for TIPS is one who has relatively preserved hepatic synthetic function and renal
function and who is free of encephalopathy.
A peritoneovenous shunt has been used for the treatment of refractory ascites. It also
increases the central volume and induces diuresis. Although as effective as repeated
LVP, it does not improve survival. Enthusiasm for this procedure has waned because
of the increased risk of complications, such as disseminated intravascular coagula-
tion, infection, and occlusion of the subclavian vein and superior vena cava, which
can preclude a liver transplant.
Management of uncomplicated ascites
This is usually managed with a combination of general measures and the use of
diuretics and sodium restriction. Initially when patients present with moderate or
severe ascites, a LVP is done to make the patients comfortable and then sodium
restriction and diuretics are used to maintain an ascites-free state. Although this is
effective in most patients, some patients progress to the refractory ascites state.
Only liver transplant provides long-term survival in patients with end-stage liver
disease, and in the absence of an obvious contraindication to transplantation, such
patients should be referred expeditiously to a transplant center.
Management of refractory ascites
Diuretic refractory ascites (see Table 1), that is, ascites that does not respond to
maximal diuretic doses, is associated with increasing systemic vasodilation and acti-
vation of systemic sodium and water retentive mechanisms. Increasing vasodilation
decreases effective circulating volume and renal perfusion. This decreases delivery
of the drugs to their site of action (the distal tubule for spironolactone and the lumen
of the loop of Henle for furosemide) and causes diuretic resistance.
Repeated LVP or TIP are the most commonly used modalities for the treatment of
refractory ascites. Although they relieve ascites immediately, they are associated
with recurrence of ascites in most patients and do not improve survival. Repeated
LVP is associated with discomfort, protein loss and malnutrition, the need for repeated
hospitalization, and health care resource utilization. It is, however, a relatively inexpen-
TIPS is substantially superior to LVP for long-term control of ascites (Fig. 2).86This
does not translate to improved survival and the decrease in ascites-related health care
resource utilization is offset by increased encephalopathy-related morbidity. Also, for
the same survival outcomes, TIPS is less cost-effective than LVP. Hyperbilirubinemia,
severe hypoprothrombinemia, and renal failure are risk factors associated with a poor
Ascites: Diagnosis and Management
outcome after TIPS.87Although TIPS is better than LVP for control of ascites in
general,88the outcomes of TIPS for refractory ascites are best in those who havefailed
repeated LVP and who have relatively preserved liver and renal function (ie, a creati-
nine less than 1.5 mg/dL, INR less than 1.5, and bilirubin less than 2 mg/dL). Ideally, it
should be used as a bridge to liver transplant.
Hyponatremia is another major management challenge in patients with cirrhosis and
ascites. It occurs in about 50% of patients with cirrhosis and is associated with
increased mortality (Fig. 3).89It is generally managed by volume restriction. There
are anecdotal reports of improvement with albumin infusions as well. The role of aqua-
retic drugs in the management of ascites-related hyponatremia is under investigation.
Spontaneous bacterial peritonitis
SBP should always be considered in the differential diagnosis when a patient with
cirrhosis and ascites develops fever, abdominal pain, altered mental status, variceal
hemorrhage, or azotemia. It is diagnosed by a diagnostic paracentesis and treated
with a third-generation cephalosporin.90A 5-day course has been found to be as
effective as a 10-day course for uncomplicated SBP.91Typically, treatment is
switched to oral quinolone therapy after 3 to 5 days of intravenous antibiotics. SBP
recurs frequently and secondary prophylaxis with oral quinolones is effective in
Fig. 2. A meta-analysis of TIPS versus LVP shows TIPS to be superior to LVP for control of
ascites as shown by the odds ratio in individual studies. (Data from Albillos A, Ban ˜ares R,
Gonz? alez M, et al. A meta-analysis of transjugular intrahepatic portosystemic shunt versus
paracentesis for refractory ascites. J Hepatol 2005;43:990–6.)
Fig. 3. A pathophysiology-based approach to the treatment of hyponatremia associated
Hou & Sanyal
preventing recurrence and is therefore recommended.92Primary prophylaxis for SBP
should be considered in those with low-protein ascites and severe hepatic synthetic
1. Hernandez-Guerra M, Garcia-Pagan JC, Bosch J. Increased hepatic resistance:
a new target in the pharmacologic therapy of portal hypertension. J Clin Gastro-
2. Gines P, Cardenas A, Arroyo V, et al. Management of cirrhosis and ascites. N Engl
J Med 2004;350:1646–54.
3. Moller S, Bendtsen F, Henriksen JH. Pathophysiological basis of pharmaco-
therapy in the hepatorenal syndrome. Scand J Gastroenterol 2005;40:491–500.
4. Schrier RW, Arroyo V, Bernardi M, et al. Peripheral arterial vasodilation hypothesis:
a proposal for the initiation of renal sodium and water retention in cirrhosis. Hep-
5. Arroyo V, Gine `s P, Gerbes AL, et al. Definition and diagnostic criteria of refractory
ascites and hepatorenal syndrome in cirrhosis. Hepatology 1996;23:164–76.
6. Oben JA, Diehl AM. Sympathetic nervous system regulation of liver repair. Anat
Rec A Discov Mol Cell Evol Biol 2004;280:874–83.
7. Arroyo V, Gines P. Mechanism of sodium retention and ascites formation in
cirrhosis. J Hepatol 1993;17:24–8.
8. Jimenez-Saenz M, Soria IC, Bernardez JR, et al. Renal sodium retention in portal
hypertension and hepatorenal reflex: from practice to science. Hepatology 2003;
9. Iwakiri Y, Groszmann R. The hyperdynamic circulation of chronic liver diseases:
from the patient to the molecule. Hepatology 2006;43:S121–31.
10. Iwakiri Y, Groszmann R. Vascular endothelial dysfunction in cirrhosis. J Hepatol
11. Wiest R, Das S, Cadelina G. Bacterial translocation in cirrhotic rats stimulates
eNOS-derived NO production and impairs mesenteric vascular contractility.
J Clin Invest 1999;104:1223–33.
12. Wiest R, Garcia-Tsao G. Bacterial translocation (BT) in cirrhosis. Hepatology
13. Riordan SM, Williams R. The intestinal flora and bacterial infection in cirrhosis.
J Hepatol 2006;45:744–57.
14. Runyon BA, Montano AA, Akriviadis EA, et al. The serum-ascites albumin
gradient is superior to the exudate-transudate concept in the differential diag-
nosis of ascites. Ann Intern Med 1992;117:215–20.
15. Runyon BA. Management of adult patients with ascites due to cirrhosis. Hepatol-
16. Moore KP, Wong F, Gine `s P. The management of ascites in cirrhosis: report on the
consensus conference of the International Ascites Club. Hepatology 2003;38:
17. Moore KP, Aithal GP. Guidelines on the management of ascites in cirrhosis. Gut
18. Runyon BA. Care of patient with ascites. N Engl J Med 1994;330:337–42.
19. Sakai H, Mendler MH, Runyon BA. The left lower quadrant is the best site for par-
acentesis: an ultrasound evaluation. Hepatology 2002;36:525A [abstract].
20. Runyon BA. Paracentesis of ascetic fluid: a safe procedure. Arch Intern Med
Ascites: Diagnosis and Management
21. Webster ST, Brown KL, Lucey MR, et al. Hemorrhagic complications of large
volume abdominal paracentesis. Am J Gastroenterol 1996;92:366–8.
22. Caly WR, Strauss E. A prospective study of bacterial infections in patients with
cirrhosis. J Hepatol 1993;18:353–8.
23. Bac D-J, Siersema PD, Mulder PGH, et al. Spontaneous bacterial peritonitis:
outcome and predictive factors. Eur J Gastroenterol Hepatol 1993;5:635–40.
24. Rimola A, Garcia-Tsao G, Navasa M, et al. Diagnosis, treatment and prophylaxis
of spontaneous bacterial peritonitis: a consensus document. J Hepatol 2000;32:
25. Pinzello G, Simonetti RG, Craxi A, et al. Spontaneous bacterial peritonitis:
a prospective investigation in predominantly nonalcoholic patients. Hepatology
26. Gine `s P, C? ardenas A. The management of ascites and hyponatremia in cirrhosis.
Semin Liver Dis 2008;28:43–58.
27. Hooefs JC. Serum protein concentration and portal pressure determine the as-
citic fluid protein concentration in patients with chronic liver disease. J Lab Clin
28. Ghassemi S, Garcia-Tsao G. Prevention and treatment of infections in patients
with cirrhosis. Best Pract Res Clin Gastroenterol 2007;21:77–93.
29. Castellote J, Lopez C, Gornals J, et al. Rapid diagnosis of spontaneous bacterial
peritonitis by use of reagent strip. Hepatology 2003;37:893–6.
30. Runyon BA. Strips and tubes: refining the diagnosis of spontaneous bacterial
peritonitis. Hepatology 2003;37:745–7.
31. Runyon BA, Canwati HN, Akriviadis EA. Optimization of ascetic fluid culture tech-
nique. Gastroenterology 1988;95:1351–5.
32. Runyon BA, Antillon MR, Akriviadis EA, et al. Bedside inoculation of blood culture
bottles is superior to delayed inoculation in the detection of spontaneous bacte-
rial peritonitis. J Clin Microbiol 1990;28:2811–2.
33. Castellote J, Xiol X, Verdaguer R, et al. Comparison of two ascetic fluid culture
methods in cirrhotic patients with spontaneous bacterial peritonitis. Am J Gastro-
34. Akriviadis EA, Runyon BA. The value of an algorithm in differentiating spon-
taneous from secondary bacterial peritonitis. Gastroenterology 1990;98:
35. Wu SS, Lin OS, Chen Y-Y, et al. Ascitic fluid carcinoembryonic antigen and alka-
line phosphatase levels for the differentiation of primary from secondary bacterial
peritonitis with intestinal perforation. J Hepatol 2001;34:215–21.
36. Runyon BA. Amylase levels in ascetic fluid. J Clin Gastroenterol 1987;9:172–4.
37. Polak M, Mattosinho Francis LC. Chronic pancreatitis with massive ascites.
38. Schindler SC, Schaefer JW, Hull D, et al. Chronic pancreatic ascites. Gastroenter-
39. DeSitter L, Rector WG Jr. The significance of bloody ascites in patients with
cirrhosis. Am J Gastroenterol 1984;95:1351–5.
40. Molina R, Filella X, Bruix J, et al. Cancer antigen 125 in serum and ascitic fluid of
patients with liver diseases. Clin Chem 1991;37:1379–83.
41. Zuckerman E, Lanir A, Sabo E, et al. Cancer antigen 125: a sensitive marker of
ascites in patients with liver cirrhosis. Am J Gastroenterol 1999;94:1613–8.
42. Sari R, Yildirim B, Sevinc A, et al. Sensitity of cancer antigen 125 in patients with
liver cirrhosis in the presence of ascites [Erratum in Am J Gastroenterol 2001; 96:
1319]. Am J Gastroenterol 2001;96:253–4.
Hou & Sanyal
43. Runyon BA, Hoefs JC, Morgan TR. Ascitic fluid analysis in malignancy-related
ascites. Hepatology 1988;8:1104–9.
44. Hillebrand DJ, Runyon BA, Yasmineh WG, et al. Ascitic fluid adenosine deami-
nase insensitivity in detecting tuberculous peritonitis in the United States. Hepa-
45. Cappell MS, Shetty V. A multicenter, case-controlled study of the clinical presenta-
tion and etiology of ascites and of the safety and efficacy of diagnostic abdominal
paracentesis in HIV seropositive patients. Am J Gastroenterol 1994;89:2172–7.
46. Gines P, Fernandez-Esparrach G, Arroyo V, et al. Pathogenesis of ascites in
cirrhosis. Semin Liver Dis 1997;17:175–89.
47. Ring-Larsen H, Henriksen JH, Wilken C, et al. Diuretic treatment in decompen-
sated cirrhosis and congestive heart failure: effect of posture. BMJ 1986;292:
48. Salo J, Gines A, Anibarro L, et al. Effect of upright posture and physical exercise
on endogenous neurohumoral systems in cirrhotic patients with sodium retention
and normal supine plasma renin, aldosterone, and norepinephrine levels. Hepa-
49. Salo J, Guevara M, Fernandez-Esparrach G, et al. Impairment of renal function
during moderate physical exercise in cirrhotic patients with ascites: relationship
with the activity of neurohormonal systems. Hepatology 1997;25:1338–42.
50. Eisenmenger WJ, Ahrens EH, Blondheim SH, et al. The effect of rigid sodium
restriction in patients with cirrhosis of the liver and ascites. J Lab Clin Med
51. Stiehm AJ, Mendler MH, Runyon BA. Detection of diuretic-resistance or diuretic-
sensitivity by the spot urine Na/K ratio in 729 specimens from cirrhotics with
ascites: approximately 90% accuracy as compared to 24-hr urine Na excretion.
Hepatology 2002;36:222A [abstract].
52. Descos L, Gauthier A, Levy VG, et al. Comparison of six treatments of ascites in
patients with liver cirrhosis. Hepatogastroenterology 1983;30:15–20.
53. Gauthier A, Levy VG, Quinton A, et al. Salt or no salt in the treatment of cirrhotic
ascites: a randomised study. Gut 1986;27:705–9.
54. Gerbes AL. Medical treatment of ascites in cirrhosis. J Hepatol 1993;17:S4–9.
55. Runyon BA. Ascites and spontaneous bacterial peritonitis. In: Feldman M,
Friedman LS, Sleisenger MH, editors. Sleisenger and Fordtran’s gastrointestinal
and liver disease. 7th edition. Philadelphia: Saunders; 2002. p. 1517–42.
56. Veldt BJ, Laine F, Guillogomarc’h A, et al. Indication of liver transplantation in
severe alcoholic liver cirrhosis: quantitative evaluation and optimal timing. J Hep-
57. Aqel BA, Scolapio JS, Dickson RC, et al. Contribution of ascites to impaired
gastric function and nutritional intake in patients with cirrhosis and ascites. Clin
Gastroenterol Hepatol 2005;3:1095–100.
58. Henkel AS, Buchman AL. Nutritional support in patients with chronic liver
disease. Nat Clin Pract Gastroenterol Hepatol 2006;3:202–9.
59. McCullough AJ. Nutrition and malnutrition in liver disease. In: Wolfe MM,
Davis GL, Farraye F, et al, editors. Therapy of digestive disorders. 2nd edition.
Philadelphia: Saunders; 2006. p. 67–83.
60. Kearns PJ, Young H, Garcia G. Accelerated improvement of alcoholic liver
disease with enteral nutrition. Gastroenterology 1992;102:200–5.
61. Cabre ` E, Gonzalez-Hux F, Abad-Lacruz A. Effect of total enteral nutrition on the
short-term outcome of severely malnourished cirrhotics: a randomized controlled
trial. Gastroenterology 1990;98:715–20.
Ascites: Diagnosis and Management
62. Perez-Ayuso RM, Arroyo V, Planas R, et al. Randomized comparative study of effi-
cacy of furosemide vs. spironolactone in nonazotemic cirrhosis with ascites.
63. Sungaila I, Bartle WR, Walker SE, et al. Spironolactone pharmacokinetics and
pharmacodynamics in patients with cirrhotic ascites. Gastroenterology 1992;
64. Daskalopoulos G, Laffi G, Morgan T, et al. Immediate effects of furosemide on
renal hemodynamics in chronic liver disease with ascites. Gastroenterology
65. Angeli P, Pria MD, De Bei E, et al. Randomized clinical study of the efficacy of
amiloride and potassium canrenoate in nonazotemic cirrhotic patients with
ascites. Hepatology 1994;19:72–9.
66. Karim A. Spironolactone metabolism in man revisited. In: Brunner HR, editor.
Contemporary trends in diuretic therapy. Amsterdam: Excerpta Medica; 1986.
67. Pockros PJ, Reynolds TB. Rapid diuresis in patients with ascites from chronic
liver disease: the importance of peripheral edema. Gastroenterology 1986;90:
68. Stanley MM, Ochi S, Lee KK, et al. Peritoneovenous shunting as compared with
medical treatment in patients with alcoholic cirrhosis and massive ascites. N Engl
J Med 1989;321:1632–8.
69. Gines P, Arroyo V, Quintero E, et al. Comparison of paracentesis and diuretics in
the treatment of cirrhotics with tense ascites: results of a randomized study.
70. Moreau R, Lec SS, Hadengue A, et al. Hemodynamic effects of a clonidine-
induced decrease in sympathetic tone in patients with cirrhosis. Hepatology
71. Esler M, Dudley F, Jennings G, et al. Increased sympathetic nervous activity and
the effects of its inhibition with clonidine in alcoholic cirrhosis. Ann Intern Med
72. Roulot D, Moreau R, Gaudin C, et al. Long term sympathetic and hemodynamic
responses to clonidine in patients with cirrhosis. Gastroenterology 1992;102:
73. Lenaerts A, Van Cauter J, Moukhaiber H, et al. Treatment of refractory ascites with
clonidine and spironolactone. Gastroenterol Clin Biol 1997;21:524–5.
74. Lenaerts A, Codden T, Van Cauter J, et al. Interest of the association clonidine-
spironolactone in cirrhotic patients with ascites and activation of sympathetic
nervous system. Acta Gastroenterol Belg 2002;65:1–5.
75. Lenaerts A, Codden T, Meunier JD, et al. Effects of clonidine on diuretic response
in ascitic patients with cirrhosis and activation of sympathetic nervous system.
76. Lee FY, Lee SD, Hwang SJ, et al. A randomized controlled trial of quinidine in the
treatment of cirrhotic patients with muscle cramps. J Hepatol 1991;12:236–40.
77. Angeli P, Albino G, Carraro P, et al. Cirrhosis and muscle cramps: evidence of
a causal relationship. Hepatology 1996;23:264–73.
78. Peltekian KM, Wong F, Liu PP, et al. Cardiovascular, renal and neurohumoral
responses to single large-volume paracentesis in cirrhotic patients with
diuretic-resistant ascites. Am J Gastroenterol 1997;92:394–9.
79. Runyon BA. Patient selection is important in studying the impact of large-volume
paracentesis on intravascular volume. Am J Gastroenterol 1997;92:371–3.
Hou & Sanyal
80. Tito L, Gines P, Arroyo V, et al. Total paracentesis associated with intravenous
albumin management of patients with cirrhosis and ascites. Gastroenterology
81. Quintero E, Gines P, Arroyo V, et al. Paracentesis versus diuretics in the treatment
of cirrhotics with tense ascites. Lancet 1985;16:611–2.
82. Ruiz-Del-Arbol L, Monescillo A, Jimenez W, et al. Paracentesis-induced circula-
tory dysfunction: mechanism and effect on hepatic hemodynamics in cirrhosis.
83. Gines P, Tito L, Arroyo V, et al. Randomized comparative study of therapeutic par-
acentesis with and without intravenous albumin in cirrhosis. Gastroenterology
84. Gines A, Fernandez-Esparrach G, Monescillo A, et al. Randomized trial
comparing albumin, dextran 70, and polygeline in cirrhotic patients with ascites
treated by paracentesis. Gastroenterology 1996;111:1002–10.
85. Wong F, Sniderman K, Liu P, et al. The mechanism of the initial natriuresis after
transjugular intrahepatic portosystemic shunt. Gastroenterology 1997;112:
86. Sanyal AJ, Genning C, Reddy KR, et al. The North American Study for the Treat-
ment of Refractory Ascites. Gastroenterology 2003;124:634–41.
87. Malinchoc M, Kamath PS, Gordon FD, et al. A model to predict poor survival in
patients undergoing transjugular intrahepatic portosystemic shunts. Hepatology
88. Albillos A, Banares R, Gonzalez M, et al. A meta-analysis of transjugular intrahe-
patic portosystemic shunt versus paracentesis for refractory ascites. J Hepatol
89. Londono MC, Guevara M, Rimola A, et al. Hyponatremia impairs early posttrans-
plantation outcome in patients with cirrhosis undergoing liver transplantation.
90. Garcia-Tsao G. Bacterial infections in cirrhosis. Can J Gastroenterol 2004;18:
91. Runyon BA, McHutchison JG, Antillon MR, et al. Short-course versus long-course
antibiotic treatment of spontaneous bacterial peritonitis. A randomized controlled
study of 100 patients. Gastroenterology 1991;100:1737–42.
92. Tito L, Rimola A, Gines P, et al. Recurrence of spontaneous bacterial peritonitis in
cirrhosis: frequency and predictive factors. Hepatology 1988;8:27–31.
93. Fernandez J, Navasa M, Planas R, et al. Primary prophylaxis of spontaneous
bacterial peritonitis delays hepatorenal syndrome and improves survival in
cirrhosis. Gastroenterology 2007;133:818–24.
Ascites: Diagnosis and Management