Updated: May 16, 2022
  • Author: Gil Z Shlamovitz, MD, FACEP; Chief Editor: Vikram Kate, MBBS, MS, PhD, FACS, FACG, FRCS, FRCS(Edin), FRCS(Glasg), FFST(Ed), FIMSA, MAMS, MASCRS  more...
  • Print


Paracentesis is a procedure in which a needle or catheter is inserted into the peritoneal cavity to obtain ascitic fluid for diagnostic or therapeutic purposes. [1, 2, 3] Ascitic fluid may be used to help determine the etiology of ascites, as well as to evaluate for infection or presence of cancer. With regard to differentiation of transudate from exudate, the preferred means for characterizing ascites is the serum-ascitic albumin gradient (SAAG). [4]

The SAAG is calculated by subtracting the albumin concentration of the ascitic fluid from the albumin concentration of a serum specimen obtained on the same day. The SAAG correlates directly with portal pressure. Transudative ascites occurs when a patient's SAAG level is greater than or equal to 1.1 g/dL (portal hypertension). Exudative ascites occurs when patients have SAAG levels lower than 1.1 g/dL. (See the Ascites Albumin Gradient calculator.)

Causes of transudative ascites include the following:

Causes of exudative ascites include the following:

An alternative way of differentiating ascites due to portal hypertension from that due to other causes is to measure ascitic fluid viscosity with a cutoff of 1.65. [5] Ascitic fluid viscosity has also been demonstrated to predict renal impairment in hepatic patients at a cutoff of 1.35 and a long intensive care unit (ICU) stay at a cutoff of 1.995.

A noninvasive method of differentiating exudative from transudative ascites by using B-mode gray-scale ultrasound histogram analysis has been described and appears to be effective. [6]  In this method, the ascites–to–rectus abdominis muscle echogenicity ratio (ARAER) is measured. A value higher than 0.002 is regarded as exudative ascites, whereas a value lower than 0.002 is regarded as transudative ascites.

In patients with malignant ascites related to ovarian cancer, an ascites symptom mini-scale has been developed that serves as a simple and easy tool for assessment. [7]  This scale includes five parameters: shortness of breath, distended abdomen, reduced mobility, fatigue, and loss of appetite. It can be used to identify patients with mild symptoms who may benefit from paracentesis.

Spontaneous bacterial peritonitis

Infection of ascitic fluid without intra-abdominal infection usually occurs in patients with chronic liver disease due to translocation of enteric bacteria. Common pathogens include Escherichia coli, Klebsiella pneumoniae, enterococcal species, and Streptococcus pneumoniae. [8] Patients with renal failure who use abdominal peritoneal dialysis are also at increased risk, as are children with nephrosis or systemic lupus erythematosus (SLE). Anaerobic bacteria are not associated with spontaneous bacterial peritonitis (SBP).

An ascitic fluid polymorphonuclear leukocyte (PMN) count higher than 250/μL (neutrocytic ascites), with the percentage of PMNs in the fluid usually greater than 50%, is presumptive evidence of SBP. Patients whose ascitic fluid meets these criteria should be treated empirically, regardless of symptoms. Secondary bacterial peritonitis is defined as infected ascitic fluid associated with an intra-abdominal infection.

A report by Lutz et al demonstrated that the relative PMN count, as compared with the absolute PMN count, is a less expensive marker associated with bacterascites and can be used to predict future episodes of SBP. [9]  This may be useful for the purposes of risk stratification.



Diagnostic tap is used for the following:

  • New-onset ascites - Fluid evaluation helps to determine etiology, differentiate transudate versus exudate, detect the presence of cancerous cells, or address other considerations
  • Suspected SBP or secondary bacterial peritonitis
  • Refractory ascites [10]

Therapeutic tap is used for the following:

A report by Huang et al found that abdominal paracentesis drainage brought about clinical improvement in patients who had non-hypertriglyceridemia-induced severe acute pancreatitis with triglyceride elevation and pancreatitis-associated ascitic fluid. [11]

Large-volume paracentesis (LVP) is often required in patients with refractory ascites. A report by Bureau et al described the use of a low-flow pump system that moves the fluid from the abdominal cavity into the bladder, from which it is removed via micturition. [12] This was shown to improve patients' quality of life and reduced the need for repeated LVP. 

Definitive management for abdominal compartment syndrome (ACS) usually consists of performing emergency surgical decompression by means of a laparotomy. However, when a patient has tense ascites that is leading to ACS, LVP can successfully reduce the intra-abdominal pressure. [13]

It is well known that liver cirrhosis, when advanced, can cause moderate-to-severe ascites leading to impairment in the respiratory pattern. Wittmer et al reported beneficial effects of paracentesis on a series of 30 patients with liver cirrhosis and ascites. [14]  They observed that there was an increase in predominantly the abdominal breathing pattern and an improvement of ventilatory variables in these patients. Paracentesis also improved thoracoabdominal mobility, with a reduction of dyspnea and fatigue level, leading to increased peripheral oxygen saturation. 



An acute abdomen that requires surgery is an absolute contraindication.

Severe thrombocytopenia (platelet count < 20 × 103/μL) and coagulopathy (international normalized ratio [INR] >2.0) are relative contraindications.

Patients with an INR greater than 2.0 should receive fresh frozen plasma (FFP) prior to the procedure. One strategy is to infuse one unit of FFP before the procedure and then perform the procedure while the second unit is infusing.

Patients with a platelet count lower than 20 × 103/μL should receive an infusion of platelets before the procedure.

In patients without clinical evidence of active bleeding, routine laboratory tests such as prothrombin time (PT), activated partial thromboplastin time (aPTT), and platelet counts may not be needed before the procedure. [15] In these patients, pretreatment with FFP, platelets, or both before paracentesis is also probably not needed.

A study of 608 patients (72% with alcohol-related liver disease) found a low overall rate of complications that required transfusions (0.2%) and a higher incidence of significant hemoglobin drop among those with severe renal failure (creatinine > 6 mg/dL). [15]

A prospective study of 171 patients undergoing paracentesis found that "major" complications occurred in 1.6% of procedures and included five episodes of bleeding and three infections, resulting in death in two cases. Major complications were associated with therapeutic but not diagnostic procedures and tended to be more prevalent in patients with low platelet counts (< 50 × 109/L), patients who were Child-Pugh stage C, and patients with alcoholic cirrhosis. [16]

Other relative contraindications include the following:

  • Pregnancy
  • Distended urinary bladder
  • Abdominal-wall cellulitis
  • Distended bowel
  • Intra-abdominal adhesions

Technical Considerations

Best practices

Depending on the clinical situation, fluid may be sent for the following laboratory tests:

  • Gram stain - In a retrospective review of 796 peritoneal fluid samples, the evaluation of Gram stain results rarely provided clinically useful information for the detection of SBP [17]
  • Cell count (elevated counts may suggest infection)
  • Bacterial culture
  • Total protein level
  • Triglyceride levels (elevated in chylous ascites)
  • Bilirubin level (may be elevated in bowel perforation)
  • Glucose level
  • Albumin level, used in conjunction with serum albumin levels obtained the same day (used to calculate SAAG; see the Ascites Albumin Gradient calculator)
  • Amylase level (elevation suggests pancreatic source)
  • Lactate dehydrogenase (LDH) level
  • Cytology

After proper antiseptic preparation and local anesthesia, a diagnostic tap can be performed with a 10- to 20-mL syringe and an 18-gauge needle.

After proper antiseptic preparation and local anesthesia, a therapeutic tap can be performed with an intravenous (IV) catheter over the needle connected to drainage tubing.

In patients who are afebrile, alert, and have no other signs of bacterial peritonitis, ascitic fluid labs are often not necessary to rule out SBP. [18, 19]

To minimize the risk of persistent leak from the puncture site, use a small-gauge needle or take a "Z" track during insertion of the needle. (During removal of the needle, the subcutaneous tissue seals on itself.)

In a retrospective review of 796 peritoneal fluid samples, the evaluation of Gram stain results rarely provided clinically useful information for the detection of SBP. [17]

Dietary sodium restriction and diuretics do not often provide symptomatic relief of refractory ascites in patients in advanced stages of cancer. Although paracentesis does effectively drain ascitic fluid, the condition invariably recurs, and repeated procedures are necessary. A 2008 study reported that a permanent peritoneal catheter to drain abdominal fluid greatly reduced the symptoms of ascites in these patients and avoided the costs and complications of frequent paracentesis procedures. [20]

A meta-analysis suggested that the use of albumin in cirrhotic patients undergoing paracentesis reduces paracentesis-induced circulatory dysfunction and reduces death and renal impairment. [21]

The ascites index, which is estimated with basic ultrasound equipment, is defined as the sum total of ascites extent in all four external quadrants of the abdomen. It appears to be a promising tool for estimating and monitoring ascites following paracentesis as reported in ovarian hyperstimulation syndrome. [22]

Complication prevention

In cases with a persistent leak, a single skin suture might solve the problem. The application of an ostomy bag around the puncture site keeps the leak contained until it is eventually sealed off.

Postparacentesis hypotension is a delayed complication that may occur more than 12 hours after a procedure in which large volumes are taken off. Patients can be pretreated with a colloid solution, such as albumin, to decrease the frequency of this complication, though no difference in survival has been noted relative to other plasma expanders. [23]

In a systematic review and meta-analysis addressing the use of serum albumin treatment during paracentesis in patients with cirrhotic ascites, Shrestha et al found that albumin reduced the odds of paracentesis-induced circulatory dysfunction (by 60%) and hyponatremia. [24]  On subgroup analysis, however, albumin was better than other volume expanders but not as good as vasoconstrictor therapy. Albumin infusion provided no significant benefit with respect to mortality, readmission rate, recurrence of ascites, hepatic encephalopathy, or GI bleeding.

In a retrospective analysis of 49 patients undergoing transjugular intrahepatic portosystemic shunt (TIPS) placement, of whom 19 received LVP, Li et al found that vasopressor use and packed red blood cell (PRBC) transfusion were significantly higher in the LVP group than in patients who underwent small-volume paracentesis (SVP). [25]  Paracentesis volume was an independent predictor of the phenylephrine dose and of the need for crystalloid and colloid. If patients are managed carefully, LVP may be tolerated as well as SVP is.

Timing of paracentesis

Whereas studies have suggested that early paracentesis (< 12 hours from admission) improves short-term survival of hospitalized patients with SBP in comparison with delayed paracentesis, earlier administration of appropriate antibiotics in the early paracentesis group may have been a major contributing factor to these findings. [26, 27]

At this time and in the absence of data from prospective randomized controlled studies, the authors recommend early diagnostic or therapeutic paracentesis and early empiric antibiotic administration in patients with suspected SBP. When it appears that LVP is likely to be required, the authors suggest consideration of early SVP (ultrasonography [US], syringe, and needle technique) followed by delayed and planned LVP (during hospitalization) under appropriate monitoring and hemodynamic support to minimize the risk of circulatory dysfunction induced by LVP.

Ultrasound guidance

The process of peritoneal drainage can be carried out effectively and safely by employing a specific US-guided technique for the following [28] :

  • Confirm the presence of ascites
  • Quantify and qualify the ascites
  • Determine trajectory and depth
  • Avoid superficial blood vessels
  • Perform US-assisted paracentesis
  • Perform US-guided paracentesis
  • Check the position of the wire prior to dilating

Automated low-flow ascites pump

Paracentesis-induced circulatory dysfunction is one of the complications of moderate-volume paracentesis or LVP. The automated low-flow pump system for the treatment of refractory ascites (Alfapump) is an alternative for repeated large-volume paracentesis in patients with a contraindication for TIPS placement or liver transplantation. This pump is an implantable device that drains ascites directly into the urinary bladder in a regulated manner.

Wong et al evaluated safety, efficacy, and quality of life (QoL) in 30 patients in whom an automated low-flow ascites pump had been inserted. [29] They found that the pump removed a mean ascites volume of 230.6 ± 148.9 L/patient at 12 months and reduced the mean paracentesis frequency from 2.4 ± 1.4/patient/month before pump implantation to 0.2 ± 0.4/patient/month afterward. All surviving patients had improved QoL and a better biochemical index of nutritional status (prealbumin 87.8 ± 37.5 vs 102.9 ± 45.3 mg/L at 3 months). Complications included bacterial infections (15 events; 13 patients), electrolyte abnormalities (11 events; six patients), and renal complications (11 events; nine patients).

Other studies have also shown the benefits of using the automated low-flow ascites pump. [30]