Paracentesis is a procedure in which excess fluid in the abdomen is sampled by aspiration through a needle. The fluid may be called ascites fluid, abdominal fluid, or peritoneal fluid.
Paracentesis is commonly performed to identify the cause of newly diagnosed ascites (excess fluid in the abdominal cavity); to diagnose changes in the condition of a patient already known to have ascites; and to relieve pressure from severe distention due to increased fluid in the abdomen. A sample of the fluid withdrawn from the abdominal cavity is nearly always sent for laboratory analysis to determine the presence or absence of infection, and/or to learn more about the cause of ascites if necessary. Ascites forms for a variety of reasons, including infection, diseases of various organs, and conditions which result in abnormal blood flow. The most common cause in the United States is alcoholic cirrhosis.
Ascites is difficult to diagnose by physical exam, although with experience health care practitioners can note "shifting dullness" by percussion. Generally at least 17 oz (500 mL) of fluid must accumulate before the effusion is detected by x ray, and 3.2 pt (1500 mL) before ascites is easily detected on physical exam. Ultrasound may be necessary to differentiate ascites from obesity and other reasons for abdominal distention. Ultrasound may even be used to guide the needle for paracentesis. When performing this procedure, the physician should observe universal precautions for the prevention of transmission of bloodborne pathogens.
Consent should be obtained for the procedure after discussion of the possible complications (discussed below). The area beneath the umbilicus is cleansed with betadine or other antibacterial solution, and local anesthetic administered. A long thin needle or trochar with a stylet is inserted about 2 in (5 cm) below the umbilicus, and the appropriate amount of fluid withdrawn. Usually a syringe is used, but for very large amounts of ascites, polyethylene tubing may be attached to vacuum bottles and the excess fluid aspirated. A minimum of 1 oz (30 mL) of fluid should be collected by sterile technique in two or three sterile syringes. One portion should be transferred
Cirrhosis of the liver and malignant abdominal masses are the two most common causes of ascites. Cirrhosis is usually associated with a transudative fluid, a fluid of low cellularity and protein, while malignancy causes an exudative (inflammatory) fluid of high cellularity and protein. Transudative fluids result from changes in blood flow, and are typically seen in persons with cirrhosis, congestive heart failure, and a few other conditions that disrupt normal hemodyndamics. An explanation of ascites formation in cirrhosis serves well to explain some principles common to transudative fluid formation. Blood entering the portal vein from the intestines passes through the liver on its way back to the heart. When progressive disease such as alcohol damage or hepatitis destroys enough liver tissue, the scarring which results compresses the hepatic sinusoids and vessels and restricts the blood flow. The blood bypasses the liver and enters the splenic, gastric, and esophageal veins causing very high hydrostatic pressure. This pressure causes fluid to escape the vessels and enter the abdominal cavity. Slowly the fluid accumulates in the areas with the lowest pressure and the greatest capacity. The free space around abdominal organs receives most of it. This space is called the peritoneal space because it is enclosed by a thin membrance called the peritoneum. The peritoneum wraps around nearly every organ in the abdomen, and lines the entire abdominal cavity, providing many folds and spaces in which fluid can gather. Normally, only 1–1.7 oz (30–50 mL) of fluid is found in the peritoneal cavity. The fluid itself is essentially an ultrafiltrate of plasma. Any condition that causes an increase in peritoneal fluid is called an effusion or ascites. Kidney disease can contribute to this process, since the kidneys have a critical role in fluid balance. Nephrotic syndrome in particular is associated with ascites formation. In this condition the kidneys lose large amounts of protein into the urine causing a drop in plasma oncotic pressure. Since proteins hold fluid in the vascular bed, loss of protein (albumin) causes fluid to enter the tissue spaces. Heart failure also can cause ascites, because decreased cardiac output causes blood to accumulate in the return circulation. The increased venous pressure results in fluid leaking from the circulatory system. First edema is noticed in the legs, due to the effect of gravity, then in ascites formation in the abdomen.
Malignancy, infection, pancreatitis, bowel obstruction, and several other conditions produce an exudative effusion. These conditions cause inflammation that results in increased blood vessel permeability. The fluid that accumulates typically contains white blood cells and if cancer is the cause, malignant cells from the tissue of origin. Malignancy may result from cancerous transformation of the cells that line the peritoneum, called mesothelial cells. Mesotheliomas may difficult to distinguish from reactive mesothelial cells that occur whenever the lining of the abdomen is traumatized. The two most common metastatic cancers invading the abdomen are ovarian and breast cancer, but lymphoma, leukemia, lung, and many other cancers can also infiltrate the abdomen. Bacterial peritonitis is an infection of the peritoneum, and is a life-threatening cause of exudative ascites. It can result from intestinal perforation, leakage through a deseased bowel wall, ruptured appendix or gall bladder, or septicemia (infection in the blood). Inflammation of the abdominal wall can also result from blunt trauma, pancreatitis, intestinal obstruction, and other conditions.
Normal ascites fluid is clear and straw colored. Turbid fluid occurs in bacterial peritonitis, malignancy, and pancreatitis. Green fluid occurs when bile is present. This can be caused by a ruptured bowel or perforated bile duct. Bloody fluid occurs in trauma, malignancy, and pancreatitis. Milky fluid contains chyle from the intestinal lymphatics and occurs when lymphatic vessels rupture.
The WBC count is performed using a hemacytometer. Normal fluid has a very low WBC count (less than 300 per microliter) and does not have to be diluted. Counts above 1000 indicate an exudative process. The differential is performed on a cytocentrifuged sample to concentrate the cells. Macrophages predominate in normal fluid and together with mesothelial cells account for about 70% of the nucleated cell population. Lymphocytes are normally less than 20% and neutrophils less than 10% of nucleated cells. Neutrophils accounting for 50% or 500 per microliter are most often associated with bacterial peritonitis. Lymphocytes will predominate in lymphoma, nephrotic syndrome, and congestive heart failure and may be abundant along with macrophages in tuberculosis. Red cell counts are also performed manually. Red cells often enter the fluid during sample collection, a process referred to as a traumatic tap. In this case, the red count will be low, the supernatant fluid will be pale yellow (normal), and the fluid will clear as more is collected. In the absence of a traumatic tap, red blood cells are most often encountered in malignancy and trauma. It is especially important to examine the fluid for the presence
Ascites—Abnormal quantity of peritoneal fluid, which is basically an ultrafiltrate of plasma.
Edema—Fluids that have shifted outside of the circulatory system and are temporarily trapped in soft tissues.
Gram stain—A common laboratory test in which a specimen on a glass slide is subjected to a series of stains and rinses to visualize micro-organisms.
Lymphocyte—A specific type of white blood cell (leukocyte) involved in fighting atypical, fungal and viral infections.
Neutrophil—A specific type of white blood cell (leukocyte) involved in fighting bacterial infections. Also called a polymorphonuclear leukocyte.
of malignant cells. As mentioned, metastatic cancer cells from ovarian and breast cancer are the most commonly seen infiltrates. Malignant mesothelial cells are difficult to distinguish from reactive mesothelial cells. Cytology should be evaluated with both Wright and Papanicilaou stains. Cytochemical tests and flow cytometry may be needed to identify malignant mesothelial cells, leukemic blasts (immature white cells), and lymphoma cells.
Chemical tests are performed on ascites fluid by the same methods used for plasma. Total protein, lactate dehydrogenase (LD), and glucose levels should be measured and compared to blood levels. Fluid to serum total protein and LD ratios are used to help distinguish exudative from transudative fluids. The serum albumin minus the fluid albumin is now considered the most sensitive single test to distinguish cirrhosis from malignancy as causes of ascites. Most transudative fluids are associated with cirrhosis and have a difference above 1.1. Most exudative fluids result from malignancy and have a difference less than 1.1. The fluid glucose is normally the same as the plasma glucose. Distinctly lower levels are seen in bacterial peritonitis, peritoneal tuberculosis, and malignancy. Lactate dehydrogenase is increased in bacterial peritonitis and malignant diseases. A fluid:serum ratio of 0.6 or higher has a sensitivity of about 80% in identifying exudative fluids. Amylase is very useful in diagnosing exudates caused by pancreatitis. Levels are usually in excess of three times the upper limit of normal. Fluid amylase testing can detect pancreatitis in approximately 90% of cases, and is also positive in the majority of persons with bowel obstruction, proliferation and intestinal cancer. Alkaline phosphatase is elevated in exudates associated with bowel injury, obstruction, and some malignancies such as hepatoma.
Tumor markers may be useful to help distinguish the tissue of origin and to increase the sensitivity of cancer detection. Both carcinoembryonic antigen levels and CA 125 levels in abdominal fluid have been found elevated in some persons with malignant infiltration of the abdomen but negative initial cytology. Creatinine may be measured when it is suspected that inadvertent puncture of the urinary bladder occurred during paracentesis. Creatinine in urine is about 100 times higher than in ascites fluid.
Bacterial cultures are usually performed on ascites fluid, but recovery of organisms is low when the fluid is frankly transudative. Gram stain detects about 25% of cases of bacterial peritonitis. The sensitivity can be increased by fluorescent microscopy using acridine orange stain. Cultures are positive in about 75% of cases that are eventually documented as infections. Detection of bacterial peritonitis is more sensitive when blood culture bottles containing tryptic soy broth are inoculated immediately after collection of the fluid rather than plating the fluid after transport to the lab. Regardless of the media used, cultures should be performed under both aerobic and anaerobic conditions. Spontaneous bacterial peritonitis which usually occurs in cirrhosis following sepsis typically grows a single organism, usually E. coli or Streptococcus pneumoniae. Peritonitis resulting from bowel sources usually grows several different intestinal organisms.
A hematocrit, prothrombin time, and platelet count should be obtained within 48 hours of paracentesis. This will identify which patients may be at risk for bleeding complications, and provide a baseline hematocrit to estimate blood loss should bleeding occur. In addition, blood should be collected for glucose, total protein, lactate dehydrogenase, and albumin at the time of paracentesis. These results are compared to those of ascites fluid as a diagnostic aid. Abdominal girth and vital signs should be documented. The patient should be asked to empty his or her bladder, and should be warned about very brief discomfort as the needle goes through the peritoneum, in spite of local anesthesia. If very large amounts of peritoneal fluid are to be removed, it may need to be done very slowly to avoid large fluid shifts and a rapid fall in blood pressure. In extreme cases, a central venous pressure
Vital signs are documented several times, perhaps even hourly for several hours if a large volume has been removed. The site of needle puncture is covered with a simple sterile dressing, or closed with a stitch if a trochar was used, and the dressing observed for possible continued leakage or bleeding.
Serious intra-abdominal bleeding is possible, although not very frequent. Puncture of the bladder or bowel are also possible. If good sterile technique is not used, infection could be introduced into the abdomen, resulting in peritonitis.
Results of laboratory tests on ascites fluid are dependent upon the method of analysis used. Most studies of normal persons are performed with very small sample sizes, and cell counts are performed manually. This results in greater interlaboratory variation in normal ranges than usually is seen for measurements performed on blood. Representative values for commonly measured analytes are shown below:
A physician collects the ascites fluid using sterile technique. The physician is usually assisted by a nurse. Nursing staff are responsible for documenting the patient's status before and after the procedure; educating and preparing the patient for the procedure; and observing for complications. Samples must be clearly labeled and sent to the laboratory. Clinical laboratory scientists/medical technologists perform blood counts, biochemical, and microbiological tests. A histologic technician prepares and stains slides for cytological review by a pathologist.
Glickman, Robert M. "Abdominal Swelling and Ascites." In Harrison's Principles of Internal Medicine, edited by Anthony S. Fauci. New York: McGraw-Hill, 1998, pp.256-257.
Malarkey, Louise M., and Mary Ellen McMorrow. Nurse's Manual of Laboratory Tests and Diagnostic Procedures, 2nd ed. Philadelphia: W. B. Saunders Company, 2000, pp.457-461.
Tierney, Lawrence M., Stephen J. McPhee, and Maxine A. Papadakis. Current Medical Diagnosis and Treatment 2001. New York: Lange Medical Books/McGraw-Hill, 2001, pp. 578-580.
Erika J. Norris