When Do You Need to Do Pericardiocentesis?

Morton Kern, MD, Clinical Editor; Chief of Medicine, Long Beach Veterans Administration Health Care System, Long Beach, California; Associate Chief Cardiology, Professor of Medicine, University of California Irvine, Orange, California

Morton Kern, MD, Clinical Editor; Chief of Medicine, Long Beach Veterans Administration Health Care System, Long Beach, California; Associate Chief Cardiology, Professor of Medicine, University of California Irvine, Orange, California

We were called to the intensive care unit to see a 52-year-old man with end-stage renal disease on weekly hemodialysis who had a large pericardial effusion and increasing dyspnea with activity. He had a respiratory rate of 18/minute, heart rate of 77/minute and blood pressure of 130/72 without pulsus paradoxus. There was mild elevation of jugular venous pressure, clear lungs, normal heart sounds, no ascites or edema. The chest x-ray showed minimal lung congestion and a large cardiac silhouette. An electrocardiogram (ECG) demonstrated sinus rhythm with left ventricular hypertrophy (LVH).  An echocardiogram showed a large pericardial effusion, but no evidence of right ventricular (RV) or right atrial (RA) diastolic collapse. Do we need to do a pericardiocentesis in this patient?  

Before answering this question, let’s review what causes a pericardial effusion, and how, when, and where to do pericardiocentesis.  Everyone working in the cath lab should become familiar with this lifesaving procedure for the known complication of cardiac perforation that may accompany coronary or structural heart interventions.

Why does a pericardial effusion occur?

The pericardium has 2 layers. The space between the outer (visceral) and inner (parietal, on the surface of the heart) pericardial membranes normally contains about 10-20 ml of fluid, and acts to lubricate the heart in motion and keep it roughly centered in the chest. Inflammation, injury, trauma, and infection can produce an outpouring of fluid or blood from the pericardial membranes. The normal pressure in the pericardial space is close to zero.  The rate of pericardial fluid accumulation and the compliance (stiff or expandable) of the pericardium result in a pressure buildup in the pericardial space (Figure 1).  When pericardial pressure equals or exceeds right atrial (RA) pressure, the RA chamber may collapse during inspiration. Similarly, when pericardial pressure exceeds RV diastolic pressure, inward collapse instead of the normal outward expansion of the RV is easily recognized on 2D echo.    

The symptoms of associated with a pericardial effusion include dyspnea, atypical chest discomfort, dizziness (from low blood pressure), palpitations, and peripheral edema (from high right heart pressures). Heart failure symptoms suggest a compromised cardiac output due to inability to fully fill the ventricles and hence, a reduced stroke volume. As the effusion becomes larger, impaired filling becomes worse and both cardiac output and blood pressure fall, eventually to levels that are insufficient to sustain life. This condition is called tamponade, a clinical, life-threatening diagnosis and is denoted clinically by tachycardia, hypotension, narrow pulse pressure, and elevated neck veins.  At this point in the course of pericardial effusion, pericardiocentesis is a lifesaving technique.

Not all pericardial effusions produce tamponade. Pericardiocentesis is also useful to make the correct diagnosis and direct management of less acute or chronic pericardial effusions without frank tamponade. Every cath lab team and operators have treated such patients, and should have sufficient training and experience to perform pericardiocentesis quickly and prevent the consequences of hypoperfusion due to tamponade.   

Where should we perform pericardiocentesis?

Pericardiocentesis can be performed at bedside on the wards, in the critical care unit, emergency department, operating room or in the cath lab.  My preference is always the cath lab, unless the patient is critically ill and hypotensive, and there is no time to call in the cath lab. I prefer the cath lab because we have all our equipment, a skilled team, good fluoroscopy, and a familiar working environment with room for echocardiographic imaging equipment. In addition, we can easily measure pericardial, RA and arterial pressures not only to confirm the diagnosis, but also demonstrate the resolution of the effusion and its abnormal hemodynamics. Although monitoring of pericardial pressure is not essential, it is important to document evidence of cardiac tamponade and to show resolution of pericardial pressure, and return of arterial pressure and cardiac output.

Steps of pericardiocentesis

1. Prepare your access site.  As with most procedures in the cath lab, there is more than one way to do it, but we prefer the commonly used method of the subxyphoid approach (Figure 2). Other labs and operators favor other access routes such as 3rd left intercostal space or apex, depending on their experience, and the location and volume of the pericardial effusion. The subxyphoid approach avoids the potential of coronary and internal thoracic artery laceration by the needle.  

2. Position the patient in the catheterization laboratory at a 30 to 45 degree, head-up angle to permit pericardial fluid to pool on the inferior surface of the heart. Palpate the subxyphoid process, about a finger-width below the edge of the rib. This location avoids difficulty in advancing the catheter through fibrous tissue closer to the lower part of the sternum itself.

3. Sterilely prepare the site and drape, covering everything but a small area around the subxyphoid process. Give local anesthesia, usually lidocaine over the anticipated needle puncture site.  

4. Insert the pericardial needle.  Advance the needle through the skin at first perpendicularly to chest, then angled lower to a plane nearly parallel with the floor, moving under the subxyphoid process toward the left shoulder. More lidocaine can be given gently through the pericardial needle as it is advanced. If the patient is obese, a longer needle and some force may be required to tip the syringe under the subxyphoid process toward the heart. To measure pericardial pressures, a stopcock on the needle is connected to a pressure line and transducer.  

For elective pericardiocentesis, right heart pressures are measured with a balloon-tipped catheter inserted through a large arm or leg vein. The RA, RV and pulmonary artery (PA) are measured to assess equalization of diastolic right-sided pressures (and later to document the change with fluid removal). Before entering the pericardium, the PA catheter is re-positioned in the right atrium for continuous monitoring of RA pressure during pericardial puncture and effusion drainage. A 5 French (F) sheath can be placed in a femoral artery for monitoring arterial pressure.

5. Advance the needle into the pericardial space. Passage of the needle through the skin may block the needle with subcutaneous tissue. Flush any tissue that may have accumulated during passage before entering the pericardium, a tough fibrous membrane. Use caution when advancing the needle through the diaphragm, as excessive forward pressure may result a sudden jump through the pericardium into a cardiac chamber.  

6. Confirm intra-pericardial position with hemodynamics or echo contrast imaging. As the needle advances, colored fluid or blood in the syringe signals likely entry into the pericardium (note that chronic effusions are often clear yellow, occasionally serosanguineous, or, less commonly, dark brown. Acute effusions resulting from trauma, cancer, or artery perforation are frankly bloody). Pericardial pressure can be seen on turning the stopcock and should be nearly the same as RA pressure in tamponade (Figure 3). Should the pressure show RV waveform, the needle has gone too far and should be pulled back into the pericardial space. Echocardiographic guidance is also confirmatory of correct positioning. Figure 4a shows echocardiographic images of a large pericardial effusion. After entering the pericardial space, an injection of 5 to 10 ml of agitated saline through the needle appears as microbubble contrast and confirms the intra-pericardial needle position (Figure 4b). If the needle tip is in a cardiac chamber (e.g., RV), the bubbles will be seen in the RV cavity and will be dispersed rapidly by RV ejection.

7. Place the pericardial drainage catheter. The needle is exchanged over a guide wire for a multiple side-hole catheter. Pericardial and RA pressures are measured again, the effusion is aspirated, and pressures are measured once more after the pericardial space is empty (Figure 4c). If the catheter will not drain or the exact position of the catheter is uncertain, a small amount of radiographic contrast medium may be injected to see the problem. Contrast medium pools in the dependent portion of the pericardial space, but rapidly washes out of a vascular space, if a cardiac chamber has been entered inadvertently. Bloody pericardial fluid may be due to chronic disease or acute trauma of the procedure. Chronic bloody effusions have a lower hematocrit value than intravascular blood and will not clot rapidly when placed in a red-top tube. 

8. Obtain serial echocardiograms before and after removal of the pericardial drainage catheter to confirm the absence of fluid re-accumulation. Remove the pericardial drain after 24-48 hours. Should fluid recur, consider a surgical pericardial window. 

A more complete explanation of pericardiocentesis can be found in the Cardiac Catheterization Handbook, 5th edition.1   

Does every pericardial effusion need to be drained?

Of course, the answer is no. Only those moderate to large effusions in which the cause of the effusion is critical to diagnosis (e.g., cancer) and remains unknown, or in which hemodynamic compromise is eminent or present, should be ‘tapped’ immediately. Stable patients with pericardial effusions in the setting of a viral illness, renal failure, or post-surgical or post-traumatic injury may be observed by serial echocardiography. The decision is more difficult in asymptomatic patients with pericardial effusions who have echocardiographic evidence of high pericardial pressure with diastolic chamber collapse. This finding is thought to be a precursor to hemodynamic collapse and is often an indication to perform pericardiocentesis. The decision to tap is easy in patients with classic findings of tamponade as noted above, with hypotension, tachycardia, and pulsus paradoxus.  

What did we do to our patient?

Our patient had dyspnea, but no tachycardia (77/minute), normal blood pressure (130/72, without pulsus paradoxus) and no peripheral sings of high, right-sided pressures. Although the echocardiogram showed a large pericardial effusion, there was no evidence of RV or RA diastolic collapse. Because the care team was concerned about the etiology of the effusion, we electively decided to perform pericardiocentesis for diagnosis. The hemodynamics showed elevated RA pressure, but the pericardial pressure was always well below the RA pressure, confirming the absence of tamponade or even pre-tamponade physiology (Figure 5). We would have moved rapidly if there were more clinical or echocardiographic signs of pre-tamponade physiology. The hemodynamics of the procedure suggested diastolic dysfunction or dialysis-related volume overload to explain dyspnea, but pericardial pressure was not a contributing factor.

As a wide variety of patients present with pericardial effusions to your institution, I hope you will consider this review and increase your perspective on when, how, and in whom to perform pericardiocentesis. 


  1. Kern MJ. (Ed.) The Cardiac Catheterization Handbook, 5th Edition. Elsevier: Philadelphia, Pennsylvania; 2011: 1-456.
  2. Holmes DR Jr, Nishimura R, Fountain R, Turi ZG. Iatrogenic pericardial effusion and tamponade in the percutaneous intracardiac intervention era. JACC Cardiovasc Interv. 2009 Aug; 2(8): 705-717. doi: 10.1016/j.jcin.2009.04.019.