contact us

Use the form on the right to contact us.

You can edit the text in this area, and change where the contact form on the right submits to, by entering edit mode using the modes on the bottom right.

Chicago, IL


The Pericardiocentesis

John Sarwark

By John Sarwark, Edited by Andrew Ketterer

By John Sarwark, Edited by Andrew Ketterer

So let's say you have a patient--maybe they have a history of malignancy or renal failure or lupus or whatever--and the patient's having some trouble. Let's say they're dyspneic and hypotensive, and round out the classic Beck's Triad with some muffled heart sounds and distended neck veins (although, like most eponymous triads and pentads, only 33% of patients actually have all three).  You astutely throw your ultrasound on the chest, and you see something that looks like this:

Let's throw out the typical Procedures Club stem of "you're at a community hospital and there's no specialist around" blah blah blah and cut to the chase.  You're doing a pericardiocentesis today!

But before you do any poking, let's review the basics for a spell.  The pericardium is made of two layers, the visceral and parietal pericardia, with the potential pericardial space between them.  Physiologically, there should be about 20ish cc's of fluid there for shock absorbancy and lubrication.

Generally speaking, once you've got about 200 cc of fluid acutely in there, you start to see some systolic dysfunction.  Chronically, the space can accommodate quite well due to elasticity of the structures, so you could theoretically see a huge effusion with no symptoms as it has been there a while.  Here's a demonstration of how you can fall off the curve, based on animal studies:


Some other things to note: first off, are they stable?  If so, you might have enough time to treat what's causing the tamponade as opposed to the tap--i.e. if they're end stage renal, get them to HD.  If they're not stable, consider fluid boluses and pressors to buy you some time.  You do want to avoid positive pressure ventilation in these cases, however, as this may cause hemodynamic collapse by diminishing the preload in a very preload-dependent physiologic state.

Diastolic RV and RA collapse are both present here, consistent with tamponade physiology. 

Diastolic RV and RA collapse are both present here, consistent with tamponade physiology. 


Second, it’s important to be able to recognize cardiac tamponade when you see it. Classically, patients will exhibit Beck’s Triad (JVP, muffled heart sounds, and hypotension), or Beck’s Chronic Triad (JVP, muffled heart sounds, and ascites). Unfortunately, as we mentioned above only a minority of patients will exhibit all three. So the bottom line is that if you see someone who you think might be worth having their heart looked at, go ahead and slap an ultrasound probe on their chest. Things you’re looking for include:

  • Fluid around the heart (which will show up as a black stripe as above)
  • Diastolic collapse of the RA (this is both very sensitive and specific)
  • Early diastolic collapse of the RV free wall
  • LA collapse (this is highly specific)
  • A small, slitlike, hyperdynamic LV
  • Swinging to and fro of the heart within the pericardial fluid
  • Dilation of the IVC and hepatic veins 


In terms of preparation, you'll more or less need the following: an 18 G spinal needle, a 10 cc syringe, ECG wires with alligator clips, a guidewire, a pigtail catheter, a dilator, a stopcock, local anesthesia, sterile drape and wipes, and your trusty bedside ultrasound.  If you're in a pinch, just grab the 18G and the syringe.


There are a few ways to approach the pericardium. The standard approach is subxiphoid, with you standing on the patient's right. After prepping and injecting some local anesthesia, you'll need to identify the left xiphocostal angle, and slide down about 1 cm, no further than approximately 2 fingerbreadths.  This will be your entry point, where you'll come in at about 30 degrees, aimed at the midpoint of the left clavicle. The heart is pretty anterior, so any angle greater than 45 degrees is too much and makes you more likely to puncture the stomach or liver. Many sources cite the left AC joint as your target, but this is wrong and makes you more likely to miss. If your first advance fails to produce any fluid, withdraw the needle to just below the skin and redirect it a little more medially.


Sagittal cadaveric section revealing position of anterior nature of the heart. 


If you're doing it blind, cadaveric studies have shown that subxiphoid is probably the safest way to go, although it is not without complications. Here's the (unauthorized?) ATLS version for your perusal:


If you have an ultrasound, however, you'll want to try a different route. Here, you'll want to position yourself to the patient's left. The most common ultrasound-guided approach is apical: After finding the apex via palpation (or simply with the ultrasound), insert the needle approximately 1 cm lateral to this, with the point directed toward the patient’s right shoulder, and use the same insertion-aspiration technique as above, visualizing the advance of your needle point with the ultrasound. Make sure to advance over the cephalad portion of the rib inferior to your insertion point, avoiding the neurovascular bundle below the rib above. Direct visualization is necessary for this approach, since the lingula lies very close to the apex, meaning that you have a high risk of causing a pneumo.



Image compliments of ALiEM. 


A third option is the ultrasound-guided parasternal approach. Academic Life in Emergency Medicine has featured a fantastic primer on this. Basically, the technique is pretty similar to the apical approach: advance at a 45-degree angle toward a well-visualized pericardial fluid collection, taking care to avoid the neurovascular bundle on the underside of the rib cephalad to your point of insertion. This technique avoids the drawback of puncturing the lingula, and in fact has minimal risk of any lung injury whatsoever. However, care must be taken to avoid the internal mammary artery, which can be visualized with color Doppler as per ALiEM (and will make you look pretty slick by doing so). Make sure to mark the site of the IMA with a sterile pen before you insert your needle.

I couldn't find a YouTube demonstration of this approach (which is a bummer as it appears to be awesome), but here is a subxiphoid approach from Ultrasound Podcast:


As with any other tap, you'll slowly aspirate as you go in. NEVER move the needle in a side-to-side fashion, as you may lacerate the underlying epicardium. Once you're in, try to get a good 60 cc out. If you've opted to attach your ECG wire, you can see if you hit epicardium by watching the waveform. If you start seeing ST elevations, back up and watch for them to resolve. After that, wire it and place your pigtail for continued drainage in an ICU setting.

If you have a legit tamponade, even a small amount of aspiration will cause a significant improvement in your patient.  Remember, how they showed the procedure in Downtown Abbey?  Dr. Clarkson did that blind, in ONE POKE?  That's how they did it back in the day.

If it's a hemorrhagic tamponade, the traditional teaching is that you can distinguish it from ventricular blood by looking for clot--pericardial blood shouldn't have any due to the fibrinolytic activity.  However, this isn't totally reliable as you may just be tapping a brisk bleed.  You'll definitely want to check a post-procedure CXR to make sure you didn't put any air somewhere air isn't supposed to be.

As always, these writings are for EDUCATIONAL PURPOSES ONLY.  Would you want a family member to unnecessarily have a needle shoved in their pericardium by some overzealous scrub because "he always wanted to do one?"  Be smart!  Talk soon.

Learn more:

Diagnosing tamponade on ultrasound at Stanford ICU echocardiography 

Follow PROCEDURES CLUB on Twitter!


Previous posts


  • Roberts and Hedges' Clinical Procedures in Emergency Medicine, 6th Edition.  Chapter 16.
  • Loukas, M., Walters, A., Boon, J. M., Welch, T. P., Meiring, J. H., & Abrahams, P. H. (2012). Pericardiocentesis: A clinical anatomy review. Clinical Anatomy, 25(7), 872–881. doi:10.1002/ca.22032
  • MPH, P. C. I. M., & MD, S. L. (2013). Techniques and Procedures. Journal of Emergency Medicine, 44(3), 661–662. doi:10.1016/j.jemermed.2012.06.019
  • MD, A. N., & MD, D. M. (2013). American Journal of Emergency Medicine. American Journal of Emergency Medicine, 31(9), 1424.e5–1424.e9. doi:10.1016/j.ajem.2013.05.021
  • Life in the Fast Lane Blog:
  • Academic Life in Emergency Medicine Blog:
  • Ultrasound Podcast YouTube Page
  • Standford Echocardiography in the ICU