Real-Time Image-Guided Pericardial Drain Placement

This brief review discusses the evolution of image-guided pericardial drainage with a specific emphasis on the real-time visualization of needle progress.

Unguided (blind) percutaneous pericardial drainage was associated with exceedingly high rates of morbidity and mortality during its years of use before the 1980s. In one study that reviewed reports of cases performed between 1976 and 1981, the mortality among 341 patients was 8.5%.1 With this technique, an electrocardiogram with a precordial lead wire attached to the needle can be used to indicate needle contact with the myocardium;2 however, this technique would not demonstrate the distribution of the effusion, thus hindering the operator's ability to choose an appropriate skin entry site. Distribution of the effusion is often affected by its size. Small pericardial effusions are often located posteriorly at or below the atrioventricular groove. With moderate effusions, a uniform distribution is seen, and in large effusions, more fluid is seen apically, posteromedially, laterally, and anteriorly.3

Cardiac ultrasound (also called echocardiography), which was developed in the 1970s, allows for the identification of the size and distribution of pericardial effusions. Echocardiography-targeted technique was then developed to allow the operator to select the optimal needle entry site at a point on the chest wall where the largest pericardial fluid accumulation is closest to the skin.4 At the selected site, the needle is inserted while continuous aspiration is maintained; insertion is then stopped at the first site of fluid aspiration. However, the advancing needle is not visualized in real time on the monitor. Nevertheless, in large pericardial effusions, this technique was reported to have high procedural success rates and a low incidence of minor (3.5%) and major (1.2%) complications. As a result, continuous visualization of the needle was considered unnecessary.4

Echocardiographic visualization of the needle was reported as early as 1978.3 In a report on 11 patients with cardiac tamponade, real-time continuous visualization of the needle insertion using parasternal medial-to-lateral in-plane access was successful in draining the pericardial effusion within 4 to 6 minutes without complications.5 In spite of such reports and the routine use of echocardiography to evaluate effusions, needle insertion continues to largely be performed blindly.6 As an adjunct to nonvisualization of needle entry into the pericardial space, several techniques have been developed to identify and prevent entry into cardiac chambers before subsequent tract dilation and catheter insertion. Administering agitated saline into the pericardial space and using a digital pressure transducer connected directly to the needle have been reported to confirm pericardial entry of the needle tip without actual visualization of the needle.7,8 It has also been reported that visualization of the needle is not always certain during real-time guidance, with failure rates as high as 56% reported during earlier practice; however, these numbers do improve with experience.9

Real-time visualization of the needle progression can be improved by the choice of ultrasound probe. Curved general ultrasound probes with frequencies ranging from 2 to 5 MHz provide the benefit of wider near field of view (similar to linear probes) and deeper penetration (similar to vector echocardiographic probes) at the expense of reduced line density with increasing distance from the transducer. Unlike a vector probe, in which near field conspicuousness is restricted, a curved or linear probe offers continuous needle tip visibility from the skin to the pericardial space. A linear probe is used to access anterior pericardial effusions seen closer to the probe, whereas a curved probe can offer access to posterolaterally located effusions seen further from the probe. The long axis of these probes is positioned between or below the ribs to avoid artifacts caused by the ribs.

In a "free-hand" approach to real-time guidance, years of practice may be necessary to perfect the hand-eye coordination needed to control the ultrasound probe with the nondominant hand and insert the needle using the dominant hand while making fine adjustments to ensure that the needle tip and shaft are continuously visualized on the monitor during the procedure. Certain practitioners, such as interventional radiologists, commonly use such skills to access collections and lesions, but this largely remains a novel technique among cardiologists. To overcome some of these limitations, a multiangle bracket mounted on the ultrasound probe can be used to support the needle. This ensures that the needle trajectory always remains within the image. After the needle angle is adjusted to ensure that the tip will enter the pericardial space at the desired site within the field of view, the needle is locked to the probe in the selected angle. While the nondominant hand is held steady to display the effusion, the dominant hand pushes the needle into the effusion. This probe-mounted needle technique was found to be effective in a series of 161 pericardiocentesis procedures among 141 patients admitted to 1 of 3 centers between 1993 to 2015; the success rate was very high (>99%), and no cases of cardiac laceration were reported.10

Ultrasound guidance can be limited by the narrow near field visibility offered by echocardiographic vector probes and further restricted by the presence of ribs, lungs, and artifacts.11 Hyperinflated lungs in patients with chronic obstructive pulmonary disease and increased subcutaneous and mediastinal fat in obese patients can also complicate ultrasound guidance. In patients recovering from a recent open heart surgery, limited patient positioning options, the presence of chest wall bandages, and the presence of mediastinal hematoma and gas can restrict evaluation and obscure the needle. Sonolucent fluid that is easily seen in spaces adjacent to the heart (e.g., pleural effusions and ascites) can be mistaken for pericardial effusion, and loculated pericardial effusion can be difficult to identify.12 Additionally, posteriorly and laterally located effusions cannot be accessed using transthoracic ultrasound guidance; one-third of postoperative pericardial effusion cases located posterolaterally thus cannot be drained using ultrasound guidance.13

In such cases, computed tomography (CT) provides a wide field of view, generates images at high spatial resolution, and can be used to identify a safe access window to reach the pericardial effusion without direct contact with the chest wall, unlike transthoracic echocardiography. CT can visualize anatomy beyond air-filled and bony structures and can differentiate mediastinal anatomy, especially in a postoperative setting. In one study, this CT-targeted technique was used to select the optimal needle entry site in 319 pericardiocentesis cases, resulting in a very high success rate (98.4%) and low complication rates (major, 0.3%; minor, 6.9%); however, needle insertion was not monitored with CT.14 To provide real-time guidance similar to that provided with ultrasound (i.e., allowing the operator to visualize and track needle insertion and adjust the technique as and when needed), CT fluoroscopy (CTF) can be used.

With continuous CTF, CT scanner technology and image reconstruction algorithms allow for the acquisition and display of multiple cross-sectional images continuously in real time on a monitor next to the examination table during a procedure. Although standard protective gear (lead apron, thyroid shield, lead goggles) offers protection from the scattered radiation that is seen during such procedures, excessive radiation to the operator's hand (which is exposed directly to the CT beam) may require the operator to use needle holders.15 However, such needle holders offered limited tactile sensation. Several techniques have been suggested to reduce radiation dose during CT-guided procedures.16 In our practice, continuous CTF is seldom used to drain collections. Instead, intermittent CTF is performed. We use CT across the cardiac silhouette for preprocedural planning and switch to intermittent CTF during the procedure. Once the needle has been inserted to 1 to 2 cm at the chosen needle entry site, a single trigger on the acquisition pedal activated by the operator in the room or the technologist outside the room generates 3 contiguous axial images centered on the needle entry site. Upon confirmation of satisfactory trajectory, the needle is advanced incrementally with periodic intermittent CTF to enable the fine adjustments required to ensure safe entry into the pericardial space (Figure 1). After a 21G needle is inserted into the space, a wire is advanced into the largest pocket of the effusion and confirmed with CTF; once the tract is dilated, a pigtail catheter (6-8F) is placed over a wire. A final helical diagnostic CT scan is taken across the cardiac silhouette to assess for complications such as pneumothorax and hematoma, to ensure satisfactory positioning of the pigtail tip of the catheter, and to document reduction in the effusion.

CT-guided pericardial drainage is effective. In retrospective studies (with at least 20 patients each) reported in the last decade, a total of 195 CT-guided pericardial drainage procedures performed in 5 different countries had technical success rates higher than 95%, minor complication rates ranging from 4.6% to 5.1%, major complication rates lower than 2.4%, and no mortality cases.11,17-20 In addition, reimbursement for CT-guided pericardial drainage is similar to that for echocardiography-guided procedures, and both methods offer up to 89% savings over the cost of surgical pericardial window.11

CT offers many advantages. Unlike the dominance of left parasternal and subxiphoid accesses in ultrasound-guided procedures, CT guidance offers a range of additional options. In studies, approximately 25% to 46% of CT-guided accesses were obtained through an intercostal approach, 12% to 26% were through a right parasternal approach, and 15% were through a left subcostal approach.14,17 CT offers improved needle identification versus ultrasound, with CTF-based procedure times approaching those of ultrasound-based techniques.21 CT-guided pericardial drainage also allows for concomitant drainage of other collections. In a study of 36 patients with symptomatic pericardial effusions who underwent CT-guided pericardial drainage, 21% of the patients also required pleural drainage, which was accomplished at the same time.11 In addition, CT facilitates better detection and delineation of complications. For instance, when a postprocedural pneumothorax seen on CT is large, it can be drained immediately. CT results can also offer reassurance regarding minor complications such as small asymptomatic pneumothoraces, which are typically managed conservatively.11

Although ultrasound offers the advantage of portability and the potential for bedside intervention (especially in unstable patients), ultrasound also requires contact and manipulation of the probe with the chest wall, which may be poorly tolerated in a patient recovering from cardiac surgery.13 CT requires planning and mobilization of a patient to a dedicated scanner, but the scan may be better tolerated by patients. In my practice, procedures are frequently performed using both ultrasound and CT guidance. The room setup is shown in Figure 2. When feasible, real-time ultrasound guidance is used to visualize needle insertion continuously; subsequent wire and catheter placements and evaluation for complications are performed using CT. When ultrasound guidance is not feasible, intermittent CTF is used during needle insertion.

In summary, continuous visualization of needle insertion into the pericardial space using real-time ultrasound guidance is a useful technique that should be familiar to imaging specialists tasked with percutaneously draining pericardial effusions. When a safe window cannot be visualized on ultrasound, CT-guided pericardial drainage should be considered before surgical intervention.

Figure 1: Images from a 64-year-old man with increasing pericardial effusion after David procedure.

Figure 1
A: CT fluoroscopy (CTF) image demonstrates the appropriate skin entry site of the access needle, but the trajectory of the needle, if continued, would likely fail to reach the pericardial space.
B: CTF image confirms an appropriate course after the trajectory of the access needle was changed at the same skin site.
C: CTF image confirms needle tip entry into the pericardial space.
D: CTF image demonstrates the wire in the posterior pericardial effusion.
E: CT image reveals resolution of pericardial effusion after pigtail (red arrow) catheter is inserted and 220 mL of serosanguineous fluid is removed.

Figure 2: Procedure room setup. When available and feasible, ultrasound is used to visualize needle insertion continuously, and CT is used to confirm wire and catheter placements.

Figure 2


I thank Megan Griffiths for her help with revising the text.


  1. Callahan JA, Seward JB, Nishimura RA, et al. Two-dimensional echocardiographically guided pericardiocentesis: experience in 117 consecutive patients. Am J Cardiol 1985;55:476-9.
  2. Bishop LJ Jr., Estes EH Jr., McIntosh HD. The electrocardiogram as a safeguard in pericardiocentesis. J Am Med Assoc 1956;162:264-5.
  3. Martin RP, Rakowski H, French J, Popp RL. Localization of pericardial effusion with wide angle phased array echocardiography. Am J Cardiol 1978;42:904-12.
  4. Tsang TS, Enriquez-Sarano M, Freeman WK, et al. Consecutive 1127 therapeutic echocardiographically guided pericardiocenteses: clinical profile, practice patterns, and outcomes spanning 21 years. Mayo Clin Proc 2002;77:429-36.
  5. Osman A, Wan Chuan T, Ab Rahaman J, Via G, Tavazzi G. Ultrasound-guided pericardiocentesis: a novel parasternal approach. Eur J Emerg Med 2017. [Epub ahead of print]
  6. Gluer R, Murdoch D, Haggani HM, Scalia GM, Walters DL. Pericardiocentesis - how to do it. Heart Lung Circ 2015;24:621-5.
  7. Ainsworth CD, Salehian O. Echo-guided pericardiocentesis: let the bubbles show the way. Circulation 2011;123:e210-1.
  8. Visweswaran GK, Lightfoot J, Gilchrist IC. Novel use of disposable digital pressure transducer to increase the safety of pericardiocentesis. Catheter Cardiovasc Interv 2013;81:E68-71.
  9. Chandraratna PA, Reid Cl, Nimalasuriya A, Kawanishi D, Rahimtoola SH. Application of 2-dimensional contrast studies during pericardiocentesis. Am J Cardiol 1983;1120-2.
  10. Maggiolini S, Gentile G, Farina A, et al. Safety, efficacy, and complications of pericardiocentesis by real-time echo-monitored procedure. Am J Cardiol 2016;117:1369-74.
  11. Palmer SL, Kelly PD, Schenkel FA, Barr ML. CT-guided tube pericardiostomy: a safe and effective technique in the management of postsurgical pericardial effusion. AJR Am J Roentgenol 2009;193:W314-20.
  12. D'Cruz IA, Constantine A. Problems and pitfalls in the echocardiographic assessment of pericardial effusion. Echocardiography 1993;10:151-66.
  13. Duvernoy O, Larsson SG, Persson K, Thurén J, Wikström G. Pericardial effusion and pericardial compartments after open heart surgery. An analysis by computed tomography and echocardiography. Acta Radiol 1990;31:41-6.
  14. Klein SV, Afridi H, Agarwal D, Coughlin BF, Schielke LH. CT directed diagnostic and therapeutic pericardiocentesis: 8-year experience at a single institution. Emerg Radiol 2005;11:353-63.
  15. Froelich JJ, Wagner HJ. CT-fluoroscopy: Tool or gimmick? Cardiovasc Intervent Radiol 2001;24:297-305.
  16. Sarti M, Brehmer WP, Gay SB. Low-dose techniques in CT-guided interventions. Radiographics 2012;32:1109-19.
  17. David S, Mansoubi H, Hoey E, Agrawal B, Tasker A. CT guided pericardiocentesis: A UK experience. ECR 2010. Poster No. C-0494.
  18. Eichler K, Zangos S, Thalhammer A, et al. CT-guided pericardiocenteses: clinical profile, practice patterns and clinical outcome. Eur J Radiol 2010;75:28-31.
  19. Syed F, Slaughter R, Tjahjadi C, Hansen M, Raffel O. Computed tomography guided pericardiocentesis in adults: a single centre experience of 44 procedures. Heart Lung Circ Abstracts S165 2011;20:S165.
  20. Neves D, Silva G, Morais G, et al. Computed tomography-guided pericardiocentesis - A single-center experience. Rev Port Cardiol 2016;35:285-90.
  21. Sheafor DH, Paulson EK, Kliewer MA, DeLong DM, Nelson RC. Comparison of sonographic and CT guidance techniques: does CT fluoroscopy decrease procedure time? AJR Am J Roentgenol 2000;174:939-42.

Clinical Topics: Cardiac Surgery, Invasive Cardiovascular Angiography and Intervention, Noninvasive Imaging, Pericardial Disease, Interventions and Imaging, Computed Tomography, Echocardiography/Ultrasound, Nuclear Imaging

Keywords: Pericardiocentesis, Pericardial Effusion, Cardiac Tamponade, Thoracic Wall, Pneumothorax, Transducers, Pressure, Ascites, Thyroid Gland, Dilatation, Echocardiography, Pericardium, Electrocardiography, Tomography, X-Ray Computed, Fluoroscopy, Myocardium, Pleural Effusion, Pulmonary Disease, Chronic Obstructive, Hematoma, Image Processing, Computer-Assisted, Tomography

< Back to Listings