EP Tips & Techniques

Integrating Long-Axis and Short-Axis Views With a Twist for Ultrasound-Guided Vascular Access, Part II: Axillary Approach

Jonathan Salcedo, MD, FACC, FHRS, Cardiac Electrophysiology, Palo Alto Medical
Foundation – Sutter Health; Silicon Valley Cardiology, Sequoia Hospital, Redwood City, California

Jonathan Salcedo, MD, FACC, FHRS, Cardiac Electrophysiology, Palo Alto Medical
Foundation – Sutter Health; Silicon Valley Cardiology, Sequoia Hospital, Redwood City, California


As a follow-up to Part I of this article,1 I will describe a similar ultrasound-guided approach to medial axillary vein access for all transvenous cardiac implantable electronic device (CIED) needs, including single- or dual-chamber pacemakers (PPMs), implantable cardioverter-defibrillators (ICDs), and cardiac resynchronization therapy (CRT-D or CRT-P) devices. Ultrasound-guided axillary access for CIEDs has been described for several years,2 and more recently, a prospective randomized trial has further validated its efficacy and safety in operators who had never used ultrasound guidance for device implants prior to the study.3 Furthermore, there has been no evidence to suggest that using ultrasound to gain access, especially before making a subcutaneous pocket, increases the risk of infection or significantly lengthens procedural time.4 Using an integrated approach of long-axis and short-axis views with a twisting motion of the needle should further enhance the safety and efficiency of ultrasound-guided medial axillary vein access.

Equipment and Field Preparation

As with femoral access, a standard linear array probe (5-15 mHz) will suffice in providing adequate imaging fidelity. A smaller footprint probe is frequently available from each manufacturer and may prove fairly easier to use in the more confined deltopectoral space, but is not necessary for success if unavailable. Needle choice can be left to the operator’s discretion, but as mentioned in the prior article, I have found that an 18-gauge echo-visible needle provides the best visualization and cutting bevel to enter the anterior wall of medial axillary vein, even in the most collapsible or dehydrated vein anatomies. 

With all procedures, preparation is paramount for a successful outcome. This clearly involves having all wires, syringes, and device access equipment within reach on a Mayo stand nearby or directly on the field (Figure 1A). However, even more importantly, the operator must correctly drape the field to set oneself up for success with ultrasound-guided axillary access. I have frequently witnessed colleagues wanting to abandon the attempt due to poor visualization, when just moving the field of view led them to success. Thus, it is highly recommended to drape the field yourself, at least for the first few procedures, as you train your staff for what landmarks need to remain uncovered. When laying the initial blue towels or drapes, a wide area should be made available, including the upper border of the clavicle superiorly, the acromioclavicular (AC) joint laterally, the mid sternum medially, and the nipple line inferiorly. This helps avoid the frustration of having to readjust your window if you find that a towel or drape is obscuring your view of the axillary vein and its orientation to the axillary artery, the clavicular shadow, and the posterior lung field (Figure 1B). 


The first step involves the initial survey of the vessels in short-axis view. The figures and videos provided are from the left axillary vein approach. In this scenario, since the axillary vein and artery run laterally (ie, perpendicular to the spine), a short-axis view of the vessels involves holding the probe parallel with the spine and perpendicular to the clavicle. This should demonstrate the axillary vein anterior (closer to the probe) and more inferior (toward the feet) than the artery. Posterior to the vessels, if the field depth is adjusted adequately, the lung pleura and a few overlying rib shadows should also be visible. Starting laterally and sliding the probe medially, you will frequently see the cephalic vein drain into the axillary vein (Figure 2A) and then finally encounter the clavicular shadow as the axillary vein becomes the subclavian vein. The target for medial axillary vein access lies in the region between the clavicular shadow and the cephalic vein branch.

At this point, turn the probe clockwise 90 degrees for long-axis orientation. Unlike with femoral access, my approach for axillary access involves the needle being guided entirely in long-axis view. Since the axillary artery is posterior and cranial, there is no risk of a glancing injury to the artery with the needle during its approach to the vein. In addition, there are no superficial crossing arterial branches (at least none that I have ever encountered or seen reported) in this space (Figure 2B). After numbing the entry point and track with lidocaine, the echo-visible needle is inserted under the skin about 0.5 cm lateral to the edge of the probe at a shallow angle of 45 degrees or less (Figure 3A). The needle should be visible on the top right of the screen, and can be tracked down to the anterior wall of the vein (Figure 3B, Videos 1 and 2). 

Once the maximally tented needle tip is visualized after making fine adjustments to the probe and/or needle tilt, a rapid twisting motion of the needle with consistent forward pressure will facilitate the sharp beveled edge to cut through and enter the front wall without the needle jumping forward and risking a back wall injury (or ‘through-and-through’ puncture). After verifying the needle tip is in the middle of the vessel, a standard guidewire can be inserted with or without ultrasound visualization, depending on the operator’s preference. A quick look on fluoroscopy should be performed at this point to verify the wire has tracked to the inferior vena cava (IVC) and that there is no persistent left superior vena cava (SVC), which could affect whether the operator chooses to switch to the patient’s right side. If the latter, then the wire can be pulled under a five-minute pressure-hold without much consequence, since no pocket has been created at this point. If the intended device involves more leads, then subsequent punctures can easily be made 2-3 mm lateral to the prior stick along the same trajectory as the first. The prior inserted wire(s) does not affect seeing the needle entry for the second or third sticks (Figures 3B-3D, Videos 1 and 2). 

After achieving the intended number of access sites, the wires should be secured to the drape with a clamp (mosquito or hemostat) while the operator creates the device pocket. After additional lidocaine, an appropriate-sized incision is made just inferior (0.5-1 cm) to the wire entry tracts, which should approximate where the lateral border of the incision ends (Figure 4A, Video 2). After dissecting down to the prepectoral fascia, the pocket can be created with blunt dissection or electrocautery inferiorly and then slightly superiorly to find the wires, which should be at about the mid-point of the incision line at this level. Metzenbaum scissors can be used to blunt dissect and completely expose the wires visually (Figure 4B). The closed scissors tip can then be placed behind the wires to serve as a backstop (Figure 4C) as the wires are pushed in from the surface to harvest them into the pocket (Figure 4D, Video 2). At this point, the implant can proceed as usual with an appropriately-sized sheath insertion over a wire for the first lead, while the other wire(s) are re-secured to the drape. After wound closure, the wire entry sites can easily be covered with skin glue (Figure 5A) or a Steri-Strip layer.

Limitations, Challenges, and Troubleshooting

Compared with using this technique for femoral access, the learning curve for axillary access tends to be slightly steeper and longer. The two main reasons are (1) there is a smaller space within which to maneuver the ultrasound probe, and (2) the axillary vein tends to be much more collapsible and pliable. Thus, it will likely take 15-20 cases before feeling fully proficient. I would recommend first getting used to long-axis visualization of needle tip puncture for femoral vessels before attempting axillary access. Occasionally, the second and/or third needle punctures can be slightly harder to visualize, but surrounding signs of tenting are usually visible. Using the same pressure and needle-twisting technique will still result in the front wall releasing tension, which is easier to detect even if the needle tip is not perfectly visible.

If you find the vein is very collapsible, the usual maneuvers of starting intravenous (IV) fluid boluses, lifting the patient’s legs on a wedge, and/or having the patient breath-hold for a few seconds help immensely in ‘plumping’ up the vein. For very pliable vein walls, after reaching the vein surface, taking a steep angle with the needle can aid in the tip of the needle ‘catching’ the vein wall. Then, leveling the needle out and performing the twisting motion will usually result in the front wall ‘giving’ and allowing the needle to enter cleanly into the central lumen. As a last resort, the first rib shadow can usually be visualized posteriorly (lower on the ultrasound screen) underneath the vein and can be used as a buttress for a more forceful needle entry into the vein. However, in over 500 cases, I have never had to use this, and only learned of its use from other ultrasound users.

As with femoral access, 5-10% of patients will pose visibility challenges due to a variety of factors, including larger body habitus with deeper vessels, renal disease, or poor tissue visualization characteristics. Using Doppler visualization can aid in the location of the vessels in these situations. If the vein is too deep, and IV contrast injection confirms patency, then the pocket could be made first, and the probe can be placed in the pocket to see the vessel more clearly. This assumes that the footprint of the available vascular probe is small enough to fit or that the pocket is made slightly larger to accommodate. 

For device upgrade cases, preexisting leads do not usually obscure the view for additional access into the axillary vein. Actually, the operator can often decide on whether to enter the vein medial or lateral to the entry sites of the prior leads. Regardless, a pre-procedural peripheral contrast venography is highly recommended for all upgrade cases to ensure central venous system (subclavian and brachiocephalic) patency before axillary venous puncture attempt.


Short- and long-axis ultrasound guidance for axillary vein puncture offers a safe, effective, and reproducible technique for any CIED implantation. Visualization of the target vessel, structures the operator hopes to avoid (axillary artery and lung pleura), and entry of the needle through the anterior wall before making any incision makes this technique unique and potentially advantageous compared with other non-ultrasound based approaches. This approach also obviates the need for IV contrast, limits fluoroscopy exposure, and provides an avenue for precise planning of where and how (the angle) to land your access attempts. As with many things, ‘seeing is believing’, which in this circumstance makes a world of difference in comfort, confidence, and reliability in access for all device cases.  


I would like to thank Dr. Arnold Seto, who introduced me to ultrasound access for CIED procedures, and Dr. Elizabeth Turner for enrolling UCI Health fellows in a formal bedside ultrasound course that taught me the long access ultrasound technique. Thank you also to my former UCSD attendings (Drs. Gregory Feld, Jon Hsu, David Krummen, Thomas McGarry, and Ulrika Birgersdotter-Green) for allowing me to develop this technique, and to my Silicon Valley Cardiology partners (Drs. Roger Winkle, Hardwin Mead, Edward Anderson, Michael Ruder, Nellis Smith, Bruce Benedick, Donald St. Claire, Rob Patrawala, Gregory Engel, Melissa Kong, Shalini Bhambani, and Chad Brodt; PA-Cs Kaya Moore, Eric Windoffer, Amy Fenton, Marie Goldberg, and Jacqueline Hunter) for trusting me to perform this technique on our patients. Additionally, thank you to Michelle Isonio, Karen Pring, and Chip Voelker for managing my equipment needs, and to all of my nursing staff at Sequoia Hospital (LaVern Colgate, Nancy Olsen, Vem Acosta, Yasmine Zapata, Michele Kermani-Sutton, Angela Boncutter, Arnette Asbury, Helena Lee, Gina Fang, Edwin Madelo, Mike Alvarez, Tony Hanni, Damon Fong, Lourdes Gueco, Vi Hetrick, Dana Weber, and Kelly Sweesy) as well as to the Medtronic representatives (Dan Faria, Jason Daum, Kyle Kretchmer, Bernie Huzar, Tara Larson, and Taylor Fujimoto) who assist me with patients and who participated in the filming and photography of this technique. Finally, thank you to my loving wife, Gwyn, and daughter, Dylan, for their patience and support while I worked on the manuscript, video editing, and narration at home. 

To see Part 1 of this article, please visit https://www.eplabdigest.com/approach-high-risk-lead-extraction-community-setting

Contact the author on Twitter: @50wattdoc

Disclosures: Dr. Salcedo has no conflicts of interest to report regarding the content herein.

Video 1:


Video 2:



1. Salcedo J. Integrating long-axis and short-axis views with a twist for ultrasound-guided vascular access, part I: femoral approach. EP Lab Digest. 2020;20(5):44-46. 

2. Seto A, Jolly A, Salcedo J. Ultrasound-guided venous access for pacemakers and defibrillators. J Cardiovasc Electrophysiol. 2013;24(3):370-374.

3. Tagliari AP, Kochi AN, Mastella B, et al. Axillary vein puncture guided by ultrasound vs cephalic vein dissection in pacemaker and defibrillator implant: a multicenter randomized clinical trial. Heart Rhythm. 2020 Apr 29;S1547-5271(20)30361-1. doi: 10.1016/j.hrthm.2020.04.030. 

4. Esmaiel A, Hassan J, Blenkhorn F, Mardigyan V. The use of ultrasound to improve axillary vein access and minimize complications during pacemaker implantation. Pacing Clin Electrophysiol. 2016;39(5):478-482. doi: 10.1111/pace.12833. Epub 2016 Mar 23.