Short versus Long Axis for Internal Jugular Cannulation

By Michael Prats

SAX vs LAX graphic

Short versus long axis ultrasound guided approach for internal jugular vein cannulations: A prospective randomized controlled trial

Am J Emerg Med EPub June 2019 - Pubmed Link

Take Home Points

1. Using a two-operator technique for US guided internal jugular vein cannulation, there was no significant difference in procedural time between short axis and long axis groups.

2. This study showed that the long axis group was superior with regard to first attempt success and complications, but these results should be interpreted with caution.


Short vs long - this is a question that comes up in many ultrasound-guided procedures. Short axis refers to visualizing your needle in an out-of-plane approach - so that your needle appears as a single point. Long axis is a 90-degree rotation from the short axis view, where the needle is now seen in-plane, both the needle point and the shaft of the needle. The theoretical benefit of short axis is that it allows you to see important structures lateral or medial to your target - you know, arteries, a trachea, etc. On the other hand, the long axis view allows you to more precisely know the location of your needle tip, ensure that you don’t go too far and poke something deep to your target. The superior option has been debated a lot, especially with regard to placing central venous catheters. This study wanted to put a rest to the debate. They randomized patients who actually needed central lines to either short or long axis and looked to see which was faster and safer. Will a clear winner surface to put an end to this debate?! Read on to find out.

Check out The Evidence Atlas - Procedures for a review of prior evidence on point of care ultrasound in procedures.


Is the long or short axis approach to ultrasound guided internal jugular cannulation superior with regard to procedural time?

Is one superior with regard to first time success and complication rate?


Study from IMS and SUM Hospital in Bhubaneswar, Odisha, India

Single center, 9 bed ICU.


  • Patients who needed a CVC


  • Emergent placement

  • Did not consent

  • Study operators were not available

  • Any site besides right internal jugular vein

  • Obese patients? (they don’t mention this until the very end of the paper)


Prospective, randomized, convenience sample

An ICU physician caring for the patient decided they needed a central line.

Patients meeting inclusion/exclusion criteria were randomized to either short or long axis.

An anesthesiologist (not otherwise caring for the patient) was called to do the procedure.

The procedure was performed with TWO PERSON technique - a resident doing the ultrasound, an attending anesthesiologist performing the procedure.

An observer measured the following:

  • Insertion Time: Time it took from skin puncture to wire insertion

  • Procedure Time: Time between start of US scanning to fixing catheter to skin

  • Number of skin punctures

  • Rate of complications (arterial puncture, hematoma, catheter misplacement, pneumothorax, hemothorax)

Primary outcome defined as difference in insertion time between the two methods.

Secondary outcomes were difference in first puncture success, number of skin punctures, and rate of complications.

Power calculation estimated 49 patients were needed in each group to detect a significant difference in time between the groups.

Clinical Trials NCT03130660

✳✳ There are a few differences between what they set out to do and what actually happened:

  • Based on clinical trials, it seems that the authors hypothesized that it was the single person technique that led to inferior performance of long axis in prior trials. Therefore, they stated they wanted to compare single operator short axis to double operator long axis. Ultimately, the trial compared double operator short vs double operator long.

  • Initial exclusion was only no consent available. Therefore, they may have added the additional ones later (sites aside from right IJ, obesity, operators not available, emergent placement).

This is a little bit fishy, but none of these changes are deal-breakers. The whole one versus two-person game is somewhat strange since I do not think many people are doing that currently. This is discussed more in the Limitations section below.

Who did the ultrasounds?

Anesthesiology residents with 3 months of ultrasound training performed the ultrasounds.

Anesthesiologists with 3 years of experience with US-guided central access - 50% short and 50% long, did the actual cannulation procedure. It seems odd that they do an even number of both.

The Scan

Linear transducer


In this study, the internal jugular vein was identified in either long or short axis.

The wire was placed using usual technique.

Then the vessel was rescanned to confirm correct placement of wire in vessel. We also advocate for this technique - it is great to know your wire is good before dilating!

Short axis - the IJV appears as a circle (wire seen as white dot in center)

IJ in Short

Long axis - the IJV appear as two parallel lines (wire seen as curving white line entering vessel)

IJ in Long

Learn how to do Ultrasound for Central Line Placement from 5 Minute Sono!


Check out Vascular Pathology on the POCUS Atlas!



232 patients met inclusion

  • 45 excluded for emergent access

  • 87 (37.5%) excluded because none of the study people were available

✳✳ Important Note: There is an error in the manuscript. Their CONSORT diagram does not add up. It reports excluding 187 patients (instead of the actual 87) due to unavailability of investigators. This would lead to 0 patients left after exclusions. I contacted the corresponding author and he informed me of the actual number listed above. This information was not available prior to recording the podcast so we mention that concern.

N = 100

  • Mean age was 57

  • No significant differences between characteristics of groups

Primary Outcome

Mean Time of Insertion

Short Axis 74.2 ± 11.1 s

Long Axis 70.3 ± 10.5 s

No statistical difference. P = 0.71

Secondary Outcomes

Total Procedure Time

Short Axis 369 ± 193.1 s (this is about 6 minutes)

Long Axis 332 ± 111.8 s

37 second difference. No statistical difference. P = 0.35

Number of Needle Sticks

Short Axis 2.0 ± 0.5

Long Axis 1.1 ± 0.9

P = 0.04

Percentage of First Time Success

Short Axis 80%

Long Axis 96%

P = 0.02


Short Axis 26%

Long Axis 8%

P = 0.03

There were 2 or 3 hematomas and arterial punctures in both groups but the main difference in complications was driven by extravasation and posterior wall puncture:

  • Extravasation: Short Axis 10% vs Long Axis 0%

  • Posterior Wall Puncture: Short Axis 6% vs Long Axis 0%

Other Findings

No significant difference in timing or complications between the two operators.


Single center in India, well trained operators, small population

Single vs Two-person operator. Initially, per their clinical trials registration, the plan was to do a single operator short axis view (vs double operator long axis). In my opinion, a two-person approach is at best awkward, and at worst dangerous. It does not take much additional skill to operate both at once and this seems to be standard now. If the investigators set out to proclaim the superiority of long axis approach, perhaps this was a ploy to make the two approaches equivalent. In my correspondence with one of the authors, this seems to be true. He informed me that they also practice mostly with single operator technique. They used two operators because they felt the advantage of long axis may be seen only when using two. It is my opinion that the long axis image might be harder to maintain during the procedure; therefore, this would give long axis the advantage and might explain the results seen in this study. This is especially suspicious in light of previous trials that have shown that long axis can take longer and that short axis is associated with higher first-time success. A metaanalysis showed there might just not be a significant difference between any of these.

Primary vs Secondary Endpoints. Remember that there was no difference in primary outcome of timing. That seems about right. If you are doing this by someone who is well trained, it probably wouldn’t take much longer to succeed. There was a difference in complication rates and first-time success, but we will have to be cautious in interpreting this because it was a secondary outcome and seems contrary to some other studies.


Why choose? You can easily and quickly switch back and forth between short and long axis, and this might be the best option for gaining the benefits of both techniques. A third idea is to use an oblique approach, it might be the best of both worlds. It is probably best to be comfortable and understand the advantages and disadvantages of each technique. Then practice and get good at one or both of them.

Is it possible that there is a difference based on experience? Perhaps earlier learners should be taught short axis to avoid complications, but more experienced practitioners should do long axis to reap the potential advantages? The jury is still out on this one.

Take Home Points

1. Using a two-operator technique for US guided internal jugular vein cannulation, there was no significant difference in procedural time between short axis and long axis groups.

2. This study showed that the long axis group was superior with regard to first attempt success and complications, but these results should be interpreted with caution.

More Great FOAMed on this Topic

Ultrasound Podcast - Microcast: IJ Placement

Our score

3 Probes

Expert Reviewer for this Post


Catherine Nix @NixLimerick

FFARCSI; FJFICMI; MRCSEd A&E; Dip Health Econ, Anaesthetist Intensivist, University Hospital Limerick, Adjunct Senior Lecturer UL GEMS

Reviewer's Comments

I have worked in Ireland and Canada & met many anaesthetists from many jusidictions – I have yet to speak to someone who regularly uses a 2-person technique to insert CVCs. I agree it is likely dangerous – please note I am not talking about teaching a novice – where the 2nd operator is ensuring the safety of the learner and the patient. The 2001 Iohom study and 2003 Blavias studies (around the time the NICE guidelines came in – 2002) favoured the SA approach as safer and quicker.

Cite this post as

Michael Prats. Short versus Long Axis for Internal Jugular Cannulation. Ultrasound G.E.L. Podcast Blog. Published on February 03, 2020. Accessed on April 14, 2021. Available at
Published on 02/03/20 05:00 AM
comments (2)
By Peter Kumasaka on 02/05/20 03:03 PM
What are your thoughts on peripheral venous access, eg basilic in the upper arm? I tend to do the long axis or oblique in plane for IJ and short axis for peripheral due to small diameter of the target and any tilting of the plane in long axis will lose the vessel.
By Mike Prats on 02/10/20 02:31 PM
Great question Peter. The long versus short discussion changes a lot depending on the target. For peripheral veins, I agree that short axis is often easier due to the difficulty of maintaining a small target in long axis. That being said - as we mention in the podcast, using a hybrid approach here can also be helpful. Starting short axis and then switching to long to confirm trajectory or success can be useful.