Subclavian Lines

[txterp98]

So this is directed at anyone who can help - attendings, residents, Anesthesiology colleagues browsing this forum, etc.

I was in the ICU today tending to a patient just moved from the floor to the ICU secondary to presumed early sepsis. I attempted a right subclavian line twice and both times I kinked the wire after going in about 3 cm. I then prepped the right IJ, used the ultrasound (which I didn't realize they had up there....probably should've known, though), and put in the sepsis catheter without problem....start to finish about 3 minutes.

So a couple of questions about what I could've done differently on the subclavian line (and I know the patient could've had some weird anatomic variant or anomaly, but I'm going to assume he didn't and chalk it up to my technique).

1) When I put in subclavians, I follow the bicipital groove and insert about 1cm inferior to the clavicle. It tends to hit the junction of the middle third/lateral third of the clavicle. Do you all go here or do most of you go for the medial third/middle third junction where the subclavian vein is most superficial and you get protection from the lung from the first rib?

2) How many of you do the supraclavicular approach (i.e. the pocket shot)?

I swear that I've done a bunch of subclavians before in both residency and as attending....I just want to learn from my failures to try and avoid them the next time. And yeah, I know the obvious answer above all is I should've pulled out the ultrasound first and spared myself all this. I'll know better next time.

---

Members don't see this ad.

 

[BADMD]

1) When I put in subclavians, I follow the bicipital groove and insert about 1cm inferior to the clavicle.

I find that if I use the standard landmarks, then back off about 2-3 cm I have a much shallower angle and a I have to do less needle and skin "manipulation" to get a good hit. I also find that when the skin recoils after I place the wire, the angle isn't as bad and thus I don't kink the wire with the dilator. Same thing with feeding the catheter; it goes down more smoothly. Although I do have to place it a few cm deeper and I end up suturing to the shoulder.

2) How many of you do the supraclavicular approach (i.e. the pocket shot)?

I've had mixed success with it, which likely comes from patient selection. I can hit it easy as can be in cardiac arrests, but start having troubles with actual live patients. My default line is an infraclavicular, so if I'm trying a supraclav, something has prevented me from going to my normal spot. I've tried it on the upright CHFer with some success, but not enough to feel good about it.

 

[bulgethetwine]

1) When I put in subclavians, I follow the bicipital groove and insert about 1cm inferior to the clavicle. It tends to hit the junction of the middle third/lateral third of the clavicle. Do you all go here or do most of you go for the medial third/middle third junction where the subclavian vein is most superficial and you get protection from the lung from the first rib?

2) How many of you do the supraclavicular approach (i.e. the pocket shot)?

I swear that I've done a bunch of subclavians before in both residency and as attending....I just want to learn from my failures to try and avoid them the next time. And yeah, I know the obvious answer above all is I should've pulled out the ultrasound first and spared myself all this. I'll know better next time.

Can you explain to me about "following" the bicipital groove? I hadn't heard this, but it might help us and others!

I use the supraclavicular approach occassionally, though similar to the above poster it's when my "go-to" line (infraclav) is failed/contraindicated. The key with supraclavs is that at no time should the needle actually be directed downward. Press the skin down so you get the angle right -- the needle tip is actually pointing slightly up over the horizontal as you skirt the underside of the clavicle.

As far as why you kinked 3cm in... if not the anatomical variant that you referred to in your other post, I suppose I might wonder if you either got unlucky and the bevel was actually not in the vessel lumen (I think this unlikely as you didn't say you had any problems with flash) or you just got unlucky and the wire was heading north. I use the technique described where you put pressure in the supraclav fossa with your non-dominant hand as you're feeding the wire -- and with the bevel opening facing toward the pateint's feet. I've not threaded one north since I started doing this, though I'm sure my 'n' is still too small (about 70?) for it to be truly powered to make the claim that it definitively works to prevent this problem.

Oh, and I hate using the ultrasound for infraclaviculars -- anyone else avoid it? The argument that ultrasound is now becoming the de facto standard for IJs notwithstanding (although I would point out that among the 4000 emergency departments in the country, outside the 150 that are academic centers, it is not anywhere NEAR the *standard of care*. Yet).

 

[southerndoc]

One of the surgeons taught me to go more medial when I couldn't get a wire to pass (twice). Both times it worked. Anecdotal, but successful in my experience.

 

[txterp98]

Whoa....my bad...I totally didn't mean bicipital groove....I meant deltopectoral groove. That's what I get for posting as soon as I get off a night shift.

When first learning about infraclavicular subclavian line placement, someone once taught me an approach to look at the deltopectoral groove, follow it cephalad to the skin folds at the axilla, and then follow this to where it transects the clavicle. This typically takes me to a point at or just a touch lateral to the junction of the lateral and medial thirds of the clavicle. From there, I'll start at a position 1cm caudal to where this imaginary line hits the clavicle. With this approach, I've never had a problem locating the vein on the first pass without aspirating air (yikes!) or encountering arterial blood. This is all anecdotal, though, and my n is approximately 70. Maybe this technique has been successful purely because it makes me consider landmarks carefully before making any kind of puncture.

Even last night, I had venous return easily both times but kinked the wire both times. Thanks for the early feedback so far. You know, it's very possible that I didn't pay attention to the bevel on my needle and that I threaded the wire directly toward the lumenal surface attached to the clavicle. Next time, I'll pay better attention to the bevel and if the same thing ever happens, pull the wire back about 1cm and rotate the finder needle 90 degrees before attempting to advance the wire again.

 

[bartleby]

In addition to anatomical variants, if the patient has had an indwelling SCV catheter before, there may also be a stricture in the vessel. The suggestion about trying a site slightly more medial when having trouble passing the wire has worked for me as well in the past.

While I can appreciate the benefits of the supraclavicular approach on an intellectual level, I think that there are better alternative sites which are less risky, particularly if ultrasound is available. If you do the infraclavicular approach correctly and keep the angle of the needle shallow (steep enough to go under the clavicle but shallow enough not to violate the pleura), you can stay out of trouble most of the time. With the supraclavicular approach, regardless of pre-existing causes of hyperinflation, all it takes is a cough, jostle, or a few seconds of distraction on the part of the operator for the needle to be headed right for the structures you want to avoid.

In regard to ultrasound being the "standard of care", it is an example of one of the things in medicine that "should be done" routinely but is not done enough yet, as the movement of ultrasound to the bedside is still an evolving process. Particularly in a busy ED when putting in a difficult central line can entail and hourlong detour to TimeKill Central, u/s can be so helpful. However, there are a ton of hoops to be jumped through (some flaming!) in order to secure an u/s machine outside the radiology department if your hospital is not used to this being so. With the myriad literature documenting the risk mitigation u/s guided line insertion allows and the attention the Institute of Medicine and the NHS have been giving, I wouldn't be shocked to see Medicare and JCAHO start to jump on the bandwagon in this regard in the next couple of years. This might have a miraculous dissolving effect on some of the administrative and institutional barriers.

 

[Winged Scapula]

One of the surgeons taught me to go more medial when I couldn't get a wire to pass (twice). Both times it worked. Anecdotal, but successful in my experience.

I would echo that...almost always when I can't hit the vessel or thread the wire its because I'm too far lateral. Most people (especially in central PA where I traine) are smaller inside than they look on the outside! 😉

 

[USAF MD '05]

The approach I learned was from a trauma surgeon. I go to the medial 1/3rd of the clav and attempt to palpate the ligament that joins the 1st rib and the clav. I go just lateral to that ligament (or where I guess it is on a fatty), drop my angle about 20-30 degrees and go right underneath. It has been money on getting the flash every time, and no dropped lungs yet. The way the surgeon drew it out for me showed that at this location and angle you are actually traveling parallell to the pleura at this point. The only snags I've hit is on a crash cordis in CPR was actually kinking the cordis within the vessel. (not the wire) It corrected itself by pulling back a little. I think that the curve that it has to make rounding the bases towards home can be a little much. No probs ever with the TLC.
Steve

 

[txterp98]

Thanks for all the feedback, guys. It's all been very helpful.

Hey Steve, I've heard about that approach, but have never done it. In your experience, roughly how far in does your finder needle go in before getting a flash? About 1 inch, maybe?

 

[davbrucas]

A lot of times when you have trouble passing the wire you are either not in the vessel completely or you are in too far and the wire is hitting the opposite side of the lumen and bending. What I do in that situation is if the wire does not EASILY pass I will put the syringe back on and ensure that I am still in the vessel (I always spin the bevel 90* right when I hit the vessel) and then I will push the external part of the needle cephalad. This usually fixes the problem if my angle into the vessel is too sharp or if I am in too far. Sometimes the vessel is so small and if you are not within a certain degree of parallel to the vessel, you just arent going to pass the wire unless you go more medial and hit the vessel behind the head of the clavicle. Another thing to try when the wire does not thread is to have someone pull the pts arm toward the feet. When I just cannot get the wire to feed, I will switch to the angiocath setup that is in the TLC kit. It basically is a sheath over the needle that you thread into the vessel like a radial art line...then you feed the wire through this. Never had one of those fail to date.

 

[droliver]

I think in general you're better off avoiding subclavicular sticks when able. The subclavian vein has a propensity for stenosis after line placement, which can effectively eliminate the ability to use that arm for dialysis access later on. In patient groups whom would be higher risks for that down the road (think...african-amercans, diabetics, young hypertensives, people with lupus, etc..) it's something to consider

 

[kungfufishing]

how often do you work in the ED?

This point about SCVL is accurate, especially when referring to bigger/stiffer catheters, but as far as typical multi lumen lines go, if the subclavian is your best bet and they need the line, I'd say do it.