Subclavian central lines

[goodoldalky]

The NEJM study referenced above, after a more thorough consideration, is nonsense. They are presenting the case, through a study subjected to a lot of different types of bias, that SC lines are "safer" than IJ and Fem lines, which is not the case, even based on their data.

They define their "primary outcome event" as a CLABSI or DVT to which they state occurred in 8 SC lines, 20 IJ lines, and 22 femoral lines. Based on this statement alone you may agree that the subclavian seems to be the safest choice. This is a convenient selection of primary outcome event, but what I am most concerned about placing these lines are non compressible arterial punctures and pneumothoraces, which are clearly as dangerous if not moreso on an occurrence basis than a CLABSI or DVT.

They do mention pneumothorax incidences of 13 in SC lines, 4 IJ, and 0 femoral. The authors view this as little more than a foot note, when to me, this is relevant. The combined event rate including pneumothorax is 21 subclavian, 24 IJ, and 22 femoral. A complete non-difference between the three. Further, if you include the number of arterial punctures or placement, subclavians become even less safe, being that it is the one non-compressible site.

In addition, the failure rate for placement of a subclavian in this study is 15%, as compared to 8% and 5% IJ and femoral. Not to mention, in 540 patients the clinicians deemed the subclavian too risky, probably "randomizing" (i.e.: selecting) the highest risk patients to receive IJ and femoral lines. Finally, this study did not mandate real-time ultrasound for IJ placement nor use of chlorhexidine cleansing and dressings for central lines, both widely accepted as standard-of-care today.

This is simply an instance of a study that despite a heavily biased design intended to demonstrate subclavian superiority, is not supported by the data. This is also in the context of the the 2012 meta-analysis of all studies previously done, including two such randomized trials and over 20,000 catheters, which shows no difference in infection rates. http://www.ncbi.nlm.nih.gov/pubmed/22809915

---

Members don't see this ad.

 

[TheBlueBlazer]

but what I am most concerned about placing these lines are non compressible arterial punctures and pneumothoraces, which are clearly as dangerous if not moreso on an occurrence basis than a CLABSI or DVT.

From the NEJM article, it appears that this is a theoretical fear more than one borne out of reality. There were equal or fewer arterial injuries of the SC than either the IJ or the Fem, even with ultrasound, and there is no mention of any bad outcomes associated with the lack of compressibility in their study of almost 3500 lines placed. I could make a similar theoretical argument that an arterial injury to the carotid is in able to cause an intimal injury that could lead to a stroke.

I don't necessarily think that any of these lines has a lower risk of complications than the others, but the reality is that subclavian lines are far easier to take care of and are more comfortable for patients than IJ or femoral lines. I think that any of the 3 is a reasonable choice, and would not criticize anyone for choosing either of the 3 at this point. To say that we should avoid subclavians is without merit, unless we are talking about lack of comfort with the procedure, in which case, that is a personal problem and not something wrong with the line.

 

[RustedFox]

... And this stuff is great. Who says this forum is dead?! This thread has gotten me to think more and argue with myself and colleagues more than any other this year thus far.

 

[deleted245139]

I found that they downplayed the mechanical complications designated secondary events) vs the primary events (dvt or infections). The thing was, they author never commented on the downstream effects of the ptx consequence. Chest tubes in and of themselves are fraught with their own set of complications, infection rates, etc. It warranted mention I thought.

Each line has its place at different times in different clinical situations. I thought the article read as an advertisement for subclavians. Don't get me wrong, I love em... But right line, right patient, right clinical context.

 

[Janders]

Excellent point with pnuemothoraxes evening out the lines. My gut tells me all three have a place, and all three have specific risks and benefits and a wise provider should be allowed to choose the correct line.

After this thread I got all excited to do a subclavian two days ago, then the patient got hypoxia and started bumping peak pressures on the vent... decided that wasn't who I wanted to give a pneumothorax to and did a femoral instead (worked great, got pulled at 36 hours when she stabilized...)

 

[Cerberus]

I do either IJs or SCs (probably 75/25), if neither of those sites are available to get a supraclavicular or a IO. I haven't done a femoral in recent memory. I have had 1 resident drop a lung doing an SC (it was one of the first few he'd done) but otherwise I've never had a complication. I think the fear is somewhat overblown given the rarity of pneumos and the potential benefit as it is a much cleaner line. I do generally avoid the SC in patient's with respiratory distress or already on a vent if I can help it.