We were in the ICU working on a intubated septic patient and had just managed to stabilize his hemodynamics.  Then one of the  nurses noticed how the tidal volumes on the ventilator were decreasing. Blood pressure and sats dropped gently too. A couple of hours before, I tried to insert a subclavian central line but failed. Pneumothorax? 

Breath sounds were possibly more quiet on the right side of the chest. We had an ultrasound standing by for a new attempt at a central line. I put the probe to the chest. There were no ‘comet tails’, no ‘beads on a string’, no ‘sliding sign’ and, most definitely, no ‘waves on a beach’. Pneumothorax.

The patient was taken off the ventilator and was carefully bagged. I then proceeded with a needle-decompression. Immediately air started hissing out of the needle. I asked the nurse to call the surgical registrar for a chest tube.

By coincidence the radiographer was in the ICU for another patient.  Hemodynamics at this stage had deteriorated somewhat, but had also plateaued. We had time to get a chest-x while waiting for the surgical reg. This is what came up. As you can tell, we were already a long way up a nasty creek.

Tension pneumothorax. This is where everything went to hell. Rapidly. Within 10-20 seconds the patients blood-pressure, sats and pulse went crashing through the floor – despite the pressure draining through the decompression needle. It became very obvious to me that my patient was dying a lot faster than I could decompress him through the B.S. needle I had in his chest.

I went for the scalpel, made the deep skin incision and pushed a finger through the chest wall between the ribs. I was rewarded with a massive long gush of air. Immediately, the patients vitals improved and within a minute or so normalised.

Here´s some stuff that went through my head afterwards.

1. Tension pneumothoraces, especially on ventilated patients, are sneakier and more potent than we tend to think. This patient must have compensated the initial build-up for a long time. There was no clear gradual deterioration of his vitals, no visible tracheal shift, no clearly absent breath sounds, none of that, until the patient suddenly crashed, and almost died within a time-span of 10-30 seconds.

2. Needle decompression is, at best, to be regarded as a diagnostic tool. It can’t be relied on to decompress a proper tension pneumothorax. Learn how to do a finger thoracostomy. Forget about the tube, tube placements, sutures and what not. The big hole in the chest is the fix you want for cases like this one. That’s what matters. You will have infinite amounts of time to sort out the rest after saving the patients life.

That’s all.  These are my recollections of a quite intense situation that only lasted half a minute or so.


This entry was posted in Cases, Code Brown, Emergency Medicine, Intensive Care, Prehospital Medicine. Bookmark the permalink.


  1. Minh Le Cong says:

    well done and thanks for sharing your near miss case. In Norway, do you need to call a surgeon to insert a chest drain?
    Another action I have done in that same situation is to insert a second needle laterally, like where you would insert a chest drain. But finger thoracostomy is almost as quick and more effective. Perhaps next time you try a subclavian line you check with the USS afterwards to look for pneumo?

    • K says:

      Hi Minh,
      Normally the surgical regs will insist on inserting the chest drains. At least in the places I worked in. Hospitals here are smaller so they need the volume.

      Didn’t think of a second or more needles!

      Next time I do a subclavian, I might use the US from the start. Up until now I have been lucky with my blind subclavians. Another reason is how I’m thinking it’s a skill I need to maintain, for retrieval in places where I don’t have access to US. Need to think about it!

  2. Good description – I’ve decompressed many tPTXs…I’ve never seen tracheal deviation and it’s mostly been done on a low threshold to suspect in the appropriate clinical context (IPPV/thoracic trauma). More like profound hypotension, desaturating. Breath sounds and even percussion can be heard in a noisy resus or at the roadside.

    I do wonder if we should be teaching finger thoracostomy in EMST/ATLS – the meme is taking a while to get ‘out there’. And even though I prefer EMST candidates to leave with a good understanding of tube thoracostomy, they may be the receiving registrar in ED/ICU finger-throacostomy resuscitated patients are dropped off.

    Also intrigued that the surgical reg does chest tubes (I mean, of course he/she does…but I reckon if I’ve caused a PTX I should fix it [and lets be honest, if you’ve done lots of subclavians you probably have. Or is it just me?])

    Here’s a question – do you use the seldinger chest tubes in your prehospital/trauma bay…or just a good old-fashioned scalpel-blunt-dissect-finger sweep.

    I prefer the latter, but I am a medical dinosaur!

    • Minh Le Cong says:

      just one note of caution for finger thoracostomies. they should only be done in intubated and ventilated patients. Somehow the increasing trend to use them in prehospital medicine has lead some to believe they are alternative to tube thoracostomy, in the spontaneously breathing patient. be very careful here. do not iatrogenically create a sucking open pneumothorax! Might sound obvious to experienced ears but I know of one case where an enthusiastic early adopter decided to use it on spont breathing patient…with anticipated problems!

      • Thomas D says:

        A very important note! Thanks for stressing that point, Minh.
        You can do a finger thoracotomy in spontaneously breathing patients, but need to cover up the wound fairly quickly afterwards, preferrably with a one-way-valve dressing that would work something like this: – hoping the thoracostomy will stay open. Ultimately it needs to be converted to a chest tube.

  3. …and I’m with you on USS vs Blind. I’ve seen problems with USS too. It worries me that there is a generation of doctors coming through who will NOT place a central line blind…even when it needs to happen now. Are we deskilling people?

    Used to be that would whack in a IJV line in a critical patient in ED in under 60 secs. Now everyone seems to fanny around with the bloody probe.

    USS is good for learning, it’s good for controlled lines in the ICU. But sometimes you just have to get on with it. And ‘doing blind’ and with an appreciation of anatomy (and how it doesn’t always follow the book) is golden.


  4. Frederic D. says:

    Hi K. I just wanted to let you know that I really appreciate this ICU / Emergency medecine blog.

    Kinds regards from Montreal, Canada.

  5. seth says:

    great discussion!
    there’s a video on emcrit on how to place a subclavian –

    if you keep the needle parallel to the floor, then the chance of pneumothorax is minimal. not zero, but definitely smaller. (compared to “traditional” method of pointing the needle down to get under the clavicle)

  6. Pingback: |

Leave a Reply

Your email address will not be published. Required fields are marked *