Clinical Case 119: Trauma & the Sensitive New Age Ultrasound

OK tonight I have a trauma case for you.

Billifred is a 24 year old man whom lives with his partner.  There relationship took a turn for the worse this evening when she found her sister’s underpants in his car.

Billifred is a “lover, no a fighter…” but tonight he was forced to defend himself against a series of stabbings.  He is brought in by taxi with multiple bleeding wounds.

Editorial note:   Presentation by taxi carries a significantly increased risk of serious badness in my world.  Ambulance services are essentially free, so they often get used as a public transport service.  However, if you are sick / worried enough to fork out $20 for a cab fare – there is a good chance that you are actually pretty sick.  Auto triage is a real thing IMHO.

On arrival he is bleeding from cuts to both hands and his ulnar forearms.  He has blood on his leg from wounds on his lateral calf.  There is a small spot of blood on his left shirt pocket.  He is not really very cooperative.  When asked what happened he says he “fell whilst sharpening his hunting knife…”

Obs:  HR 120/min,   RR 28/min,  SpO2 on RA = 90% but up to 98% on 6L/min HM,  BP140/100.  He looks sweaty and anxious.

After stripping off his shirt you see a 1 cm stab wound just below the clavicle on the (L) anterior chest, midclavicular line…

So – control the bleeders – he gets a tourniquet to his arm to stem the loss from the cuts there and then needs a chest US….  bugger, the O&G doc has taken the machine to labour ward.

After securing IV access x2 and giving him some analgesia the radiographer pops in and does a quick portable CXR.  It is a rough shot with him breathing fast, not really vertical and taking shallow breaths.

Whilst you are waiting for the films to be processed ( yes, we still use actual plastic films!!) the Bedside US machine reappears.  So you do a quick chest scan looking for pneumothorax or haemothorax on the left.

You scan from the clavicle down the anterior chest and…  no sliding.  A static pleural line with no sliding.  His right chest is normal with good sliding easily seen.  The heart looks good and there is no pericardial effusion.

Being a super-sleuth sonowarrior, you scan now across the chest laterally towards the bed looking for a lung point [contact point] – and at about the posterior axillary line you see this:   CLICK TO SEE US CLIP

Now – that is golden – a good sized pneumothorax with the whole anterior chest showing no sliding and a lung point around laterally.  The evidence suggests 100% specificity for US when a clear lung point is seen.

A quick look at the left lung base shows a small effusion ~ 1cm deep, looks like partially clotted blood with mixed echo texture .

Meanwhile the friendly radiographer has returned with the plain film and the verbal report is ” ALL Clear, no pneumothorax…”  Here is the CXR…

CXR norm

Alright then – we will pause the case at this point.

This is the “le moment decisif” the point in time where you need to decide what are you going to do next.

Do you:

(A)  Crack on a put in a chest tube

(B)  Sit on the patient, admit and observe.

(C)  Fly them to the closest CT scanner (1000 km away) for a CT chest.

We covered the pneumothorax / Airplane transfer debate a while back in Clinical Case 108.

(D) Something else / smarter??


You have about 20 minutes of suturing time to think about it as you close his lesser injuries.

You even decide to repeat the chest US after sewing up his limbs – and it looks the same – pneumothorax to the posterior axillary line with a small effusion – stable in size.

OK let me know how you will play this one.



  1. A)
    With those vital numbers, oxygenation impairment and an obvious diagnosis of pneumo = tube
    No other reason for low Sp02?
    Other possible option, monitor closely in the short term and if any deterioration ‘crack on’ with the tube

    Then maybe transfer for CT if concerned about damage to other structures.

  2. I say place the tube now. With the hemopneumothorax, you really want to know how much blood there is – and how fast it’s accumulating. And, I agree with the above comment that a symptomatic PTX needs to be relieved. As far as the transfer goes, that’s kinda outside of my realm. I’ve never worked 1) outside a trauma center or 2) without a CT scanner.

  3. Great case Casey!
    Few points to raise:

    1) Sensitivity of US vs CxR for detection of PTX/HTX in trauma:
    The best summary of this is the new NICE Major Trauma draft guidelines, available here:

    In a nutshell – US is more sensitive than CxR, specificity is similar
    There is the issue of operator dependence for US, time to complete the study (everyone says “it’s faster than a CxR”, but that ’s hugely patient and operator dependent – see below) and availability – it’s almost ubiquitous but not available everywhere yet, and don’t forget that things like obesity and subcutaneous emphysema can seriously bugger up your US beams!
    CxR can pick things that US can miss – for example in this case – a broken off blade in the chest – that can change management.

    On speed of obtaining US vs CxR:
    Where I work I can get a CxR taken & up on the computer screen in resus before anyone can complete a full chest US, especially if there’s technical issues with US (e.g. inexperienced operator, obese body habitus, sub cut emphysema – which can all drastically slow an US down), and if the CxR is diagnostic & the patient is compromised, I’d stop the US & decompress the chest. If not, we can carry on with the US.

    So my take home message is – both have pros and cons, neither carry significant risks, and you should just do both, being aware of the sensitivity & specificity issues mentioned in the NICE guidelines.

    Here’s a case I saw that covered some of these issues:

    2) Causes for abnormal vital signs
    It’d be unusual to get a completely normal erect CxR with that degree of clinical compromise (dyspnoea, tachypnoea and marked hypoxia). Supine CxR – yes, erect – I’d be very surprised.

    Causes of tachycardia?
    Any or all of:
    Hypovoloaemia (from open wounds)
    ? Drugs (e.g. Amphetamines)
    +/- Tension/Tamponade

    Cause for the hypoxia?
    Can only really be PTX/HTX, and at 90%, one would assume that there is tension occurring, regardless of the normal BP.
    Here’s an interesting study that describes the pathophysiology of tension PTX – in summary hypoxia occurs early, and CVS collapse occurs later.

    3) What to do in this case?
    It’s a no brainer:
    You have a high risk mechanism, diagnostic US and clinical compromise, and we know that the “normal” CxR lacks sensitivity
    1) Relieve actual or potential obstructive shock – place a 28-32Fr ICC
    2) Give volume if ongoing shock despite relief of obstruction
    2.1) Replace blood with blood
    2.2) Minimise saline use
    2.3) Serial US to look for tamponade

    The question is actually more about what to do after the drain goes in, regarding need for transfer/CT and/or cardiothoracics input!

    Do you:
    a) Just keep ‘em in Broome and do serial US/CxR for 24-48hrs and decide what to do based on clinical course/imaging.
    b) Wait and see what comes out of the drain – i.e. lots of blood = transfer for CT? (Given the completely normal CxR I’d expect minimal blood in the drain in this case).
    b) Transfer regardless as all stabbings with significant PTX need a CT and specialist input as they’re high risk for haemothorax, empyema and/or persistent air leak.



    • Awesome comments
      Thanks Andy
      I never cease to be surprised by how complicated apparently simple trauma cases can be.
      No wonder we need to be slow, thorough and systematic.


  4. He’s hypoxic, his vitals are poor. Something bad is happening in his respiratory system. I’m not convinced on the info given that it will be helped by a chest tube. The pneumothorax is small, there’s no evidence of tension. Is the primary problem intra pulmonary haemorrhage? He needs a CT. He will get a chest tube in case he deteriorates en route, but it’s a risky procedure.

    The next couple of hours before the plane arrives will prove decisive for him. Possibly, if he stabilizes and improves you may not decide to put a tube in. If he deteriorates significantly, he will not be getting to a cardiothoracic unit anyway.

  5. Aaron Cummings says

    How about heading back to basics and doing clinical assessment / auscultation:
    ? Decreased breath sounds left compared to right.
    ? Could the pneumothorax be anterior in the chest cavity if he is lying supine – if so, would a lateral or oblique CXR show the PTX.

    • HI Aaron
      I guess you do what you are comfortable with doing.
      Sure – if the patient is tanking and there is a left sided stab and you think its a pneuma / haemotx clinically – then I would continue to insert a drain / thoracostomy without any imaging.
      Having said that – I am quite comfortable in US for chests – and it takes me lest to do the scan than to boot up the Turbo in a thin young guy.
      Recall Dr Dave did a FAST scan in 13 secs at the recent Broome Sonowars!!
      So I think you go with your own level of comfort.

      The negative CXR can provide a source of bias in this setting if we place too much weight on its diagnostic power despite the clinical picture – i.e. stab with resp compromise. CXR lack the power to “rule out” in such a high pretest probability scenario.
      Likewise clinical exam also has pretty appalling diagnostic characteristics – so I would be reluctant to sit on this case because he “sounds OK”. In a noisy trauma room this is even harder.


  6. Hi! I’m Kat.. And whilst I am not a Trauma/Cardiothoracics Surgeon, I have a good understanding of working in a resource-limited environment.

    In the South African setting our penetrating chest trauma proximity often presents via private vehicle to 24 hour Primary Healthcare Clinics (called Community Health Centres – no wards, just one/two beds for emergencies).. They present via private vehicle for identical reasons to yours (except usually pay a Neighbour/passer-by rather than a cab).

    Whilst some of these facilities have X-rays during the daytime, they are usually closed at night (which, let’s be honest, is when the penetrating chest trauma occurs).
    There is no access to Ultrasound at these facilities.

    Huge volumes of patients with stab wounds to the chest can be seen on Saturday nights. These facilities are usually staffed with a community service doctor (the 1 year after our internship) and having spent 9months working as such facility myself the decision to insert an intercostal drain had to be entirely clinicial, the patient you describe would have definitely received an ICD in our type of setting: anxious, sweating, tachypnoeic, low sats, tachycardic. Especially with our inability to monitor the patients we felt patient was safer with drain in (they would usually sit on the “ICD bench” by the ambulance entrance awaiting transfer to hospital unless very unstable & would be taken 4+ patients per ambulance if stable – often with no EMS provider qualified to decompress tension if patient was sent without ICD) My personal record was 20 ICDs in 16 patients in 1 nightshift! Whilst I aknowledge we work in different resource settings, the ability for you to monitor patient plays a large role, if unable to monitor, probably safer to have drain in rather than patient tension later whilst not watched.

    On a funny side-note: As a hospital intern we would have to hunt down our patients.. we learnt from our physiotherapists the best way to do “intercostal-drain-chest physio-en-mass” was the Monday & Tuesday morning game of soccer on the grass patch outside the surgical ward.. By Wednesday there usually weren’t enough players left in the hospital to form teams.

    I look forward to hearing some responses from the specialists in high resource settings.

  7. Hi Casey,

    CXR is less sensitive for PNX than USS. Lots of data to suggest that and in this setting where clinical pre test probability is high then you are now in a position that you have a patient who very likely has a PTX.

    The HTX looks minimal and is unlikely to be a cause of the physiological abnormalities seen.

    You also need to look at the abdomen as I’m still not sure where the end of that knife may have ended up.

    So what?

    Well, they are also symptomatic and physiologically compromised so that would push me towards putting a chest drain in.

    In UK we put this chap through the CT~ scanner within 30 mins of arrival so much of the dilemmas you face are not faced over here.

    Without the ability to definitively investigate in your hospital I would prob go for an ICC based on the information you have, then rescan post drain (and keep an eye on the belly too).



  8. Ehm.. Right I agree with Andy Buck in that the physiological markers (sao2&RR) don’t fit with a fit and health chap and that CXR (I couldn’t get the USS clip to play). I regularly sleep with my USS and we move in together next week, so I don’t need to be sold on its bedside use.

    I do have a left field random question based only on one stabbing I delt with in Sunderland: can we lie this chap on his right lateral side and take a PLAX view of his LV (or his left side if you don’t believe in bubbles rising – have no feelings either way) and make sure his left hilum wasn’t stabbed? Might see air in LV suggesting the difference between his CXR and the markers might be down to Air Embolisms. (Remember when we use to take CVC out on the ward- always had to lie them flat for precisely 4hrs)
    Then if that was the case he’d need some Cardiothoracic intervention. Nice case.

  9. Sooo…
    Stab wound, left anterior chest, midclav.line
    Exam : Tachycardia, tachypnoe, sweaty, anxious, hypoxia
    US positiv for Pneumotorax (operator-dependent)
    CXR – if there is an anterior pneumothorax, it will be missed
    => Chest tube insertion is the next step, so A

Speak Your Mind