Searching for Smaug

My armour is like tenfold shields, my teeth are swords, my claws spears, the shock of my tail is a thunderbolt, my wings a hurricane, and my breath death!


Smaug, the most powerful, the most cunning, and some would even say the greatest, dragon of the Third Age, had a story steeped in mystery.  By the time an unexpected journey began, Smaug had not been seen for 150 years, but legend had it that he had laid waste to the Lonely Mountain in Erebor, and was feared as a ruthless killer.

The obvious questions that ensue from this frightening tale are:

  1. Does Smaug really exist, or is he some terrifying myth?
  2. If he does exist, is he a swift and definitive killer, feared by man, dwarf and elven folk alike?
  3. Does he have a weak spot, which is amenable to piercing by a single needle-like arrow, to smote him still and dead?

Just for a moment, as we take the Middle out of Earth, we consider the dragon of the current age.

The spontaneous tension pneumothorax

We shall look at the 3 questions posed.

  1. Firstly – does it exist, or is it some terrifying myth?

We have all had the experience of the tension phenomena in the setting of trauma, or in the iatrogenic setting, particularly in conjunction with positive pressure ventilation.

There are some that would say that the occurrence of a primary spontaneous tension pneumothorax in a spontaneously ventilating patient is incredibly rare.  There are very few reported cases in the literature

  • Spontaneous Pneumothorax:  Is it under tension? [J Accid Emerg Med 2000]
  • Spontaneous Tension Pneumothorax. [NEJM 2010]

This post arose from 2 curious happenstances.

The first was a clinical case of a young man, a non-smoker, a non-partaker of illicit substances.  He presented to ED several hours after the onset of right pleuritic chest pain and shortness of breath.  His observations showed a persistent and worsening tachycardia and developing hypotension.  At the time of his CXR he had a HR of 125 and a systolic BP of 90mmHg. Sao2 on air were 93%.


Is this a tension?  Not much mediastinal shift on the CXR I hear you say.  But what about the tachycardia and hypotension?  Blunt tools I admit, but pretty impressive in a young fit man.

So, we did something a little crazy. Whilst preparing to put in a narrow bore intercostal catheter, we put a needle into the 5th ICS AAL and hooked him up to the invasive pressure monitor, to see if we could answer the question!  The monitor showed a very steady 17mmHg, with only minor fluctuations with the respiratory cycle.   What does this mean? We don’t know, as there is next to no literature about this.  Several animal studies have looked at intrapleural pressures in induced porcine pneumothoraces, whereby discreet volumes of air are induced (apparently the swine model most closely resembles the human condition, because the intact mediastinal pleura is similar, but this seems a long bow) – (and I haven’t seen mentioned made of a dragon model).  At 0, 10 , 20 cmH20 increases in intrapleural pressures, there is little change in SVC, oesophageal and CO pressures/parameters, but at 10 and 20 cm H20 the HR starts to increase.

  • The Pathophysiology of tension pneumothorax in ventilated swine.  [J Emerg Med. 1997]

We plan to look out for further cases to see how reproducible this is (and perhaps put the call out to FOAMites to consider doing the same – to consider transducing a pressure from a needle inserted at the time of preparing to drain a PTx)

As soon as his ICC was placed, his HR dropped to <90, and his SBP went up to >110mmHg.


The other unhappy circumstance that led to this post was a case of which I became peripherally aware, in the pre-hospital setting, whereby the adherence to an algorithm in a patient who had an out of hospital PEA arrest, led to a series of events whereby a patient had bilateral needle thoracostomies performed.  The patient recovered from their PEA arrest, which was probably, in retrospect, anaphylactic in nature. They then, however, expired from a massive haemothorax, probably as a result of a needle thoracostomy.

It is not appropriate to dive into the details of this case, but it does serve to illustrate a case for reviewing a risk: benefit analysis.  If primary spontaneous pneumothorax is incredibly rare, are we well served by seeking it out so aggressively?

Thus to ponder the second question.  Is a primary spontaneous tension pneumothorax a ruthless, swift and silent killer, or is it, in fact, urban legend, dogma passed down through the ages like tales around a flickering fire in the dark.

The third question leaves us on far more solid ground.

Smaug slumbered for centuries on his plundered treasure, allowing his soft and vulnerable underbelly to be impregnated with gold and gems, protecting him from attack, where his scales finished.  The nominated ‘burglar’, one Bilbo Baggins, discovered a single bare patch over Smaug’s left breast, whilst pilfering a small gold goblet. This information was leaked, allowing Smaug to meet his demise at the hands of a single man, wielding a needle-like arrow, which pierced this very spot.

Even if we do believe that the primary spontaneous tension pneumothorax exists, does a lone angiocath, plunged deep into the flesh of the 2nd ICS MCL redeem the situation?

The needle is too short, kinks too freely, and has unclear endpoints. For references for this, I direct you, once again, to the Kings and Queens of the FOAM realms, who have published much, and referred to the original evidence.

So what do we know?

Thus, we are left with few definitive and certain points.

There have been documented cases of primary spontaneous tension pneumothorax.

The case that I have brought before you is undefined.  It is certainly not clear whether this was truly a tension – although he had clear cardiovascular (albeit mild) compromise, his CXR changes were unexciting, and his measured intrapleural pressures were undecipherable in the face of a lack of comparable evidence.

Needle thoracostomy is a procedure, steeped in dogma, but without much proven benefit, and high potential risks, in this subset of patients.  There is no question that its utility ought to be severely questioned if following an algorithmic course for management of PEA arrest.

If you have any experience, either with primary spontaneous tension pneumothoraces, or indeed, with dragon-slaying, I would encourage you to share them in this new FOAMed realm.  Perhaps we can achieve what literature, both fiction, and non- has been unable to do, and definitively answer this question.

Print Friendly


  1. Malcolm McKenzie says

    Great post and links, thanks. Seen once which had begun as a non-serious chest pain for which help was not sought and no analgesia required, after the patient had run a short distance. Pain and SOB got suddenly worse several hours later and the patient presented peri-arrest. Needle decompression performed on resonant side on arrival (no radiology as looked too ill). Immediate rush of escaping air and rapid return of normal obs. Check CXR nicely expanded lung, small Ptx. Then deteriorated in 20 mins, found to have haemothorax from the needle decompression. Sucks, huh?
    Would this count as ‘spontaneous’ given it likely started with exercise? Suppose even the spontaneous ones have to have *some* precipitant, even if it’s just a big cough!
    Great post, thanks again.

  2. dlpthomas says

    Back in the last century when I was an intern I saw a patient who had presented to ED with right sided pleuritic chest pain and mild shortness of breath. She had a normal BP and HR, was able to walk from triage to a bed and gave a detailed history without any real signs of clinical distress. Clinically, I felt she had a pnumothorax (PTX) and odrered a chest x-ray. I had just finished explaining to the consultant (a FACEM examiner) why it was not a tension PTX when the radiographer came sprinting round the corner with chest x-ray in hand (“Thomas’s Sign”). The x-ray showed a large right sided PTX with tracheal displacement and significant / massive / enormous (you get the idea) mid-line shift. I was shunted to the side whilst the consultant and registrar “needled the chest” (loud hiss) and placed a chest drain.

    Based on this n of 1, I have conlcuded that whilst you can get a spontaneous tension pneumothorax,
    a radiological tension PTX is not the same as a clinical tension PTX. Perhaps, unlike in a ventilated patient, a tension PTX in a spontaneously breathing patient develops slowly and thus in some cases can be well tolerated for a surpisingly long period of time despite a dramatic CXR.

    I met the FACEM again several years latter at the VIVA for my primary exam. His station seemed to go OK but as I walked to the next table I thought I heard some-one make a loud hissing noise.

  3. says

    Great post as only Elytherius can put into words.
    I absolutely agree with the premise that tension is extremely rare in spontaneous pneumothoraces and it all relates to the pressure driving any increased intra-pleural pressure within the pneumothorax. This can only be significantly above atmospheric pressure if the patient coughs or valsalvas and then is momentary and slow to accumulate.

    My two favourite papers looking at the issue are :
    Simpson, G. Spontaneous pneumothorax: time for some fresh air. Intern Med J. 2010 Mar ;40 (3):231-4 By an Aussie Respiratory Physician.

    And required reading for anyone managing thoracic trauma: (And I think my favourite paper ever…. from my HEMS London colleagues)
    Leigh-Smith, S, Harris, T. Tension Pneumothorax. Time for a Rethink. Emerg Med J 2005;22:8-16

    I have a video posted on youtube (from our HEMS inductions) on the issue of management of traumatic pneumothoraces which goes over some of the same issues:

  4. T says

    I saw one last year: Patient came in extremely breathless Sats 92% on 15L. No hx of trauma, no underlying pathology, no chest pain, just progressively breathless over hours. I initially assumed either PCP, massive TB pleural effusion (it was in South Africa), or PE. Examination revealed poor entry on both sides. Unfortunately, I mistook the hyperesonance on the right for normal resonance on the right and dullness on the left. Portable radiographs revealed my mistake: massive right sided tension pneumothorax. Needle decompressed: which wasn’t particularly useful; chest drain did the trick. Spontaneous tension pneumothorax does exist.

  5. Aaron Cummins says

    I had a 25yr male who had taken MDMA Friday evening. Thoughout Saturday, he complained to his sister that he was SOB and his pulse was racing. She dosed him with some Valium and propanol( part of her Come Down recipe). Self presented on Sunday and was streamed to Urgent Care in view of his normal vitals and looking well.
    . CXR full on tension PTX; no pneumomediastinum ( have seen several of those with similar aetiology)
    Thoughts on large bore Seldinger v Surgical Drain?
    After decompression I placed a 20 Seldinger and Medical team insisted I swapped it for a surgical drain. I consulted Thoracic Surgery and they were OK with my initial intervention

  6. says

    yes it does exist, i myself suffered one and i suppose thankfully survived. personally i want to know how long recovery is gonna take!

    Circumstances -- Asleep in bed -- awoke in pain and unable to breath. result very ill. Completely spontaneous.

  7. graham simpson says

    Can I add my other paper to the one referred to above?(and thanks for the kind comments)

    Intern Med J. 2012 Oct;42(10):1157-60.
    Spontaneous tension pneumothorax: what is it and does it exist?
    Simpson G, Vincent S, Ferns J.
    Author information
    Tension pneumothorax is variously defined but is generally thought of as a pneumothorax in which the pressure of intrapleural gas exceeds atmospheric pressure, producing adverse effects, including mediastinal shift associated with cardiovascular collapse, often attributed to reduced venous return and kinking of the great vessels. The mechanism of tension pneumothorax is said to be a valvular defect in the visceral pleura such that air enters the pleural space in inspiration but cannot exit in expiration, leading to a progressive increase in pressure. However, as the driving pressure forcing air into the pleura in inspiration is atmospheric pressure, the pleural pressure can never exceed 1 atm during inspiration in a spontaneously breathing subject. Furthermore, all pneumothoraces must have pressures greater than atmospheric during expiration, or conventional treatment with intercostal tube drainage would not work. Pilot experiments have failed to show any re-entry of pleural gas into the lung in patients with persistent air leaks but no evidence of tension, suggesting these behave as valvular pneumothoraces. Case reports of tension pneumothorax in spontaneously breathing patients are rare, and most patients have other explanations for clinical deterioration. Although a large and rapidly expanding pneumothorax may require urgent intervention, it is unlikely that the effects are mediated by high intrapleural pressures. The term tension pneumothorax in spontaneously breathing patients should be reconsidered.
    PMID: 23227475 [PubMed -- indexed for MEDLINE]

    PS I’m actually a Geordie from the north of England,though I work in Australia

  8. Clare says

    Another case of clinical but not radiological tension.

    Previously well 76 year old gentleman presented to small regional hospital. Spend day in garden chopping trees, generally working hard. Developed pleuritic chest pain and dyspnoea but attributed it to muscular pain. Didn’t seek any treatment or take any pain relief until he collapsed unconscious in front of his wife.

    Arrived looking unwell, hypoxic SaO2 84% RA, 92-4% on 10L and tachycardic 120-130 but not hypotensive.

    From history concerned about possibility of PE/disection and planned urgent transfer to referral hospital for additional investigations and management.

    Reduced breath sounds right side compared to left. Urgent CXR -- large pneumothorax with increased lung markings at left base but without mediastinal shift.

    No hiss on initial insertion of needle in placement of 14 F chest tube saldinger technique.

    Obs normalised. Hr 86. SaO2 99% on 4 L O2.