Sometimes it’s worth wondering if the things we hear, see and feel are quite as we thought they were. Dr Alan Garner has a look at your senses when you get into the chest and wonders whether it’s all as straight forward as we like to think?
Let’s start this post by stating right upfront that this is about chest wounds. If that is not what you were thinking then time to look elsewhere.
What I want to discuss is the clinical diagnosis of tension pneumothorax in the field. The reason for the discussion is that I believe it is way over-diagnosed. When I worked in the UK 6 years ago it seemed tension was being diagnosed frequently and the reason given was the sound as they breached the pleura with the forceps. As the patient was positive pressure ventilated at the time then the sound must have been air rushing out of the pleural space as their intrathoracic pressure was positive throughout the respiratory cycle right?
Remember how we can’t rely on the sounds involved in clinical examination in the prehospital environment because they’re too unreliable? Well I was being told this one was always right. ‘Always’ is a big word in medicine
I’m also aware of at least one case where a patient with a single epigastric gunshot wound from a low velocity weapon had intubation and then bilateral finger thoracostomies. The comment at the time was that the prehospital doctor, who no doubt went into it all in good faith, stated that at the time of the thoracostomies they found a pneumothorax on one side and a tension on the other.
However on imaging and surgery the projectile went straight back into the pancreas and nowhere near either hemithorax or the diaphragm. Indeed the only injuries identified to any part of the chest were the thoracostomy wounds themselves. Again an intubated patient so the intrathoracic pressure must have been positive right? If the lung felt down then it had to be a pneumothorax? And if there was a sound on breaching pleura it must have been a tension?
Clearly in the second case the signs were misleading so what is happening here? Let’s put aside for a second the challenges of the initial diagnosis of pneumothorax and focus on the feel with the finger and the sound to the ears. Could it be that some of the evidence we’ve been lead to believe tells us we’re dealing with a pneumothorax can mislead experienced, well trained clinicians?
Perhaps I have done a few more chest drains than most. Partly that is due to more than 20 years in the prehospital space but I probably did even more when I was a registrar 25 year ago. I spent 6 months working for a couple of respiratory physicians and I put lots of drains (mainly for malignant effusions) in patients who certainly did not have a pneumothorax before I started. It was common to hear a noise as the pleura was breached as the air rushed in. But this of course was in spontaneously ventilating patients and that is different right?
Obviously we need to go back to the physiology to see what is driving the movement of air either into or out of the hole we have made to determine whether the sound we are hearing is air going in, or air going out.
Back to Basics
Ptp = Palv – Pip. Where Ptp is transpulmonary pressure, Palv is alveolar pressure, and Pip is intrapleural pressure.
(If you’d like a little more on this the excellent Life in the Fast Lane has a bit on transpulmonary pressure here.)
Also it turns out that you can get a google preview of John West’s classic textbook on respiratory physiology. Take a moment to go and enjoy Figure 4-9 on page 59.
You can see from panel B (I meant it, go and have a look) that intrapleural pressure varies between about -5 and -8 cmH2O at the mid-lung level during normal respiration. It is always negative and that’s due to elastic recoil of the lung which is being opposed by the chest wall. It is less negative at the dependent regions of the lung (reducing alveolar size) and more negative at the apex (increasing alveolar size).
Let’s Add Air
In the situation of a small pneumothorax the air in the pleural space makes the intrapleural pressure less negative and the driving pressure difference for ventilation is therefore reduced. If the pneumothorax is completely open to the air such as with an open thoracostomy wound the intrapleural pressure is equal to atmospheric pressure, the elastic recoil of the lung causes complete collapse and ventilation by chest expansion is impossible – positive airway pressure has to be applied.
It is not the situation of the pneumothorax that particularly concerns me. If they are hypoxic or hypotensive and the patient has a pneumothorax the chest should be decompressed – a complete no-brainer. The question is why are good clinicians decompressing normal chests and thinking there was a pneumothorax or even a tension when there was not? Does the physiology lead us there?
First let’s consider the non-intubated patient with normal respiration and no pneumothorax. This is the situation with the patients with malignant effusions I was putting drains in years ago. Here the alveolar pressure is never more than a cmH2O or two positive or negative. The intrapleural pressure however is -5 to -8 cmH2O. Therefore it does not matter what phase of respiration you breach the pleura, the pressure gradient between the pleural space and atmosphere is negative and air will rush in.
The gradient is bigger in inspiration when alveolar pressure is negative (and therefore the total pressure is around -8 cmH2O) and less negative during expiration when it is more like -5 cmH2O. It is however always negative. It does not matter which part of the respiratory cycle you breach the pleura, air is going to flow into the pleural space and the elastic recoil of the lung will drive it to collapse. If you hear a noise as I often did, it is air rushing in, the classic sucking chest wound. An iatrogenic one.
I don’t think anyone would have an issue with things so far. So let’s move on to the intubated patient who does not have a pneumothorax. I am going to assume here that there is not a lot of airway resistance in our trauma patient (which is not to say they don’t have underlying obstructive pulmonary disease, anaphylaxis to the induction drugs you gave or a clot sitting in a big bronchus/ETT) as it makes the discussion a bit easier to assume that resistance is minimal (futile according to the Daleks) and the pressure you are seeing on your ventilator gauge is largely transmitted directly to the alveoli.
Looking at our transpulmonary pressure equation, unless the airway pressure and hence alveolar pressure is higher than about 5 cmH2O then the gradient at the time you open the pleura means air is going to enter the pleural cavity. (If they have significant airway resistance this could happen with much higher airway pressures).
Just have a quick eyeball of this time pressure chart of a standard volume cycled ventilator with no PEEP (and a self-inflating bag will provide a similar though more variable trace). And I deliberately have no PEEP in this chart. PEEP is not likely to be the first thing we reach for in the hypotensive trauma patient we have just intubated where we are concerned about the possibility of a pneumothorax.
With normal lungs the peak pressure here is probably about 20 cmH2O. What proportion of the total respiratory cycle is the airway pressure (and hence the alveolar pressure in our patient with low airway resistance) likely to be below 5 cmH2O? If your little prehospital ventilator has a roughly 1:2 I:E ratio as most do, then the answer is most of it.
In other words unless you have PEEP of at least 5 cmH2O even in your intubated patient the transpulmonary pressure is negative for a good half of the respiratory cycle. During at least half the respiratory cycle, if you hear a noise as you breach the pleura you are hearing air rushing IN.
The elastic recoil of the lung is the reason that you feel the lung has collapsed by the time you pull the forceps out and put your finger in unless you have some PEEP in play.
Now I’m not saying there has never been a time when the air wasn’t rushing in. I don’t think much of the word “always” in medicine, remember? I’m just suggesting that what we know of physiology would argue that there is at least a solid proportion of the time where that transpulmonary pressure gradient is negative when you breach the pleura, which means that there’s likely to be a good proportion of cases where those “certain” clinical signs become less reliable.
For a demonstration of this with the mother of all open thoracotomies (in a cadaver) check out this video.
The cadaver is intubated, a “generous” pleural decompression wound has been created, and on each expiration the lung collapses right down unless PEEP is applied. And note the collapse is complete on each expiration.
As long as the thoracostomy is big enough to freely communicate with the air (and if you are relying on the open “finger” technique rather than putting in a drain it needs to be large or they may re-tension), when you put your finger in during expiration the lung will be collapsed unless there is a reasonable amount of PEEP splinting things open pretty impressively.
It will be collapsed whether it already was before you made the wound or whether it happened as you spread the forceps and made the communicating hole. The time between making the hole and getting that sense of lung up or lung down with the finger is ample time for the lung to collapse down. It seems like this particular clinical sign probably tells you nothing about the state of play prior to the wound being made.
So noises can be deceptive and feeling a collapsed lung just means that the lung recoiled as the pleura was opened. Can you even guarantee which phase of the respiratory cycle the patient was in when you made that hole? Unless you had at least 5 cmH2O (and maybe more) PEEP on at the time you breached the pleura neither of these signs necessarily means anything.
Again, I’m not really into saying things like “always” or “never”. What I’m suggesting is that there might be a lot more grey around these clinical signs than might first seem to be the case.
So how do you know if they had a pneumothorax? For me that is almost always by ultrasound now. I don’t know how I managed for 15 of those 20+ years of prehospital care without one. Sometimes of course the scan is equivocal and you need to make a call based on the signs you see and the condition of the patient but I find this to be very infrequent with a good high frequency linear probe.
And as for tension the hallmark is abnormal physiology, particularly blood pressure. If decompressing the chest fixes the physiology then they had a tension. If it does not then they had a simple pneumothorax – or none at all. Because the noise you heard as you breached the pleura may have been air either entering or leaving the building, hearing a noise does not help you either way. Was Elvis ever in the building at all?
I had the brilliant Dr Blair Munford review a heap of the physiology here to make sure it matched up.
After that link to the LITFL bit on transpulmonary pressure again? Then go right here.
And John West’s masterpiece (well at least the page mentioned) is here.
That image of Nahni with the big ears was posted to the Creative Commons part of flickr by Allan Henderson and is unaltered here.
Oh, and in case you didn’t know the truly amazing John West, Adelaide boy made good, has recorded his whole lecture series for you to go and watch. Because when you’re in your 80s you’ll probably be contributing to medical education like that too, right?