Intrinsic PEEP and positive pressure ventilation

Though dynamic hyperinflation and intrinsic PEEP are not mentioned in the 2023 CICM Primary Syllabus, there is the expectation that the mid-program trainee "demonstrates methods to measure iPEEP" in the college WCA document "Ventilation". From this, one may infer that the college also expects their trainees to have a sound understanding of what iPEEP is, how it develops, and what its consequences may be. One would certainly get that impression very late in one's training, where in the Part II exam multiple detailed questions ask about the diagnosis and management of intrinsic PEEP. At that stage, the trainees would probably be better served by a brief point-form summary for the purposes of revision, and this is available in the Required Reading section for the Part II exam.

In summary:

In terms of peer-reviewed sources, Laurent Brochard published an article in 2002 which subsequently became a chapter for Applied Physiology in Intensive Care Medicine (2006), which is somehow available for free via ESICM even though Springer paywalled the rest of the textbook. Another excellent resource is Blanch et al (2005), which is available for free.

Definition of "intrinsic PEEP" and dynamic hyperinflation

Intrinsic PEEP has many synonyms, including autoPEEP (Pepe et al, 1982),  iPEEP and PEEPi. Some people refer to this phenomenon as"dynamic hyperinflation", and others use this the term for the pathological process which produces intrinsic PEEP (whereas the intrinsic PEEP is just the increased intrathoracic pressure). It is occasionally referred to as "air trapping", or "gas trapping" presumably wherever you are trapping something other than air; that's what Norman Bergman originally called it when he first described it in 1972 . These terms appear to be used interchangeably in the literature and on this site. As far as is possible to tell, nobody feels strongly enough about this to demand that we use a single description for this phenomenon. Whatever you call it, it is defined as

"Unintentional or un-measured end-expiratory hyperinflation"  with "positive end-expiratory alveolar pressure associated with a higher than resting lung volume" - Brochard, 2002

"The increase in end-expiratory lung volume (EELV) that may occur in patients with airflow limitation when minute ventilation increases" - Agusti & Soriano, 2006

"when expiratory flow has not emptied alveoli to their resting FRC values by the end of exhalation... [intrinsic PEEP is] the residual positive pressure within the lungs referenced to atmospheric pressure or to PEEP" - Brander & Slutsky, 2012

In general, though many authors do seem to use the terms interchangeably, it would appear that dynamic hyperinflation and intrinsic PEEP are not synonymous. Authors who make that distinction usually point to the fact that dynamic hyperinflation implies a change in end-expiratory volume, whereas lung volumes may actually remain normal while pressure rises (eg. with vigorous expiratory muscle activity). It is therefore possible to be dynamically hyperinflated and have a relatively normal alveolar end-expiratory pressure, or to have wildly elevated pressure at relatively normal lung volumes. To be clear, both of these possibilities are probably pretty unlikely, and by the time the patient ends up in ICU the two phenomena become so confluent that the pragmatic intensivist would view them as a single clinical entity. It is also possible to develop auto-PEEP and hyperinflation without any airway flow limitation, in the presence of a sufficiently rapid respiratory rate; Bergman originally described the phenomenon in elective surgical patients being ventilated with excessively fast rates, none of whom had bronchospasm.

Pathophysiology of intrinsic PEEP

In summary:

• Expiration usually occurs as a passive process
• If expiration is interrupted before its natural end by the next inspiration, some un-expired residual gas remains in the chest
• This gas exerts a pressure onto the respiratory circuit
• As a result, the alveolar pressure at the end of expiration is higher than zero (zero being the atmospheric pressure).
• This process of incomplete emptying is called dynamic hyperinflation, and the positive alveolar pressure is called intrinsic PEEP or auto-PEEP.

The main pathophysiological factors involved in this process are an insufficient time to return to FRC and dynamic collapse of the airways. 

Insufficient time to reach FRC causes dynamic hyperinflation

If there is not enough time to get back to FRC, dynamic hyperinflation and intrinsic PEEP will occur. This might be because the airways are very obstructed and the expiratory flow rate is very slow, or it might happen with a normal expiratory flow rate where the respiratory rate is so fast that the alveoli don't get time to empty. 

In either case, there is still flow at the end of expiration. The most important feature of this to internalise is this: because for flow to occur there must be a pressure gradient, in these cases there must be raised alveolar pressure at the end of expiration.  Otherwise, there would not still be expiratory flow. 

Dynamic airway collapse

When the pressure of alveolar gas exceeds the compression pressure of the small airways, those airways can collapse. The gas becomes trapped. With each subsequent breath, more gas is entrained into the alveoli, and alveolar pressure increases, contributing to further airway collapse.

This is actually less important in severe asthmatics. Those small airways are muscular and resistant to compression, and besides that, they are young and their lung units still have some intrinsic elastic recoil, which resists hyperinflation. Lung interstitial tissue pulls the airways open, resisting pressure from surrounding alveoli. In contrast, in the COPD patient, all that tissue has burned away in cigarette smoke and in these people the dynamic collapse of the airways contributes more to the dynamic hyperinflation. There is probably no actual data to support this - it is something asserted by Ranieri et al (1993), who derive their information from a chapter of a defunct textbook rather than from direct experimental observation.

Physiological consequences of intrinsic PEEP

Firstly, it's still PEEP. It has all the benefits and problems of actual machine-generated PEEP. It is an end-expiratory positive pressure and therefore has some expiratory and haemodynamic consequences. These are discussed in full elsewhere. There is one specific and unique matter is the effect of intrinsic PEEP on work of breathing.

Consider the normal breath.

In order to generate a pressure gradient, the patient needs to expend work (X) to generate an negative leural pressure of -2 cm H2O, which translates into a negative alveolar pressure (the numbers are approximate and unrealistic for the purpose of illustration, but in general the intrapleural pressure is always slightly more negative than alveolar pressure).

Now, let's add some airway obstruction and intrinsic PEEP.

Now, the alveolar pressure is increased to 5 cm H₂O,  and therefore the pleural pressure is also higher. This is the reason why at the end of expiration flow is out of the lungs in these patients. In order to generate inward flow, the pleural pressure needs to become negative. Thus, the respiratory muscles need to generate a pressure of -10 cm H₂O (to get the same negative pressure of -1 cm H₂O in the alveoli). 

This has been demonstrated experimentally by oesophageal manometry in vivo. Blach et al (2005) had this diagram recorded from experimental animals being manually bag-ventilated. Here, "Peso" stands for oesophageal pressure, which is a reasonable surrogate for pleural pressure.  Notice how it increases during rapid manual ventilation.

Do Intrinsic PEEP and extrinsic PEEP add to each other?

This question comes up a surprisingly large number of times. For instance, a classic remembered past paper MCQ from the examination process of another much wealthier specialist program reads, 

"A patient undergoing robotic prostatectomy with controlled mandatory volume ventilation has the following measurements: 

• AutoPEEP 4
• PEEP 8
• Peak pressure 38
• Plateau pressure 32
• Vt 600mL

The static compliance is…

a.   30 mL/cmH2O
b.   23 mL/cmH2O
c.   25 mL/cmH2O
d.   20 mL/cmH2O
e.   38 mL/cmH2O
 

The sneaky examiners of the Other College clearly tried to climb the pyramid of Bloom's taxonomy here by making the trainees decide whether they would add the "AutoPEEP" to the PEEP, as all of them (by this stage in their training) will surely have a clear understanding of what the definition of static compliance is. One also wonders about the utility of the history and patient details, as this question would probably lose none of its content validity if they were omitted. Thus, the stem should really read, 

"Do intrinsic PEEP and extrinsic PEEP add to each other?"

A) Yes.

B) No.

The answer is B).  You don't add them. Consider: static compliance is the change in volume per unit of change in pressure; specifically transpulmonary pressure (the difference in pressure between the alveoli and the pleural space). Alveolar pressure is all we are interested in here, as we will assume that the pleural pressure is not going to be fluctuating wildly; and the interaction of alveolar pressure and ventilator pressure can be explained in a series of logical steps:

• We do not usually contemplate the effect of intrinsic PEEP on lung compliance because the extrinsic PEEP is all there is, i.e. it is the pressure measured by the ventilator at the end of the ventilator circuit. We usually assume this pressure to be the same as the pressure in the alveoli. 
• The only time the ventilator circuit pressure is not the same as the pressure in the alveoli is when the airways are collapsed and gas is trapped in the alveoli, in which case the airway pressure is not transmitted to the alveoli and the transpulmonary pressure is the difference between intrinsic PEEP and pleural pressure.
  
• If the extrinsic PEEP and intrinsic PEEP are the same, it obviously does not matter which value you use to compare to pleural pressure.
  
• If the extrinsic PEEP is higher than the intrinsic PEEP, then the extrinsic PEEP is what is being transmitted to the alveoli (and the normal method of measuring intrinsic PEEP will not reveal it, as holding a higher pressure in expiration with the expiratory hold manoeuvre will not result in any increase in the circuit pressure if the alveoli do not have any pockets of pressurised gas to share with the circuit).
  

This could be explained in a series of painful diagrams, but the waterfall analogy from Tobin & Lodato (1989) is timeless and effective, making it a much better choice:
 

As such, the measurement of plateau pressure in the MCQ would not need to incorporate the "AutoPEEP", because the inspiratory pressure hold used to calculate the plateau pressure (32 cmH₂O) is well above the "AutoPEEP" and absorbs it completely.