Auto-PEEP (reviews here and here) occurs when the lung has not finished exhalation at the start of the next inspiration. Because of elastic recoil, this generates a pressure. The volume that remains (above the FRC) doesn’t go anywhere, and with subsequent breaths this results in an increasing lung volume i.e. dynamic pulmonary hyperinflation (DPH or breath-stacking). DPH doesn’t result from extrinsic PEEP (set by us) because there is no flow at end expiration – the lung has returned to FRC. Auto-PEEP has been said to occur in all ventilated patients with COPD, and about 1/3 of those without. The adverse effects of auto-PEEP are those of excessive PEEP as outlined above, but also those caused by DPH.
Auto-PEEP should be suspected whenever the flow seen to be not returning to the baseline (i.e. zero), but also in any patient observed to be ‘fighting the ventilator’ or to have higher than expected airway pressures. Thankfully, ventilators can measure auto-PEEP (Drager calls it intrinsic PEEP or PEEPi); they this by occluding the expiratory limb at end expiration and measuring the rise in preasure until it plateaus.
The causes of auto-PEEP are essentially anything that limits exhalation. These include; mucus plugs causing a ball valve effect, a fixed resistance such as a partially blocked tube, a change in compliance (such as in ARDS), or a set expiratory time that is unrealistically short. The most important pulmonary cause however is airway collapse during exhalation (expiratory flow limitation, EFL), the reason being that the application of extrinsic PEEP is beneficial if that is the cause (by ‘holding the airways open’). In all other cases, the addition of extrinsic PEEP could be expected only to add to the problem by reducing the pressure gradient for expiration and increasing the total pressure.
The level of PEEP that should be applied in the case of EFL is less than the level of auto-PEEP. Because the latter is not fixed, 80% of it is often mentioned. If the level of auto-PEEP is exceeded, the theory goes that total PEEP will increase thereby reducing any beneficial effect (described as the waterfall effect – the article is here).
Other treatment options are to:
- Reduce inspiratory time or respiratory rate – both will increase expiratory time.
- Treat pain, anxiety, fever etc. in the spontaneously breathing patient to reduce tachypnoea.
- Use bronchodilators if there’s a responsive element (although not empirically)
- Check your circuit for obstruction (secretions etc.)
Anyhow, after all that the paper that triggered looking at auto-PEEP is this one:
I’m not convinced I know what the authors were expecting from this study, but it illustrates some of the points above. 100 patients with auto-PEEP of 5 or higher were ventilated with an extrinsic PEEP set at of 80% their auto-PEEP. They tested for EFL to look closer at the concept that, if present, total PEEP should not rise but if absent it should. What they found was that patients with EFL were as likely to increase their total PEEP as not, but that patients without EFL were very likely to increase their total PEEP. A respiratory rate of >20 was also predictive of an increase in total PEEP.
None of the tests described to identify EFL seem to be clinically reliable to my me, and in any case I think this study again shows that lungs are not either one or the other. Artificial ventilation is just that, and also potentially harmful. PEEP is a simple manoeuvre with complex and contradictory effects just within the lungs of one patient; I don’t think we will ever have a unifying recipe for how to use it and if we do I think we’d need to be sceptical. I also don’t think we give auto-PEEP enough credit in its milder forms; the consequences of that I’m not sure of though.
I’d recommend having a look at the reviews – they go into a lot more detail if you’ve got this far…