The thinking about ventilation over the last 20 years has two distinct polarities… academics with PhDs in ventilation science who really understand the physiology, the trials and are right up to date with the latest gadgets; and the protocol followers, those who realize that understanding what might be going on beyond the ETT is not going to happen, so are content with turning this or that knob when the blood gas result is put in front of them.
This short article is for those who would like to be somewhere in between these two positions. The principles it covers are applicable to any mode or type of ventilator ever used or that has yet to be invented.
There are just a few very basic principles governing decision making for ventilation. These are mostly true in all situations:
1. Ventilation’s only purpose is to keep the patient alive while something good can happen to allow them to breathe for themselves.
2. Although ventilation cannot cure the patient, it can damage.
3. Oxygenation occurs because there is adequate alveolar-capillary gas transfer.
4. Carbon dioxide is removed by establishing adequate minute ventilation.
Everything to do with ventilation lives within these four principles, which can now be examined in turn.
PRINCIPLE #1: Ventilation’s only purpose is to keep the patient alive while something good can happen to allow them to breathe for themselves.
This isn’t quite true, although there is no evidence that pneumonia, bronchiolitis, aspiration or any other respiratory problem resolves faster when the patient is ventilated. The reverse is probably true – poor secretion clearance, and a suppressed cough – necessary evils accompanying ventilation – will impair clearance of secretions. This might be mitigated to some extent by good physiotherapy, good nursing care, but its hard to find anything but personal experience and small trials looking at this.
That said there are some exceptions – ventilation allows bronchoscopy, physical secretion clearance, sample collection and instillation of agents such as DNase, which can be used for persistent collapse.
The big exception to the principle #1 is cardiac causes for respiratory distress. PEEP physically pushes pulmonary alveolar oedema into the capillaries and so improves work of breathing. It also off-loads the left ventricle (at the expense of the right). This is because, when the thorax is exposed to positive pressure, the left ventricle is ‘above’ the rest of the body by an amount similar to the mean airway pressure. The left ventricle can therefore pump blood ‘downhill’. For a patient on the edge of cardiac failure, this is a significant boost.
Despite these exceptions, mostly ventilation merely allows the patient to live while time, or the clinical team, deal with the problem (if they can). As ventilation complications roughly relate to the days of ventilation, getting on with whatever can be done is key. This might be antibiotics, steroids, an investigation or an operation.
C. Take home message
Ventilate for as short a time as possible – use the time to work out what is causing the deterioration and do something about it.
PRINCIPLE #2: Although ventilation cannot cure the patient, it can damage.
It doesn’t have to harm the patient, but this requires a skilled and well drilled workforce, as well a bit of luck. And for these reasons, intensive care has become specialised into age and disease categories.
B. Classification of damage
Being ventilated is uncomfortable, especially invasive ventilation using an endotracheal tube (ETT). Here, the cough reflex must be suppressed in order for the ETT to be tolerated. Mask ventilation is also uncomfortable, but as no ETT is involved, there is no cough reflex to deal with.
A plethora of sedative agents and muscle relaxants have been used. All have unwanted effects. Without getting into the non-CNS unwanted effects (eg hallucinations, mast cell degranulation) the CNS effects themselves cause enough problems, in a dose-dependent, sedation proportional manner.
CNS depression reduces blood pressure, and reduces movement, leading to peripheral oedema. It reduces respiratory drive and therefore makes the patient more dependent on the ventilator. The cough suppression assists the development of ventilator associated pneumonia.
Lastly, if the ETT becomes dislodged, the airway is then usually unsafe and the patient without respiratory drive. This is a big problem – so unplanned extubation is a thing to avoid!
All in all, the less sedation you can get away with, the better. Ideal sedation is just enough to make the patient safe, not trying to pull their ETT out, and enough to allow the patient to let the ventilator do its job.
If an ETT is used, this is an alien piece of plastic in the location where there should be nothing but air. This irritates and inflames the airway leading to oedema. Damage can also be mediated by ischaemia, caused by a tight fitting tube, an inflated cuff or trauma during intubation. This leads to worse oedema and in some cases, fibrosis. This can cause post-extubation stridor, or occasionally a longstanding fixed inspiratory obstruction.
The aetiology of long term damage associated with ventilation has been much studied without any clear answers. From a physiologic point of view it is likely that there are at least two causes for the inflammation and fibrosis seen in the lungs of patients who have undergone long term (<2 weeks) ventilation. These are mechanical causes and oxygen toxicity. The lung is neither designed to cope with high pressure ventilation nor high concentration oxygen, and both will damage small airways and alveoli, and will set up an inflammatory reaction.
As there is limited evidence that anti-inflammatory agents (steroids and others) do anything to outcome, the best option is to therefore limit pressure, particularly peak pressure, and FiO2. And try to balance the toxicity levels of each, as both are likely to be exponentially more toxic as exposure increases.
From this also comes the conception of permissive ventilation – accepting non-physiologic pCO2 and pO2 levels in order to reduce damage. As a patient will not usually suffer (exception – raised intracranial pressure) with a CO2 of 8-9KPa or a pO2 of 7-9KPa, ventilator settings can be reduced to only achieve these numbers.
C. Take Home Message
Ventilate as gently as you can. Get the tube right, get the sedation right. These things really matter.
PRINCIPLE #3: Oxygenation occurs because there is an adequate alveolar-capilliary gas transfer.
Having discussed the role of ventilation in the treatment of a patient and its problems, now we are onto the basics of how to actually do it. The first step is to establish sufficient surface area for gas exchange and then to move gas to and from this surface – and this is covered in principle #4.
There have been two types of respiratory failure described. Of course life isn’t so simple, but it does highlight the two main problems and strategies consequently required.
Type 1 respiratory failure is characterized by poor oxygenation, and this is caused by one or more of the following: alveolar or lobar collapse; alveolar or interstitial oedema or inflammation; V/Q mismatching. Apart from the V/Q mismatching, all show lots of white on the chest radiograph (CXR).
The aim of ventilation in this situation is to drive gas nearer to capillaries in the alveoli. This is by a mixture of inflating pressure (PIP) opening up airspaces and PEEP keeping them open. Practically, you can see you are making progress because the FiO2 required to adequately oxygenate the patient slowly falls to below 0.4.
Achieving adequate oxygenation means overcoming type 1 failure. There isn’t much a ventilator can do about vascular causes of V/Q mismatching, but it is good at dealing with alveolar collapse and oedema, and okay at dealing with interstitial oedema and lobar collapse. You can see that the ventilator is addressing these problems as the required FiO2 will fall to <0.4. If it doesn’t you haven’t done enough with the ventilator or its another problem.
Practically, there are just 3 knobs to twiddle – PEEP, PIP and Ti (inspiratory time). There is no relevant research to tell us what ‘baseline’ settings ought to be, hence the vast number of protocol variations across units.
So this is what is reasonable:
PEEP – start this at 5 cmH2O but be ready to increase this if oxygenation not rapidly improving and especially if there is lots of white on the CXR. You can increase this up to 10 if oxygenation remains a problem.
PIP – there should be enough PIP to visibly inflate the chest similar to that of a normal breath. The machine might give you a number, and 5-10 ml/Kg is appropriate, but check the visible inflation of the chest rather than believe the numbers. A neonate with normal lungs will only need 12-14 cmH2O and serious lung disease 25 + cmH2O. Once you are getting to 30+ cmH2O you are likely to be doing substantial damage to the lungs if this is prolonged.
- If you are doing PIP properly, the CO2 will be under control in most situations.
- If you are not inflating the chest at 30 cmH2O then you are heading for a different mode of ventilation.
- If you are not oxygenating well at 30/10, you are also heading for another mode.
Ti – this is just the amount of time that the PIP is being applied. It needs to be long enough to allow the gas to go into the chest and inflate airspaces. Too short and the gas will be going out before it has had a chance to do anything. Too long and it will reduce the number of breaths possible in a given time, reducing the minute ventilation.
Typically a neonate will need a rate of 30+, limiting the Ti to <0.4 sec. Older children need Tis in the region 0.7-1.0 sec, allowing a rate of 20-30/min.
Practically you can look to see that inflation of the chest is adequate and pretty much finished at the end of the Ti.
Take Home Message
Start with PEEP 5 and PIP to inflate the chest normally. Increase PIP until you are inflating properly. Once this is happening, increase PEEP until the oxygenation is OK with a falling FiO2.
PRINCIPLE #4: Carbon dioxide is removed by establishing adequate minute ventilation.
While the big problem for oxygen transfer is oxygen’s relative insoluabilty, leading to a need for a large surface area and short alveolar-blood distances, CO2 is very soluble in blood and at low concentration in (inspired) air. Consequent to this, CO2 moves easily from blood to airspaces, even in the context of substantial alveolar or interstitial lung pathology. Therefore the big limiter for CO2 elimination is the removal of gas from the alveoli and replacement with fresh gas (with a low CO2). This is termed minute ventilation and can be thought of as rate x tidal volume.
Patients with type 2 ventilatory failure have a high CO2 because of the difficulty in maintaining minute ventilation, and the most usual cause for this is expiratory obstruction, itself caused by small airway disease. Typically the CXR is black, and there are xray and clinical features of hyperexpansion. Its worth looking carefully at the patient’s chest as they are ventilated – is there a delay in the chest falling in expiration? Is it complete before the next breath goes in? If not, your patient probably has an expiratory lung problem.
If there is a CO2 problem, then the options are to improve tidal volume, to increase the rate or tolerate the problem (its hard to die from a high CO2 alone). This gets more complex as the rate affects tidal volume and vice versa, simply because it takes more time to move air out.
Practically, patients with expiratory lung problems (eg bronchiolitis, asthma, COPD) cannot tolerate a rate above 25 (big patient) or 30 (small patient). In severe disease the maximum rate may be much lower. This is because the expiratory phase will not have finished before the next breath begins. This is a phenomenon known as stacking and has the effect of increasing hyperexpansion and further obstructing expiration.
The best approach in type 2 problems is to start with inspiration – find enough PIP at the shortest Ti to get a tidal volume of 5-7 ml/kg, or normal visual chest rise. Typically this Ti will be 0.6-0.9 sec. Then increase the expiratory time (Te) until the chest has stopped falling at the end of it, or you can see on the ventilator that the air is no longer coming out. At this stage you have got the chest moving as fast as it will go.
If the CO2 is still too high, you might get benefit from increasing the PIP a bit and shortening the Ti (allowing more TV, or more Te), but you may also be looking at another mode of ventilation, such as HFOV, which is excellent at removing CO2.
Take home message
Get enough tidal volume, and make sure all that gas has time to get out!
Conclusions, where next?
This short piece is not intended to be an exhaustive treatise on ventilation. Ventilation is a mix of experience, trial and error and looking at risk/benefit. Although there is plenty of science, so is there quite a bit of art. These comments above are intended to give the reader a head start in practical understanding of how to address common situations and some wisdom around the major trade-offs integral to invasive ventilation.
Dr Jonathan Round – PICU Consultant