The Board of Management and Trustees of the British Journal of Anaesthesia 2007. All rights reserved. For Permissions, please e-mail: journals.permissions@oxfordjournals.org
Just a little oxygen to breathe as you go off to sleep...is it always a good idea?
Administration of
oxygen 100% to patients be
fore inducing anaesthesia provides a reserve of
oxygen, mostly in the patient's functional residual capacity (FRC), to extend the time be
fore hypoxia occurs, should there be difficulties achieving adequate ventilation after induction. Pre
oxygenation is now widely used, and in the operating theatres in which I work, the practice has extended beyond the anaesthetists to nursing staff and operating department practitioners, who now automatically apply a face mask to the patient while I administer the induction agents. An editorial in 2004 argued that
routine pre
oxygenation ‘could be recommended to the profession regardless of the experience, expertise or grade of the practitioner, and mandated
for trainees'.
1 The author of this recommendation, and the enthusiastic theatre staff, all make the assumption that administering
oxygen 100% is harmless. In his translation of the Hippocratic oath, Galen put great emphasis on the phrase
primum non nocere (first, do no harm) and this tenet should be applied to the use of
oxygen 100% be
fore it becomes an accepted practice in all patients.
Oxygen toxicity has been known to occur
for many decades. Despite ubiquitous and multiple cellular defence mechanisms, all mammals are sensitive to
high concentrations of
oxygen,
with death occurring
within a few days of exposure to
oxygen 100%,
2 although among mammalian species humans tolerate hyperoxia relatively well. The likelihood of toxicity is a function of both
oxygen partial pressure and duration of exposure. Breathing
oxygen 100% at one atmosphere absolute pressure
for <12 h has no known detrimental effects in humans. Beyond 12 h, the classic symptoms of an urge to take deep breaths, chest pain, and cough occur, and after 24 h
forced vital capacity is reduced, indicating early lung injury. In terms of causing pulmonary
oxygen toxicity, pre
oxygenation in the anaesthetic room
for 3–5 min is, there
fore, harmless.
Indirect adverse effects of breathing
oxygen 100% are far more applicable to anaesthetic practice.
Atelectasis, or collapse of small regions of lung, occurs in a majority of patients having a general anaesthetic involving muscle relaxation and artificial ventilation.
2 This results from changes in the shape of the chest wall, spine, and diaphragm, causing a reduction in FRC and the volume of specific areas of the chest, particularly in the dependent areas of lung and behind the diaphragm. Three mechanisms contribute to lung collapse. First, compression
atelectasis occurs when lung regions are reduced in volume to such an extent that the air is effectively squeezed out. Secondly, absorption
atelectasis when airway closure is followed by absorption of the gases distal to the airway leading to complete collapse of the alveoli. Thirdly,
atelectasis may occur when airways are extremely narrow, but not closed, when the rate at which alveolar gas is absorbed into the blood exceeds the rate at which gas can flow through the narrow airway to replace it, accelerating airway closure and alveolar collapse. When breathing
oxygen 100%, this is likely to occur in lung regions
with a
/
ratio of <0.05.
3 Using the multiple inert gas elimination technique, areas of lung
with /
ratios this low are easily demonstrated in the elderly when awake, and during general anaesthesia in patients of all ages.
In vivo it is likely that all three mechanisms are at work simultaneously in different lung regions. In dependent regions behind the diaphragm, compression
atelectasis will occur as a result of the weight of the abdominal contents in the absence of diaphragmatic muscle tone, a situation more likely to occur in obesity and in the presence of increased intra-abdominal pressure. In other dependent areas of the lung, the reduced FRC during anaesthesia will lead to resting lung volumes falling below the closing capacity, leading to airway closure and absorption
atelectasis. In regions of lung bordering these dependent areas, airway narrowing will reduce the
/
ratio below the threshold needed
for collapse to occur.
Atelectasis during anaesthesia can be detected using computerized tomography (CT) scans, usually involving a single lung slice taken immediately cranial to the dome of the right diaphragm. The amount of
atelectasis is quantified by measuring the cross-sectional area of the
atelectasis, expressed as a percentage of the total cross-section of lung on that CT slice. The percentages obtained by this technique seem small, but it must be remembered that each 1% of
atelectasis on a cross-sectional CT scan represents around 3% of normally expanded lung volume. Surrogate measures of the amount of
atelectasis during anaesthesia are often used, such as calculating the alveolar-arterial
Po2 difference or
PaO2:
F[SIZE=-2]I[/SIZE]O2 ratio.
When airway closure or narrowing occurs, absorption
atelectasis is greatly influenced by the gas mixture present in the alveolus. When breathing air, the partial pressure of all gases present in mixed venous blood is about 87 kPa, compared
with 95 kPa in the alveolar gas.
2 This difference of 8 kPa arises because water vapour is only present in alveolar gas, carbon dioxide carriage in the blood is mostly as dissolved bicarbonate rather than carbon dioxide gas in solution, and because of the small alveolar to arterial
P[SIZE=-2]O[/SIZE]2 difference caused by venous admixture. To keep the alveoli open in the normal healthy lung, this difference of 8 kPa must be countered by the elastic recoil of the respiratory system. When breathing
oxygen 100%, the alveolar
P[SIZE=-2]N[/SIZE]2 will fall quickly, and
for a short time nitrogen will diffuse from the blood into the alveolus and so mitigate against alveolar collapse. However, once blood
P[SIZE=-2]N[/SIZE]2 becomes negligible, the total partial pressure of gas in the mixed venous blood, even when breathing
oxygen 100%, decreases to just 12.5 kPa (
P[SIZE=-2]O[/SIZE]2 of 6.4 kPa and
P[SIZE=-2]CO[/SIZE]2 of 6.1 kPa) whereas that of the alveolus remains unchanged, so introducing a pressure gradient of more than 80 kPa resulting in rapid transfer of
oxygen across the alveolar-capillary barrier and alveolar collapse. Mathematical modelling of absorption
atelectasis during anaesthesia has been used to predict the time taken
for an area of unventilated lung to collapse after induction of anaesthesia.
4 This model supports the physiological principles already described by predicting that the rate at which collapse occurs is related to the
F[SIZE=-2]I[/SIZE]O2 during anaesthesia and that pre
oxygenation
for 3 min substantially reduces the time taken
for collapse to occur irrespective of the
F[SIZE=-2]I[/SIZE]O2 used after induction.
Do these physiological principles impact on clinical practice, in particular the role of
oxygen 100% and absorption
atelectasis? It is now more than a decade since CT studies first demonstrated that pre
oxygenation leads to greater areas of
atelectasis after induction.
5 For example, if
F[SIZE=-2]I[/SIZE]O2 be
fore induction is 0.3, 0.6, 0.8, or 1.0, the mean percentage of
atelectasis seen on CT scans post-induction is 0.2%, 0.2%, 1.3%, and 5.6% respectively.5
6 Re-expansion of
atelectasis during anaesthesia (discussed later), usually provoked by falling
oxygen saturation, is another time when
oxygen 100% is often used. In a study, again using CT scanning, use of an
F[SIZE=-2]I[/SIZE]O2 of 1.0 during the re-expansion manoeuvre led to recurrence of the
atelectasis in 5 min whereas the lung remained expanded
for more than 40 min when
F[SIZE=-2]I[/SIZE]O2 was 0.4.
7 Finally, administering
oxygen 100% be
fore extubation also worsens
atelectasis. CT scans per
formed 20 min post-extubation in groups randomly assigned to be ventilated
with an
F[SIZE=-2]I[/SIZE]O2 of 0.4 or 1.0 be
fore extubation were found to have 2.6% and 8.3%
atelectasis, respectively.
8 The presence of more than 8%
atelectasis immediately after operation is clinically very significant as re-expansion of this collapsed lung after major
surgery may take some days.
Not all studies of
atelectasis and anaesthesia have given such a clear link between
F[SIZE=-2]I[/SIZE]O2 and
atelectasis. Maintenance of anaesthesia
with an
F[SIZE=-2]I[/SIZE]O2 of either 0.3 or 0.8 found no significant difference in the amount of
atelectasis 24 h after operation (2.5%
vs 3.0%, respectively).
9 However, despite the lack of statistical difference between the groups in this small study, only four of 14 patients with an intraoperative FIO2 of 0.3 had more than 2% atelectasis after operation compared with 10 of 14 in the FIO2 of 0.8 group. These results also provide some reassurance that the atelectasis seen so commonly during anaesthesia may be partially resolved 24 h after the anaesthetic.
Collectively, these studies offer good evidence that the amount of atelectasis during anaesthesia increases significantly with increasing FIO2 and that the use of oxygen 100% at any stage of an anaesthetic is associated with significant pulmonary collapse. Reducing FIO2, even by a small amount to 0.8, seems to be substantially better than using oxygen alone.
Either nitrogen or nitrous oxide may be used to reduce the
F[SIZE=-2]I[/SIZE]O2. Mathematical modelling predicts that the two gases should have similar effects on the time taken
for gas to be absorbed from an unventilated lung unit.
4 However, this prediction over-simplifies the clinical situation, which will be influenced by the timing of the closure of the airway during the anaesthetic. If nitrous oxide is used immediately after induction and airway closure occurs in the first few minutes of the anaesthetic, when the alveolar to arterial
P[SIZE=-2]N[/SIZE]2[SIZE=-2]O[/SIZE] gradient is large, then absorption of N2O from the alveolus will be rapid and faster than the diffusion of any remaining nitrogen from the blood into the alveolus. Under this combination of circumstances,
atelectasis is likely to occur. Should airway closure occur later in the anaesthetic when alveolar and arterial
P[SIZE=-2]N[/SIZE]2[SIZE=-2]O[/SIZE] are similar, then little gas exchange will occur between the blood and alveolus beyond the closed airway, and the alveolus should remain expanded. Clinical support
for these observations is sparse,
with only one study comparing
F[SIZE=-2]I[/SIZE]O2 of 0.4 in nitrogen or N2O.
10 This study, which used
PaO2:
F[SIZE=-2]I[/SIZE]O2 ratio to indirectly estimate the amount of
atelectasis 30 min after induction, found that nitrous oxide at this early stage of an anaesthetic did indeed behave in a similar fashion to
oxygen 100%.
Thus, it seems that if N2O is part of the anaesthetic technique from the outset, then atelectasis may be more frequent than when ventilation is with oxygen and air.
Re-expansion of
atelectasis is possible
for a patient who has a tracheal tube, and two techniques are described. The first involves increasing positive end-expiratory pressure (PEEP) to 15 cm H2O, followed by an increase in tidal volume until peak inspiratory pressure reaches 40 cm H2O. This pattern of ventilation is then maintained
for 10 breaths, be
fore returning to standard ventilator settings.
11 The second involves a vital capacity manoeuvre to a sustained airway pressure of 40 cm H2O, which in the original studies was maintained
for either 15 or 25 s.12
13 On the basis of subsequent CT scan studies, when using this technique half the
atelectasis is re-expanded after just 2 s, and in three-quarters of patients all the
atelectasis is re-expanded in 8 s.
13 At these
high inflation pressures, there are benefits to minimizing the duration, particularly to reduce the cardiovascular effects of this prolonged and severe Valsalva manoeuvre and to minimize the small risk of pulmonary barotrauma. Prevention of
atelectasis can be achieved
with modest levels of PEEP,
with 10 cm H2O preventing
atelectasis formation even when
high F[SIZE=-2]I[/SIZE]O2 is used.
14 Continuous positive airway pressure (CPAP) of 6 cm H2O applied via a tight fitting facemask be
fore induction is also effective at preventing
atelectasis formation, again despite using
oxygen 100%, although this is a rather invasive technique to be used
routinely.
15
Use of
oxygen 100% be
fore and during anaesthesia will always be necessary in some patients. These include patients
with a known difficult airway, a reduced FRC and there
fore
oxygen reserve (term pregnancy, obesity, abdominal distension, and lung pathology), an increased
oxygen consumption (pregnancy, paediatrics, and sepsis), or pre-existing hypoxia from lung pathology. In these situations, an effective technique of pre
oxygenation
16 should continue to be used and should always be followed, whenever possible, by a properly administered re-expansion manoeuvre and PEEP then used to prevent
atelectasis re
forming. In patients who are hypoxic be
fore induction, the use of CPAP be
fore and during induction should be considered.
In other groups of patients, where the reasons for using oxygen 100% are less compelling but the anaesthetist wants the security provided by greater oxygen reserves than found when breathing air, use of FIO2 of 0.8 or 0.6 should be considered. Several minutes of protection from desaturation will still be obtained, and the possibility of atelectasis during anaesthesia and into the postoperative period will be reduced. In practice, the casual pre
oxygenation referred to at the start of this editorial usually involves a short exposure to an inadequate flow of
oxygen with an ineffective seal between the mask and the patient. This type of pre
oxygenation will rarely achieve an
F[SIZE=-2]I[/SIZE]O2
high enough to contribute to
atelectasis formation, but neither will it significantly prolong the time to hypoxia if ventilation should prove impossible. If breathing additional
oxygen is considered desirable be
fore induction and the anaesthetist is content to avoid
oxygen 100%, then the required
F[SIZE=-2]I[/SIZE]O2 should still be delivered using the same technique as
for pre
oxygenation
1617 but
with some added air.
The same considerations should be applied to the use of
oxygen 100% during re-expansion manoeuvres and be
fore extubation. Un
fortunately, the groups of patients in whom
atelectasis may be particularly detrimental are the same groups as listed above in whom use of
oxygen 100% is more strongly indicated, so as usual clinicians must compromise between two opposing requirements.
A. B. Lumb St James's University Hospital Leeds UK
