3. Failing Lung

traditional methods of ventilating patients with ALI maximized oxygenation by using normal tidal volumes (10–12 ml kg1), which in non-compliant lungs were associated with very high peak and plateau airways pressures. The ventilatory mode was usually volume-controlled with synchronized intermittent mandatory ventilation (SIMV). A major concern was barotrauma. It has more recently become apparent that barotrauma is much less of a problem than volutrauma (caused by over distension of the lung), atelectrauma (owing to cyclical shearing forces generated by alveoli closing and reopening), and biotrauma (so-called because of surfactant reduction and cytokine release in response to this repetitive injury)

it has now become standard practice to try to minimize ventilator-associated lung injury (VALI) by using ‘lung-protective’ ventilation in which plateau airways pressures are limited to 30 cmH2O by means of much reduced tidal volumes, typically of 6 ml kg1. There are two consequences of this technique: the minute ventilation may be insufficient for adequate removal of CO2, and low tidal volumes will predispose to closure of alveoli and gas trapping. The first problem is dealt with by allowing the PaCO2 to rise: this is ‘permissive hypercapnia’. The second is addressed by adding PEEP to maximize the recruitment of alveoli.

this is a key part of current ventilatory strategies, and there are experimental data to suggest that it is safe (up to a PaCO2 of ~9.0 kPa and pH of ~7.2) and that it might confer some protection in the context of lung injury and associated systemic organ damage. Hypercapnic acidosis (as opposed to metabolic acidosis) appears to attenuate VALI, particularly that associated with volutrauma rather than atelectrauma. It also has some myocardial protective effects, and although a PaCO2 of >10 kPa does depress myocardial contractility, cardiac output can still increase as a result of a decrease in systemic vascular resistance. In other tissues, hypercapnic acidosis attenuates reperfusion brain injury and delays hepatocyte cell death.

PEEP increases airways pressures and may contribute to a fall in cardiac output, most clinicians consider it essential for alveolar recruitment and prevention of atelectrauma. It does not appear that outcomes are influenced by the use of ‘high’ (~13 cmH2O) rather than ‘low’ PEEP (~8 cmH2O). Typically PEEP is set at 5–10 cmH2O, but ideally this should be done with reference to the static pressure–volume curve The upper inflection point represents probable encroachment on total lung capacity, and so the distending pressure should be kept below this point to avoid overexpansion lower inflection point is where small airways and alveoli open (and is effectively the closing volume), and the inflation pressure should be just above this point to avoid de-recruitment of alveoli. Pressure-controlled ventilation on the steep linear part of the curve midway between the two points reduces the peak airway pressure for a given mean airway pressure and minimizes intrinsic PEEP. In practice, however, although modern ventilators will produce pressure–volume curves, the inflection points are often difficult to identify.

High frequency ventilation: ventilation at very high rates with low tidal volumes is theoretically ‘lung-protective’. High frequency jet ventilation (HFJV) uses rates of between 60 and 300 min1, while high frequency oscillation (HFO) uses still higher rates of 300–1800 min1. HFJV is used for the management of ARDS in some units and can be useful in differential lung ventilation (via a double-lumen tube) and in patients with bronchopleural fistulae. HFO, in which there is considerable experience in children, is probably used more widely The OSCILLATE trial, however, terminated early because of higher mortality in the group receiving high frequency oscillatory ventilation (47% v 35%).

: the practice waxes and wanes in popularity, but meta-analyses of the numerous trials that have been performed suggest that in patients with severe ARDS, it confers a survival benefit of around 10%, and that in most, although not all patients, the PaO2 will improve. Any positive response is usually observed within the first hour. Prone ventilation reduces shunt and improves oxygenation by mechanisms which are thought to include better distribution of ventilation to previously dependent areas of lung, perfusion of less oedematous areas of lung, a rise in endexpiratory volume and an increase in diaphragmatic excursion. These improvements, however, are not explained by the traditional gravitational theories of improved perfusion of dependent areas and are thought to relate more to the geometry of the pulmonary airways and vasculature. This is discussed in more detail under ‘The Prone Position in Anaesthesia’.

: changing the I:E ratio from 1:2 to 2:1 or even 3:1 will increase the inspiratory time sufficiently to allow ventilation of lung units with prolonged time constants. In effect, this may just be a way of increasing PEEP.

This is a technique that provides continuous positive airway pressure (CPAP) with a brief release of minimal duration. In effect it is inverse ratio, pressure controlled, intermittent mandatory ventilation throughout which the patient is able to breathe spontaneously. It is a means of providing ‘open lung’ ventilation. Two levels of PEEP, high and low, are set, with the time spent in high PEEP set, for example, at around 4.5–6.0 seconds and in low PEEP at a brief 0.5–0.8 seconds. The airway pressure release time is usually set at around 1 time constant (the time that it takes to empty 63% of the lung volume). As complete emptying requires 4 time constants, this short release time results in a degree of auto PEEP which further reduces alveolar collapse. The technique recruits alveoli and improves oxygenation while allowing spontaneous ventilation, but this may be at the expense of increased transpulmonary pressure, elevated work of breathing and the risk of dynamic hyperinflation. Although APRV improves oxygenation, there is no evidence to show that it improves mortality in ARDS because there have been no defined standard settings, not least because the time constants referred to previously will vary substantially according to the degree of acute lung injury.

inhaled NO is delivered to better-recruited alveoli where it dilates the associated pulmonary vessels and reduces shunt fraction. It improves oxygenation, but no study has established that this is mirrored by better outcomes

: these include nebulized prostacyclin PGI2 (less effective than NO in improving oxygenation), artificial recombinant protein C-based surfactant (evidence is awaited of its benefit in adult patients), partial liquid ventilation with perfluorocarbons which preferentially fill and recruit dependent atelectatic areas of lung (there is no evidence as yet of improved outcomes), and interventional lung assist membrane ventilator devices (such as the Novalung).

the indications for ECMO have wideNed from its use in neonates with respiratory distress syndrome to adults who require respiratory support and to those who need cardiorespiratory support after, for example, acute myocardial infarction. Evidence suggests that ECMO for respiratory support is safe, but its efficacy has not been established. It was hoped that the CESAR trial would do so (Lancet 2009, 374: 1351–630), but because ECMO was carried out in a single centre after referral from hospitals which in contrast had very heterogeneous management strategies, the interpretation of the results was equivocal. There are, however, data from the Extracorporeal Life Support Organization registry which indicate that the recent survival rate for patients with ARDS and treated with ECMO is 60–70%. The technique is relatively straightforward. Venovenous ECMO is appropriate for patients who do not need cardiac assistance, and involves passing desaturated blood from the vena cava through a membrane oxygenator across which gas exchange can take place. Venoarterial ECMO is used if myocardial support is also necessary.