The cardiovascular effects of positive pressure ventilation

Abstract

Key points•Right ventricular afterload increases when positive intrathoracic pressure is added, caused by increased pulmonary vascular resistance.•Both left and right ventricular pressures are increased relative to atmospheric pressure during positive pressure ventilation (PPV). Left ventricular transmural (or transaortic) pressure gradient is reduced in systole: this is key in the reduction of left ventricular stroke work.•The addition of PEEP and pressure above PEEP with inspiration means that supra-atmospheric pressure is present throughout the respiratory cycle and affects all intrathoracic structures equally.•The clinical response to the effect of PPV and PEEP depends on the patient's baseline cardiovascular status and disease state.•Interventions to mitigate the physiological effects of PPV include giving fluids, drugs, and the choice of ventilator settings.Learning objectivesBy reading this article you should be able to:•Explain that altered vascular driving pressure rather than absolute pressure brings about many of the effects of positive pressure ventilation•Recall the interactions between intrathoracic volume and haemodynamic changes.•Explain the different or exaggerated haemodynamic effects of positive pressure ventilation in disease states.•Describe approaches to mitigating these effects. •Right ventricular afterload increases when positive intrathoracic pressure is added, caused by increased pulmonary vascular resistance.•Both left and right ventricular pressures are increased relative to atmospheric pressure during positive pressure ventilation (PPV). Left ventricular transmural (or transaortic) pressure gradient is reduced in systole: this is key in the reduction of left ventricular stroke work.•The addition of PEEP and pressure above PEEP with inspiration means that supra-atmospheric pressure is present throughout the respiratory cycle and affects all intrathoracic structures equally.•The clinical response to the effect of PPV and PEEP depends on the patient's baseline cardiovascular status and disease state.•Interventions to mitigate the physiological effects of PPV include giving fluids, drugs, and the choice of ventilator settings. By reading this article you should be able to:•Explain that altered vascular driving pressure rather than absolute pressure brings about many of the effects of positive pressure ventilation•Recall the interactions between intrathoracic volume and haemodynamic changes.•Explain the different or exaggerated haemodynamic effects of positive pressure ventilation in disease states.•Describe approaches to mitigating these effects. Positive pressure ventilation (PPV) causes a radical departure from the physiology of spontaneous breathing where a positive intrathoracic pressure is only generated transiently by coughing or performing a Valsalva manoeuvre.1Soni N. Williams P. Positive pressure ventilation: what is the real cost?.Br J Anaesth. 2008; 101: 446-457Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar This article considers the haemodynamic consequences of PPV and the immediate changes to heart–lung interactions. Other adverse effects of PPV exist, for example the longer-term impact of shear forces on lung tissue and inflammatory reactions secondary to mechanical ventilation and extrathoracic sequelae.2Acute Respiratory Distress Syndrome NetworkVentilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the acute respiratory distress syndrome.N Engl J Med. 2000; 342: 1301-1308Crossref PubMed Scopus (9376) Google Scholar Researchers have investigated the haemodynamic consequences of PPV over the past 70 yrs, more recently using computational models of cardiovascular and pulmonary components to better understand complex interactions in different diseases.3Cournand A. Motley H.L. Werko L. et al.Physiological studies of the effects of intermittent positive pressure breathing on cardiac output in man.Am J Physiol. 1948; 152: 162-174Crossref PubMed Scopus (398) Google Scholar,4Das A. Haque M. Chikhani M. et al.Development of an integrated model of cardiovascular and pulmonary physiology for the evaluation of mechanical ventilation strategies.Conf Proc IEEE Eng Med Biol Soc. 2015; 2015: 5319-5322Google Scholar Heart–lung interactions were first described in animals, for example Hales in 1733 measured BP changes during the respiratory cycle in horses.5Sette P. Dorizzi R.M. Azzini A.M. Vascular access: an historical perspective from Sir William Harvey to the 1956 Nobel prize to Andre F. Cournand, Werner Forssmann, and Dickinson W. Richards.J Vasc Access. 2012; 13: 137-144Crossref PubMed Scopus (16) Google Scholar One of the earliest mechanical ventilators in routine human use was the Drinker respirator or ‘iron lung’, developed for use during the 1928 polio epidemic. This ventilator created a subatmospheric pressure around the thorax, expanding it, and therefore drawing gas into the lungs via the resulting negative pressure gradient. The first positive pressure ventilator was invented in 1949 by John Emerson and PPV has been adopted into many areas of medical practice since the 1950s.6Colice G.L. Historical perspective on the development of mechanical ventilation.in: Tobin M.J. Principles & practice of mechanical ventilation. McGraw-Hill, New York2006Google Scholar There are now many ventilators and ventilation modes, though all rely on the principle of a positive pressure gradient driving gas into the lungs. To bring about inspiration, supra-atmospheric pressure must be applied, either at fixed time intervals or be triggered by a patient's own effort. In pressure-controlled ventilation (PCV), peak, mean and plateau pressures are independent variables which are set by the user. In volume-controlled ventilation (VCV), these are dependent variables as a result of the tidal volume (VT) set by the user, the compliance and the resistance of the patient's respiratory system. It is important to note that the most important factor affecting haemodynamic changes throughout the respiratory cycle is mean airway pressure. Exhalation is passive in PPV and relies on the combined elastance of the lungs and chest wall. Elastance is the reciprocal of compliance and is a measure of the ability of a tissue to recoil on removal of a distending pressure. The proportion of elastance arising from lung and chest wall is approximately equal in health but can vary greatly in disease, for example, primary acute respiratory distress syndrome (ARDS) increases lung elastance, whereas circumferential chest wall burns increase chest wall elastance. To mitigate the tendency of small airways and alveoli to collapse, positive pressure can be maintained during the expiratory phase. PEEP is defined as the ‘artificial maintenance of supra-atmospheric pressure at the end of passive exhalation during controlled ventilation’.7Wild M. Alagesan K. PEEP and CPAP.BJA CEPD Rev. 2001; 1: 89-92Abstract Full Text PDF Google Scholar As the haemodynamic effects of PPV are related to the increased mean airway pressure it is interesting to note that even a PEEP of 5 cmH2O leads to a doubling of mean airway pressure. Heart–lung interactions and haemodynamic changes during spontaneous breathing have been described recently and will not be considered in detail here.8Madhivathanan P.R. Corredor C. Smith A. Perioperative implications of pericardial effusions and cardiac tamponade.BJA Educ. 2020; 20: 226-234Abstract Full Text Full Text PDF PubMed Scopus (2) Google Scholar Three concepts important in understanding the haemodynamic effects of PPV are:(i)Changes in intrathoracic pressure created by ventilation are transmitted to the heart, pericardium, intrathoracic arteries and veins. PEEP and pressure greater than PEEP impose supra-atmospheric pressures on these structures throughout the respiratory cycle. Measured pressures within the thorax will differ depending on the compliance of the chest wall and lung parenchyma.9Lansdorp B. Hofhuizen C. van Lavieren M. et al.Mechanical ventilation-induced intra thoracic pressure distribution and heart-lung interactions.Crit Care Med. 2014; 42: 1983-1990Crossref PubMed Scopus (49) Google Scholar,10Pinsky M.R. Why knowing the effects of positive pressure ventilation on venous, pleural and pericardial pressures is important to the bedside clinician.Crit Care Med. 2014; 42: 2129-2131Crossref PubMed Scopus (3) Google Scholar(ii)The heart can be considered a pressure chamber inside another pressure chamber, the thorax.11Mahmood S. Pinsky M.R. Heart-lung interactions during mechanical ventilation: the basics.Ann Transl Med. 2018; 6: 349-357Crossref PubMed Google Scholar PPV increases intrathoracic pressure and so the increase in pressure is applied to all intrathoracic structures. The increase in pressure is therefore an absolute increase: pressures are raised in all structures rather than altering pressure gradients between them. However, pressure gradients between intrathoracic and extrathoracic structures are affected, for example the pressure gradient for venous return.(iii)Lung volume affects pulmonary vascular resistance (PVR) and therefore the afterload of the right ventricle (RV). A U-shaped curve of changes in PVR with lung volume is described,4Das A. Haque M. Chikhani M. et al.Development of an integrated model of cardiovascular and pulmonary physiology for the evaluation of mechanical ventilation strategies.Conf Proc IEEE Eng Med Biol Soc. 2015; 2015: 5319-5322Google Scholar with the nadir of PVR occurring at functional residual capacity (FRC). This is true for both PPV and spontaneous breathing. The three main factors causing immediate haemodynamic effects in PPV are shown in Figure 1. Right ventricular preload is decreased; PVR is increased, which increases right ventricular afterload; and a transaortic pressure gradient (also referred to as a left ventricular transmural pressure gradient) is created in systole, which reduces left ventricular work. The vena cava, right atrium and RV form part of the low-pressure, high-capacitance venous system. Right atrial pressure during spontaneous ventilation decreases during inspiration, increasing the vascular driving pressure between the vena cava and right atrium, thus enhancing venous return. By contrast, right atrial pressure is increased during PPV; this means that venous return is impaired no matter where atrial diastole occurs in the respiratory cycle. Using PEEP with the associated increase in mean airway pressure, exacerbates these haemodynamic effects on the right side of the heart. As PEEP is increased, the inferior vena cava diameter increases in both inspiratory and expiratory phases, indicating an increase in transmural pressure and a decrease in venous return.12Miranda D.R. Klompe L. Mekel J. et al.Open lung ventilation does not increase right ventricular outflow impedance: an echo-Doppler study.Crit Care Med. 2006; 34: 2555-2560Crossref PubMed Scopus (25) Google Scholar In one study, the cardiac output of anaesthetised subjects breathing spontaneously was compared with the same subjects during PPV and PEEP; relative to spontaneous respiration, cardiac output was reduced by 10% with PPV and zero PEEP, 18% with 9 cmH2O PEEP and 36% with 16 cmH2O PEEP.13Bindslev L.G. Hedenstierna G. Santesson J. et al.Ventilation-perfusion distribution during inhalational anaesthesia.Acta Anaesthesiol Scand. 1981; 25: 360-371Crossref PubMed Scopus (74) Google Scholar The effect of reduced preload on RV function is more pronounced in those with pre-existing RV impairment, although animal studies have found no effect on RV contractility even at a PEEP of 25 cmH2O.14Berglund J.E. Halden E. Jakobson S. et al.Echocardiographic analysis of cardiac function during high PEEP ventilation.Intensive Care Med. 1994; 20: 174-180Crossref PubMed Scopus (41) Google Scholar Positive pressure ventilation affects pulmonary blood flow and distribution with consequences for haemodynamics and gas exchange. The effects of lung volume on PVR are well documented and arise because the alveolar unit applies an extrinsic pressure to the intrathoracic vessels with resulting axial stretch and radial compression. Given the lung circulation is within the thorax, there is no pressure effect from PPV on the pulmonary circulation itself, as flow depends on a pressure gradient that is consistently applied to all components of the pulmonary circulation. Adding PEEP during PPV enables the recruitment and maintenance of some collapsed lung units. PVR is lowest at FRC, so if the effect of PEEP is to recruit collapsed lung units, PVR should be reduced up to a point.15Cronin J.N. Crockett D.C. Farmery A.D. et al.Mechanical ventilation redistributes blood to poorly ventilated areas in experimental lung injury.Crit Care Med. 2020; 48: e200-e208Crossref PubMed Scopus (6) Google Scholar Investigating individual lung units using a computational model has shown that higher levels of PEEP (up to 30 cmH2O) increases PVR. The greatest effects are on the lower-pressure right heart system, with RV afterload increased, thus opposing the ejection of blood during systole.4Das A. Haque M. Chikhani M. et al.Development of an integrated model of cardiovascular and pulmonary physiology for the evaluation of mechanical ventilation strategies.Conf Proc IEEE Eng Med Biol Soc. 2015; 2015: 5319-5322Google Scholar Increased RV end-diastolic volume, end-systolic volume and reduced RV stroke volume have all been described, affecting the left ventricle (LV) downstream. At FRC during spontaneous ventilation, alveolar pressure (Pa) is equal to atmospheric pressure and the intrapleural pressure is approximately −5 cmH2O. The difference between alveolar and intrapleural pressures is the recoil pressure. Downward deflection of the diaphragm results in a greater negative intrapleural pressure of −8 cmH2O and a negative Pa of 1–2 cmH2O (relative to atmospheric pressure) allowing the lung to inflate. On expiration, intrapleural pressure returns towards the recoil pressure at FRC. Positive pressure ventilation reverses this with high intrathoracic pressures, typically +15–30 cmH2O transmitted to the alveoli and the interstitial tissues. Intrapulmonary and interstitial pressures remain positive throughout inspiration but will return towards recoil pressure at FRC on expiration, unless PEEP is added, when the pressures remain positive throughout the respiratory cycle. These effects occur regardless of the means of delivery of the PPV, PEEP or CPAP (e.g. via facemask or tracheal tube) and whether or not it is timed by the ventilator or initiated by a patient's own respiratory effort. In 1965, West described the interactions of Pa, blood flow and vascular resistance within the lung in three zones: Zone 1 being towards the apices of the lungs, Zone 2 the middle and Zone 3 at the bases. The zonal distribution is largely gravitational but PPV appears to increase the proportion of the lung in Zones 1 and 2. In Zone 1, pulmonary arterial pressure (Pa) is below Pa and so no blood flow occurs; this area contributes to physiological dead space. In Zone 2, the alveoli and vessels behave like Starling resistors; Pa is greater than Pa (for some or all of the cardiac cycle), but Pa exceeds pulmonary venous pressure (Pv). In Zone 3, both Pa and Pv exceed Pa, and the collapsible vessels are held open.16West J.B. Dollery C.T. Distribution of blood flow and the pressure-flow relations of the whole lung.J Appl Physiol. 1965; 20: 175-183Crossref Google Scholar Mean pulmonary artery pressure in health is approximately 13 mmHg (∼17.5 cmH2O) and mean pulmonary capillary pressure is 7–10 mmHg (9.5–14 cmH2O); even if gravitational effects are discounted, higher PEEP and mean airway pressures contribute to an increase in Zones 1 and 2. Given that all structures within the thorax are subject to the same pressure and so one would expect Pa, Pa and Pv to increase equally, the reason for changes in West's zone distribution rests on changes in lung unit volumes. Soni and Williams suggest that ‘compliant inflatable lung regions readily transmit pressure and in these areas the pressure [and therefore hyperinflated alveoli] will impede capillary flow, whereas noncompliant damaged or infected lung will dampen pressure transmission and may have better perfusion’.1Soni N. Williams P. Positive pressure ventilation: what is the real cost?.Br J Anaesth. 2008; 101: 446-457Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar In patients with ARDS, Fougeres and others hypothesised that pulmonary microvessels were collapsed by positive pressure, shifting affected lung units from Zone 3 to Zones 1 or 2 with an overall increase in PVR.17Fougères E. Teboul J.-L. Richard C. et al.Hemodynamic impact of a positive end-expiratory pressure setting in acute respiratory distress syndrome: importance of the volume status.Crit Care Med. 2010; 38: 802-807Crossref PubMed Scopus (105) Google Scholar This effect is likely to be more pronounced when the central blood volume is depleted, as may be the case in disease states and with the current trend of a more conservative approach to giving fluids in the ICU and the perioperative period. In a more recent model of experimental lung injury, PEEP was also shown to redistribute blood away from well-ventilated areas worsening ventilation/perfusion matching and a poorer Pao2/Fio2 ratio.15Cronin J.N. Crockett D.C. Farmery A.D. et al.Mechanical ventilation redistributes blood to poorly ventilated areas in experimental lung injury.Crit Care Med. 2020; 48: e200-e208Crossref PubMed Scopus (6) Google Scholar In lungs with >20% atelectatic mass in expiration, using a VT of 10 ml kg−1 and a PEEP of 10 cmH2O was shown to redistribute blood to dependent, atelectatic lung regions. As mentioned, the clinical effects of the application of PEEP depend on whether the result is recruitment of collapsed lung units or hyperinflation. The latter becomes more likely at higher levels of PEEP, increasing PVR. At even greater lung volumes, the heart can become compressed in the cardiac fossa with the ventricles compressed into each other, reducing the volume of both RV and LV, although cardiac output may be restored by giving i.v. fluids (up to a point).18Pinsky M.R. Cardiovascular issues in respiratory care.Chest. 2005; 128: 592S-597SAbstract Full Text Full Text PDF PubMed Scopus (208) Google Scholar Where lung units are collapsed, hypoxic vasoconstriction where regional alveolar Po2 is below approximately 8 kPa contributes to high PVR. Therefore, if recruitment manoeuvres are successful and re-expand collapsed lung units, hypoxic vasoconstriction is reduced and with it, PVR. It is therefore possible for PEEP to both raise and lower PVR. The pulmonary arterial to left atrial pressure gradient is unaltered by changes in intrathoracic pressure because that entire circuit is within the thorax. By contrast, the systemic circulation may be greatly affected by changes in intrathoracic pressure, especially in the presence of cardiorespiratory disease.11Mahmood S. Pinsky M.R. Heart-lung interactions during mechanical ventilation: the basics.Ann Transl Med. 2018; 6: 349-357Crossref PubMed Google Scholar A reduction in venous return leads to a reduced left ventricular preload and cardiac output after two to three beats; this is illustrated by the Valsalva manoeuvre and is more significant in the presence of hypovolaemia.15Cronin J.N. Crockett D.C. Farmery A.D. et al.Mechanical ventilation redistributes blood to poorly ventilated areas in experimental lung injury.Crit Care Med. 2020; 48: e200-e208Crossref PubMed Scopus (6) Google Scholar A decline in MAP closely follows the decrease in cardiac output precipitated by increasing PEEP in patients with acute lung injury, and although some increase in systemic vascular resistance (SVR) occurs, this is only about half that required to maintain systemic arterial pressure.19Marini J.J. Dynamic hyperinflation and auto-positive end-expiratory pressure. Lessons learned over 30 years.Am J Respir Crit Care Med. 2011; 184: 756-762Crossref PubMed Scopus (71) Google Scholar PPV can reduce LV stroke work and to an extent this can mitigate the effects of decreased LV preload. In diastole, the LV transmural pressure gradient (the difference between the pressure inside the LV and the pressure around it) is unaffected by the raised intrathoracic pressure. However, in systole it is thought that there is a reduction in LV transmural pressure and therefore afterload because the increased intrapleural pressure does not affect the extrathoracic arterial system. However, because the pressure difference occurs between the intra- and extrathoracic aorta, it is really a transaortic pressure gradient that is created. Whichever term is used, this reduction in afterload improves the LV ejection fraction and reduces LV work (Fig. 2), even with an increase in arterial pressure.20Vieillard-Baron A. Matthay M. Teboul J.L. et al.Experts' opinion on management of hemodynamics in ARDS patients: focus on the effects of mechanical ventilation.Intensive Care Med. 2016; 42: 739-749Crossref PubMed Scopus (119) Google Scholar In health, this effect is not important because the reduction in preload reduces cardiac output to a greater extent. However, in disease, especially LV systolic failure, the reduction in afterload can be of benefit in increasing the cardiac output and reducing the myocardial oxygen demand, although the effects may be diminished in elderly patients if the compliance of the aorta is also low.18Pinsky M.R. Cardiovascular issues in respiratory care.Chest. 2005; 128: 592S-597SAbstract Full Text Full Text PDF PubMed Scopus (208) Google Scholar During spontaneous ventilation, negative intrathoracic pressure promotes drainage into the lymphatic system. Although PPV increases the hydrostatic pressure gradient across the alveolar capillary membrane causing fluid to move from the alveoli into the interstitium, it also leads to a collapse of the thin-walled lymphatic vessels, reducing flow. In PPV, central venous pressure is increased, which further impedes lymphatic drainage from the interstitium into the venous system and results in interstitial fluid retention.1Soni N. Williams P. Positive pressure ventilation: what is the real cost?.Br J Anaesth. 2008; 101: 446-457Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar In lung injury models of PPV, lymph production has been found to increase, whilst drainage decreases, although the mechanism is unclear.1Soni N. Williams P. Positive pressure ventilation: what is the real cost?.Br J Anaesth. 2008; 101: 446-457Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar PPV has additional effects on extrathoracic organs, largely resulting from the reduced pressure gradient between arterial and venous systems as a result of reduced cardiac output and impaired venous return.1Soni N. Williams P. Positive pressure ventilation: what is the real cost?.Br J Anaesth. 2008; 101: 446-457Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar Renal perfusion and glomerular filtration reduce and a PEEP-mediated increase in antidiuretic hormone secretion occurs. Reduced hepatosplanchnic perfusion and lymphatic drainage has been found to impair hepatic function in sepsis when PEEP is >10 cmH2O.18Pinsky M.R. Cardiovascular issues in respiratory care.Chest. 2005; 128: 592S-597SAbstract Full Text Full Text PDF PubMed Scopus (208) Google Scholar Although the precise correlation between PEEP and increased ICP is unclear, a study of 340 patients with acute brain injury suggested that PEEP could be safely applied in the initial period of brain injury.21Boone M.D. Jinadasa S.P. Mueller A. et al.The effect of positive end-expiratory pressure on intracranial pressure and cerebral hemodynamics.Neurocrit Care. 2017; 26: 174-181Crossref PubMed Scopus (26) Google Scholar However, common practice is to reduce PEEP to the minimum required to reduce ICP, especially once cerebral autoregulation becomes impaired within the first day after brain injury. In addition to the effects of PPV in patients with RV or LV failure described above, studies have investigated the effects of PPV in patient groups with altered lung mechanics. Lung mechanics may be affected by global or regional obstructive and restrictive changes and the interactions with PPV are mediated by alterations in both compliance and resistance. Animal studies have indicated that a reduced pulmonary compliance limits the effect of raised intrathoracic pressure on cardiac output because the transmission of airway pressure to other thoracic structures is limited. Patients with ARDS have reduced lung compliance which limits the increase in intrapleural pressure and their cardiovascular systems are therefore better ‘protected’ against the adverse effects of PPV and PEEP.1Soni N. Williams P. Positive pressure ventilation: what is the real cost?.Br J Anaesth. 2008; 101: 446-457Abstract Full Text Full Text PDF PubMed Scopus (66) Google Scholar In acute bronchospasm, narrowed airways lead to a greater measured pressure in the upper airways than is transmitted to the smaller airways distally. A greater delivered pressure may therefore be required to ventilate such patients' lungs, but this has fewer haemodynamic effects. In the presence of expiratory airflow obstruction, expiratory flow may occur throughout all of the expiratory time and the set PEEP will underestimate the true PEEP to which the alveoli are exposed. The additional PEEP is known as auto or intrinsic PEEP (PEEPi) and may be encountered in patients with acute and chronic obstructive airways disease, particularly asthma. Iatrogenic causes for expiratory outflow obstruction can also lead to the development of PEEPi and include obstructed ventilator tubing or a waterlogged humidifying filter in the circuit.22Conacher I.D. Dynamic hyperinflation—the anaesthetist applying a tourniquet to the right heart.Br J Anaesth. 1998; 81: 116-117Abstract Full Text PDF PubMed Scopus (16) Google Scholar There may be a marked decrease in systolic arterial BP as a result of PEEPi under these circumstances. The process of dynamic hyperinflation (DHI) develops if inspiration is initiated before the end of complete exhalation of the previous breath. The DHI causes a progressive increase in end-expiratory lung volume and subsequent worsening venous return, cardiac output and systolic BP, and increases PVR. The ventilator should be set to allow adequate expiratory time to minimise the development of DHI and the effects of PEEPi. In extremis, disconnection of the tracheal tube from the ventilator to rapidly decrease lung volume may be required to avoid cardiovascular collapse. The impact of PPV in restrictive lung disease varies. Where there is upper airway collapse such as in patients with obstructive sleep apnoea, higher levels of PEEP are required to maintain airway patency. In patients with global restriction such as scoliosis, modest increases in PEEP can result in a marked reduction in VT (or an increase in peak pressure [Ppeak], plateau pressure, or both, depending on ventilatory mode) because of poor chest wall compliance. In these patients a careful balance is required when considering ventilator settings to minimise haemodynamic compromise; in the attempt to optimise ventilation, the mean pressure and Ppeak must be kept to the minimum tolerable. The mainstays of mitigating the effects of PPV are judicious fluid management and selection of ventilator settings. Whilst in practice the addition of vasopressors is common, it should be stressed that these often offset the low SVR that accompanies the patient's need for PPV rather than the effects of PPV itself. The use of fluids to optimise venous return is well established, but a detailed discussion of fluid optimisation is outside the scope of this article. As described above, impaired LV contractility may be improved and cardiac output restored despite the continued presence of PEEP when i.v. fluids are given to restore LV end-diastolic volume to normal.18Pinsky M.R. Cardiovascular issues in respiratory care.Chest. 2005; 128: 592S-597SAbstract Full Text Full Text PDF PubMed Scopus (208) Google Scholar However, despite this, PPV may be tolerated poorly by patients with established acute or chronic right heart failure. Vasopressor, positive inotropic, or inodilator drugs may also be needed to support cardiac output after optimising fluids and ventilator settings. Adrenaline (epinephrine), dobutamine or milrinone can be used to enhance myocardial contractility at the expense of increasing myocardial oxygen demand. Vasopressor drugs such as metaraminol or low-dose noradrenaline (norepinephrine) can be useful to restore normal systemic vascular resistance and increase mean arterial pressure. The risks and benefits of using vasoactive drugs must be considered: no one agent is superior in all cirumstances. The haemodynamic effects of PPV during anaesthesia for routine surgery are usually minor because our modern ‘standard’ settings for ventilation are close to ideal for an average healthy person. In addition, although initially developed in critical care during the 1990s, the use of ‘lung-protective ventilation’ is also becoming widespread in anaesthesia. Lung-protective ventilation aims to reduce the incidence of ventilator-induced lung injury (VILI), include limiting VT to 6–8 ml kg−1 ideal body weight and limiting the plateau airway pressure. The haemodynamic effects of positive pressure ventilation are demonstrated, with a VT of 10 ml kg−1 or greater known to significantly increase RV work, mediated by an increase in PVR.11Mahmood S. Pinsky M.R. Heart-lung interactions during mechanical ventilation: the basics.Ann Transl Med. 2018; 6: 349-357Crossref PubMed Google Scholar Although atelectrauma can be minimised by using PEEP, adverse haemodynamic effects usually become too great when PEEP exceeds 10 cmH2O and certainly when more than 15 cmH2O. In unwell patients, in particular those requiring critical care, the factors that independently affect mortality and the weighting of these factors should be examined. Ideal ventilation utilises the least possible energy to deliver the goals of ventilation (Table 1) whilst minimising harmful effects. Ventilator manufacturers have strived to innovate and a large number of machines with more than 170 PPV modes are now available.23Mireles-Cabodevila E. Hatipoğlu U. Chatburn R.L. A rational framework for selecting modes of ventilation.Respir Care. 2013; 58: 348-366PubMed Google Scholar No longer delivering a simple square wave pressure or volume control, modern ventilators use measurements to vary flow throughout inspiration in an attempt to reduce adverse effects or automatically reduce support as the patient's condition improves. For example, ‘automode’ ventilation weans from PCV to pressure support once the patient starts triggering their own breaths aiding synchrony with the ventilator without manually changing mode. It should be remembered that any ‘mode’ can be dangerous if used incorrectly with or without appropriate alarms.Table 1The goals of mechanical ventilationGoalDependent variablesOxygenation•Fraction of inspired oxygen (Fio2)•PEEP•Alveolar ventilation•Ventilation:perfusion (V̇/Q̇) matching•Lung diffusion capacity•Blood transit time through the lung•Mixed venous saturationVentilation•Alveolar ventilation•V̇/Q̇ matching•Lung diffusion capacity•Blood transit time through the lungMinimise VILI:•Volutrauma•Biotrauma•Barotrauma•Atelectrauma•Alveolar overdistension•Proinflammatory response to injury•High transpulmonary pressure•High shear forces from cyclical collapseSafety and minimising compromise to other organs•Evidence-based ventilator configuration•Appropriate alarm settings Open table in a new tab Each mode has constants and variables (Table 2). It is important to consider the variation in haemodynamic effects over time. An evidence-based ventilatory strategy can become suboptimal or unsafe if the mode or alarm settings are selected inappropriately. For example hypotension caused by direct pressure effects or DHI can lead to increasing requirements for vasopressor drugs and ultimately cardiovascular collapse.Table 2Modes of ventilation with their dependent and independent variables and important alarms to set. APRV, airway pressure release ventilation; NIV, non-invasive ventilation; Pe′co2, end-tidal CO2; P-high, ‘high’ CPAP pressure; Pinsp, inspiratory pressure; P-low, ‘low’ CPAP pressure; PS, pressure support; SIMV, synchronised intermittent mechanical ventilation; T-high, time at P-high; T-low, time at P-low; VF, ventilatory frequencyModeIndependent variablesInitiatedTerminatedFlowPressureVTVFCritical alarmsVCVVT, VF, I:E ratio, PEEP, Fio2ControlledVT achieved or alarm limit reachedConstant, decelerating in newer modesRises or rises and then fallsFixedFixedPpeak, Pe′co2PCVPinsp, VF, PEEP, Fio2ControlledInspiratory timeDeceleratingConstantVaries with complianceFixedVT, Pe′co2PSPinsp, PEEP, Fio2PatientOnce 25% peak flow reachedDeceleratingConstantDependent on compliance and driving pressurePatient-controlledVF, Pe′co2, VT, consider driving pressureSIMVCan be VCV or PCV, PEEP, Fio2Patient or controlled (synchronised with patient)VariableVariableVariableVariableVariableAs VCV/PCVPe′co2NIVPS, PEEP, Fio2PatientPatientPatient-dependentConstantDependent on compliance and driving pressurePatient controlVF, consider driving pressureAPRVP-high, P-low, T-high, T-low, PEEP, Fio2ControlledAfter T-highZero usually for 90% time dependent on T-highConstantFixedFixedVT, Pe′co2 Open table in a new tab Once a mode and alarm settings have been selected and the independent variables set, subsidiary settings and static pressure measurements can be taken. Subsidiary settings include inspiratory:expiratory ratio (I:E ratio) and if not in a controlled mode, ramp and trigger. I:E ratio is the ratio within one respiratory cycle between inspiratory and expiratory time. Therefore, if ventilatory frequency is set to 15 bpm, the respiratory cycle will be 4 s. If the I:E ratio is 1:3 then inspiration will take 1 s and expiration 3 s. This variable is important in patients with bronchoconstriction to help prevent breath stacking, DHI and their haemodynamic consequences. An expiratory hold will measure total PEEP (PEEPtot): if this is greater than the set PEEP then breath stacking may be occurring. PEEPtot may be greater than PEEP but be static over time and in this instance, the respiratory system is likely to have found a static state that is safe. However, if the PEEPtot continues to increase, the I:E ratio should be increased, for example to 1:4 or 1:5. The effect of this change is to shorten the inspiratory time and therefore in a VCV mode the Ppeak increases. In a PCV mode the VT would decrease. Although this would alter Ppeak and plateau pressure or decrease ventilation, the haemodynamic consequences are usually small and outweighed by alleviating the high PEEPtot. All changes should still deliver lung-protective ventilation. Pressure ramp percentage or time is the percentage or time of inspiratory time taken to reach maximum pressure/flow. This is of benefit in patient-initiated supported modes and can aid ventilator synchrony, especially when bronchoconstriction is present. Some home non-invasive ventilation machines have an expiratory ramp to aid flow dynamics in patients with chronic obstructive pulmonary disease. In clinical practice it has little effect in a mandatory mode and little effect on haemodynamics. In the age of goal-directed therapies it is appealing to consider the concept of ‘optimum PEEP’. There are several methods of determining optimum PEEP including adjusting PEEP based on Fio2 requirements, using the inflection point on the pressure-volume loop or measuring the static compliance, although the latter is impractical in clinical practice. The goal of optimum PEEP is to maximise oxygenation, and minimise both end-expiratory atelectasis and end-inspiratory overdistension. Measuring these goals is not currently possible nor is it possible to achieve all goals all of the time. Therefore, as is the case with the subject of this paper, compromise must be sought. An extension of optimum PEEP is the idea of an 'open lung strategy', where PEEP is used after a period of 'recruitment'; this concept is appealing but a recent trial in patients with moderate to severe ARDS indicated harm with the initial recruitment procedure (with a maximum PEEP of 45 cmH2O) and the protocol was modified following multiple incidences of cardiac arrest.24Cavalcanti A.B. Suzumura É.A. Laranjeira L.N. et al.Effect of lung recruitment and titrated positive end-expiratory pressure (PEEP) vs low PEEP on mortality in patients with acute respiratory distress syndrome: a randomized clinical trial.JAMA. 2017; 318: 1335-1345Crossref PubMed Scopus (412) Google Scholar Caution is advised in patients at risk of DHI and those with hypovolaemia of any cause: selection of patients is key. Close attention should be paid if recruitment manoeuvres are carried out and plans should be in place to correct profound change in arterial pressure. Given the potential for haemodynamic compromise, it is logical that optimum PEEP for recruitment may require additional drug therapies and therefore may not be the overall best PEEP for the patient. The haemodynamic effects of PPV are: reduced venous return from increased intrathoracic pressure; increased PVR causing increased RV stroke work; and decreased intrathoracic to extrathoracic aortic pressure gradient, which reduces LV afterload and LV stroke work. The net effect is a decreased cardiac output proportional to mean airway pressure.