Pleural pressure is transmitted to the pericardium and therefore any increase in Pes due to passive ventilation changes the pericardial pressure and, consequently, the right heart transmural pressure. Venous return is driven by the gradient between mean circulatory filling pressure (Pmcf) and the right atrial pressure (Pra), which is usually only 4–8 mmHg (Guyton et al., 1957). During positive pressure ventilation, increased Pes can impact Pra and therefore venous return and cardiac output in two ways: during the respiratory cycle (increase in Pes related to tidal volume) or at the end of expiration (increase in baseline Pes by application of PEEP). Higher tidal volumes and PEEP will increase Pra, therefore reducing venous return, the end-diastolic volume and the stroke volume of the right ventricle.

Nevertheless, the negative effects of increased Pes are not linear and subject to compensatory mechanisms. Studies found that, in passively ventilated patients, higher PEEP and therefore higher end-expiratory Pes, can increase both Pra and Pmcf, without impacting the gradient for venous return (Van Den Berg et al., 2002). This could be related to the downward displacement of the diaphragm induced by PEEP, which could increase abdominal pressure and thus Pmcf, or to the presence of feedback by baroceptors that increase systemic venous tone (Fessler et al., 1991).

The pulmonary circulation is affected by changes in lung volume and thus in P_L. Lung inflation, which is associated with an increase in P_L, has opposing effects on extra-alveolar and alveolar capillaries, expanding the first and compressing the latter. Thus, passive inspiration is associated with collapse of alveolar capillaries in the middle vascular segment (Marini et al., 2003).

In 1965, West described the interaction between pulmonary artery pressure (PAP), alveolar pressure (PA) and pulmonary venous pressure (Pv) identifying 3 zones with different driving pressure for blood flow and approximating its behavior to a Starling resistor (West and Dollery, 1965). A Starling resistor can be imagined as a compressible tube inside a box where flow is possible only when the entrance pressure (PAP) is higher than the box pressure (PA) and of the output pressure (Pv) (Permutt et al., 1962). In zone 1, PA is higher than PAP and therefore flow is zero; in zone 2, PAP is higher than both PA and Pv, flow is preserved but dependent on PA; in zone 3, PA is lower than both PAP and Pv, flow is present and independent of PA. In the presence of high lung elastance, elevated P_L can increase the amount of zone 1 and 2 (non-zone 3) conditions, thereby increasing right ventricular afterload during inspiration (Slobod et al., 2022). Although the normal RV is well adapted to handle cyclic small increases in afterload (Magder, 2021), many patients admitted to the ICU have RV contractile dysfunction or limitation of filling and cyclic tidal increases in RV afterload can have dramatic hemodynamic consequence (Magder et al., 2023).

Both high Ppl and P_L affect RV performance, but there is no consensus on which has the greater impact and computational modeling showed inconclusive results (Magder and Guerard, 2012). During controlled mechanical ventilation, the inspiratory increase in Ppl is responsible for the decreased preload effect, while the increase in P_L is associated with an increased afterload effect (Slobod et al., 2022) (Repessé et al., 2018) (Table 1). Both these effects generate reversed pulsus parodoxus (Massumi et al., 1973). In experimental models (Webb and Tierney, 1974) (Hotchkiss et al., 2001) (Katira et al., 2017) of severe pulmonary edema due to extremely high driving pressure (Webb and Tierney, 1974), right heart failure developed (Katira et al., 2017), likely because of combined preload and afterload effects, causing dramatic oscillations of pulmonary vascular resistance and blood flow with each breath.

Positive pressure ventilation can impact LV performance through both direct and indirect effects. Indirectly, the effects of increased Pes and P_L on RV preload and afterload can reduce LV preload and stroke volume (Corp et al., 2021). RV dilation caused by an increase in afterload will also impact LV performance via ventricular interdependence (Weber et al., 1981).

Higher pleural pressure can also directly affect LV outflow. An increase in Pes simultaneously acts on the LV and aorta resulting in higher LV wall compression. Despite unchanged ventricular-aortic gradient, the absolute ventricular and aortic pressure will increase, as well as the gradient between thorax and peripheral vessel, leading to improved cardiac output. Moreover, the increased aortic wall pressure triggers autoregulation by peripheral baroreceptors causing a reduction of systemic vascular resistance and LV afterload (Rankin et al., 1982), further improving cardiac output. Finally higher Pes by positive pressure ventilation reduces the transmural LV gradient, decreasing oxygen consumption for the same absolute arterial pressure, thus protecting the stressed LV.

In summary, sustained increase of Pes during passive ventilation can improve or depress cardiac output (Table 1), depending on the patient’s characteristics. In patients with RV failure or preload dependence, cardiac output will decrease. On the contrary, in afterload dependent states (e.g., LV failure), higher pleural pressure can improve LV function and cardiac output (Alviar et al., 2018).

Another important aspect to be considered is the duration of exposure to a change in pleural pressure. Different components that play a role in influencing cardiac output do not have the same kinetics in response to sustained pleural pressure change. For example, it is well known that left ventricular outflow is affected by an inspiratory hold (Valsalva maneuver) only when pleural pressure is increased for a relevant length of time (e.g., >10 s) (Gavelli et al., 2019).

Finally, the effect of intrathoracic pressures on LV may not automatically influence the RV. Computational models (Magder and Guerard, 2012) showed that pulmonary vessels can act as a buffer, cyclically increasing and decreasing their blood content. This phenomenon may be dampened by large change in pleural pressure since increased transpulmonary pressure impacts transmural vascular pressure and reduce the capacitance for lung blood volume.

Increased pleural pressure due to PEEP during controlled mechanical ventilation may impact perfusion of distal organs, such as brain, liver and kidneys. For example, experimental studies showed that intrathoracic pressure can increase intracranial pressure (ICP), but the effect also depends on the relative elastances of the chest-wall and the lung. In conditions of high chest wall elastance, an increase of intrathoracic pressure may increase ICP (Chen et al., 2018). Increase in intrathoracic pressure has been shown to correlate also with a decrease in renal plasma flow, glomerular filtration rate (GFR) and urine output during mechanical ventilation. This is associated to a reduction of cardiac output (Annat et al., 1983) and an increase of the venous and parenchymal pressures. Further studies are needed to investigate the relative impact of pleural and/or transpulmonary pressure on distal organ function.

During an inspiratory effort, the Pes surrounding the RV falls, causing an increase in transmural pressure. This does not increase RV afterload because the downstream vascular bed into which the RV ejects (pulmonary arteries) is subject to the same change in pleural pressure. Moreover, the inspiratory fall in Pes increases the pressure difference between the Pmcf in the upstream systemic veins and the Pra (Magder, 2018). This favors venous return and will augment stroke volume if the right ventricle has not reached a state of limitation (Magder, 2018; Magder et al., 2023). Indeed, spontaneous breathing during mechanical ventilation has been associated with an increase in RV end diastolic volume and cardiac output, suggesting that preload is increased compared to passive ventilation (Rankin et al., 1982) (Masmoudi et al., 2017).

Compared to passive ventilation, it must be emphasized that although Pes falls during the inspiratory phase of a spontaneous breath, P_L must increase to generate lung inflation. While the “preload effect” of changes in Ppl has differing consequences during passive ventilation (negative) and spontaneous breathing (beneficial), the cardiovascular effects of the inspiratory increase in P_L are similar, regardless of whether inspiratory efforts are present or not. Therefore, the increase in non-zone 3 conditions also occurs during spontaneous breathing, leading to increased afterload opposing RV ejection (Figure 1) (Hughes et al., 1968) (Brower et al., 1985) (Mekontso Dessap and Vieillard-Baron, 2022). This relationship between increasing inspiratory P_L and RV afterload has been demonstrated to be linear in patients with and without lung injury (Slobod et al., 2022) (Vieillard-baron et al., 1999).

The net effect of spontaneous breathing on cardiac output will again reflect the balance of these changes in RV preload and afterload. This balance will ultimately depend on the relative inspiratory change in Pes vs. P_L and on lung and chest wall mechanics, tidal volume, inspiratory effort, and volume status (Spinelli et al., 2020) (Repessé et al., 2018). Vigorous inspiratory effort in the setting of increased lung elastance results in increased P_L that will increase RV afterload due to increased non-zone 3 conditions. Moreover, unlike passive ventilation, large deflections in pleural pressure will augment venous return (Table 1). During subsequent RV ejections, increased stroke volume will be ejected into a pulmonary vasculature that is more extensively in non-zone 3, opposing RV ejection. In this way, vigorous efforts, coupled with low lung compliance, may increase shear forces and inflammation within the pulmonary vasculature.

Unlike for the RV, decreases in Pes directly increase LV afterload due to the increase in transmural ventricular pressure. The inspiratory decrease in Pes represents an additional load that the LV must overcome to eject blood through the aortic valve. Increased inspiratory effort can thereby increase left ventricular end-diastolic pressure and facilitate cardiogenic pulmonary edema. Furthermore, a significant inspiratory reduction in Pes lowers the pressure surrounding most of the pulmonary vasculature, causing an increase in the transvascular pressure across the pulmonary capillary endothelium, worsening the risk of hydrostatic edema. Given that Pes deflections also increase venous return that can cyclically increase pulmonary perfusion and intravascular pressure, spontaneous breathing with increased inspiratory effort can favor fluid extravasation from pulmonary vessels, causing pulmonary edema (Spinelli et al., 2020) (Repessé et al., 2018).

In patients with ARDS, respiratory drive is frequently increased due to several mechanisms (Brochard et al., 2017), and vigorous inspiratory efforts can perpetuate a vicious cycle of impaired gas exchange, increased respiratory drive, and worsening lung edema. This process has been termed patient self-inflicted lung injury (Spinelli et al., 2020). Another clinical consequence of adverse heart-lung interactions in the context of high respiratory effort is “weaning induced pulmonary edema” for patients failing spontaneous breathing trials or extubation (Lhéritier et al., 2013; Liu et al., 2016).