Among mechanically ventilated patients admitted to an intensive care unit after an out-of-hospital cardiac arrest (OHCA), less than 40% will survive to hospital discharge [1]. This high mortality rate is in part attributable to the multi-organ dysfunction caused by post-reperfusion syndrome [2]. Up to 50% of patients successfully resuscitated from cardiac arrest develop lung injury, fulfilling acute respiratory distress syndrome (ARDS) criteria during the stay in the intensive care unit (ICU) [3]. Lung injury occurs because of the systemic inflammation caused by the post-reperfusion syndrome and as a direct consequence of chest compression-induced lung damage [4]. In ARDS patients, limiting tidal volumes (VT), plateau and driving pressure (ΔP) represents the mainstay of respiratory support management, with the aim of reducing ventilator-induced lung injury (VILI) [5, 6]. In mechanically ventilated patients, VILI is caused by excessive stress and strain in the aerated lung, the volume of which is markedly reduced by alveolar flooding, edema and inflammation. This releases cytokines (biotrauma) that lead to multi-organ dysfunction, the most frequent cause of death in patients with ARDS [7].

Data on the ventilator management of critically ill patients resuscitated from cardiac arrest are scarce and limited to observational studies: high VT (8 ml/kg of predicted body weight on average) and ΔP (15 cmH2O on average) are typical, presumably because of a perceived need for a tight CO2 control [8]. However, there is some evidence that VT lower than 8 ml/kg is associated with improved functional outcome among OHCA patients [9], although an observational study of in-hospital cardiac arrest patients showed no such association [10]. Thus, low-certainty data advocate a possible need for strategies to prevent VILI in critically patients who receive mechanical ventilation after resuscitation from cardiac arrest.

Since the beginning of the low-VT, low-plateau pressure era, research has been focusing on identifying the optimal ventilatory settings to prevent VILI during ARDS. The ΔP represents VT normalized to respiratory system compliance (a good estimate of aerated lung size) rather than to predicted body weight: ΔP can be easily calculated at the bedside as plateau pressure minus positive end-expiratory pressure (PEEP) and is the final mediator of the effect of VT-lowering on clinical outcome in ARDS [11, 12]. The ΔP represents a good estimate of lung stress (increase in pressure) and strain (lung deformation) induced by each breath, but does not take into account respiratory rate, which can itself contribute to VILI [13]. This aspect is of utmost clinical relevance, as lowering VT to reduce the ΔP requires increases in respiratory rate to maintain a constant PaCO2. For these reasons, mechanical power, which is a more inclusive formula incorporating VT, flow, respiratory rate, airway pressure, respiratory system resistive and elastic properties and PEEP has been proposed to best assess the risk of VILI at the bedside [14]. Subsequently a pooled analysis of 4549 ARDS patients was conducted to discriminate the respective roles of ΔP, respiratory rate and mechanical power in determining the outcome of ARDS patients. High mechanical power was strongly associated with mortality, but ΔP and respiratory rate combined in a simpler model based on the formula (4*ΔP + respiratory rate) were at least as informative [15].

In this issue of the journal, Robba and coworkers [16] report the results a pre-planned secondary analysis of a large trial investigating targeted temperature management in critically ill patients resuscitated from OHCA, which aimed to assess the association between ventilatory variables and clinical outcome (6-month mortality and neurological status). They studied 1848 patients and showed that respiratory rate, ΔP and mechanical power were independently associated with 6-month mortality; respiratory rate and ΔP were also independently associated with poor neurological outcome, and their combination in the formula (4*ΔP + respiratory rate) had the strongest association with both 6-month mortality and poor neurological outcome. With a rigorous and sound methodology, this is the largest investigation addressing the effect of ventilatory settings on the clinical and functional outcome of cardiac arrest survivors.

These results have relevant clinical implications.

First, they indirectly suggest that VILI contributes to the outcome of patients resuscitated from cardiac arrest, which would warrant interventions to enhance lung protection. This task may be particularly challenging in this population, given the possible need to maintain specific oxygenation and PaCO2 targets to avoid secondary brain damage. This represents a call for further prospective studies on the topic, especially because most recent guidelines on post-resuscitation care do not provide evidence-based recommendations on the optimal ventilator settings to apply after cardiac arrest, given the paucity of available data [17].

Finally, this is the first study outside the context of ARDS demonstrating the relevance of the formula (4*ΔP + respiratory rate), which was associated with both mortality and poor neurological outcome. This formula can easily guide ventilator settings at the bedside: clinically, reducing VT to lower ΔP by 1 cmH2O is worthwhile only if the PaCO2 can be kept constant by increasing the respiratory rate by less than 4. Conversely, it may be worthwhile reducing the respiratory rate by four breaths if the increase in VT needed to maintain a constant PaCO2 results in an increase in ΔP less than 1 cmH2O. For a given patient condition (PaCO2, VCO2, and respiratory system compliance) there is an optimal value of (4*ΔP + respiratory rate) that can be obtained through a simple nomogram (Fig. 1).

Fig. 1
figure 1

Impact of changing tidal volume and respiratory rate (not shown) on the variable (4*ΔP + respiratory rate) under simulated isocapnic conditions using three different compliances of the respiratory system. The black dot indicates average values of ΔP and respiratory rate in the TTM-2 trial. As tidal volumes approach the dead space, the efficiency of ventilation decreases with disproportional increases in respiratory rate being required to compensate for the decrease in tidal volume. As a result, under these isocapnic conditions, there is always a combination of ΔP and respiratory rate that minimizes the variable (4*ΔP + respiratory rate). Note that the location of this minimum depends on the compliance of the respiratory system with higher compliances associated with minimum values located at higher tidal volumes. For further details, please refer to [15]. ΔP: driving pressure

Full size image

As the authors have indicated, causal inference is limited due to the observational nature of the study. The associations found could alternatively be due to reverse causality or residual confounding. Although not reported in the primary TTM-2 manuscript, the presumed cause of death in the TTM-1 study was cerebral in 58%, cardiovascular in 24% and multiple organ failure in just 12% of cases [18]. The distribution of these causes of death is likely similar in the TTM-2 study. Thus, the influence of biotrauma-induced multiple organ failure on outcome overall may be limited.

The authors must be commended for conducting this important investigation. Studies such as the one performed by Robba and coworkers represent a step ahead in the understanding of the complex physiology underlying the role of mechanical ventilation in critically ill patients, including those resuscitated from cardiac arrest.