Abstract
Ventilation during chest compressions can lead to an increase in peak inspiratory pressure. High inspiratory pressure can raise the risk of injury to the respiratory system and make it challenging to deliver the required tidal volume. The utilization of mechanical devices for chest compression has exacerbated this challenge. The aim of this narrative review was to summarize the different mechanical ventilation strategies applied during mechanical cardiopulmonary resuscitation (CPR). To this end, we searched the PubMed and BioMed Central databases from inception to January 2020, using the search terms “mechanical ventilation,” “cardiac arrest,” “cardiopulmonary resuscitation,” “mechanical cardiopulmonary resuscitation,” and their related terms. We included all studies (human clinical or animal-based research studies, as well as studies using simulation models) to explore the various ventilation settings during mechanical CPR. We identified 842 relevant articles on PubMed and 397 on BioMed Central; a total of 38 papers were judged to be specifically related to the subject of this review. Of this sample, 17 studies were conducted on animal models, 6 considered a simulated scenario, 13 were clinical studies (5 of which were retrospective), and 2 studies constituted literature review articles. The main finding arising from the assessment of these publications is that a high 
must be guaranteed during CPR. Low-grade evidence suggests turning off inspiratory triggering and applying PEEP ≥ 5 cm H2O. The analysis also revealed that many uncertainties persist regarding the ideal choice of ventilation mode, tidal volume, the ventilation rate setting, and the inspiratory:expiratory ratio. None of the current international guidelines indicate the “best” mechanical ventilation strategy to apply during mechanical CPR. We propose an operating algorithm worthy of future discussion and study. Future studies specifically addressing the topics covered in this review are required.
Introduction
International guidelines recommend ensuring effective ventilation for cardiac arrest patients;1,2 however, a “best” invasive ventilation strategy for cardiopulmonary resuscitation (CPR) has yet to be established. The European Resuscitation Council guidelines suggest a protective ventilation approach derived from the management of other types of critically ill patients, such as patients with ARDS or acute respiratory failure,2 but the conditions that arise during cardiopulmonary resuscitation can be very different from the clinical models investigated to date. Following cardiac arrest, the thoracic system's compliance declines, leading to an increase in pressure against the mobilized volumes.3 Furthermore, asynchronous ventilation during the delivery of chest compressions can increase the risk of a rise in peak inspiratory pressure.4 In addition to increasing the potential risk of injury to the respiratory system, it may also make it challenging to deliver the required tidal volume (VT). A recent retrospective study reported that nonsurvivors of cardiac arrest had received a higher mean plateau pressure and higher driving pressure, which suggests that ventilation plays a central role in determining survival following a cardiac arrest.5 The continued development of mechanical devices for chest compression and the rapid spread of their use underscore the uncertainty that continues to exist in this area. Although mechanical CPR has not been proven superior to manual CPR, the former seems to be useful, particularly when maintaining high-quality chest compressions is difficult (eg, during ambulance transport or a coronary angioplasty procedure).6 The adoption of mechanical chest compressions in clinical practice has led to problems in the management of invasive mechanical ventilation, and the guidelines have yet to address this issue adequately.
Our narrative review aims to summarize the different strategies of mechanical ventilation presented in the literature. In particular, we discuss the following topics: the choice of ventilation mode; the challenge of achieving the predetermined VT; the PEEP setting and ventilation rate; the role of the inspiratory-expiratory ratio and the most appropriate inspiratory trigger threshold; and adequate 
. Finally, future and advanced perspectives are addressed.
Search Strategy and Study Selection
We searched the MEDLINE (PubMed) and BioMed Central databases, from inception to January 2020, in accordance with the PRISMA guidelines, using following the search terms: “mechanical ventilation,” “cardiac arrest,” “cardiopulmonary resuscitation,” and “mechanical cardiopulmonary resuscitation,” as well as their possible variations or other closely related terms. We also searched actual citations of relevant primary and review articles. We included a wide range of study types, including those on human or animal models or simulation models, as well as literature review articles and observational studies. Studies conducted within the hospital setting as well as out-of-hospital cardiac arrest scenarios were included. We excluded studies that considered pediatric subjects only, mechanical ventilation after the return of spontaneous circulation, extracorporeal cardiopulmonary resuscitation, case reports, conference abstracts, and articles in languages other than English. Although we considered a broad spectrum of different study designs, our review attempts to classify the evidence collected: meta-analyses before randomized clinical trials, followed by observational studies, and finally animal or preclinical studies.
The articles were screened and read independently by three authors (DO, LV, NF). Each author made an independent judgment regarding the degree of relevance of the study in question. These judgments were compared, and a majority criterion was used to include the research in the review.
Data Extraction and Synthesis
The following data were extracted from the selected studies: year of publication, study design, ventilation variable(s) studied, clinical context, the aim of the study, measured parameters, and the main findings. Because the included studies were all very different from each other in terms of design and aims, we summarized the results in the form of a narrative review according to the following scheme: ventilation mode, VT, PEEP, ventilation rate set, inspiratory/expiratory time ratio, 
, and new research directions.
Literature Review
We identified 842 articles on PubMed and 397 on BioMed Central; of these, 38 were judged to be related to the subject of this review (see Fig. 1 and the search strategy section in the supplementary materials at http://www.rcjournal.com). Seventeen articles were conducted on animal models, 6 were considered a simulated scenario, 13 were clinical studies (5 of which were retrospective), and 2 constituted literature review articles.
Flow chart.
Ventilation Modes
The ventilation mode chosen during CPR is not irrelevant in determining the outcome. In pressure control continuous mandatory ventilation, the provider can control the pressure level applied, but this modality runs the risk of not achieving sufficient VT. By contrast, in volume control continuous mandatory ventilation, the volume delivered by the ventilator is established a priori, but this runs the risk of exceeding safe peak inspiratory pressure levels. The main problem in administering high inspiratory pressures is related to the risk of overdistention of the alveolar structures (ie, barotrauma).7 Depending on the mechanical ventilation settings used, asynchronous ventilation combined with chest compressions may result in high positive pressure and rapid changes in chest wall compliance secondary to chest compressions. Furthermore, it is not possible to preset the maximum pressure limit on all transport ventilators, so achieving the target volume can be challenging. Additionally, positive pressure mechanical ventilation causes some alterations in hemodynamic physiology, the most relevant of which is the reduction in venous return and, therefore, the reduction in ventricular preload.8,9
A recent survey reported that the most commonly used ventilation mode is volume control continuous mandatory ventilation.10 However, no clinical studies have yet defined the “best” mode of invasive ventilation to use during mechanical CPR. Only experimental bench simulation studies have been proposed to evaluate the effects of different mechanical ventilation modes during mechanical CPR. Speer and colleagues11 compared pressure modes with volumetric modes and established that both permitted adequate VT to be achieved without increasing the peak inspiratory pressure.
Some alternative strategies to administering oxygen with positive pressure without the risk of high peak pressures have been studied in the literature. These approaches are discussed in the section on future and advanced perspectives, given their limited diffusion in clinical practice.
Tidal Volume
The VT range most frequently cited as being administered in actual practice is 6–8 mL/kg10; however, it is essential to note that the studies contributing to the deduction of this range mainly concern protective ventilation in patients with ARDS. Indeed, the debate about the best VT remains very much open. For example, although the evidence correlating the magnitude of VT with the probability of spontaneous recovery of circulation tends to be weak,12 VT has been shown to correlate positively with the level of neurological recovery in the context of in-hospital cardiac arrests.13 Moreover, a reasonable VT value might never be established in cardiac arrest patients, especially if the patient is subjected to mechanical chest compressions. Mechanical chest compression devices can deliver compressions in “synchronous” mode (ie, 30 compressions alternating with 2 ventilations) or in “asynchronous” mode (ie, continuous compressions that are not synchronized with ventilation). The latter method allows the pauses between compressions to be minimized, and the reduction in the pause time between one compression and another is a known factor associated with better survival.1,2 However, how this mode affects the delivery of an adequate VT is uncertain. In fact, a recent retrospective study reported that the asynchronous mode, and not the synchronous mode, correlated with a higher survival rate.14 However, it is difficult to determine whether this correlation is related to more effective chest compression or to better ventilation in terms of VT. Moreover, in the survey conducted by Cordioli et al,10 the most frequently observed complications of invasive mechanical ventilation during cardiopulmonary resuscitation were activation of the high-pressure alarm and delivery of an insufficient VT. Thus, the broad range of possible intrathoracic pressure changes and the compliance of the respiratory system make it difficult to predict the behavior of a cardiac arrest patient's ventilatory system during chest compressions.10
Moreover, the only direct evidence available has been derived from animal studies. One recent study reported favorable consequences following the use an ultra-low VT (ie, 2–3 mL/kg) in terms of both adequate ventilation and a reduced risk of iatrogenic damage.15 By contrast, another study demonstrated a high VT ventilation strategy (ie, 10 mL/kg vs 7 mL/kg) to increase the probability of return of spontaneous circulation.12 Further studies are required to define the most appropriate VT to apply in different clinical contexts.
PEEP
The application of PEEP can bring about different effects. First of all, PEEP is known to improve oxygenation by increasing the VT and keeping the alveoli open (the so-called “open the lung and keep it open” concept).16 A recent Canadian study on cadavers showed that changes in intrathoracic pressure are related to the PEEP levels applied rather than the inspiratory pressure (at least within certain limits). Moreover, the authors were able to establish that while intrathoracic pressure is generally stable during CPR, the pressure in the airway can vary, even dropping below the alveoli closure limit. This effect has important implications in terms of oxygenation during cardiac arrest.17 The same group also reported that end-tidal CO2 influences proper alveolar ventilation, as it decreases in cases of alveoli closure despite a satisfactory hemodynamic effect of chest compressions.18 On the other hand, PEEP increases the risk of dynamic hyperinflation (and is associated with important hemodynamic effects); this can, in turn, cause a reduction in venous return and, therefore, cardiac output (under conditions of preload-dependent cardiac stroke).19
As long ago as 1980, Babbs and co workers noted that applying positive airway pressure during chest compressions increased oxygenation without deteriorating cardiac function.20 Data gathered from animal models seem to point in the same direction: the application of PEEP brings about an improvement in survival independently of other parameters.21,22 Considering the actual evidence, application of at least 5 cm H2O PEEP seems to be beneficial; however, the optimal PEEP still needs to be adequately investigated. Excessive PEEP may worsen the outcome of cardiopulmonary resuscitation, but more evidence is required to establish the validity of this statement. For instance, Van der Touw et al23 pointed out that, during conditions of hyperinflation, the increase in intrathoracic pressure resulting from chest compressions reduces the cardiac output and the mean arterial pressure.
Ventilation Rate
In a small case study, Maertens et al24 reported that subjects in cardiac arrest, even when intubated, are ventilated at a higher frequency than prescribed by the guidelines (ie, 10–12 breaths/min). Vissers and colleagues25 systematically reviewed the literature to investigate whether the optimal set ventilation rate during cardiopulmonary resuscitation was indeed ∼ 10 breaths/min. Their results were inconclusive, as was the issue of whether rates lower or higher than 10 breaths/min are able to influence outcomes (Table 1). Note that rates < 10 breaths/min run the risk of not achieving the target minute volume, whereas higher rates are more likely to cause dynamic hyperinflation and bring about a deterioration in hemodynamic parameters. In an animal model, a high ventilation rate was associated with a reduction in coronary perfusion.26 However, a recent prospective observational study reported that subjects who reached the return of spontaneous circulation received faster ventilation compared to subjects who did not get the return of spontaneous circulation.14 The effect caused by the ventilation frequency is probably indirect and related to changes in the patient's intrathoracic pressure, volume state and the normal range for the patient's body structure. However, studies specifically aimed at this issue are required to obtain clearer data.
Summary of the Reviewed Studies
Inspiratory-Expiratory Ratio
The relationship between inspiratory time and expiratory time is fundamental for the complete replacement of the respiratory system's anatomical deadspace. If the expiratory time is not long enough, the phenomenon of dynamic hyperinflation can occur, causing an increase in intrinsic PEEP. The mechanisms through which this generates hemodynamic impairments have been clearly demonstrated in the literature.27,37 Fitz-Clarke28 highlighted the relationship between target VT and the duration of the inspiratory phase by means of a physiological model. The model showed that the length of the inspiratory phase correlates inversely with the pressure regime, such that an inspiratory time that was too short could result in gastric insufflation. However, this study was conducted using an unprotected airway model. Von Goedecke et al,29 considering a bag-mask ventilation simulation model, assessed the possibility of reducing the inspiratory time from 2 s to 1 s. They reported that, although the target VT was continuously met, it was detrimental to peak inspiratory pressure, which increased.29 However, the comparability of this simulated model to invasive mechanical ventilation of a patient during mechanical CPR is debatable. No clinical studies have investigated the inspiratory-expiratory ratio in the invasive ventilated patient during cardiac arrest.
Inspiratory Trigger
One of the most frequently encountered problems during mechanical cardiopulmonary resuscitation is the auto-triggering or the inappropriate activation of ventilator delivery due to the incorrect setting of the ventilator's inspiratory trigger.38 Ventilation with pressure- or flow-triggering can lead to hyperventilation and deteriorating gas exchange and hemodynamics during CPR. Indeed, the results obtained from small animal model studies have encouraged physicians to turn the inspiratory trigger off during CPR (or to increase the threshold to at least 20 cm H2O).30
FIO2
Guidelines regarding cardiac arrest patients usually recommend 
close to 1.0 to improve oxygen delivery.2 However, this physiological assumption has never been proven. Furthermore, the harmful role of hyperoxemia in the development of post-cardiac arrest syndrome is being increasingly acknowledged.39-41 For example, a 2017 study using an animal model investigated whether 
significantly < 1.0 (in this case, 0.50) would permit comparable cerebral oxygenation and reduce mitochondrial oxidative stress.31 The results obtained were not straightforward: on the one hand, a 
of 0.50 resulted in a reduction in cerebral oximetry values compared with those achieved with 
values nearing 1.0; on the other hand, invasive methods of measuring cerebral oxygenation showed no significant differences31 (Table 1). However, how hyper- and hypoxemia affect survival and neurological outcomes is controversial.32,33,42 A recent meta-analysis, which pooled the data of the only 2 clinical studies present in the literature,34,35 concluded that whereas hyperoxemia in the post-arrest period is associated with a worse outcome, during CPR it appears to be related to a higher rate of return of spontaneous circulation.36 The effects induced by hyperoxemia seem to correlate with the timing of the pathological process, rather than a simple on/off effect.43,44
Future and Advanced Perspectives
This review has focused on the problems primarily encountered when invasively ventilating a patient in cardiac arrest with mechanical CPR. However, other strategies that permit respiratory homeostasis to be maintained in cardiac arrest patients subjected to mechanical ventilation have been explored in the context of clinical research (Table 2). Data in the literature suggest that ventilation involving air solely mobilized by the mechanical chest compressor is not sufficient to meet the organism's metabolic needs. Deakin and colleagues45 observed that the volume of air mobilized by an automatic device for chest compressions was ∼ 40 mL/breath. This value cannot meet the metabolic needs of an adult patient of standard size; indeed, the eliminated CO2 quota was well below the regular quota (only 20 mL/min were observed vs the standard value of > 150 mL/min).
Summary of the Reviewed Studies for Future and Advanced Perspectives
Some research groups have explored the possibility of not ventilating cardiac arrest patients at all but only oxygenating them; this is known as “apneic oxygenation,” which exploits the displacement of air caused by the chest compressions themselves.46,58 Both animal and clinical studies seem to demonstrate the feasibility of this oxygenation strategy.47,48 Steen et al,49 using an animal model, reported a higher coronary perfusion pressure in the group of animals treated with continuous passive oxygenation compared with the group treated with standard intermittent ventilation. In the early 2000s, Saïssy et al50 conducted a clinical study involving subjects in out-of-hospital cardiac arrest. The authors noted better oxygenation levels and significantly more CO2 elimination in the group treated with passive oxygenation alone.50 However, the size of the study was not adequate to detect a statistically significant difference in terms of survival rate. Bobrow and colleagues,51 in their case history, reported that passive oxygen insufflation is superior to bag-valve-mask ventilation in terms of survival outcome and hospital discharge with preserved neurological status. Bertrand and colleagues52 found the application of a passive oxygen flow of ∼ 15 L/min to be associated with the same survival rates as obtained with conventional invasive ventilation.
It is essential to distinguish between passive oxygen insufflation, achieved mainly using dedicated devices (such as the Boussignac tube), and pure apneic oxygenation, in which the complete denitrogenation of the alveoli is sought through the administration of pressure flow (usually 20 cm H2O). From a theoretical point of view, this distinction is essential for determining the risk of atelectasis from oxygen reabsorption; however, from a practical point of view (at least in relation to the studies conducted on animal models), no substantial difference has been identified between the 2 methods in terms of side effects, or between these methods and invasive ventilation performed at zero PEEP.9,53,54
Although, from a theoretical point of view, the application of PEEP seems to be beneficial for achieving an adequate oxygenation target (see the section on PEEP), other clinical studies have focused on more specifically adapting ventilation to the patient, even during the inspiratory phase. Indeed, the risk of barotrauma during ventilation occurring simultaneously with chest compressions is well documented in the literature.59 The arrival of new portable ventilator devices in the market capable of activating flow delivery through reverse inspiratory triggers (which work when the airway pressure rises fast enough above an absolute threshold after a minimal time of expiration) has opened a new line of research into intermittent ventilation synchronized with chest compressions. Animal studies have shown greater efficiency in terms of achieving (and maintaining) certain threshold levels of 
, with beneficial effects also on the hemodynamic conditions (eg, maintenance of adequate mean arterial pressure) and acid/base homeostasis (Table 1).55,56 However, as previously mentioned, the studies conducted to date have only involved animal models. Thus, no conclusions can yet be drawn about such devices in terms of achieving return of spontaneous circulation or better neurological performance.
Therefore, 2 opposing strategies exist: passive oxygenation versus the synchronization of ventilation with chest compressions. Until now, the only study to compare these 2 strategies in an animal model is that by Soltsez and colleagues.57 They found a higher coronary perfusion pressure and compression phase aortic pressure when positive-pressure ventilation was applied.
Limitations
The priority of this review was to consider all of the literature assessing mechanical ventilation during mechanical CPR. In doing so, we have been largely inclusive, as demonstrated by the wide variety of study designs involved. Most investigations are preclinical studies (animals or laboratory models), so their direct relevance to clinical practice is limited. In addition, the literature contains controversial results regarding the need for ventilation during cardiopulmonary resuscitation.60-64 Indeed, doubts persist regarding the need to intubate cardiac arrest patients at all65-68; although a discussion of this issue is beyond the scope of our review, we can affirm that the benefits of protecting the airways and mechanical ventilation continue to be widely recognized and prioritized.69,70 This review has only addressed the evidence regarding mechanical ventilation during mechanical CPR. Any evidence that relates to data gathered outside this time frame (including after return of spontaneous circulation) has not been covered. The reader is instead referred to the recent publication by Holmberg and colleagues,71 which discusses oxygenation and ventilation targets after cardiac arrest.
Summary
Research into the optimal way to ventilate a patient in cardiac arrest using mechanical chest compressions is ongoing. At present, very few clinical studies have explored the best ventilation strategy for cardiac arrest patients during mechanical CPR. According to the evidence published to date, a high 
during CPR must be guaranteed in these patients. Low-grade evidence suggests deactivating the inspiratory trigger and applying PEEP (at least 5 cm H2O). Uncertainties remain about the ideal ventilatory mode, VT, the ventilation rate setting, and the inspiratory-expiratory ratio. Current international guidelines do not provide any indications about the “best” mechanical ventilation strategy to use during mechanical CPR. Here, we put forward an operating algorithm based on the current state of knowledge (Fig. 2). Studies specifically addressing the topics covered in this review would be required to investigate its validity.
Proposed operating algorithm for invasive mechanical ventilation during mechanical CPR. CPR = cardiopulmonary resuscitation; VT = tidal volume; I:E = inspiratory-expiratory ratio; ROSC = return of spontaneous circulation; 
= end-tidal carbon dioxide pressure.
Footnotes
- Correspondence: Luigi Vetrugno, Department of Anesthesia and Intensive Care Clinic, ASUFC University Hospital “Santa Maria della Misericordia,” p. le S. Maria della Misericordia 15, 33100 Udine, Italy. E-mail: luigi.vetrugno{at}asufc.sanita.fvg.it
Supplementary material related to this paper is available at http://www.rcjournal.com.
The authors have disclosed no conflicts of interest.
- Copyright © 2021 by Daedalus Enterprises
