Should We Use Driving Pressure To Set Tidal Volume? PDF
Should We Use Driving Pressure To Set Tidal Volume? PDF
Should We Use Driving Pressure To Set Tidal Volume? PDF
CURRENT
OPINION Should we use driving pressure to set
tidal volume?
Domenico L. Grieco a,b,c,d, Lu Chen a,b, Martin Dres a,b,e,f,
and Laurent Brochard a,b
Purpose of review
Ventilator-induced lung injury (VILI) can occur despite use of tidal volume (VT) limited to 6 ml/kg of
predicted body weight, especially in patients with a smaller aerated compartment (i.e. the baby lung) in
which, indeed, tidal ventilation takes place. Because respiratory system static compliance (CRS) is mostly
affected by the volume of the baby lung, the ratio VT/CRS (i.e. the driving pressure, DP) may potentially
help tailoring interventions on VT setting.
Recent findings
Driving pressure is the ventilatory variable most strongly associated with changes in survival and has been
shown to be the key mediator of the effects of mechanical ventilation on outcome in the acute respiratory
distress syndrome. Observational data suggest an increased risk of death for patients with DP more than
14 cmH2O, but a well tolerated threshold for this parameter has yet to be identified. Prone position along
with simple ventilatory adjustments to facilitate CO2 clearance may help reduce DP in isocapnic conditions.
The safety and feasibility of low-flow extracorporeal CO2 removal in enhancing further reduction in VT and
DP are currently being investigated.
Summary
Driving pressure is a bedside available parameter that may help identify patients prone to develop VILI and
at increased risk of death. No study had prospectively evaluated whether interventions on DP may provide
a relevant clinical benefit, but it appears physiologically sound to try titrating VT to minimize DP, especially
when it is higher than 14 cmH2O and when it has minimal costs in terms of CO2 clearance.
Keywords
extracorporeal CO2 removal, plateau pressure, respiratory mechanics, stress and strain, tidal volume,
ventilator-induced lung injury
In the present manuscript we will discuss the easily available from large datasets [8 ]. It is also
physiological meaning of DP and its possible appli- important to keep in mind that in some situations
cation in titrating VT in patients with ARDS. like airway closure, the airway pressure may not
represent alveolar pressure [10]. We do not know
the prevalence of this problem in ARDS, but it has
PHYSIOLOGIC MEANING OF DRIVING been well described in obese patients. Finally, during
PRESSURE pressure controlled ventilation, the peak pressure is
Because the ratio of respiratory system compliance frequently used as a surrogate for the plateau pres-
(CRS) to the healthy lung available for ventilation sure, which, however, is only an approximation [11].
1070-5295 Copyright 2017 Wolters Kluwer Health, Inc. All rights reserved. www.co-criticalcare.com 39
&
[17 ,1820]. The use of partially assisted mechan- VILI at the beginning of ARDS. Moreover, it can be
ical ventilation is frequent in the recovery phase of considered as simple and bedside tool to reliably
ARDS, but very few data thoroughly described assess the effectiveness of interventions and to
respiratory mechanics in spontaneously breathing monitor the course of the disease.
patients with ARDS. It seems physiologically reasonable to hypoth-
Notwithstanding that airway pressure during esize that strategies to limit DP may provide a
assisted mechanical ventilation is usually lower than relevant clinical benefit, but no study has prospec-
during controlled mechanical ventilation, dynamic tively assessed whether systematic interventions
transpulmonary pressure (PLdyn), defined as the titrated to DP reduction improve clinical outcome.
swing in transpulmonary pressure during inspiration Nevertheless, it must be noted that evidence con-
and computed as the difference between airway and cerning a safe level of DP to achieve when adjusting
esophageal pressure, may reach very high values ventilator settings is lacking: currently, limiting DP
because of intense inspiratory effort [21]. to values equal or lower than 14 cmH2O seems to be
&&
However, being measured when flow is not zero, the wisest approach [1 ].
PLdyn reflects not only the elastic but also the resistive
properties (because of airway resistance) of the respir-
atory system. In addition, airway resistance signifi- DRIVING PRESSURE LIMITING DURING
cantly varies with flow, making difficult to assess to CONTROLLED MECHANICAL VENTILATION
what extent PLdyn reflects changes DP or DPL [22]. Different strategies can be used to limit DP during
Georgopoulos et al. reported the results of a study ARDS. As suggested by Amato, PEEP setting can
comparing DP during controlled and proportional significantly modify DP, as it affects the amount
&&
assist ventilation in a mixed cohort of intubated of aerated lung and hence CRS [8 ]; however, this
&&
patients [23 ], a ventilator mode that continuously topic goes beyond the purposes of the present
measures CRS and thus allows the calculation of DP. manuscript and will not be discussed further.
The authors observed that critically ill patients during
spontaneous breathing controlled DP by sizing the VT Prone position
to individual respiratory system compliance. Thus, Prone positioning has been convincingly shown to
DP was similar during control and assisted mechan- improve survival of patients with ARDS. Changes in
ical ventilation and mostly kept below 15 cmH2O, both lung and chest wall mechanics contributing to
whereas VT was not. Interestingly, the authors a more uniform gas insufflation have been addressed
suggested DP as a possible target of feedback mech- as possible mechanism [24,25]. Cornejo et al. [26]
anisms aiming at limiting lung injury. Whether this is showed that, when high PEEP is applied, prone
true in the specific subgroup of patients with ARDS position may reduce tidal hyperinflation, alveolar
needs confirmation. cyclic recruitment/derecruitment and slightly
Bellani et al. recently showed the feasibility of decrease DP, leading to the idea that changes in
Pplat measurement (2-s inspiratory hold, aiming to DP may contribute to the effects of prone position
obtain a period of no muscle activity) and reported on survival.
the behavior of alveolar pressure both during con-
trolled and assisted mechanical ventilation [22]. Muscle paralysis
With the same PEEP applied, no difference was Muscle paralysis in the early phase of the disease has
found in airway and lung Pplat, CRS, CL, and VT been shown to improve patients outcome [27]. The
between controlled and assisted ventilation at mechanism hypothesized to explain this evidence is
similar volumes and flow, indicating that both DP a lower transpulmonary pressure during muscle
and DPL are similar in the two conditions. The study paralysis, along with improved patientventilator
was conducted in patients with mild severity interaction, thanks to the avoidance of high-strain
(mean PaO2/FiO2 ratio of 224 mmHg, mean CRS of double cycled breaths or other dyssynchrony [28].
43 ml/cmHO). Given the feasibility of Pplat measure- Whether this may be reflected by changes in DP or
ment during pressure support ventilation, further DPL is unknown, but sedation and paralysis remain a
studies are warranted to investigate the behaviour of crucial instrument to enhance efficient and rigorous
DP in spontaneously breathing patients with ARDS. protective and ultra-protective ventilation in the
very early phase of the disease.
1070-5295 Copyright 2017 Wolters Kluwer Health, Inc. All rights reserved. www.co-criticalcare.com 41
available procedures leading to lower dead space can Extracorporeal membrane oxygenation
allow reducing VT and DP in isocapnic conditions. Extracorporeal membrane oxygenation (ECMO) is
Heated humidifiers, as compared to heat and moist- increasingly being used as a rescue therapy for
ure exchangers, decrease instrumental dead space patients with most severe oxygenation impairment.
and improve CO2 clearance. Moran et al. conducted Theoretically, given that ECMO allows oxygenation
a crossover study showing that heated humidifiers along with full extracorporeal CO2 clearance, the
allow to reduce VT from 7.3 to 6.1 ml/kg PBW and ventilatory approach should aim at minimizing
Pplat from 25 to 21 cmH2O without CO2 changes the risk of VILI without the need of providing any
[29]. Because PEEP was stable (average value CO2 washout.
9 cmH2O) over the entire course of the study, we However, ventilator settings during ECMO are
may hypothesize that heated humidifiers may lead still matter of debate and the management signifi-
to a decrease in DP (i.e. from 16 to 12 cmH2O) similar cantly varies across countries and institutions [34].
to Pplat reduction. Despite observational studies indicating that VT less
Some authors have suggested that a longer end- than 4 ml/kg/PBW and Pplat less than 1922 mmHg
inspiratory pause enhances diffusion between during ECMO are associated with improved sur-
inhaled VT and resident alveolar gas, thus facilitat- vival, the latter is often hardly achievable if high
ing the transfer of CO2 from alveoli toward the PEEP is used [35]. Serpa Neto et al. recently con-
airways [30]. Accordingly, Aguirre et al. recently ducted a pooled individual patient data analysis to
reported the results of a study on 13 patients with investigate whether different ventilator settings
ARDS, demonstrating that a longer end-inspiratory during ECMO can affect patients outcome.
pause (from 0.17 to 0.7 s) reduces dead space frac- Initiation of ECMO was associated to lower VT, Pplat,
tion and enhances CO2 washout, finally allowing to and improved CRS but, again, lower DP during the
lower VT and DP (13.6 to 10.9 cmH2O) with stable treatment was the only ventilator variable associ-
&
CO2 and no development of auto-PEEP [31 ]. ated to improved survival; also in patients under-
going ECMO, the effects of CRS, VT, and PEEP setting
on mortality were fully mediated by changes in DP,
Ultra-protective ventilation with CO2 removal
finally suggesting a possible role of such parameter
As previously highlighted, some patients may be at &&
in this specific context too [36 ].
risk of overinflation even though VT is 6 ml/kg PBW.
Bein et al. showed that an ultra-protective ventilation DRIVING PRESSURE LIMITING DURING
strategy providing VT as low as 3 ml/kg PBW and ASSISTED MECHANICAL VENTILATION
permitted by veno-venous extracorporeal CO2
Although sedation and paralysis is strongly recom-
removal (ECCO2-R) can lower the driving pressure
as compared to the standard 6 ml/kg PBW, but the mended in the early phase of ARDS to minimize the
clinical benefit (time to successful weaning) seemed progression of lung damage from a form of patient
self-inflicted lung injury [37], assisted mechanical
to be limited to a post hoc subgroup of patients with
ventilation is often used in the recovery phase of the
PaO2/FiO2 ratio lower than 150 mmHg [32].
disease. Data concerning DP in spontaneously
Nonetheless, to achieve such relevant VT
breathing patients with ARDS are lacking. Mauri
reduction, high blood flows (1.3 l/min) in the
et al. showed that inspiratory effort, PLdyn and VT
ECCO2-R system were necessary and this aspect
can be controlled through the use of extracorporeal
can limit the clinical application of the strategy.
Recently, the feasibility and safety of new devi- ECCO2-R while patients are recovering from ARDS
ces allowing low-flow ECCO2-R to enhance ultra- [22]. Whether this can be associated to a lower DP
& is unknown and further clarifying studies are
protective ventilation have been tested [33 ]. In a
warranted.
pilot study, 15 patients with moderate ARDS under-
went VT reduction to 4 ml/kg and low-flow ECCO2-R
was initiated when respiratory acidosis eventually CONCLUSION
developed. Mean ECCO2-R flow of 420 ml allowed Driving pressure allows identifying patients that are
to significantly reduce VT and DP, with no hyper- burdened by an increased risk of VILI and by a lower
capnia nor other side effects. A larger study with survival. Despite not demonstrated in clinical stud-
similar design is currently ongoing and will provide ies, targeting ventilatory interventions and VT to
more definite results (NCT02282657). achieve lower DP appears physiologically reason-
Given that ECCO2-R may not be available for all able. It is wise to suggest that DP values higher than
patients with ARDS, identifying patients that may 14 cmH2O should be avoided, but a really well tol-
most benefit from an ultra-protective ventilation erated individual threshold to achieve in patients
strategy is a research priority. with ARDS is yet to be identified.
1070-5295 Copyright 2017 Wolters Kluwer Health, Inc. All rights reserved. www.co-criticalcare.com 43
33. Fanelli V, Ranieri MV, Mancebo J, et al. Feasibility and safety of low-flow 36. Serpa Neto A, Schmidt M, Azevedo LCP, et al. Associations between
& extracorporeal carbon dioxide removal to facilitate ultra-protective ventilation && ventilator settings during extracorporeal membrane oxygenation for
in patients with moderate acute respiratory distress sindrome. Crit Care refractory hypoxemia and outcome in patients with acute respiratory
2015; 20:36. distress syndrome: a pooled individual patient data analysis:
Pilot study addressing the safety and feasibility of ultra-protective ventilation Mechanical ventilation during ECMO. Intensive Care Med 2016;
(VT ml/Kg IBW) enhance by low-flow CO2 removal. Driving pressure was 42:16721684.
significantly lower during ultra-protective ventilation. Pooled individual data analysis. Among mechanical ventilation parameters during
34. Schmidt M, Stewart C, Bailey M, et al. Mechanical ventilation management ECMO, the driving pressure is the one most strongly associated to changes in
during extracorporeal membrane oxygenation for acute respiratory distress survival.
syndrome. Crit Care Med 2015; 43:654664. 37. Brochard L, Slutsky A, Pesenti A. Mechanical ventilation to minimize progres-
35. Marhong JD, Munshi L, Detsky M, et al. Mechanical ventilation during extracorporeal sion of lung injury in acute respiratory failure. Am J Respir Crit Care Med 2016.
life support (ECLS): a systematic review. Intensive Care Med 2015; 41:9941003. [Epub ahead of print]