Stress, strain and mechanical power: let's not forget the shape of the flow

We read the work of Buiteman-Kruzinga et al. [1] and believe that it adds to the evidence regarding the relevance of respiratory rate in the energy transfer of the ventilatory system. For viscoelastic bodies, cyclicity is probably the most important factor to explain the disruption of their initial conformation, which has already been extrapolated to clinical studies [2], and this study confirms this. However, we believe it is necessary to clarify some points.

Despite the rigour of the design and the robustness of the clinical trials included, the dichotomisation of initially continuous variables and imputation of the data (even when reducing the variance) can generate biased estimates, even using advanced techniques. The authors assert that the higher the peak pressure (P_peak), the greater the mechanical power in patients without acute respiratory distress syndrome, except in the subgroup of patients with low V_T and low programmed respiratory rates. Under this premise, it can be inferred that P_peak may not be so important when gentle ventilation is ensured. However, this work does not specify the inspiratory flow delivery form, and despite the fact that in the original mechanical power equation this variable has no place, from a rheological and thermodynamic perspective, we know that it has relevance [3].

Bodies that exhibit viscoelastic behaviour have the capacity to store or dissipate energy when deformed by stress. Under this argument, the strain is modified by the cyclicity, but also by the time and the way in which it is exposed to the stress. Anisotropic strain, defined as the rate of deformation under non-uniform stress, changes as a function of lung geography but is also sensitive to variations in flow. Thus, the higher the flow delivery velocity, the higher the P_peak, which translates into a higher pulmonary viscoelastic rate with the resulting decrease in the slope of the strain-volume or strain-pressure curve. Several authors have already explored this hypothesis, and although there are no randomised clinical trials, the existence of its association with mechanotransduction cannot be ignored in healthy lungs, acute respiratory distress syndrome, or in paediatric patients [4].

We performed a post hoc analysis of the “mechanical power day” [5] by means of multiple frequentist linear regression. We observed that ventilation with decelerated flow (coefficient 3.4, 95%CI 2.4–4.5, p < 0. 001) correlated independently with a higher probability of high energy mechanical power (> 17 J.min^-1) in patients without acute respiratory distress syndrome (R² = 71%), when compared with constant flow ventilation (coefficient difference -3.4, 95%CI -4.4 to -2.3, p < 0.001). Figure 1 shows the correlation between decelerated flow with high energy mechanical power as a function of driving pressure. To corroborate the above, using Betabinomial Bayesian linear regression modelling with adaptive algorithm (Markov chain Monte Carlo sample size = 10,000, Burn-in 2000, Random-walk Metropolis-Hastings sampling) and using the a priori data obtained from the mechanical power day, we observed that ventilation with decelerated flow presented a higher a posteriori probability of correlation with high energy mechanical power (mean 3.3, 95% credible interval 2.2–4.3). Post-estimation showed that this model has a higher a posteriori probability (96%).

Figure 1

Open in figure viewerPowerPoint

Correlation between decelerated and constant flow with mechanical power, as a function of driving pressure. Pink, decelerating flow; green, constant flow.

In conclusion, the ability to measure mechanotransduction is contingent on the use of mechanical power as a potential gold standard. However, it is likely that some variables, such as the flow delivery form, play a relevant role in energy transfer and thus in the development of VALI. The results of our analysis support the hypothesis of the relevance of the flow delivery form, and thus, P_peak may be overestimated.