A Statistical Model of Cellular Injury Heterogeneity During the Progression of Ventilator-Induced Lung Injury
Year: 2021
Type: conference-abstract
Abstract: Rationale: Lung injury in acute respiratory distress syndrome and ventilator-induced lung injury (VILI) is spatially heterogeneous. Understanding the spatial interdependence between new and existing injury may help us elucidate the mechanisms of VILI pathogenesis, allowing for development of protective ventilation protocols that stymie the spread of VILI from injured regions. We hypothesize that initial regions of injury create stress foci which cause injury to preferentially develop in neighboring areas. Methods: Mice were injuriously ventilated with 30 cmH 2 O plateau pressure and 0 cmH 2 O positive end-expiratory pressure to a minimal increase (1.1xH), doubling (2xH), or quadrupling (4xH) in respiratory elastance (n=5 per group). The location of injured (propidium iodide-labeled) cells in 15-20, 5m lung lobe sections per mouse were determined with custom software. In order to understand how the injury forms over time, we interpolate between these time points using a 3D statistical model of injury progression. At each time step in the model every cell can either become injured or remain healthy. The heterogeneous spread of cellular injury is based on the interdependence between injured regions. The likelihood of a cell becoming injured increases with the local concentration of injury within a radius (r). This area of increased injury probability is intended to capture the effects of edema and alveolar collapse on septal tissue strain in neighboring alveoli. The statistical model is fit to the distribution of injury density at the three experimental observation timepoints. Results: As VILI progresses, the density of experimental (colored dots) and model-predicted (black line) cellular injury increases at an accelerating rate. The injury variability between sections in the experimental data (error bars) and planes in the model predictions (grey shaded region) also increases with time. For the applied ventilation, simulated interdependence effects extend r=500m from injured regions and injury is 1200x more likely to occur within that region of interdependence. Conclusions: We have created a novel statistical model that recapitulates our experimental observations of injury progression. The model shows that injury heterogeneity increases with time in VILI. The 1200x increase in injury likelihood within the region of interdependence suggests a rich-get-richer mechanism whereby injured regions tend to attract further injury. The predicted range of interdependence (r=500m) is consistent with that of finite element simulations, suggesting that heterogeneous VILI propagation is due to parenchymal tethering. Fitting this model to multiple ventilation strategies may help elucidate what settings minimize r, and therefore minimize the spread of VILI.
Authors: C. Mattson, B.J. Smith
Language: English
Subfield: Pulmonary and Respiratory Medicine
Field: Medicine
Domain: Health Sciences
Open Access status: gold
OpenAlex ID: W4253186651