Worsening of lung perfusion to tissue density distributions during early acute lung injury

Lung perfusion magnitude and distribution are essential for oxygenation and, potentially, lung inflammation and protection during acute respiratory distress syndrome (ARDS). Yet, perfusion patterns and their relationship to inflammation are unknown preARDS. We aimed to assess perfusion/density ratios and spatial perfusion-density distributions and associate these to lung inflam-mation, during early lung injury in large animals at different physiological conditions caused by different systemic inflammation and positive end-expiratory pressure (PEEP) levels. Sheep were protectively ventilated (16-24 h) and imaged for lung density, pulmonary capillary perfusion ((13)Nitrogen-saline), and inflammation (F-18-fluorodeoxyglucose) using positron emission and computed tomography. We studied four conditions: permissive atelectasis (PEEP = 0 cmH(2)O); and ARDSNet low-stretch PEEP-setting strategy with supine moderate or mild endotoxemia, and prone mild endotoxemia. Perfusion/density heterogeneity increased pre-ARDS in all groups. Perfusion redistribution to density depended on ventilation strategy and endotoxemia level, producing more atelectasis in mild than moderate endotoxemia (P = 0.010) with the oxygenation-based PEEP-setting strategy. The spatial distribution of F-18-fluorodeoxyglucose uptake was related to local Q/D (P < 0.001 for Q/D group interaction). Moderate endotoxemia yielded markedly low/zero perfusion in normal-low density lung, with (13)Nitrogen-saline perfusion indicating nondependent capillary obliteration. Prone animals' perfusion was remarkably homogeneously distributed with density. Lung perfusion redistributes heterogeneously to density during pre-ARDS protective ventilation in animals. This is associated with increased inflamma-tion, nondependent capillary obliteration, and lung derecruitment susceptibility depending on endotoxemia level and ventilation strategy.

Automated summary

Lung perfusion redistribution is associated with increased inflammation, nondependent capillary obliteration, and lung derecruitment susceptibility depending on endotoxemia level and ventilation strategy during pre-ARDS protective ventilation in animals.