Validation of a Recruitment Model with Eit Imaging

Mechanical ventilation is a standard and necessary treatment at lung injuries like the acute respiratory distress syndrome. But ventilator induced injuries are possible consequences especially at patients with long time ventilation or mislead settings. The prevention of such injuries can be optimized by finding an optimal model for the patients lung and identify the patients specific parameters of such models. In focus is the patients’ resistance, compliance and threshold opening pressure (TOP which represents the ideal pressure for opening and holding collapsed alveoli open for better oxygenation). In this study the so called “pressure recruitment model” (PRM) which is a combination of alveoli recruitment and static as well as dynamic effects of lung tissue implemented by C. Schranz, should be pre-validated. The algorithm is based on a combination of Hickling’s lung layer model and Salazar’s pressurevolume relationship of alveoli and here tested with six patient datasets. Every patient underwent a LowFlow maneuver with 4ml/min which is the necessary input for the PRM-calculations. Identified parameters of the PRM are compared with the first order lung model (FOM) and an electrical impedance tomography (EIT) method. The PRM is showing a up to ten times better fitting quality compared to the FOM. Although the results in part of lung compliance is illogically higher in the PRM than in the FOM. Accordance between the EIT method and the algorithm from Schranz in case of TOP could not be found. In both comparisons further studies must be done to exclude influence factors for the different models and improve the performance. Keywords— Validation, recruitment, lung model, EIT, ARDS