A55-13 Protective Lung Ventilation: What Is the Most Important Parameter?

Abstract Background Mechanical ventilation (MV) strategies are pivotal in critical care, but their relationship with mortality varies across patient conditions. Current guidelines suggest that low tidal volume/ideal weight is the most important aspect of protective lung ventilation. This study investigates the association between key ventilatory parameters—tidal volume per kilogram (TV/kg), lung compliance, and driving pressure—and in-hospital mortality in a large general ICU cohort. Methods We analyzed data from 2,025 ICU patients, including 906 in-hospital deaths (44.7%). Patients were stratified by quartiles of TV/kg and lung compliance (24-hour average). Driving pressure was calculated as the difference between 24-hour average PIP and PEEP. Univariate and multivariate logistic regression models were used to evaluate associations with mortality. ROC curves and contour plots visualized the predictive and interactive effects of these variables. Results Lower TV/kg and lower compliance were both significantly associated with higher mortality (p 0.01). Driving pressure also showed a strong positive correlation with mortality (r = +0.21). In multivariate logistic regression, TV/kg (β = -0.13, p = 0.03) and driving pressure (β = +0.04, p = 0.009) remained independent predictors of mortality, while compliance (β = -0.015, p = 0.12) lost significance when modeled alongside its interaction with TV/kg. The interaction term (TV/kg × compliance) was not statistically significant (p = 0.62). Stratified analysis revealed the highest mortality (67.5%) occurred in patients in the lowest quartiles for both compliance and TV/kg, with an average driving pressure of 19.1 cmH2O. The lowest mortality (37.4%) was seen in patients in the highest quartiles of both parameters, with driving pressures averaging 13.1 cmH2O. AUC for the logistic regression model was 0.66, indicating fair predictive value. Visualizations confirmed a consistent mortality gradient: mortality increased with decreasing compliance and increasing driving pressure, with TV/kg modifying the risk within those strata. Driving pressure was most strongly influenced by compliance, especially at higher tidal volumes. Conclusion In this ICU cohort, low compliance and high driving pressure were associated with increased mortality, independent of tidal volume. Lower tidal volumes were observed in patients with lower compliance, and these combinations were linked with the highest mortality risk. These findings highlight the importance of individualized ventilator settings, particularly in patients with poor lung mechanics. Further work is needed to assess causal relationships and optimal strategies in specific patient subgroups. Changing the guidelines for managing ventilation from emphasis on tidal volume to controlling driving pressure may be appropriate. This abstract is funded by: None