(1665) Impact of Ventilator Mechanical Power on Outcomes in Patients With Cardiogenic Shock

Introduction: Cardiogenic shock (CS) patients requiring positive-pressure mechanical ventilation (MV) face high mortality rates. Elevated ventilatory mechanical power (MP), a marker for ventilator-induced lung injury, is linked to increased mortality in ARDS patients. However, its impact on CS outcomes has not been studied. We hypothesize that MP predicts disease severity and mortality risk in these patients.

Methods: We used the MIMIC-III database, a large de-identified single-center database from Beth Israel Deaconess Medical Center from 2001-2012. We selected adult CS patients requiring invasive positive-pressure mechanical ventilation (MV). We calculated mechanical power as [MP = 0.098 x RR x TV x (∆Pinsp + PEEP)]. Patients were categorized into survivors and non-survivors and mean MP was compared between the two groups using a Student's t-test. Data were also analyzed by race (Asian, Black, Declined to answer, Unable to obtain, Multi-Race, White) and ethnicity (Hispanic or Latino) and mean MP was compared using a one-way ANOVA. A random forest model was used to identify mortality predictors, including MP, age, peak inspiratory pressure (PIP), and MV requirement. The accuracy of the model was calculated by comparing the predicted outcomes to the actual outcomes in the test dataset.

Results: From 857 patients with CS, there were 468 survivors and 389 non-survivors. The mean MP was 23.001 J/min in survivors and 29.948 J/min in non-survivors (p < 0.001). The mean MP among ethnicity groups was found to be different (p = 0.0074), with an F-value of 2.772, demonstrating that the mean MP was similar within the groups but different among different ethnic groups. The model predicted with 69.7% accuracy that MP was the most important feature in identifying higher mortality compared to other variables alone.

Conclusions: MP > 29 J/min identifies CS patients at the highest mortality risk. The model’s accuracy of 69.7% has significant implications for 1 in 3 CS patients. Prospective randomized controlled trials are needed to show if maintaining MP < 23 J/min can improve outcomes in patients with CS.