Abstract 503: Predictors of 30-Day Mortality in Patients with Refractory Cardiogenic Shock Following Acute Myocardial Infarction Despite a Patent Infarct Artery
Background: While there is compelling evidence that early coronary revascularization improves outcomes in patients with cardiogenic shock complicating acute MI, little is known about predictors of survival in patients with persistent shock despite infarct-related artery (IRA) patency.
Methods: We examined data from patients in TRIUMPH, a multicenter randomized clinical trial of the nitric oxide synthase inhibitor, L-NMMA, in patients with persistent vasopressor-dependent cardiogenic shock complicating acute MI at least one hour after established IRA patency. Patients who died within 30 days were compared with those who survived. Continuous variables were assessed using the Wilcoxon Rank Sum and categorical variables using the χ2 test. Pre-specified variables were included in an initial multivariable logistic regression model to predict mortality. Creatinine clearance was estimated using the Cockcroft-Gault formula. A second model incorporating baseline vasopressors and dosages, and a third model including the change in systolic blood pressure at two hours were also developed. Bootstrapping was used to assess the stability of model variables.
Results: 180 out of 396 patients (45.5%) died within 30 days. Systolic blood pressure, measured on vasopressor support, and creatinine clearance were significant predictors of mortality in all models. Norepinephrine dose was associated with death in the second model, but was no longer significant when change in systolic blood pressure at two hours was added as a covariate. Significant multivariable predictors of 30-day mortality are shown in the Table⇓.
Conclusion: Systolic blood pressure and creatinine clearance are significant predictors of mortality in patients with persistent vasopressor-dependent cardiogenic shock following acute MI despite a patent IRA. These prognostic variables may be useful in selecting patients for investigation of additional therapies, including ventricular assist devices.