Abstract 15943: Phenotypes of Cardiogenic Shock Associated With Increasing In-Hospital Mortality: A Report From The National Cardiogenic Shock Working Group Registry

Document Type


Publication Date


Publication Title





Introduction: Cardiogenic shock (CS) mortality remains prohibitively high. A major limitation in the field has been the lack of large registries that include hemodynamic data for analysis. The National Cardiogenic Shock Working Group Registry is a multi-center retrospective database of CS patients in the United States. We now provide an interim update on the registry.

Hypothesis: We hypothesized that escalating drug and device use are associated with increasing in-hospital mortality.

Methods: We analyzed the Registry from 2017 to 2019. Modified versions of the SCAI Shock Classifications were employed. Class C and D shock patients were subdivided into two groups. Class C1 required up to one drug and no device therapy; C2 required one device without drug therapy; D1 required one drug and one device; D2 required greater than one drug and device therapy or multiple devices.

Results: The Registry contains 1,565 CS patients due to any cause obtained from 8 medical centers. Hemodynamic data were available for 1079 patients. SCAI Class distribution was: <1% A/B; 12% C1; 20% C2; 18% D1; 50% D2. Class E patients were excluded. Overall in-hospital mortality was 30.5% for the entire study and lower among patients with pulmonary artery catheter data available for analysis (28% vs 61%, p<0.001; Odds Ratio 0.65 [0.5-0.84]). In-hospital mortality was associated with increasing SCAI Class and drug escalation (Figure). Compared to survivors, non-survivors had significantly increased right atrial pressure, pulmonary wedge pressure, lactate, and serum creatinine and decreased mean arterial pressure.

Conclusions: Using the largest retrospective multicenter registry of CS with hemodynamic data, we report that a modified version of the SCAI CS Classification, escalating drug therapy, and increased congestion, but not cardiac output, correlate with increasing in-hospital mortality. These findings may inform the development of new management strategies and prognostic tools for CS.

Clinical Institute

Cardiovascular (Heart)




Center for Cardiovascular Analytics, Research + Data Science (CARDS)