(P028) IN-HOSPITAL CONGESTION PROFILES ARE ASSOCIATED WITH MORTALITY IN PATIENTS WITH CARDIOGENIC SHOCK

Background: Cardiogenic shock (CS) is characterized by low cardiac output and tissue hypoperfusion attributable to either impaired left ventricular (LV), right ventricular (RV), or biventricular function. While mortality in CS remains high, pulmonary artery catheterization (PAC) can improve outcomes through optimization of volume status, use of inotropes and vasopressors, and use of advanced heart failure therapies including orthotopic heart transplantation (OHT) and mechanical circulatory support (MCS). How congestion profiles identified through PAC may affect outcomes or inform clinical management is currently unknown.

METHODS AND RESULTS: Consecutive adult CS patients admitted between January 2014 and February 2023 were identified from a CICU registry and were included in this study if a complete PAC profile was measured at time of insertion, including right atrial pressure (RAP) and pulmonary capillary wedge pressure (PCWP). Patients were grouped according to the following congestion profiles: isolated LV congestion (RAP < 12, PCWP≥15), isolated RV congestion (RAP≥12, PCWP < 15), biventricular congestion (RAP≥12, PCWP≥15), or no congestion (RAP < 12, PCWP < 15). Primary outcomes of interest were in-hospital mortality, need for OHT or durable left ventricular assist device (LVAD) during index admission. Secondary outcomes included need for renal replacement therapy (RRT), CICU and hospital length of stay (LOS). Categorical variables and count data were compared with chi-square tests between groups. Continuous variables were compared using a one-way ANOVA on ranks. In-hospital survival analysis, up to 100 days length of stay, was performed with Kaplan Meier methods and a log-rank test between groups with patients censored at time of OHT or LVAD implantation.
603 CS patients were included with mean age of 58 years, 72% male, 21% presented with acute myocardial infarction. 44% of patients had biventricular congestion, 32% had isolated LV congestion, 4% had isolated RV congestion, and 19% had no congestion. In-hospital mortality was 27.53% and not statistically different between groups (p>0.05). LVAD implants were significantly different among groups (p < 0.01), with more implants in patients with LV congestion (Table 1). Hospital LOS was lower in patients with no congestion. In-hospital survival analysis up to 100 days with showed a difference among congestion profile groups (pLogRank < 0.05) (Figure 1), with worse outcomes in patients with biventricular congestion.

Conclusion: In a contemporary CICU registry, in-hospital survival was different among congestion profile groups. Congestion profiles also impacted likelihood of receiving LVAD and length of stay in hospital. Future study will involve the use of regression models to elucidate this association further.