Trends in first-time hospitalization, management, and short-term mortality in acute myocardial infarction-related cardiogenic shock from 2005 to 2017: A nationwide cohort study

Abstract

Background: Cardiogenic shock remains the leading cause of in-hospital death in acute myocardial infarction (AMI). Because of temporary changes in management of cardiogenic shock with widespread implementation of early revascularization along with increasing attention to the use of mechanical circulatory devices, complete and longitudinal data are important in this subject. The objective of this study was to examine temporal trends of first-time hospitalization, management, and short-term mortality for patients with AMI-related cardiogenic shock (AMICS).

Methods: Using nationwide medical registries, we identified patients hospitalized with first-time AMI and cardiogenic shock from January 1, 2005, through December 31, 2017. We calculated annual incidence proportions of AMICS. Thirty-day mortality was estimated with use of Kaplan-Meier estimator comparing AMICS and AMI-only patients. Multivariable Cox regression models were used to assess mortality rate ratios.

Results: We included 101,834 AMI patients of whom 7,040 (7%) had AMICS. The median age was 72 (interquartile range: 62-80) for AMICS and 69 (interquartile range: 58-79) for AMI-only patients. The gender composition was similar between AMICS and AMI-only patients (male: 64% vs 63%). The annual incidence proportion of AMICS decreased slightly over time (2005: 7.0% vs 2017: 6.1%, P for trend < .0001). In AMICS, use of coronary angiography increased between 2005 and 2017 from 48% to 71%, as did use of left ventricular assist device (1% vs 10%) and norepinephrine (30% to 70%). In contrast, use of intra-aortic balloon pump (14% vs 1%) and dopamine (34% vs 20%) decreased. Thirty-day mortality for AMICS patients was 60% (95% CI: 59-61) and substantially higher than the 8% (95% CI: 7.8-8.2) for AMI-only patients (mortality rate ratio: 11.4, 95% CI: 10.9-11.8). Over time, the mortality decreased after AMICS (2005: 68% to 2017: 57%, P for temporal change in adjusted analysis < .0001).

Conclusions: We observed a slight decrease in AMICS hospitalization over time with changing practice patterns. Thirty-day mortality was markedly higher for patients with AMICS compared with AMI only, yet our results suggest improved 30-day survival over time after AMICS.