Impact of pre-existing heart failure on mortality in acute myocardial infarction-related cardiogenic shock

Background
Acute myocardial infarction-related cardiogenic shock (AMICS) is a critical condition with high mortality. Patients with chronic heart failure constitute a particularly vulnerable population and may face an even higher risk of adverse outcomes when suffering AMICS.

Purpose
To assess both short- and long-term mortality in AMICS patients with pre-existing heart failure.

Methods
In a cohort of consecutive AMICS patients admitted to one of two Danish tertiary heart centres between 2010 and 2017, we identified patients with pre-existing heart failure as those having an ICD-10 diagnosis of heart failure within the last 10 years prior to admission combined with a prescription redemption for either beta blockers or renin-angiotensin system inhibitors within the last two years. Using national health registries, we performed a complete 5-year follow-up to assess mortality.

Results
Of 1716 patients admitted with AMICS, 154 (9 %) had a medical history of heart failure. There was no difference in sex proportions between patients with and without heart failure, neither in systolic blood pressure, heart rate or arterial lactate level at admission. Patients with heart failure were significantly older (73 vs. 67 years, p<0.001), had a lower left ventricular ejection fraction on admission (25% vs. 30%, p<0.001), and were more likely to have a medical history of hypertension (78% vs. 49%), dyslipidaemia (66% vs. 32%), diabetes (29% vs. 18%), ischaemic heart disease (74% vs. 25%), and prior myocardial infarction (51% vs. 12%) (p<0.001 for all). Furthermore, heart failure patients presented less frequently with ST-segment elevation myocardial infarction (60% vs. 71%, p=0.012) and out-of-hospital cardiac arrest (25% vs. 44%, p<0.001), were less often transferred directly to an invasive centre (56% vs. 68%, p=0.004), and underwent revascularization less frequently (75% vs. 88%, p<0.001). No differences were observed in culprit vessel, multivessel disease, post-PCI TIMI flow, nor treatment with mechanical circulatory support.

There was no difference in mortality between patients with and without heart failure at 30 days (59% vs. 53%, respectively, p=0.062), but heart failure patients had a significantly higher mortality at 5 years (77% vs. 64%, p=0.00088) (Figure). In a 30-day landmark analysis, only 56% of patients with heart failure were alive after 5 years, compared to 76% of those without (p=0.00017). In a univariate Cox regression analysis, heart failure patients had a significantly higher mortality risk at 5 years (HR 1.38, 95%CI 1.14-1.67, p<0.001), however, not statistically significant after adjustment for age and sex (HR 1.17, 95%CI 0.96-1.41, p=0.11).

Conclusion
Patients with pre-existing heart failure experience similar 30-day mortality rates after AMICS as those without heart failure, but have a significantly higher mortality at 5 years, which may be explained by older age and greater comorbidity burden.