P851Incidence and prognostic impact of acute infection among patients with type 1 and 2 myocardial infarction

Abstract

Abstract Background The incidence and impact of acute infection among patients with type 1 (T1MI) and 2 myocardial infarction (T2MI) is not well known. Inflammation and increased procoagulant activity can lead to T1MI, whereas cytokine release can cause hemodynamic alterations affecting myocardial oxygen consumption and contribute to T2MI. Purpose Determine the incidence and prognostic impact of acute infection among patients with T1MI and T2MI. Methods Post-hoc analysis of UTROPIA (NCT02060760), a prospective, observational cohort study involving 1,640 consecutive emergency department patients with serial cardiac troponin (cTn) I measurements obtained on clinical indication. The incidence of acute infection, antibiotic use, and/or laboratory evidence of infection were examined among patients with T1MI and T2MI. 5-year cumulative survival curves were plotted using the Kaplan-Meier method, frequencies were compared via Chi-Square, and a multivariate cox-proportional hazards model was used to assess the impact of infection on mortality. Results Among 217 patients with acute MI (T1MI, n=77 and T2MI, n=140), acute infection occurred in 63 (29%) of patients. Acute infection was more common in patients with T2MI than T1MI (35% vs. 18%, p=0.009). The predominant source of infection was pneumonia (59%), followed by urinary tract infections (11%). Patients with T2MI were also more likely than T1MI to have sepsis (11% vs. 1%, p=0.012), bacteremia (30% vs. 18%, p=0.057), and antibiotic use (35% vs. 17%, p=0.005). At 5-years, patients with acute MI and acute infection had a higher mortality rate than patients without infection (49% vs. 25%, p=0.0006) (Figure). Among the 31 deaths in MI patients with acute infection, most deaths were non-cardiac (65%) and occurred in those with T2MI (77%). Following adjustment for age, sex, and comorbidities in a Cox proportional hazards model, acute infection was an independent predictor of death (hazard ratio: 2.2, 95% CI: 1.3–3.5, p=0.0016). Conclusion Acute infection occurs in almost a third of patients that are diagnosed with acute MI during the index hospitalization. Most infections are due to pneumonia and occur most often in patients classified as having T2MI. Acute infection is an independent risk factor for 5-year mortality, with nearly half of all patients with acute MI in whom concomitant acute infection is present during the index hospitalization dead at 5-years. Most deaths are non-cardiac and the vast majority occur in patients with T2MI. Acknowledgement/Funding Abbott Diagnostics and Hennepin Healthcare Research Institute (formerly MMRF)e