Abstract Background/Introduction Sepsis causes widespread microvascular dysfunction. While myocardial injury in sepsis is common and associated with mortality, the extent of coronary microvascular dysfunction in affected patients is unknown. Purpose To examine the association between coronary microvascular flow indices and (i) high-sensitive cardiac troponin T (hs-cTnT) and (ii) other cardiac biomarkers, coronary atherosclerosis, and ventricular dysfunction in sepsis. Methods We prospectively enrolled sepsis patients with hemodynamic instability and myocardial injury (hs-cTnT ≥15 ng/L) admitted to the intermediate or intensive care unit at our institution. Patients underwent comprehensive echocardiography and coronary angiography, including coronary microcirculatory flow assessment in the LAD using the thermodilution technique, between days 3-9 from sepsis onset. Biventricular function was assessed using multiple echocardiographic parameters (Figure 1). Linear regression analysed associations with hs-cTnT, logistic regression assessed associations with ventricular dysfunction, and linear mixed-effects models compared coronary microvascular indices between sepsis and chronic coronary syndrome (CCS) patients. Results Of the 62 patients (31% female), 61 underwent echocardiography, 55 had coronary angiography, and 49 underwent coronary microvascular flow assessment. The median age was 75 years (interquartile range IQR 70-80), 4 (7%) had a prior myocardial infarction, and 41 (61%) required vasopressors. Median peak hs-cTnT and N-terminal pro B-type natriuretic peptide (NT-proBNP) were 102 ng/L (IQR 52-228) and 6,825 ng/L (IQR 2,530-14,025), respectively. Echocardiography revealed left ventricular (LV) systolic dysfunction in 56%, LV diastolic dysfunction in 59% and right ventricular (RV) dysfunction in 86% of patients. Elevated index of microcirculatory resistance (IMR 25) and/or microvascular resistance reserve (MRR 3) were observed in 31/49 (63%) of patients. Obstructive epicardial coronary artery disease (≥50% stenosis and/or fractional flow reserve 0.8) was present in 12/55 (22%) and six were revascularized with PCI. Acute renal failure, epicardial coronary artery disease, and microcirculatory flow measurements were not associated with hs-cTnT (Figure 1). IMR, and MRR were associated with LV systolic dysfunction, while hs-cTnT was associated with LV diastolic dysfunction. NT-proBNP was associated with RV dysfunction. Sepsis patients had lower MRR than CCS patients in the matched control analysis (Figure 2). Conclusion(s) This is the first study to examine invasive measurements of coronary microvascular function in relation to myocardial injury and biventricular dysfunction in sepsis patients. The majority of sepsis patients exhibit coronary microvascular dysfunction, which is more pronounced than in chronic coronary syndrome. Coronary microvascular dysfunction is associated with LV systolic dysfunction but not with hs-cTnT levels.
Loerstad et al. (Sat,) studied this question.