Background: Sepsis survivors are at risk of developing myocardial infarction and heart failure. It remains unclear whether coronary artery disease (CAD) is a major contributor to the development of these complications. This study sought to characterize the burden and distribution of significant CAD in sepsis survivors. Methods: Sepsis survivors who underwent computed tomography coronary angiography (CTCA) or invasive coronary angiography (ICA) in a UK tertiary cardiac center for suspected ischemic heart disease were retrospectively studied. Results: Of the 30 sepsis survivors (age 57 ± 12 years; 50% males), 21 patients underwent CTCA and 9 patients underwent ICA a median 39 days [IQR 12–152] from the sepsis episode. Eight patients (~27%) had angiographically significant CAD (n = 6 severe [>70%] stenosis; n = 2 moderate [50–70%] stenosis). The CT coronary calcium score was higher in patients with significant CAD compared to patients without significant CAD (638 [368–1015] vs. 4 [1–72]; p < 0.001). Of the 8 patients with significant CAD, 3 patients had LV systolic dysfunction (38%) on echocardiography and 8/21 (38%) patients without significant CAD had LV systolic dysfunction (p = 1.00). Long-term adverse complications (all-cause mortality and/or heart failure hospitalization) occurred 3/8 (38%) patients with significant CAD and 4/22 (18%) patients without significant CAD (p = 0.345). Conclusions: A minority of sepsis survivors have significant CAD. The presence of significant CAD cannot fully explain the occurrence of post-sepsis LV systolic dysfunction and adverse outcomes. The ischemic and non-ischemic mechanisms underlying post-sepsis cardiovascular disease require further investigation.

Background: Post-sepsis cardiomyopathy is associated with an increased risk of adverse cardiovascular outcomes. It remains poorly understood, which limits therapeutic development. This study characterised post-sepsis cardiomyopathy using cardiovascular magnetic resonance (CMR) imaging. Methods: Patients admitted with acute sepsis and suspected cardiac injury or heart failure who subsequently (47 days [IQR: 22–122]) underwent CMR at a UK tertiary cardiac centre were included. Age- and gender-matched controls (n = 16) were also included. Subjects underwent CMR at 1.5 Tesla with cines, native T1- and T2-mapping, and late gadolinium enhancement (LGE) imaging. Results: Of the 22 post-sepsis patients (age 50 ± 13 years; 64% males), 13 patients (59%) had left ventricular (LV) dilatation. Patients had significantly elevated left ventricular (LV) end-diastolic and end-systolic volume indices compared to controls (p = 0.011 and p = 0.013, respectively). Eleven patients (50%) had LV systolic dysfunction (ejection fraction < 50%), most of whom (8/11) had non-ischaemic patterns of LGE (n = 7 mid-wall; n = 1 mid-wall/patchy). In the eleven patients with preserved LV systolic function (ejection fraction ≥ 50%), three patients (27%) had significant LGE (n = 1 mid-wall; n = 1 subepicardial/mid-wall; n = 1 patchy). Compared to controls, patients had elevated septal native myocardial T1 values (p < 0.001) but similar septal native myocardial T2 values (p = 0.090), suggesting the presence of myocardial fibrosis without significant oedema. Conclusions: Post-sepsis cardiomyopathy is characterised by LV dilatation, systolic dysfunction, and myocardial fibrosis in a non-ischaemic distribution. Significant myocardial oedema is not prominent several weeks post-recovery. Further work is needed to test these findings on a multi-centre basis and to develop novel therapies for post-sepsis cardiomyopathy.

Background: Sepsis survivors can develop left ventricular systolic dysfunction (LVSD) and heart failure. These patients are often treated with guideline-directed medical therapy (GDMT) known to be effective in patients with non-sepsis-related heart failure. This study sought to assess the use of GDMT on sepsis survivors with LVSD. Methods: Sepsis survivors with suspected myocardial injury and/or heart failure diagnosed with LVSD in a UK cardiac centre were retrospectively studied. Clinical and transthoracic echocardiography (TTE) data were recorded and analysed. Results: Of the 25 sepsis survivors (age 56 ± 11 years; 52% males), 11 (44%) had LVSD (LVEF < 50%). LV end-diastolic internal diameter (LVIDd) was similar between patients with vs. without LVSD (5.2 ± 0.8 cm vs. 4.7 ± 0.8 cm; p = 0.214). Patients with LVSD had significantly greater LV end-systolic internal diameter (LVIDs) than those without LVSD (4.0 ± 1.2 cm vs. 2.8 ± 0.6 cm; p = 0.027). Tricuspid annular plane systolic excursion (TAPSE) was similar between the two groups (2.1 ± 0.5 cm vs. 2.2 ± 0.6 cm; p = 0.910). Of the 11 patients with LVSD, nine patients underwent repeat TTE scans after 6 months [IQR 3–9], most of whom were taking GDMT. The majority (8/9) of these patients demonstrated LV systolic functional recovery (>5% LVEF increase; mean LVEF improvement 16 ± 11%) after GDMT. Reductions were seen in LVIDd (5.3 ± 0.8 cm to 5.0 ± 0.7 cm) and LVIDs (4.1 ± 1.2 cm to 3.7 ± 0.8 cm) after GDMT, though these changes did not reach statistical significance (both p > 0.05). Conclusions: GDMT appears beneficial in sepsis survivors with LV dysfunction. This finding should be validated on a larger and multi-centre basis to further affirm the value of medical therapy in post-sepsis heart failure.

Background: After recovery from acute sepsis, patients can exhibit left ventricular systolic dysfunction (LVSD) and non-ischaemic myocardial fibrosis. The relationship between myocardial fibrosis and LVSD remains poorly defined. This study sought to fill this knowledge gap using quantitative late gadolinium enhancement (LGE) cardiovascular magnetic resonance (CMR). Methods: Twenty-eight sepsis survivors underwent CMR at 1.5-Tesla for the assessment of cardiac volumes, systolic function and LGE. Myocardial fibrosis burden was derived quantitatively by LGE, expressed as a percentage of LV mass. Results: Study patients (age 51 ± 16 years; 57% males) had a median LVEF of 59% (IQR: 43–64) of whom 43% had LVSD (LV ejection fraction [LVEF] < 50%). LGE was found in 64% of the study patients by visual assessment, mostly in non-ischaemic patterns. The overall myocardial fibrosis burden was 3.3% (IQR: 0.9–7.1) of LV mass. Myocardial fibrosis burden was inversely correlated to LVEF in sepsis survivors (Rho = -0.385; p = 0.043). Patients with LVSD had greater myocardial fibrosis burden than patients without LVSD (7.3 ± 6.0% vs. 3.1 ± 2.5%; p = 0.041). Myocardial fibrosis burden was not significantly influenced by the presence of major co-morbidities. Conclusions: Myocardial fibrosis burden may play a role in LV dysfunction in sepsis survivors. Further work is needed to better understand its prognostic value.

Background/Objectives: Survivors of sepsis can develop left ventricular (LV) systolic function with focal myocardial fibrosis. The relationship between diffuse myocardial fibrosis or oedema and LV systolic function remains unknown in this patient cohort. This study sought to address this knowledge gap using cardiovascular magnetic resonance (CMR) parametric mapping methods. Methods: Sepsis survivors who underwent CMR at a UK cardiac centre were included. CMR images analysed include cines, native T1-mapping, native T2-mapping, and post-contrast T1-mapping. Synthetic extracellular volume (ECV) fraction was also estimated. Native myocardial T1 values, native myocardial T2 values, and ECV values were compared against LV ejection fraction (LVEF). Results: Of the 37 sepsis survivors (age 53 ± 16 years old; 57% males), the mean left ventricular ejection fraction (LVEF) was 55% (IQR 43–62), and 43% of the patients had LV systolic dysfunction (LVEF < 50%). Mean native myocardial T1 values were 1055 ± 65 ms (septal) and 1051 ± 60 ms (global). Mean synthetic ECV values were 0.30 ± 0.04. Mean native myocardial T2 values were 52 ± 7 ms (septal) and 53 ± 6 ms (global). Septal and global native myocardial T1 values were not significantly correlated with LVEF (rho = 0.080, p = 0.637; rho = 0.036, p = 0.831, respectively). Synthetic ECV was not significantly correlated to LVEF (rho = −0.082; p = 0.723). Septal and global native myocardial T2 values were weakly correlated with LVEF (rho = 0.261, p = 0.281; rho = 0.216, p = 0.375, respectively). On ROC analysis, the performance of native myocardial T1 values, ECV, and native myocardial T2 values for predicting LV dysfunction was modest (AUC: 0.53 ± 0.10, 0.54 ± 11, and 0.68 ± 0.14; all p > 0.05, respectively). Conclusions: CMR markers of diffuse myocardial fibrosis (native T1-mapping and ECV) and myocardial oedema (native T2-mapping) have weak relationships with left ventricular systolic function in this study cohort of sepsis survivors. Further work is needed to better assess the role of diffuse myocardial fibrosis and oedema in the pathophysiology of post-sepsis cardiomyopathy.