Hot septum and strain reversus: additional echocardiographic signs associated to constrictive pericarditis and pericardial involvement detected by CMR

Abstract

Abstract Funding Acknowledgements Type of funding sources: None. Background A typical echocardiographic global longitudinal strain (GLS) pattern has been described in in patients with constrictive pericarditis (CP): a characteristic reduction in the regional longitudinal strain of the left ventricular free wall and relative sparing of the septal longitudinal strain value which creates the so-called "hot-septum (HSS)" at the bullseye plot, described by some authors as "strain reversus (SR)". Currently, there is no consensus about definition, prevalence and cut-off values of this GLS pattern. Purpose The aim of this study was to evaluate the prevalence of echocardiographic HSS or SR among pericardial diseases, their diagnostic role in constrictive pericarditis and their correlation with pericardial involvement assessed by cardiac magnetic resonance (CMR). Methods This is a retrospective, single centre cohort study. Consecutive patients admitted for pericardial diseases were enrolled if they underwent both to echocardiogram and CMR. Exams were reviewed and all signs of pericardial disease were systematically assessed, including echocardiographic GLS. Diagnosis of CP was confirmed at hemodynamic catheterization. HSS was visually assessed and defined as an area of septal sparing with reduced LS of the lateral wall at bullseye plot. SR was defined as the ratio between septal LS and lateral LS <1 and was assessed as anterolateral wall LS (ALWLS)/inferoseptal wall LS (ISWLS) and inferolateral wall LS (ILWLS)/anteroseptal wall LS(ASWLS) at basal, mid and apical level. Results After screening of 196 patients, a total of 53 were included. The study population was divided in CP (16 patients) and non-constrictive pericarditis (NCP, 37 patients). Left ventricular ejection fraction (LVEF) was similar between CP and NCP (59.5±10.0% vs 62±6.2%, p 0.190) while GLS was significantly reduced in CP (13.8±2.7% vs 17.6±3.8%). The prevalence of HSS was higher in CP than NCP (75% vs 10,8%, p <0.001) and it was strongly associated with the diagnosis of CP (OR: 3.20, CI: 1.67 – 4.74, p <0.001; c-statistic 0.82). SR was significantly lower in CP patients (0.68±0.43 vs 1.18±0.68, p 0.004). Logistic regression model showed an association between CP and a reduced basal ILWLS/ASWLS ratio (OR: 2.23, CI: 0.59 – 4.07; p 0.017; figure 1); the best cut-point for CP prediction was 0.96 (c-statistic: 0.71, sensitivity 0.54, specificity 0.88, figure 2). HSS was mainly associated to pericardial thickening (OR: 2,09, CI: 0.74–3.4; p 0.002) and less to pericardial LGE (OR: 1.78, CI: 0.16 – 3.39; p 0.031) and pericardial oedema (OR: 1.28, CI: 0.36 – 2.53; p 0.044). Conclusion Hot septum and strain reversus are echocardiographic signs associated to constrictive pericarditis and pericardial involvement detected by cardiac magnetic resonance. The use of strain echocardiography can help to distinguish between pericarditis with and without constriction.