Hypereosinophilic syndrome (HES) is characterized by persistently elevated eosinophil count (>1500 cells/ml) in peripheral blood for at least six months, and single- or multiple end-organ damage attributable to cytotoxic injury by eosinophils.1 In the literature the overall prevalence of cardiovascular involvement is 40–50% of HES patients.1–3 The early stage of cardiac involvement consists of eosinophilic infiltration, followed by an intermediate thrombotic stage, and finally a late fibrotic stage. An enlarged atrium with normal-sized left ventricle is a minor criterion for endomyocardial fibrosis.1 At present little is known about left atrial (LA) function in HES. Three-dimensional (3D) speckle tracking echocardiography (3DSTE) is a new non-invasive clinical tool for volumetric and strain analysis of the left atrium.4 The present study was designed to compare 3DSTE-derived LA volumetric, volume-based functional and strain parameters between HES patients and matched controls.

The clinical data of each patient including organ involvement are presented in Table 1. Only one patient had a history of cardiac disease (non-ST-elevation myocardial infarction).

The laboratory findings were as follows (HES vs. controls): red blood cell count: 4.2±0.5 T/l vs. 4.3±0.4 T/l (p=0.94); hemoglobin: 126.7±18.8 g/l vs. 130.1±10.2 g/l (p=0.86); platelet count: 276.3±176.7×109/l vs. 282.4±158.2×109/l; hematocrit: 36.9±5.5% vs. 37.8±4.9%; white blood cell count: 16.7±5.8×109/l vs. 6.8±1.2×109/l (p=0.02); eosinophil ratio: 49.0±16.6% vs. 3.2±2.3% (p=0.001); and absolute eosinophil count: 8.7±4.8×109/l vs. 0.4±0.1×109/l (p=0.001). No correlations could be demonstrated between any of the laboratory findings and 2D echocardiographic and 3DSTE data in this patient population.

The routine two-dimensional echocardiographic data are presented in Table 2. None of the controls or HES patients showed grade >1 mitral or tricuspid regurgitation. Only LA diameter and interventricular septal thickness showed significant differences between HES patients and controls.

The 3DSTE data are presented in Table 3. Significantly increased maximum (p=0.01) and minimum (p=0.03) LA volumes and LA volume before atrial contraction (p=0.01) and elevated total (p=0.02) and active (p=0.005) atrial stroke volumes characterizing LA reservoir and booster pump functions were found in HES patients as compared to controls (Table 3). All emptying fractions did not differ between groups. Both global and mean segmental peak CS were significantly reduced in HES patients, suggesting decreased LA reservoir function (Table 4).

Cardiac manifestations are the major cause of morbidity in HES and follow three stages. The first, an acute necrotic stage, is due to infiltration of eosinophils in the myocardium and is mostly asymptomatic.13 The initial damage is thought to be caused by eosinophilic granules. The intermediate (thrombotic) phase is characterized by thrombus formation followed by organization of the thrombus into a thick layer of granulation tissue. In the third, fibrotic stage, granulation tissue changes into fibrosis, sometimes still with a small inflammatory zone.13 Nowadays the term ‘Löffler's endomyocarditis’ is used to describe the involvement of the heart in HES in the thrombotic and fibrotic stage.14 Typical echocardiographic findings in HES are endocardial thickening, fibrothrombotic obliteration of the ventricular apices, and valvular regurgitation due to restricted motion of the posterior mitral leaflet as assessed by 2D Doppler echocardiography.1–3,13

In the present study, most HES patients had no known cardiac disease or clinical signs of thrombosis or fibrosis at enrollment (except case 5). They were presumably in the first, asymptomatic stage, therefore the alterations in LA morphology and function would be related solely to HES. Only left ventricular hypertrophy and dilated LA could be detected by conventional 2D Doppler echocardiography, without significant valvular regurgitation or thrombus formation. 3DSTE confirmed LA volume changes in all phases of LA function and found alterations in both peak global and mean segmental LA CS, suggesting reduced reservoir function and LA remodeling.

A wide variety of pathophysiological changes may lead to LA remodeling, with structural, functional, neurohormonal or other consequences.15 The real mechanism behind LA remodeling in HES is not known, but myocyte necrosis and alterations in the extracellular matrix and in the release of atrial hormones due to toxic proteins from degranulating eosinophils and diastolic dysfunction could theoretically explain our findings.15

3DSTE is a new clinical tool for non-invasive 3D cardiac chamber quantifications of the left ventricle and atrium based on a block-matching algorithm of myocardial speckles during their frame-to-frame motion.4 3DSTE has been demonstrated to be useful for LA volumetric7–10 and strain10–12 assessments, allowing more detailed assessment of LA function from the same 3D dataset. Different patterns in 3DSTE-derived volume-based and strain functional properties can be demonstrated in different disorders. In a recent 3DSTE study, peak LA RS and LA LS were found to be altered in hypertrophic cardiomyopathy, together with preserved LA CS.10 In another study, strains at all LA levels showed alterations in atrial fibrillation by 3DSTE.12 In the present study, only peak LA CS was decreased in HES patients, while RS and LS remained unchanged. The real pathological mechanism behind the reason that only LA CS shows alterations in HES is not known, but hemodynamic reasons and their relationship with LA fiber orientation cannot be excluded, beyond the above-mentioned pathophysiologic factors.

Several important limitations of the assessments should be taken into account, including lower image quality as compared to 2D echocardiography. Only a limited number of HES patients were examined, which should be considered when interpreting the results. However, HES is a relatively rare disease. Coronary angiography was performed in one HES case, with a negative result. Moreover, transesophageal echocardiography was not performed to exclude thrombi in HES patients.

Increased LA volumes and stroke volumes were demonstrated in HES patients, accompanied by reduced LA peak circumferential strain as assessed by 3DSTE. These results suggest structural and functional LA remodeling in HES patients.

The authors have no conflicts of interest to declare.