Impact of Pericardial Effusion on Cardiac Mechanics in Patients with Dilated Cardiomyopathy Francesco Scardulla1, Antonino Rinaudo1, Cesare Scardulla2 and Salvatore Pasta3 1Dipartimento di Ingegneria Chimica, Gestionale, Informatica e Meccanica, Universita' di Palermo, Viale delle Scienze Ed. 8, 90128 Palermo, Italy 2Mediterranean Institute for Transplantation and Advanced Specialized Therapies (ISMETT), Via Tricomi n.1, 90127, Palermo, Italy 3Fondazione RiMED, Via Bandiera n.11, 90133, Palermo, Italy Keywords: Finite Element Analysis, Cardiac Mechanics, Cardiomyopathy, Pericardial Effusion. Abstract: Dilated cardiomyopathy (CDM) is a degenerative disease of the myocardium accompanied by left ventricular (LV) remodeling, resulting in an impaired pump performance. Differently, pericardial effusion (PE) is a liquid accumulation in the pericardial cavity, which may inhibit blood filling of heart chambers. Clinical evidence show that PE may improve pump performance in patients with CDM. Therefore, this study aims to assess wall stress and global function of patients with CDM, PE as compared to healthy patient. These findings suggests that CDM has an important implication in the mechanical changes of LV and right ventricle by increasing wall stress and reducing pump function. Conversely, PE determines lowering myocardial fiber stress and improves global function as compared to those of CDM. 1 INTRODUCTION after surgery or can be secondary to congestive, severe heart failure. However, the mechanism of PE Dilated cardiomyopathy (CDM) is a degenerative development and its prognostic value in heart disease of the myocardial tissue accompanied by left failure remain elusive. A persistent PE at ventricular (LV) remodeling (Nakayama et al., echocardiographic follow-up was associated with 1987). The histologic characteristics of CDM unfavourable outcome when compared with patients include hypertrophy of myofibers, myofibrillar lysis, with resolved PE. Indeed, patients with PE exhibit nuclear changes and vacuolization of myocardial worse right ventricular (RV) function, larger right fibers and interstitial fibrosis of the myocardium atrial dimensions, more pronounced tricuspid (Hayashida et al., 1990). LV remodeling is a regurgitation as well as a higher prevalence of multistep process that involves acute dilation of the pulmonary hypertension. A recent study shows that infarcted area, increase of LV volume, lengthening patients with PE have significantly elevated RA of the LV perimeter, and decrease of LV curvature. filling pressures and an increased mean arterial Natural history studies show that progressive LV pulmonary pressure, whereas the left ventricular remodeling is directly related to future deterioration filling pressure and wedge pressure did not differ of LV performance and a poor clinical course (Cohn between PE and control group (Frohlich et al., et al., 2000) (Swynghedauw, 1999). 2013). Pericardial effusion (PE) is a pathological Mathematical modeling of the cardiovascular accumulation of fluid within the pericardial space system using the finite element (FE) approach is an (Mirhosseini et al., 2013). Usually, such disease do useful too to estimate the cardiac mechanics and not influence clinical decision-making as long as the wall stress, likely inhibiting CDM and PE. A few FE PE is not considered haemodynamically modeling studies of the LV have validated stress compromising cardiac functionality (Frohlich et al., calculations by showing good agreement with 2013). PE fluid accumulation can be attributed to an myocardial strain measured with implanted markers underlying systemic or local inflammatory process (McCulloch et al., 1992); (Omens et al., 1993); such as cancer or myo-/pericarditis or might occur (Vetter and McCulloch, 2000). Guccione et al. have Scardulla F., Rinaudo A., Scardulla C. and Pasta S.. Impact of Pericardial Effusion on Cardiac Mechanics in Patients with Dilated Cardiomyopathy. 639 DOI: 10.5220/0004615906390644 In Proceedings of the 3rd International Conference on Simulation and Modeling Methodologies, Technologies and Applications (BIOMED-2013), pages 639-644 ISBN: 978-989-8565-69-3 Copyright c 2013 SCITEPRESS (Science and Technology Publications, Lda.) SIMULTECH2013-3rdInternationalConferenceonSimulationandModelingMethodologies,Technologiesand Applications successfully modeled end-isovolumic systole in an and end-systole (ES), which are defined as the ovine model of myocardial infarction and images with the maximum and minimum cross- determined material parameters that reproduced sectional area, respectively. Patients with PE circumferential stretching (as measured with 2D required also segmentation of the outer pericardial tagged MRI) in the infarcted border zone (Guccione layer. After segmentation process, endocardial and et al., 2001). epicardial surfaces of LV and RV were obtained by The key role of wall stress in the progression of loft protrusion of segmented contour lines. LV remodeling was studied in DCM (Quarterman et al., 2002); (Ratcliffe et al., 1998); (Zhong et al., 2.3 FE Model 2009). An increase in wall stress is known to reduce myocardial fiber shortening, and increases in LV The space between the endocardial and epicardial wall stress have been reported in DCM. LV wall surfaces was meshed with 8-node brick elements to stress is in part determined by the local curvature of generate a volumetric mesh in ABAQUS FE code. the ventricular wall (i.e., decreased curvature will Cardiac myofiber angles at the epicardium and increase wall stress). In addition to increasing LV endocardium were assigned to be -60 degrees and 60 size, CDM can alter myocardial properties and degrees, respectively (counterclockwise positive normal LV shape curvature. The border zone will when viewed from the epicardium). have a higher stress, which makes it more Nearly incompressible, transversely isotropic, susceptible to ischemia and infarction and may hyperelastic constitutive laws for passive and active accelerate the remodeling process. myocardium was implemented in ABAQUS/Explicit Therefore, the purpose of present study was to using a VUMAT subroutine (Ratcliffe et al., 1998). assess the key role of wall stress and global function Myocardial material parameters were estimated of patients with CDM, PE. Cardiac mechanics was comparing MRI measured and computationally thus compared to that of healthy patients with derived LV and RV volumes at ED and ES, normal wall thickness. We also tested the hypothesis respectively. Manual iteration was used rather than that pump function and wall stress in CDM can be formal optimization. Similarly, PE was modeled as positively affected by the presence of a PE liquid. isotropic, hyperelastic material which mechanical properties were empirically found to match ED and ES volumes of PE. 2 MATERIALS AND METHODS The basal node of LV were constrained along long axis direction. The endocardial wall was loaded 2.1 Imaging Procedure at ED and ES pressure occurring at LV and RV. For LV, ED pressure (PED) was 100 mmHg while ES We retrospectively identified patients with CDM pressure (PES) was 25 mmHg. For RV, PED was 15 and PE who underwent magnetic resonance imaging mmHg while PES was 30 mmHg. (MRI) from radiologic records of Mediterranean Institute for Transplantation and Advanced 2.4 Pressure-volume Relationships Specialized Therapies (ISMETT) and Ospedale and Stroke Volume Riuniti Trieste. Patients underwent MRI as part of their care, and not for the purpose of our study. Chamber ED and ES volume (VED and VES) A series of long- and short-axis images of the solutions were used with the corresponding PED and heart were obtained performing MR imaging PES to plot the ED and ES pressure-volume synchronized to the R wave of the electrocardiogram relationships (ESPVR and EDPVR, respectively), signal. Short -axis slices were taken sequentially which were then fit to appropriate polynomial every 6 mm until complete scanning of heart equations. The following linear equation was used to chambers. estimate the ESPVR: P E V V (1) 2.2 Heart Reconstruction where EES is the end-systolic elastance and V0 is the Endocardial and epicardial MRI surfaces of LV and volume intercept of the ESPVR, each determined by RV were segmented by contour lines using the linear regression of the data. vascular modeling toolkit VMTK The polynomial equation used to estimate the (http://www.vmtk.org). Specifically, LV and RV EDPVR was as follows: geometries were reconstructed at end-diastole (ED) P E, E,V E,V (2) 640 ImpactofPericardialEffusiononCardiacMechanicsinPatientswithDilatedCardiomyopathy Figure 1: Representative map of average wall stress for (A) healthy patient, (B) CDM and (C) PE (solid grey indicates liquid volume); models are not at same scale. where E0,ED, E1,ED and E2,ED represent stiffness of the average fiber stress for patients with PE and CDM as LV diastolic compliance, again determined by linear compared to the healthy patient. It can be observed regression. that ES LV stress is higher than that of RV, and this To determine global changes to pump function, occurs in the lateral side of LV chamber. the stroke volume (SV)/PED and SV/VED Differently, the patient with PE show higher ES wall relationships were calculated and plotted, assuming stress at the endocardial surface a cause of the liquid that arterial elastance (EA) was constant. SV was constraining LV wall motion. calculated according to the following equation: Maximum values of ES average fiber stress was found markedly higher for patients