Repeated Echocardiographic Left Ventricular Ejection Fraction Measurements: A Strong And Accessible Tool For Detecting At High Risk Of Heart Failure Thalassemia Major Population

Abstract Abstract 3197 Background: The prognosis for thalassemia major (TM) has dramatically improved in the last two decades. However, many transfusion-dependent patients continue to develop secondary iron overloading, and eventually death, particularly from cardiac disease. The possibility of detecting easily and earliest the patients at risk of cardiac death is so far the main challenge of clinical management of these patients. Therefore, the mean reduction of Left Ventricular Ejection Fraction (LVEF), determined by echocardiography, was evaluated over the time. Methods: Among the 413 observed patients only 188 had complete records for LVEF measurements during, at least, five considered consecutive years. Included patients were divided into two cohorts: the not alive and the alive-group with 22 and 166 patients, respectively. Generalized Estimating Equations (GEE) model was used to show the reduction of the mean of LVEF (Hedeker & Gibbons, 2006). This approach was implemented in the 'xtgee' procedure of Stata 11 software (StataCorp, College Station, TX, USA). The logistic regression model was used to evaluate the risk of death (Collet D. 2003). In this analysis, the mean reduction of LVEF was categorized into three levels: the baseline category including all patients with an increase greater than 0%, the category 1 including all patients with a reduction greater than 0% but less than 7% and the category 2 including all patients with a reduction higher or equal to 7%. All of the statistical analyses were performed under code at the Department for Mathematical and Statistical Sciences 'S. Vianelli', University of Palermo (Italy) by A.V. Results: Baseline findings are shown on Table I. Figure 1 shows the proÞles of the GEE model for the mean LVEF between the two groups. The regression coefficient of Status×Time shows a statistically significant linear decrease over the time of 1,51 per year of the mean LVEF between not alive versus alive patients (Coeff. −1.51, CI (−2,31;−0,71), p-value<0,0001,Fig. 1). Patients with a mean reduction of LVEF greater or equal to 7% over the time had a statistical significant higher risk of death from heart failure (OR= 4,93,95% CI 1,61;15,11, p-value = 0,005). Discussion: Recently, Kirk et al. 2009 suggested as cardiac T2* magnetic resonance is able to detect patients at high risk of heart failure and arrhythmia from myocardial siderosis. However, other studies showed the presence of patients with abnormal heart function and normal heart T2* and did not suggest lower heart T2* for patients suffering from arrhythmia (Pepe et al., 2006; Marsella et al.,2011). Moreover, although the use of T2* is spreading, its availability is so far limited. Instead, availability of echocardiography is surely greater. Moreover, interobserver and intraobserver reliability for the visual assessment of the global LVEF measurements have been extensively shown even in comparison with Magnetic Resonace Imaging (Hoffmann et al. 2005; Gimelli et al. 2008; Blondheim et al., 2008; Sjzli et al. 2011). Therefore, repeated measurements of LVEF may be a strong and more accessible tool for detecting at risk of heart failure TM population. Disclosures: No relevant conflicts of interest to declare.