Comparación del strain longitudinal global tomado con diferentes vendor en pacientes con trasplante cardíaco

Abstract

Background: Global longitudinal strain (GLS) is a useful tool for evaluation of rejection in Heart Transplant patients (HT-P). However, controversies exist regarding its accuracy and agreement among different vendors. Methods. Stable HT-P underwent simultaneous GLS analysis with vendors (Philips® and General Electric® -GE-). We evaluated GLS agreement between Philips and GE using mean signed difference, Bland-Altman plots and a pre-specified coverage probability of 2 % points. Then, patients were divided into four groups according to rejection history. Differences between groups were performed with T student test. p value < 0.05 was considered significant. Results. In 78 studies were performed. Limits of agreement between vendors were broad, ranging from -6.52% to 6.35%. The mean signed difference between GLS measured by Phillips and GE, was closest to 0, indicating no substantial GLS over or under-estimation by either vendor. However, the percentage of observations that fell within a 2% range was 44 % between vendors. Remarkably, although the mean left ventricle ejection fraction (LVEF) was normal (55.6  7.6%), the average GLS is low as measured by either vendor. Analysis of GLS showed a significant higher GLS in patients without HTr history (Group A), with statistical significance as compared with B, and D. No differences on GLS between groups of patients with any grade of rejection history (Group B vs C). No differences on LVEF were found in any group comparison. Conclusion: Vendors must work in improving agreement in GLS analysis due to the meaningful information it provides in HT-P and other cardio-myopathies.