RESUMEN
Cardiac magnetic resonance (CMR) derived left ventricular global longitudinal strain (LV-GLS) for evaluating dilated cardiomyopathy patients has been addressed in studies with contradictory results. We therefore performed the first systematic review evaluating evidence on the prognostic value of CMR derived LV-GLS for ischaemic (IDCM) and non-ischaemic dilated cardiomyopathy (NDCM) patients. Systematic review (PROSPERO CRD42020171582) identified studies up to January 2021 that measured LV-GLS for predicting major adverse cardiac events among dilated cardiomyopathy patients. Studies were identified from MEDLINE, Embase and PubMed by two independent reviewers. 2099 studies were screened. Three prospective and three retrospective observational studies comprising of 1758 patients (29% IDCM patients; 71% NDCM patients) with a weighted mean follow up of 3 years (SD = 1 year) were identified. All six studies included mortality in the primary composite outcome. LV-GLS was associated with increase primary composite outcome among mild to moderately impaired left ventricular ejection fraction (LVEF) IDCM and NDCM patients (> 30%) in univariable and multivariable analysis. Association was lost among severely impaired LVEF patients (< 30%). From sensitivity analysis, LV-GLS showed significant association with death among NDCM patients (HR 1.27; 95% CI 1.10-1.46; p = 0.001; I2 = 59%) but insignificant for heart transplant outcome (HR 1.23; 95% CI 0.46-3.33; p = 0.68, I2 = 44%). LV-GLS threshold for effectively stratifying patients is - 12.5% to - 13.5%. LVEF in IDCM and NDCM became an insignificant prognostic marker in multivariable analysis. CMR LV-GLS shows promise as an independent predictor of mortality in IDCM and NDCM patients. However, in patients with LVEF < 30% LV-GLS may have less prognostic value.Prospero Registration: CRD42020171582.
