Impact of left ventricular remodelling on outcomes after left ventriculoplasty for ischaemic cardiomyopathy: Japanese surgical ventricular reconstruction group experience

Abstract

OBJECTIVES: Surgical ventricular reconstruction (SVR) for patients with severe left ventricular (LV) remodelling due to ischaemic cardiomyopathy is still controversial, because the Surgical Treatment for Ischaemic Heart Failure (STICH) trial demonstrated that SVR not only has no beneficial effect on survival compared with coronary artery bypass grafting (CABG) alone, but also is worse for those with a larger LV. Therefore, we assessed the impact of LV remodelling on the outcomes after SVR for ischaemic cardiomyopathy in Japan, using Di Donato's LV shape classification. METHODS: From 2000 to 2010, 627 patients underwent SVR for ischaemic heart failure in 11 Japanese hospitals. To assess the patients with an LV ejection fraction (LVEF) of ≤35% like the STICH trial, considering the severity of LV remodelling, the patients with a preoperative LVEF of >35%, no preoperative LV volume assessment and no preoperative LV shape classification were excluded. Finally, 323 patients were selected as the study subjects. The LV shape was divided into three types according to Di Donato's classification. Types 1 and 3 indicate the aneurysmal and globally akinetic LV, respectively. Type 2 is the intermediate shape. RESULTS: Type 1, 2 and 3 LV shapes were observed in 85 (26%), 104 (32%) and 134 (42%) of the patients, respectively. The preoperative LV volume and diameter increased if the LV became more akinetic (Type 3 > 2 > 1, P < 0.001). LVEF was lower in those with more akinetic LV (P = 0.002). The preoperative LV end-diastolic volume index and LVEF in Type 3 patients were 133 ± 47 ml/m2 and 22 ± 7%, respectively. Mitral valve repair was more frequently performed for patients with the Type 3 LV shape (65%) than for the others (P < 0.001). The hospital mortality rates were 2.4, 2.9 and 7.4% for Type 1, 2 and 3 patients, respectively (P = 0.16). Kaplan-Meier analysis demonstrated no significant difference in mortality among the three groups (log-rank P = 0.37). The 5-year survival rates were 81, 70 and 73% for Type 1, 2 and 3 patients, respectively. CONCLUSIONS: The severity of LV remodelling did not affect survival after SVR plus CABG. The results of SVR were acceptable even for those with globally akinetic LV due to ischaemic cardiomyopathy. © The Author 2013. Published by Oxford University Press on behalf of the European Association for Cardio-Thoracic Surgery. All rights reserved.