Mortality Associated with Proportionality of Secondary Mitral Regurgitation After Transcatheter Mitral Valve Repair: MFIRE Registry

BACKGROUNDThe association, if any, between the effective regurgitant orifice (EROA) to left ventricular end-diastolic volume (LVEDV) ratio and 1-year mortality is controversial in patients undergoing mitral transcatheter edge-to-edge repair (m-TEER) with the MitraClip system. The objective is to determine the association between EROA/LVEDV and 1-year mortality among patients undergoing m-TEER with MitraClip. METHODSIn patients with severe secondary (functional) mitral regurgitation (MR), we analyzed registry data from 11 centers using generalized linear models with the generalized estimating equations approach. RESULTSWe studied 525 patients with secondary MR who underwent m-TEER in 11 centers. Most patients were male (63%) and were NYHA class III (61%) or IV (21%). MR was caused by ischemic cardiomyopathy in 51% of patients. EROA/LVEDV values varied widely with median=0.19 mm2/mL, interquartile range [0.12,0.28] mm2/mL, and 187 (36%) patients had values <0.15 mm2/mL. Post-procedural MR severity improved substantially, being 1+ or less in 74%, 2+ in 20%, 3+ in 4%, and 4+ in 2%. 1-year mortality was 22%. After adjustment for confounders, the logarithmic transformation of EROA/LVEDV was associated with 1-year mortality (odds ratio=0.600, 95% confidence interval [0.386,0.933], p=0.023). A higher Society of Thoracic Surgeons risk score was also associated with increased mortality. CONCLUSIONSLower values of Ln(EROA/LVEDV) were associated with increased 1-year mortality in a multi-center registry. The slope of the association is steep at low values, but gradually becomes flatter as Ln(EROA/LVEDV) increases. What is KnownIn recent years, mitral transcatheter edge-to-edge repair (m-TEER) approval has expanded to select patients with functional mitral regurgitation. The paradigm of proportionate and disproportionate MR as measured by EROA/LVEDV has been proposed as a potential rationale to reconcile conflicting trial results. What the Study AddsBefore adjustment, EROA/LVEDV was not associated with mortality (0.22 {+/-} 0.13 vs 0.21 {+/-} 0.14, p=0.425, standardized difference = -0.08). However, after adjustment we found that the logarithmic transformation of the pre-procedural EROA/LVEDV, Ln(EROA/LVEDV) was associated with 1-year mortality (OR = 0.600, 95% CI [0.386, 0.933], p=0.023). Progressively higher values of pre-procedural EROA/LVEDV were always associated with a progressively lower odds ratio of death post-procedurally, without a plateau or cutoff but with an "elbow" around 0.15mm2/mL. Graphic AbstractThe Association of EROA/LVEDV with Mortality. Plot of the adjusted odds ratios associated (solid line) with 95% confidence intervals (dashed lines) with EROA/LVEDV=0.15 as the reference showing the risk of death increases as EROA/LVEDV decreases. The plot is back transformed from Ln(EROA/LVEDV). O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=118 SRC="FIGDIR/small/23287387v1_ufig1.gif" ALT="Figure 1"> View larger version (20K): org.highwire.dtl.DTLVardef@108acaeorg.highwire.dtl.DTLVardef@810440org.highwire.dtl.DTLVardef@3e1fa7org.highwire.dtl.DTLVardef@9027bc_HPS_FORMAT_FIGEXP M_FIG C_FIG