Transcatheter chordal repair enables off-pump, echocardiography-guided treatment of degenerative mitral regurgitation, but real-world outcomes remain variably reported. We performed a systematic review and meta-analysis of studies evaluating outcomes after transcatheter chordal repair in adults with primary degenerative mitral regurgitation. Pooled event rates were calculated using random-effects models. Seventeen studies ( n = 3,787) were included. Mitral regurgitation ≤ mild was achieved in 93.9% of patients at discharge and 82.5% at 30 days, while moderate mitral regurgitation occurred in 5.0% and 17.5%, respectively. Procedural success was 96.2%, although small-study effects were suggested. Early adverse events were infrequent, with 30-day mortality of 1.2%, stroke of 0.7%, and myocardial infarction of 1.2%. Reintervention occurred in 4.5%. In conclusion, transcatheter chordal repair demonstrates high technical success and favorable early safety; however, early mitral regurgitation recurrence and heterogeneity highlight the need for anatomy-guided patient selection and cautious interpretation given single-arm study limitations.
Highlights
-
•
Pooled procedural success of transcatheter chordal repair exceeds 96%.
-
•
Over 93% of patients achieved mild or less MR at discharge.
-
•
Low 30-day rates of mortality (1.2%), stroke (0.7%), and MI (1.2%).
-
•
Reintervention rates were low but varied with anatomical complexity.
-
•
Anatomy-driven patient selection is key to durable MR reduction.
Degenerative mitral regurgitation (MR) due to leaflet prolapse or flail is a common indication for intervention, with surgical repair considered the standard of care. However, many patients are not suitable for surgery because of advanced age, comorbidities, or frailty, creating a need for less invasive therapies that can address the primary leaflet pathology without cardiopulmonary bypass. Transcatheter chordal repair has emerged as a novel, minimally invasive approach, replicating the surgical principle of neochordal implantation on the beating heart under real-time echocardiographic guidance, most commonly via a transapical route. ,,, Early feasibility studies demonstrated both safety and technical success of transapical neochordal implantation, with meaningful acute reductions in MR. ,, Subsequent multicenter cohorts and institutional series reported procedural refinements, learning-curve effects, and encouraging midterm outcomes, including 5 year durability in selected anatomies. ,, Parallel platforms such as HARPOON have provided further proof-of-concept for echo-guided, off-pump neochordal implantation. ,, While these experiences suggest that transcatheter chordal repair is feasible and clinically effective, the evidence base remains fragmented across small, heterogeneous studies. Important knowledge gaps persist. Reported rates of procedural success and early MR are highly variable. Data on safety events such as stroke, tamponade, and early mortality are limited by small sample sizes. Durability signals, including reintervention rates and recurrent MR, remain inconsistent. Comparative data with surgical or transcatheter edge-to-edge repair are also scarce. ,,,, We therefore performed a systematic review and meta-analysis of contemporary studies on transcatheter chordal repair for degenerative MR. Our objectives were to (1) quantify procedural success and early MR reduction, (2) evaluate safety outcomes, and (3) summarize durability signals across available follow-up. This synthesis provides benchmark estimates to inform patient selection, guide clinical decision-making, and aid the design of future comparative trials.
Materials and Methods
This systematic review and meta-analysis was conducted in accordance with PRISMA 2020 guidelines. The protocol was predefined before literature screening and data extraction, and that all analyses were conducted according to this predefined plan.
Data sources and search strategy
A comprehensive search of PubMed/MEDLINE and Embase (Elsevier) was performed from database inception through September 9, 2025, without language restrictions. We combined controlled vocabulary and free-text terms for degenerative mitral regurgitation and transcatheter neochordal/chordal repair (e.g., “mitral regurgitation” OR “mitral valve insufficiency” AND “ NeoChord” OR “HARPOON” OR “neochord” OR “ chordal repair” OR “ transapical” ), using Boolean operators AND/OR and field tags appropriate to each database. Reference lists of eligible articles and relevant reviews were also screened. The full database-specific strategies are provided in the Supplementary File .
Eligibility criteria
Inclusion:
-
1.
Population: Adults (≥18 years) with primary/degenerative MR (leaflet prolapse or flail).
-
2.
Intervention: Transcatheter chordal repair on the beating heart (e.g., NeoChord, HARPOON), typically via transapical access.
-
3.
Design: Prospective or retrospective cohorts/registries (single-arm or comparative) reporting extractable outcome data.
-
4.
Outcomes: At least 1 prespecified endpoint (defined below).
Exclusion: Case series with <10 patients; studies focused on secondary/ischemic or rheumatic MR; procedures primarily involving transcatheter edge-to-edge repair (TEER) or surgical annuloplasty without neochord implantation; intraoperative surgical neochord-only techniques; duplicate populations (largest/most recent cohort retained).
Study selection
Two independent reviewers (PA and KM) independently screened all records using the Rayyan platform. During title and abstract screening, 5 disagreements occurred among 116 records, corresponding to an agreement rate of 95.7%. During full-text review of 20 retrieved articles, no disagreements occurred (agreement 100%). All conflicts were resolved through discussion and, when necessary, consultation with a third reviewer (DP), resulting in full consensus. Given the very small number of conflicts and the absence of the full 2×2 reviewer decision matrix from the exported screening logs, formal inter-rater reliability(IRR) statistics such as Cohen’s kappa were not calculated.
Outcomes
Primary efficacy outcomes were: (1) MR ≤ mild at discharge and 30 days and (2) procedural success (per study definition; mapped to MVARC when available ).
Secondary outcomes included: residual ≥ moderate MR at discharge/30 days; safety events (all-cause mortality, stroke, myocardial infarction, pericardial tamponade, major bleeding as reported); mitral reintervention (surgical or transcatheter); and cardiac reverse remodeling metrics (e.g., change in left ventricular ejection fraction [LVEF] or LV volumes). For multiple time points, we prioritized discharge/30-day results for acute efficacy/safety and the longest available follow-up for durability. For the purpose of meta-analytic pooling, “MR ≤ mild” (≤2+) was defined as the presence of none/trace or mild residual regurgitation on transthoracic or transesophageal echocardiography and was considered indicative of procedural or early echocardiographic success. “MR ≥ moderate” (≥3+) encompassed both moderate and severe regurgitation and was analyzed as the composite endpoint of clinically significant residual or recurrent MR.
Data extraction and risk of bias
Data extraction was independently performed by 2 authors (PA and KM) using a predesigned Microsoft Excel spreadsheet. Any discrepancies were addressed and resolved by a third author (DP). The extracted data included the first author’s name, publication year, country of study, study type, sample size, mean age, male/female, baseline LVEF (%), baseline MR grade, type of transcatheter chordal repair used, and the reported outcomes. The quality of observational studies was evaluated using the Newcastle–Ottawa Scale (NOS), an 8-item tool designed to assess the quality of nonrandomized studies in meta-analyses. The NOS assigns scores ranging from 0 to 9 stars, with scores of 7 or higher indicating high quality and those below 7 signifying low quality.
Risk of bias was assessed at the study level using an appropriate tool for single-arm observational studies. The assessment focused on key methodological domains including patient selection, outcome assessment, completeness of follow-up, and reporting transparency. Given the absence of comparator arms, domains related to randomization, allocation concealment, and blinding were not applicable. Overall, the included studies demonstrated a moderate risk of bias, primarily driven by nonrandomized designs, single-center experiences, and potential learning-curve effects inherent to early device adoption. A summary of the risk of bias assessment is provided in the Supplementary Material ( Figure S1 ).
Handling of missing data
Missing data were handled using a complete-case analysis approach. Only studies reporting extractable data for the outcomes of interest were included in the respective pooled analyses. No statistical imputation was performed, as missingness primarily occurred at the study level rather than the individual-patient level, which is typical for aggregate-data meta-analyses. Accordingly, the denominator for each pooled outcome reflects only the number of studies reporting that specific endpoint.
Statistical analysis
All analyses were performed using MedCalc Statistical Software, version 23.3.7 (MedCalc Software Ltd, Ostend, Belgium). A single-arm meta-analysis was conducted to estimate pooled event rates with 95% confidence intervals (CIs) for each endpoint.
Pooled proportions were calculated using both fixed-effects (inverse variance weighting) and random-effects (DerSimonian–Laird) models, with the random-effects model prespecified as primary. Heterogeneity was assessed with Cochran’s Q test (p <0.10) and I² statistic, categorized as low (<25%), moderate (25% to 75%), or high (>75%). Publication bias was evaluated using Egger’s regression test (p <0.05). All tests were 2-tailed, with p <0.05 considered significant. Forest plots were generated in MedCalc, and R to display pooled estimates and heterogeneity.
Small-study effects were assessed visually using funnel plots for outcomes with a sufficient number of studies. In addition, Egger’s regression test was performed where appropriate to statistically evaluate funnel plot asymmetry. Given the limited number of studies for certain endpoints and the presence of substantial between-study heterogeneity, these assessments were considered exploratory.
Results
Following PRISMA 2020 guidelines, 119 records were identified through PubMed ( n = 101) and Embase ( n = 18). After removing 3 duplicates, 116 unique records were screened and 96 excluded. About 20 full-text articles were retrieved, 1 was unavailable, and 2 were excluded (conference abstract and noneligible design). About 17 studies were included in the final meta-analysis as shown in Figure 1 . Studies comprising 3,491 to 3,787 patients with degenerative MR were included. Most patients presented with isolated posterior leaflet prolapse or flail, while anterior and bileaflet disease were less frequent. Outcome definitions varied across studies, though Mitral Valve Academic Research Consortium (MVARC) criteria were used wherever available. Outcome definitions varied across studies, though Mitral Valve Academic Research Consortium (MVARC) criteria were used wherever available. A comprehensive summary of all pooled outcomes is presented in Table 1 .
Echocardiographic Outcomes
At discharge, MR ≤ mild was achieved in 93.9% (95% CI, 91.8 to 96.0; I² = 0. 878, p <0.0001), demonstrating excellent immediate procedural efficacy.
At 30 days, MR ≤ mild persisted in 82.5% (95% CI, 74.3 to 90.6; I² = 0. 559, p = 0.0594), confirming early durability with only modest heterogeneity.
Residual moderate MR was uncommon, occurring in 5.0% (95% CI, 3.3 to 6.7; I² = 0. 711, p <0.0001) at discharge and 17.5% (95% CI, 12.0 to 23.0; I² = 0. 208, p = 0.2822) at 30 days.
These results demonstrate that while mild recurrence of MR occurred in some patients within the first month, overall valve competence remained well-preserved across studies. Both Mild and Moderate MR at discharge and 30 days respectively are shown in Figures 2-5 .
Procedural success
The pooled procedural success was 97.9% (95% CI, 97.1 to 98.7) under the fixed-effects model and 96.2% (95% CI, 93.1 to 98.4) under random-effects modeling.
Between-study heterogeneity was moderate (I² = 0.804, p <0.0001) and likely reflected differences in patient selection, operator experience, and device iteration. Egger’s regression suggested a small-study effect, indicating potential publication bias (p = 0.034). Funnel plot inspection suggested asymmetry, and Egger’s regression indicated a potential small-study effect (p = 0.034). Procedural success after transcatheter chordal repair is shown in Figure 6 .
Safety outcomes
Early adverse event rates were low. Thirty-day mortality averaged 1.2% (95% CI, 0.8 to 1.7; I² = 0.105, p = 0.331). Stroke occurred in 0.7% (95% CI, 0.3 to 1.3; I² = 0. 135, p = 0.318), and myocardial infarction in 1.2% (95% CI, 0.6 to 2.1; I² = 0, p = 0.960). All pooled safety endpoints exhibited minimal heterogeneity and high consistency across centers.
Reintervention
The pooled incidence of reintervention was 2.5% (95% CI, 2.1 to 3.1) under fixed-effects and 4.5% (95% CI, 2.8 to 6.6) under random-effects modeling (I² = 0. 784, p <0.0001).
Egger’s regression suggested a potential small-study effect (p = 0.0066), though absolute event rates remained low. These findings are presented in Table A1.
Discussion
This meta-analysis demonstrates that transcatheter chordal repair for degenerative MR is characterized by high procedural success and favorable safety, with early mortality, stroke, and myocardial infarction rates consistently below 2%. These results reaffirm the technical feasibility and reproducibility of the approach across multiple centers. ,,
Procedural success and early safety
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
