Transcatheter Repair versus Mitral-Valve Surgery for Secondary Mitral Regurgitation (MATTERHORN)
Authors, Journal, Affiliations, Type, DOI
- Authors: Stephan Baldus, Torsten Doenst, Roman Pfister, Jan Gummert, et al., for the MATTERHORN Investigators
- Journal: New England Journal of Medicine, Volume 391, Issue 19, Pages 1787–1798
- Lead affiliations: Department of Cardiology, University Hospital Cologne (Baldus); multiple German university centers
- Type: Multicenter, investigator-initiated, randomized, noninferiority trial; Germany only
- DOI: 10.1056/NEJMoa2408739
- Funded by: Unrestricted research grant from Abbott Vascular
- Published: August 31, 2024 (online); November 14, 2024 (print); ClinicalTrials.gov NCT02371512
Overview
MATTERHORN is the first RCT to directly compare transcatheter edge-to-edge repair (TEER, MitraClip) versus surgical mitral valve repair or replacement in patients with heart failure and secondary MR — the head-to-head question that ESC 2021 and ACC/AHA 2020 guidelines had left unanswered due to absence of randomized data. TEER was noninferior to surgery on the 1-year composite efficacy endpoint (death, HF hospitalization, MV reintervention, LVAD, or stroke: 16.7% vs 22.5%; P<0.001 for noninferiority) while being dramatically safer at 30 days (14.9% vs 54.8% composite major adverse events; P<0.001). The 30-day safety advantage was driven by markedly lower major bleeding (3.1% vs 24.4%), new-onset AF (3.1% vs 27.8%), and stroke (0% vs 4.4%) with TEER. MR recurrence ≥3+ at 1 year was numerically higher with TEER (8.9% vs 1.5%), flagging a durability concern requiring longer follow-up. The population was lower-risk than COAPT (mean LVEF 43%, STS-PROM 2.0%), limiting extrapolation to advanced HFrEF.
Keywords
Transcatheter edge-to-edge repair, MitraClip, mitral valve surgery, secondary mitral regurgitation, noninferiority, major bleeding, atrial fibrillation, MATTERHORN, heart failure
Key Takeaways
Background and Context
- Isolated MV surgery for secondary MR reduces symptoms in prospective registries and non-randomized studies
- TEER has become the dominant approach (up to two-thirds of TEER procedures are for secondary MR)
- Guideline positions were discordant: ESC 2021 reserved TEER for patients not eligible for surgery; ACC/AHA 2020 reserved surgery for anatomy unfavorable for TEER
- No randomized head-to-head comparison of TEER vs surgery in secondary MR existed before MATTERHORN
Study Design
- Multicenter, prospective, randomized, controlled noninferiority trial; Germany only
- Enrollment: February 2015 – December 2022; n=210 randomized (104 TEER, 104 surgery)
- Noninferiority margin: 17.5 percentage points (upper bound of 90% CI of mean difference)
- 30-day and 1-year follow-up visits; central echocardiographic core laboratory (University of Mainz)
- All TEER procedures performed with latest available MitraClip iteration; surgical technique left to surgeon's discretion (access, ablation, tricuspid surgery)
Eligibility Criteria
- Clinically significant secondary MR (≥2 of: EROA ≥20 mm², vena contracta >8 mm, regurgitant volume ≥30 mL, regurgitant fraction ≥50%, or ≥2 HF hospitalizations in prior 12 months)
- LVEF ≥20%; NYHA class II or higher despite GDMT
- Eligible for BOTH TEER AND surgery as determined by local heart team
- Exclusion: Additional severe valvular disease; PCI/CRT within 1 month prior
Baseline Characteristics
| Parameter |
Value |
| Mean age |
70.5 ± 7.9 years |
| Women |
39.9% |
| Mean LVEF |
43.0 ± 11.7% |
| MR ≥3+ |
96.0% (4+: 38.2%) |
| Median EROA |
0.22 cm² (IQR 0.17–0.28) |
| Mean regurgitant fraction |
57.0 ± 21.0% |
| Median STS-PROM |
2.0% (IQR 1.1–3.7) |
| Median EuroSCORE II |
3.0% (IQR 1.7–4.3) |
| CRT devices |
13.1% |
| NYHA III/IV at baseline |
82.4% (TEER arm) / 89.1% (surgery arm) |
Procedures
- TEER group: Attempted in 102/104 (1 crossed over due to intraprocedural chord rupture; 2 withdrew); 48.5% received 1 device, 40.6% received 2 devices
- Surgery group: 94/104 underwent assigned surgery; 72.0% MV repair, 28.0% MV replacement
- MR ≤2+ at 1 year: 96.1% TEER vs 98.6% surgery
Primary Efficacy Results (1 Year)
| End Point |
TEER |
Surgery |
Difference (95% CI) |
P for NI |
| Composite (death/HF hosp/reintervention/LVAD/stroke) |
16.7% (16/96) |
22.5% (20/89) |
−6 pp (−17 to 6) |
<0.001 |
- Relative risk ratio: 0.74 (95% CI 0.41–1.34)
- Log-rank P=0.30 — not superior to surgery
- Kaplan–Meier estimate: 16.9% TEER vs 22.7% surgery (P=0.30)
- Multiple imputation (missing at random): estimated difference −12 pp (95% CI −24 to 1) — supports noninferiority
Primary Safety Results (30 Days)
| End Point |
TEER |
Surgery |
Difference (95% CI) |
P |
| Composite major adverse events |
14.9% (15/101) |
54.8% (51/93) |
−40 pp (−51 to −27) |
<0.001 |
Key 30-day safety components:
| Component |
TEER |
Surgery |
Difference |
| Death |
2.0% |
4.3% |
— |
| Major bleeding (VARC) |
3.1% |
24.4% |
−21 pp (−31 to −12) |
| New-onset AF |
3.1% |
27.8% |
−25 pp (−35 to −15) |
| Stroke/TIA |
0% |
4.4% |
— |
| Rehospitalization |
7.2% |
11.6% |
— |
Secondary Results (1 Year)
| End Point |
TEER |
Surgery |
Result |
| MR recurrence ≥3+ at 1 yr |
8.9% (7/79) |
1.5% (1/65) |
+7 pp (0 to 14); NI met (margin 17.5 pp) |
| 6-min walk distance change |
+31 m |
+32 m |
Diff +5.2 m (−42 to 53); NI met (margin 50 m) |
| NYHA III/IV at 1 year |
23.3% |
18.2% |
No significant difference |
| Minnesota QoL score change |
−10 (IQR −20 to 0) |
−5 (IQR −19 to 4) |
Numerically favors TEER |
| Secondary safety (1 yr) |
36.5% (35/96) |
75.3% (70/93) |
−39 pp (−51 to −25) |
| Any adverse event |
41.0% |
77.3% |
— |
| Serious adverse events |
35.0% |
66.0% |
— |
As-Treated Analysis
- Primary efficacy: 13.8% TEER vs 25.3% surgery (RRR 0.55; 95% CI 0.30–1.01) — directionally favors TEER, CI crosses 1.0
- Primary safety: 12.1% vs 56.8% — consistent with ITT
Comparison with Other SMR Trials
| Parameter |
MATTERHORN |
COAPT |
RESHAPE-HF2 |
| Comparator |
Surgery |
Medical Rx |
Medical Rx |
| Mean LVEF |
43% |
31% |
31% |
| Median EROA |
0.22 cm² |
0.40 cm² |
0.23 cm² |
| STS-PROM |
2.0% |
N/A |
N/A |
| Primary event rate (control) |
22.5% |
High |
High |
| Study type |
Noninferiority |
Superiority |
Superiority (3 primaries) |
- MATTERHORN patients less sick: higher baseline LVEF, lower event rate than COAPT/RESHAPE-HF2
Limitations
- Single country: Germany only; results may not generalise to different healthcare systems or surgical volumes
- Lower-than-expected event rate: Control arm 22.5% (lower than the assumed 35%); trial may have been underpowered to detect smaller efficacy differences
- Open-label: Patients and investigators unblinded; standardized echo core laboratory mitigates echocardiographic bias
- 7-year enrollment with device drift: Latest MitraClip and latest surgical techniques not available to entire cohort
- Short 1-year follow-up: MR durability differences (8.9% vs 1.5% recurrence ≥3+) may compound over time; longer follow-up needed
- Less advanced HF: Mean LVEF 43% — less sick than COAPT/RESHAPE-HF2; less applicable to LVEF <35% population
- GDMT not maximised: Pharmacotherapy not administered as vigorously as in contemporary trials; <50% of patients had LVEF <40% (the main HFrEF population)
- Imputation for 18 deaths: Patients who died before 1-year assessment had values imputed; sensitivity analyses supportive
- Industry funded: Abbott Vascular unrestricted grant; academic investigators maintained oversight
Key Concepts Mentioned
Key Entities Mentioned
- entities/Heart-Failure — underlying condition; important note that MATTERHORN population had higher LVEF than COAPT
Wiki Pages Updated
wiki/sources/mvteer-matterhorn-nejm-2024.md — createdwiki/sourceindex.md — updatedwiki/wikiindex.md — updatedwiki/concepts/Secondary-Mitral-Regurgitation.md — updated with MATTERHORN head-to-head data