Long-Term Survival Advantage of Reoperative Surgical Mitral Valve Replacement Over Transcatheter Mitral Valve-in-Valve: A Multicenter Cohort.
Study design
Detecting…
BACKGROUND: Although transcatheter mitral valve-in-valve replacement (mViV) is safe, long-term outcomes compared with reoperative surgical mitral valve replacement (rSMVR) remains unknown. This study compared long-term outcomes between mViV and rSMVR in patients with failing mitral bioprostheses.
METHODS: This retrospective, cohort study analyzed patients undergoing mViV or rSMVR for failing bioprosthetic mitral valves without active endocarditis at 2 quaternary care centers from 2004 to 2023. Primary end points included 30-day Mitral Valve Academic Research Consortium (MVARC) procedural success, 5-year cumulative incidence of death, freedom from reintervention, and cumulative incidence of heart failure. Echocardiographic evaluation of mean mitral valve gradients was assessed at baseline and 5-year follow-up.
RESULTS: Of 229 patients (90 rSMVR, 139 mViV; mean age, 66.0 vs 68.7 years), MVARC 30-day outcomes were similar except for higher prolonged ventilation with rSMVR. At 5 years, rSMVR had significantly lower mortality (20.3% vs 40.9%; P = .01) and lower mean (standard deviation) mitral gradients (immediate postprocedure: 5.1 [2.3] mm Hg vs 7.8 [3.0] mm Hg; 5-year: 5.3 [2.0] mm Hg vs 9.8 [5.3] mm Hg; P < .001). Freedom from mitral valve reintervention (5.8% vs 6.1%; P = .97) and cumulative heart failure incidence (14.3% vs 22.9%; P = .17) were similar.
CONCLUSIONS: Although mViV demonstrated comparable 30-day MVARC outcomes, rSMVR was associated with superior long-term survival and hemodynamic durability among patients deemed appropriate surgical candidates by a multidisciplinary heart team. Treatment decisions should balance short-term risk with long-term outcomes.
💡 Tap a highlighted statistic above for a plain-language explanation.
🔬 Stat Lens — compute the numbers from the paper
Enter the count with the outcome in each arm. ARR > 0 means the intervention reduced the event → NNT; ARR < 0 means it increased it → NNH. RR/OR/NNT are point estimates — check the paper's confidence intervals for precision.
TP/FN are diseased; TN/FP are disease-free (by the reference standard). LR+ > 10 or LR− < 0.1 shifts probability a lot; near 1 barely moves it.
Bayes in odds form: pre-test odds × LR = post-test odds. Use LR+ for a positive test, LR− for a negative one.
Guided critical appraisal
Work through each domain from the paper itself. Stuck? Tap 💡 Coach me — it asks guiding questions, it won't hand you the answer.
Loading the checklist…
Clinical question & bottom line
When does PICO fit? PICO (Population, Intervention, Comparison, Outcome) suits some designs better than
others — adapt it loosely, and leave fields blank when they don't apply. Fits cleanly: RCTs, cohort
(frame the exposure as the intervention), case-control (backward: outcome → exposure), systematic reviews,
diagnostic accuracy (PIRD: Population, Index test, Reference standard, Diagnosis). Looser fit:
cross-sectional (often no comparator), qualitative (SPIDER/PICo suit better), case reports (appraise design,
bias, and applicability instead).
This is a read-only demo — start your own club to save, share, and export.