Author + information
- Juliëtte Velu1,
- Friso Kortlandt2,
- Tom Hendriks3,
- Remco Schurer4,
- Ad van Boven5,
- Ben Van Den Branden6,
- Jan van der Heyden2,
- Berto Bouma1,
- Benno Rensing7 and
- Jan Baan1
- 1Academic Medical Center, Amsterdam, Netherlands
- 2St. Antonius hospital, Nieuwegein, Netherlands
- 3University Medical Center Groningen, Groningen, Netherlands
- 4UMCG, Groningen, Netherlands
- 5Medisch Centrum Leeuwarden, Haren, Netherlands
- 6Amphia Hospital, breda, Netherlands
- 7Department of Cardiology, St. Antonius Ziekenhuis, Nieuwegein, Netherlands
Selecting patients for the MitraClip implantation remains challenging. The European Society of Cardiology guidelines prescribe a life expectancy >1 year to consider the procedure. As current clinical data are insufficient to assess the survival in the first years after implantation, it is challenging for the physician to estimate the prognosis.
Between January 2009 and January 2016, 618 consecutive patients treated with percutaneous mitral valve repair in 5 Dutch hospitals were prospectively followed. A simplified risk stratification model was obtained by applying weights to individual predictors proportional to the hazard ratios of the multivariable Cox regression. The score per patient was defined as the sum of the points from each category, with a higher score indicating a higher mortality rate. Low, intermediate, and high risk was defined based on score, number of patients, and observed survival.
Mean age was 74 ± 11 years, >35% had a European System for Cardiac Operative Risk Evaluation score >20%. Multivariable Cox regression analysis showed that age, cardiac implantable electronic devices, previous valve surgery, NYHA functional class, N-terminal pro-B-type natriuretic peptide levels, and MR grade predicted survival after MitraClip implantation (Table 1). Among the 55 high-risk patients (>6 points), the median survival was 18 months, the 1-year survival was 60%, and the 5-year survival was 8%. While the 466 intermediate-risk patients (3 to 6 points) had a 5-year survival of 38% and the 97 low-risk patients (<3 points) had a 5-year survival of 76%. The C-statistic of the risk model against the 1-year survival was 0.71 (95% confidence interval: 0.65 to 0.77).
|HR (95% CI)||p value||Risk Model Score|
|Age at procedure, 70–79 yrs||2.7 (1.3–5.3)||0.006||2|
|Age at procedure, >=80 yrs||3.1 (1.5–6.3)||0.002||2|
|Previous valve surgery||2.4 (1.5–3.9)||<0.001||1|
|NYHA functional class III||1.9 (1.0–3.4)||0.038||1|
|NYHA functional class IV||3.9 (2.1–7.4)||<0.001||3|
|NT-proBNP, 2,000–4,999 ng/l||1.7 (1.1–2.8)||0.031||1|
|NT-proBNP, >=5,000 ng/l||3.1 (1.9–5.2)||<0.001||2|
|MR grade 4||1.8 (1.3–2.5)||0.001||1|
Selecting suitable patients for the MitraClip implantation, while meeting the guidelines, remains challenging. Our risk model might enhance selection by identifying patients with a poor prognosis after MitraClip implantation.
STRUCTURAL: Valvular Disease: Mitral