Transcatheter aortic valve implantation (TAVI) is an effective alternative therapy in selected patients with severe aortic stenosis. The role and effects of coexistent moderate to severe mitral regurgitation (msMR) in patients who undergo TAVI remain unclear. Thirteen studies enrolling 4,839 patients who underwent TAVI, including patients with msMR, were considered in a meta-analysis and analyzed for all-cause-mortality; a further meta-analysis was performed to assess mitral regurgitation (MR) evolution after TAVI. In patients with msMR, all-cause-mortality after TAVI was significantly increased at 30-day (effect size [ES] −0.18, 95% confidence interval [CI] −0.31 to −0.04, I 2 = 46.51, Q = 7.48), 1-year (ES −0.22, 95% CI −0.36 to −0.08, I 2 = 56.20, Q = 11.41), and 2-year (ES −0.15, 95% CI −0.27 to −0.02, I 2 = 0.00, Q = 2.64) follow-up compared with patients with absent or mild MR, independent of baseline left ventricular ejection fraction. Interestingly, the impact of msMR on outcomes was statistically stronger when the CoreValve system was used. TAVI was also associated with an improvement in MR entity at 3- and 6-month follow-up (overall ES −0.19, 95% CI −0.37 to −0.01, I 2 = 61.52, Q = 10.39). In conclusion, the presence of preoperative msMR in patients with severe, symptomatic aortic stenosis who undergo TAVI negatively affects outcomes after TAVI. In addition, in the same group of patients, a trend toward a reduction in MR severity was observed. Whether the decrease in MR severity affects mortality after TAVI remains to be defined.
Highlights
- •
Moderate to severe MR in patients with severe AS negatively affects outcomes after TAVR.
- •
This finding is particularly true when the CoreValve system is used.
- •
TAVR is associated with a trend toward a reduction in MR severity.
- •
Further studies including direct comparisons of different valves are warranted.
- •
Whether MR recovery affects survival is still a matter of concern.
Transcatheter aortic valve implantation (TAVI) has emerged as an alternative treatment for patients with symptomatic severe aortic stenosis (AS) who are at an unacceptably high risk for conventional surgical aortic valve replacement (SAVR). Although good procedural success and favorable clinical outcomes have been reported, issues remain regarding optimal patient selection. Indeed, risk calculators commonly used for valvular surgery are not considered accurate in patients who undergo TAVI, because they do not account for all clinical characteristics that may significantly affect percutaneous procedural and postprocedural mortality. Since its introduction into clinical practice, selection criteria for TAVI have significantly broadened. As a consequence, patients currently found in TAVI registries and trials are often treated with off-label indications. TAVI in patients with mitral regurgitation (MR) is one example. The prevalence of MR reaches 20% in patients with severe AS and initially represented an exclusion criterion for TAVI. Although in current clinical practice, the degree of MR is not considered a contraindication to the procedure, its role on survival is controversial. Moreover, it is still unclear whether TAVI could exert a beneficial effect on MR severity after the procedure. Thus, the aims of this meta-analysis were to define the impact of MR severity on outcomes after TAVI and to assess whether TAVI might reduce MR.
Methods
The study was designed according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses requirements. MEDLINE, the Cochrane database, the ISI Web of Science, and SCOPUS were searched for reports published from April 2002 (the date of the first-in-human TAVI procedure) until January 2014. Studies were identified using the major medical subject heading “transcatheter aortic valve implantation OR transcatheter aortic valve replacement OR TAVI OR TAVR” combined with “mitral regurgitation.” Citations were screened at the title and abstract level by 2 independent reviewers and retrieved as full reports if they reported data on outcomes after TAVI, on the basis of MR severity. No language limitations were applied. The full texts and references lists of all potential reports also were retrieved in detail to seek additional relevant studies.
Studies were included if (1) TAVI was performed in high–surgical risk or inoperable patients (as defined by a logistic European System for Cardiac Operative Risk Evaluation score >20 or by the presence of contraindications to surgery such as porcelain aorta, severe respiratory failure, and unfavorable anatomy for sternotomy) with symptomatic, severe AS (defined as an aortic valve area <1 cm 2 or an indexed aortic valve area <0.6 cm 2 /m 2 ), (2) MR severity was classified according to guidelines, and (3) 1 of the following criteria applied: reported data of mortality outcomes on the basis of MR severity and reported data of MR at baseline and after TAVI.
Studies were excluded if any of the following criteria applied: (1) duplicate publication, (2) lack of data on MR before TAVI, (3) lack of data on MR after TAVI, and (4) the outcome of interest was not clearly reported or was impossible to extract or calculate from the published results.
Two reviewers independently screened reports for fulfillment of the inclusion criteria. Reviewers compared selected trials, and discrepancies were resolved by consensus. Baseline characteristics, MR at baseline and during follow-up after TAVI, and outcomes, including mortality outcomes and MR variations, were abstracted.
The primary end point evaluated was the incidence of all-cause mortality in patients with moderate to severe MR (msMR) who underwent TAVI. The secondary end point was MR reduction after TAVI. Of 507 reports identified by the initial search, 30 were retrieved for more detailed evaluation, and 13 trials were included in the study ( Figure 1 ).
Means, SDs, p values, and correlations were used. Overall estimates of effect (effect size [ES]) were calculated with a random-effects model. When correlations were not available, missing correlations were imputed, and changes in ejection fraction from baseline to follow-up were obtained according to the Cochrane handbook. Statistical significance was set at p <0.05 (2 tailed). Heterogeneity was assessed by the Q statistic and I 2 test. Significant heterogeneity was considered present for p values <0.10 or an I 2 value >50%. Data analysis was performed using ProMeta version 2 (Internovi, Cesena, Italy). For verification of the robustness of the results, sensitivity analyses were performed to test the influence of potential effect modifiers, including mean age, gender, baseline functional class (New York Heart Association class III or IV), left ventricular ejection fraction at baseline, atrial fibrillation (AF), type of valve implanted, diabetes, chronic kidney disease, coronary artery disease (CAD), chronic obstructive pulmonary disease, hypertension, logistic European System for Cardiac Operative Risk Evaluation score, Society of Thoracic Surgeons score, previous myocardial infarction, previous coronary artery bypass grafting, previous percutaneous coronary intervention, previous stroke and/or transient ischemic attack, and study publication year.
The significance level for all meta-regression analyses was set at p ≤0.05. Publication bias was assessed using funnel plots and Egger’s test, consisting of a linear regression of the intervention effect estimates on their standard errors, weighting by 1/(variance of the intervention effect estimate). If there was some evidence of publication bias, the trim and fill method, which estimates the number and results of potential missing studies resulting from publication bias, was applied.
Results
Of the 507 articles identified in the initial search, 30 were retrieved for more detailed evaluation. Seventeen studies were subsequently excluded, and therefore 13 studies were finally included in the analyses, enrolling 4,839 patients ( Tables 1 and 2 , Figure 1 ). No significant limitations were identified for 13 trials, 1 of which was a randomized comparison, while the other 12 were observational studies.
| Authors (Ref. #) | Male | n | Age (ys) | AF | CAD | CKD | COPD | CVs | DM | ESv | Hypertension | Logistic Euro SCORE | LVEF | NYHA Class III/IV | Prior MI | Prior PCI | Prior Stroke | STS Score |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Tzikas et al. | 46.0% | 79 | 81.0 | 27.0% | N/A | 20.0% | 26.0% | 100% | 22.0% | 0% | N/A | 16.0% | 52.0% | 88.0% | 24.0% | 24.0% | 23.0% | N/A |
| Gotzmann et al. | 46.0% | 39 | 78.5 | 26.0% | 56.0% | N/A | 36.0% | 100% | 33.0% | 0% | 92.0% | 17.7% | 57.0% | N/A | 31.0% | 46.0% | N/A | 22.7% |
| Samim et al. | 50.0% | 22 | 79.0 | 45.0% | 59.0% | 23.0% | 27.0% | 0% | 18.0% | 100% | 73.0% | 21.3% | N/A | N/A | 32.0% | 41.0% | 32.0% | N/A |
| Tamburino et al. | 44.0% | 633 | 81.0 | 16.4% | 48.0% | 23.2% | 21.3% | 100% | 26.4% | 0% | 75.1% | 23.0% | 52.1% | 71.5% | 21.6% | 28.5% | 7.2% | N/A |
| D’Onofrio et al. | 42.0% | 176 | 80.7 | 21.0% | 58.5% | 61.4% | 27.0% | 0% | 25.0% | 100% | 89.8% | 21.4% | 54.7% | 69.9% | 25.7% | N/A | N/A | 11.2% |
| De Chiara et al. | 47.0% | 58 | 82.0 | 7.0% | 47.0% | 27.0% | N/A | 100% | N/A | 0% | N/A | 24.0% | 56.0% | N/A | N/A | N/A | N/A | N/A |
| Hekimian et al. | 52.0% | 254 | 82.0 | N/A | N/A | N/A | N/A | 0% | N/A | 100% | N/A | N/A | 48.0% | N/A | N/A | N/A | N/A | N/A |
| Toggweiler et al. | 50.0% | 451 | 76.9 | 33.5% | 69.2% | 59.6% | 26.1% | 0% | 26.6% | 100% | 73.0% | N/A | 58.1% | 84.1% | 43.3% | 27.0% | 20.0% | 7.4% |
| Hutter et al. | 37.7% | 268 | 80.9 | 23.1% | 53.0% | 19.0% | 21.3% | 72.4% | N/A | 27.6% | N/A | 20.6% | N/A | N/A | N/A | N/A | 13.3% | 6.3% |
| Barbanti et al. | 53.3% | 499 | 83.5 | N/A | N/A | 18.8% | 41.7% | 0% | 39.1% | 100% | 88.2% | N/A | N/A | 93.8% | 24.4% | 31.5% | 28.2% | 11.6% |
| Bedogni et al. | 44.9% | 1007 | 81.2 | 17.0% | N/A | N/A | 22.9% | 100% | 27.8% | 0% | N/A | 23.1% | 51.5% | 69.6% | 21.6% | 30.6% | 6.9% | 8.0% |
| Giordana et al. (A) | 40.0% | 10 | 78.1 | 20.0% | N/A | N/A | N/A | 100% | 40.0% | 0% | N/A | 19.3% | 53.8% | 60.0% | 10.0% | N/A | 10.0% | N/A |
| Giordana et al. (B) | 28.0% | 25 | 83.0 | 24.0% | N/A | N/A | N/A | 0% | 32.0% | 100% | 0% | 21.8% | 59.4% | 84.0% | 20.0% | N/A | 20.0% | N/A |
| Zahn et al. | 41.5% | 1318 | 81.7 | N/A | N/A | 60.5% | 27.9% | 81.4% | 33.9% | 17.9% | N/A | 20.3% | 53.5% | 85.1% | N/A | N/A | 8.1% | N/A |
| Authors (Ref. #) | Valve Type and Model | Patients With msMR at Baseline, n | Patients With Worsening and/or Unchanged MR at Follow-Up, n |
|---|---|---|---|
| Tzikas et al. | CoreValve ReValving SystemTM | 18 | 38 |
| Gotzmann et al. | CoreValve ReValving SystemTM | 19 | 13 |
| Samim et al. | Edwards-SAPIEN | 16 | 5 |
| Tamburino et al. | CoreValve ReValving SystemTM | 42 | N/A |
| D’Onofrio et al. | Edwards-SAPIEN | 43 | 31 |
| De Chiara et al. | CoreValve ReValving SystemTM | 16 | 18 |
| Hekimian et al. | Edwards-SAPIEN | 38 | 60 |
| Toggweiler et al. | Cribier-Edwards/Edwards-SAPIEN/SAPIEN XT | 132 | 31 |
| Hutter et al. | Edwards-SAPIEN/CoreValve ReValving SystemTM | 35 | 10 |
| Barbanti et al. | Edwards-SAPIEN | 52 | 21 |
| Bedogni et al. | CoreValve ReValving SystemTM | 337 | 106 |
| Giordana et al. (A) | CoreValve ReValving SystemTM | 10 | 6 |
| Giordana et al. (B) | Edwards-SAPIEN | 25 | 6 |
| Zahn et al. | Edwards-SAPIEN/CoreValve ReValving SystemTM | 42 | N/A |
In patients with msMR, overall mortality risk after TAVI was significantly increased, as demonstrated by the pooled estimate for overall incidence of 30-day (ES −0.18, 95% confidence interval [CI] −0.31 to −0.04, I 2 = 46.51, Q = 7.48; Figure 2 ), 1-year (ES −0.22, 95% CI −0.36 to −0.08, I 2 = 56.20, Q = 11.41; Figure 2 ), and 2-year (ES −0.15, 95% CI −0.27 to −0.02, I 2 = 0.00, Q = 2.64; Figure 2 ) all-cause-mortality, compared with patients with absent or only mild MR. Among mortality outcomes, when valve type was considered as a categorical modifier (CoreValve system Medtronic, Minneapolis, Minnesota; Edwards SAPIEN valve, Edwards Lifesciences, Irvine, California), a significant influence on ES was observed. In particular, the presence of msMR at baseline significantly affected all-cause mortality only when the CoreValve system was used. These data were confirmed at short-term (1-month) (CoreValve system ES −0.23, 95% CI −0.35 to −0.11, I 2 = 0.00, Q = 0.09; Edwards SAPIEN valve ES −0.13, 95% CI −0.38 to 0.11, I 2 = 66.38, Q = 5.95; Figure 3 ) and long-term (1-year) (CoreValve system ES −0.28, 95% CI −0.39 to −0.17, I 2 = 4.81, Q = 3.15; Edwards SAPIEN valve ES −0.06, 95% CI −0.34 to 0.21, I 2 = 71.36, Q = 3.49; Figure 3 ) follow-up.
