Clinical Outcomes of Transcatheter Aortic Valve Replacement in Nonagenarians: A Systematic Review and Meta-Analysis

Objectives To compare the incidence of mortality and complications between nonagenarians and younger patients undergoing transcatheter aortic valve replacement (TAVR). Background TAVR has become an alternative treatment for nonagenarian patients with severe aortic stenosis. Previous studies have reported conflicting results regarding the clinical outcomes between nonagenarians and younger patients who underwent TAVR. Methods We searched PubMed, EMBASE, and Cochrane Library databases with predefined criteria from the inception dates to July 8, 2018. The primary clinical endpoint was 30-day and 1-year all-cause mortalities. Secondary outcomes were considered the rates of stroke, myocardial infarction, any bleeding, any acute kidney injury, any vascular complications, new pacemaker implantation, and conversion to surgical aortic valve replacement. Results A total of 5 eligible studies with 25,371 patients were included in this meta-analysis. Compared with younger patients who underwent TAVR, nonagenarians had a significantly higher mean Society of Thoracic Surgeons score (STS score) (MD, 2.80; 95%CI: 2.58, 3.30; P<0.00001) and logistic European System for Cardiac Operative Risk Evaluation (logistic EuroSCORE) (MD, 2.72; 95%CI: 1.01, 4.43; P=0.002). Nonagenarians were associated with significantly higher 30-day mortality (6.2% vs. 3.7%; OR, 1.73; 95%CI: 1.49, 2.00) and 1-year mortality (15.5% vs. 11.8%; OR, 1.39; 95%CI: 1.26, 1.53), without significant statistical heterogeneity. Nonagenarians were associated with significantly increased rates of major or life-threatening bleeding, vascular complications and stroke of 20%, 35%, and 32%, respectively. There were no significant differences in the rate of myocardial infarction, stage 2 or 3 acute kidney injury, new pacemaker implantation, or conversion to surgical aortic valve replacement. Conclusions Nonagenarians showed worse clinical outcomes than younger patients after TAVR, while the incidence of mortality was acceptable. TAVR remains an option for nonagenarian patients with severe aortic stenosis and should be comprehensively evaluated by the heart valve team.


Introduction
As forecasted by Kontis et al., the life expectancy in industrialized countries would break the 90-year barrier by 2030 [1]. Since the prevalence of aortic stenosis is increasing with lifespan, [2,3] it is urgent to address the management of aortic stenosis (AS) for nonagenarians.
Since 2002, when the first transcatheter aortic valve replacement (TAVR) was carried out by Cribier, [4] it has become an alternative treatment for intermediate-to highrisk patients with severe AS [5,6]. Several studies have shown no significant differences in the short-or long-term survival of nonagenarians between surgical aortic valve replacement (SAVR) and TAVR [7][8][9]. In consideration of frailty and inoperability, TAVR is a preferable approach for elderly patients.
However, previous studies have reported conflicting results regarding the clinical outcomes between nonagenarians and younger patients who underwent TAVR [10][11][12][13][14]. To further confirm the feasibility and safety of TAVR in nonagenarians, we performed this systematic review and meta-analysis to explore the short-to mid-term clinical outcomes of TAVR.

Methods
This meta-analysis was performed based on the Cochrane Handbook for Systematic Reviews of Interventions [15] and is presented according to the MOOSE (Meta-analysis of Observational Studies in Epidemiology) guidelines [16].
. . Search Strategy. We searched PubMed, EMBASE, and Cochrane Library databases for the current literature. The following search terms were used: ((((("transcatheter aortic valve replacement"[mesh]) OR transcatheter aortic valve implantation) OR TAVR) OR TAVI)) AND ((("Aged, 80 and over" [mesh]) OR 90 years) OR Nonagenarian * ). No language, publication date, or publication status restrictions were imposed. The last search was performed on July 8, 2018. Two investigators (YL and YD) performed the initial search separately, deleted duplicate records, screened titles and abstracts for relevance, and identified relevant articles for further full-text assessment. Reference lists from these retrieved articles were manually screened to identify additional relevant studies.
. . Inclusion Criteria. Studies were selected based on the following inclusion criteria: (1) studies enrolling nonagenarian and younger patients undergoing TAVR in current clinical practices, (2) studies comparing clinical outcomes of nonagenarians to younger patients undergoing TAVR, and (3) studies in which the clinical endpoints and adverse events were diagnosed by the Valve Academic Research Consortium 2 definitions [17]. Conference abstracts, reviews, comments, and editorials were excluded.
. . Data Extraction and Quality Assessment. Two investigators (YL and YD) independently extracted data (first author, country of origin, publication year, number of enrolled patients, and baseline patient characteristics) using a standardized data abstraction form. When the same patients were reported in several publications, only the largest study was used for the meta-analysis to avoid data duplication.
Two investigators (YL and YD) independently assessed the quality of selected studies based on the 9-star Newcastle-Ottawa Scale [18]. This scale rates studies based on eight criteria in three sources of bias. Disagreement was resolved by discussions and by consulting a third investigator (YJZ).
. . Clinical Endpoints. The primary clinical endpoint of interest was 30-day and 1-year all-cause mortality, and secondary outcomes were considered as the rates of stroke, myocardial infarction, any bleeding, any acute kidney injury, any vascular complications, new pacemaker implantation, and conversion to surgical aortic valve replacement. All definitions of clinical endpoints were based on the Valve Academic Research Consortium 2 definitions.
. . Statistical Analysis. The odds ratios (ORs) with 95% confidence intervals (CIs) for the endpoints were calculated from each study. Trial-specific ORs were combined with the Mantel-Haenszel fixed-effects model or with random effects model if heterogeneity was statistically significant or I 2 > 50%. If no events were reported for one group in a comparison, a value of 0.5 was added to both groups for each of these studies. Trials with no events in both groups were not included in the meta-analysis when the ORs were calculated.
The presence of heterogeneity among studies was evaluated with the Cochran Q chi-squared test, with P<0.10 considered to indicate statistical significance, and the I 2 test was used to evaluate inconsistencies. The I 2 statistic is derived from the Q statistic and describes the percentage of total variation across studies which is due to heterogeneity; values of 25%, 50%, and 75% correspond to low, moderate, and high heterogeneity, respectively. The funnel plot was not drawn for outcomes due to the small number of studies included in this analysis.
We did not contact the authors of the included studies to obtain raw data. All analyses were performed using Review Manager version 5.3 (Copenhagen, Denmark; Cochrane Collaboration). All tests were two tailed, and P< 0.05 was considered significant.

Results
. . Study Characteristics and Quality Assessment. From the searches for meta-analysis, 5,238 potentially eligible studies were identified. Titles and abstracts of these studies were screened for inclusion. Full-texts of 44 studies were read, and 5 studies met the inclusion criteria ( Figure 1) [10,[19][20][21][22]. The main characteristics of the included studies are described in Table 1(a). The current meta-analysis included 25,371 patients (3,929 in the nonagenarian group and 21,442 in the younger group). The mean Society of Thoracic Surgeons score (STS score), logistic European System for Cardiac Operative Risk Evaluation (logistic EuroSCORE), left ventricular ejection fraction (LVEF), mean pressure gradient, and medical history of the different studies are summarized in Tables 1(b) and 1(c). All trials reported that the clinical outcomes of interest varied from a 30-day to a 3-year follow-up period.
The assessment of quality is presented in Table 2. The total score of the 5 observational studies was >5 according to the Newcastle-Ottawa Scale for risk of bias in observational studies, representing a low risk of bias.

Discussion
To the best of our knowledge, this is the first comprehensive review of the current literature comparing the clinical outcomes of TAVR between nonagenarian and younger patients in a meta-analytic approach. In the results reported here, nonagenarians, with a higher mean STS score and logistic EuroSCORE, had an increased 30-day and 1-year postoperative all-cause mortalities compared with the younger group. In addition, the rates of major or life-threatening bleeding, vascular complications and stroke were also higher in nonagenarians. Furthermore, no significant differences were observed in the rates of myocardial infarction, stage 2 or 3 acute kidney injury, new pacemaker implantation, and      conversion to surgical aortic valve replacement between the two groups.
Although the mortality rate for nonagenarians in our study remained higher than for younger patients, considering life expectations, comparison of medical treatment, and quality of life, the mortality rate may be acceptable. It was reported in nonagenarians that age alone accounted for a predicted logistic EuroSCORE mortality risk of 6.55% for male patients, and for female patients, this risk rises to 8.89% without any other preoperative risk factors [23]. Similarly, our results showed that nonagenarians had a significantly higher logistic EuroSCORE than younger patients (MD, 2.72; 95%CI: 1.01, 4.43). Bernal et al. indicated that approximately one-third of nonagenarians with severe aortic stenosis have few comorbidities [24]. Therefore, in a particular population, patients at a very advanced age could be the primary factor making nonagenarians have a high risk for surgery. Furthermore, Bernal et al. observed that nonagenarians who underwent conservative management tended to have a higher 1-year mortality than those who underwent TAVR (58% vs. 40.7%, P=0.097) [24]. In our meta-analyses, the pooled postoperative 30-day and 1-year all-cause mortalities of nonagenarians was 6.2% and 15.5%, respectively. Despite the higher mortality than younger patients, TAVR could be a better option for severe AS nonagenarians.
Manolis et al. demonstrated that, in nonagenarians who underwent TAVR, bleeding and vascular complications ranged from 9% to 34% (average 16%), and stroke risk ranged from 2% to 18% (average 3-4%) [25]. Similarly, in this meta-analysis, we indicated that nonagenarians were associated with significantly increased rates of major or lifethreatening bleeding, vascular complications and stroke of 20%, 35%, and 32%, respectively. The higher incidence of vascular complications may be attributed to the higher rates of transfemoral access. Another possible reason could be the higher rates of vascular calcification [26] in the "oldest old" population. The higher incidence of stroke and major or life-threatening bleeding may be attributed to the higher prevalence of atrial fibrillation [11] and antiplatelet therapy after TAVR [27], respectively.
In the general population, life expectancy in nonagenarians is 2.5 and 3.5 years for men and women, respectively [23]. The rate of 1-year mortality in nonagenarians was 19.3% [28]. Accordingly, we should focus on the improvement of quality of life and try to add life to their years instead of years to their life [29]. Stanska et al. showed that the quality of life in elderly patients was significantly improved at the one-month followup after TAVR [30]. Mack [10].
Nonagenarians referred for TAVR should be evaluated carefully by the heart valve team composited of cardiovascular surgeons, interventional cardiologists, imaging specialists, cardiovascular anesthesiologists, and cardiovascular nursing professionals [31]. Generally, we use the STS score and logistic EuroSCORE to assess the risk of TAVR, and neither of these scores includes the specific geriatric conditions, which tend to generate a considerable impact on prognosis in elderly patients. Okoh et al. indicated that frailty status is independently associated with increased mortality after TAVR (hazard ratio: 1.84; 95%CI: 1.06-3.17; P=0.028) [32]. Multidimensional Geriatric Assessment (MGA) is a diagnostic process that determines the medical and functional resources and problems of elderly patients. According to Stortecky et al., risk prediction of TAVR can be improved by adding MGA-based information to global risk scores [33]. Furthermore, in nonagenarians, transapical TAVR was associated with a significantly higher risk of early mortality compared with transfemoral TAVR [34][35][36][37]. The procedure should be comprehensively evaluated by the heart valve team.
This study has several limitations. First, all of the eligible studies were observational studies, and the results may have been affected by unmeasured confounding variables. Second, due to the limited number of studies included in this analysis, we did not conduct sensitivity and meta-regression analysis for outcomes. Third, our analysis did not include individual patient data. Finally, these data are mostly from highly experienced TAVR centers and may not be generalizable to other hospitals with less experience.
In conclusion, the results reported here suggest that nonagenarians showed higher short-to mid-term mortalities and higher rates of major or life-threatening bleeding, vascular complications and stroke compared to younger patients. However, the rate of mortality in nonagenarians is potentially acceptable. TAVR remains an optional therapy for nonagenarian patients, which should be comprehensively evaluated by the heart valve team.

TAVR:
Transcatheter aortic valve replacement STS score: Thoracic Surgeons score logistic EuroSCORE: Logistic European System for Cardiac Operative Risk Evaluation AS: Aortic stenosis SAVR: Surgical aortic valve replacement MOOSE: Meta-analysis of Observational Studies in Epidemiology OR: Odds ratio CI: Confidence interval LVEF: Left ventricular ejection fraction KCCQ: Kansas City Cardiomyopathy Questionnaire MGA: Multidimensional Geriatric Assessment CAD: Coronary artery disease COPD: Chronic obstructive pulmonary disease.

Data Availability
The data supporting this article are from previously reported studies and datasets, which have been cited. All relevant data are within the article.

Conflicts of Interest
The authors declare that they have no conflicts of interest.