Abstract
Objective
Transcatether aortic valve replacement (TAVR) has emerged as an acceptable alternative to surgical aortic valve replacement (SAVR) in patients with high-risk surgical profile. In this analysis, we compare both approaches in non-high surgical risk severe aortic stenosis.
Methods
Only studies comparing SAVR and TAVR and enrolling patients with low-intermediate risk were included from January 2000 through May 2016. Primary endpoints were assessed at 1 and 12 months including: mortality, cerebrovascular accident (CVA) and myocardial infarction (MI). Secondary endpoints at 1 month included: bleeding, acute kidney injury, new permanent pacemaker implantation and vascular access site complications.
Results
A total of 5223 patients from 8 studies were included (4 prospective and 4 retrospective). There were no significant differences between TAVR and SAVR at one month in terms of mortality (risk ratio RR 0.91, 95% CI: 0.68 to 1.20), or CVA (RR 0.91, 95% CI 0.68 to 1.21). However, MI was lower in the TAVR group (RR 0.58, 95% CI 0.34 to 0.99). At 12 months, there was no significant difference between strategies in terms of mortality (RR 0.98, 95% CI 0.84–1.13), CVA (RR1.07, 95% CI 0.85–1.33) or MI (RR 0.78, 95% CI 0.53–1.15). With regard to secondary outcomes, TAVR was associated with lower rates of bleeding (RR 0.44, 95% CI 0.22–0.88) and acute kidney injury (RR 0.54, 95% CI 0.31–0.93) but higher need for new permanent pacemaker implantation (RR 2.99, 95% CI 1.51–5.94) and high rate of vascular access site complications (RR 9.08, 95% CI 2.03–40.66).
Conclusion
In severe AS patients with non-high surgical risk, TAVR has lower risk of MI at one month, but both TAVR and SAVR yielded similar outcomes at one month in terms of mortality and CVA and at 12 months in terms of mortality, CVA and MI. In regard to secondary outcomes at one month, TAVR has lower rates of bleeding and acute kidney injury, and higher rates of need for new permanent pacemakers and high rate of vascular access site complications.
Highlights
- •
The prevalence of severe aortic stenosis is 3.4% of population 75 years or older.
- •
Eligible candidates for valve replacement increase each year.
- •
TAVR is currently proved for high surgical risk patients.
- •
The primary outcomes of TAVR for non-high surgical risk patients are comparable to surgical replacement.
- •
TAVR has lower risk of bleeding and kidney injury, and higher risk for permanent pacemaker implantation and vascular access site complications than surgical valve replacement.
1
Introduction
Aortic stenosis (AS) is the most common valvular disease in developed countries; the prevalence of this disease increases with age from 0.2% for ages 50–59 years to up to 10% at 80–89 years. The prevalence of severe AS is estimated to be 3.4% of the elderly population aged 75 years or older . Severe AS associated with symptoms and/or left ventricular dysfunction is fatal without relief of the obstruction by valve replacement .
Based on the American Heart Association/American College of Cardiology 2014 valvular heart diseases guidelines, the standard treatment for severe AS with indication for treatment is surgical aortic valve replacement (SAVR). Trans-catheter aortic valve replacement (TAVR) is recommended for those who are at high surgical risk; which is estimated by different risk scores like Society of Thoracic Surgeons (STS) score > 8% or Logistic EuroScore I > 20%, who have a predicted post-TAVR survival of more than 12 months of duration . A significantly higher number of patients are expected to be eligible for TAVR each year .
Several studies comparing TAVR and SAVR in high surgical risk patients have revealed comparable results. However, when looking into strategies in non-high surgical risk groups it remains largely unexplored. Some early studies suggested worse outcomes with TAVR for none-high surgical risk AS . So we sought to systematically compare TAVR to SAVR in non-high surgical risk patients with severe AS based on available literature.
2
Methods
The aim of this meta-analysis was to examine the 1- and 12-month outcomes associated with TAVR versus SAVR in patients with AS who are classified as intermediate and low to intermediate surgical risk. Primary endpoints assessed at 1 and 12 months included: mortality, cerebrovascular accident (CVA) and myocardial infarction (MI). Secondary outcomes assessed at 1 month included: bleeding, acute kidney injury, need for new permanent pacemaker and vascular access site complications.
The study was performed following procedures recommended by the Cochrane collaboration and is reported in accordance with the recommendations set forth by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement .
2.1
Information sources and search methods
A comprehensive literature search was conducted through the electronic databases MEDLINE™, EMBASE and the Cochrane Central Register of Controlled Trials (CENTRAL) for abstracts using various combinations of the terms severe aortic stenosis, low and intermediate risk trancatheter aortic valve replacement, low and intermediate risk TAVR and non-high risk TAVR
Two reviewers (JO, TE) identified articles eligible for further review by performing a screen of abstracts and titles. If a study met the inclusion criteria, the manuscript was obtained and reviewed. Inclusion criteria were (1) studies comparing TAVR to SAVR for severe AS in non-high (i.e. low and intermediate) surgical risk and (2) double-armed studies (TAVR versus SAVR). Exclusion criteria were (1) high surgical risk or mixed data (high and non-high) severe AS and (2) single-armed studies (no SAVR group). In addition, bibliographic references of identified randomized clinical trials and review articles, in order to find randomized clinical trials not identified by the electronic searches, were reviewed.
2.2
Studies identification
The previously described data sources were searched for possible studies irrespective for dates of publications. The search was limited to English-language literature. We included original papers irrespective of randomization. The initial search identified 250 citations. 155 citations were excluded by identifying abstract/title. The final search identified 8 original papers that fulfilled the criteria for inclusion ( Fig. 1 ).
2.3
Data collection and extraction
Two independent reviewers (JO and TE) extracted data from the included studies by using pre-specified data elements. We abstracted data on patient demographics and baseline characteristics, study design, sample size, intervention type (TAVR vs SAVR) and type of outcome measures (primary outcomes: mortality, CVA and MI and secondary outcomes: bleeding (minor and major), kidney failure, need for new permanent pacemaker and vascular access site complications) when available.
One review author extracted the data from included studies and a second author verified the extracted data. The number of events in each trial was extracted when available. Table 1 depicts patients’ baseline characteristics as well as studies’ description and outcomes.
| Trial name | Study type | Risk score (%) | Surgical risk | Type ( N ) | Mean age | Gender | |
|---|---|---|---|---|---|---|---|
| Age | ± | Percent male | |||||
| NOTION (2015) | RCT | STS 1.3–4.8 | Low (82%) Intermediate (18%) | TAVR (139) | 79.2 | 4.9 | 53.8 |
| SAVR (135) | 79 | 4.7 | 52.6 | ||||
| ITALIAN OBSERVANT (2015) | Observant | Euroscore I score 1.0–19.4 | Low to intermediate | TAVR (650) | 80.5 | 6.2 | 41.1 |
| SAVR (650) | 80.3 | 5.1 | 40.5 | ||||
| PARTNER 2A (2016) | RCT | STS 3.7–7.9 | Low to intermediate | TAVR (1011) | 81.5 | 6.7 | 54.2 |
| SAVR (1021) | 81.7 | 6.7 | 54.8 | ||||
| Latib (2012) | Observant | STS 2.3–7.2 | Low to intermediate | TAVR (111) | 80 | 7.6 | 52.9 |
| SAVR (111) | 79.4 | 3 | 44.1 | ||||
| STACCATO (2012) | RCT | STS <3.0–5.0 | Low to intermediate | TAVR (34) | 80 | 3.6 | 26.5 |
| SAVR (36) | 82 | 4.4 | 33.3 | ||||
| Piazza (2013) | Observant | STS 3.0–8.0 | Low to intermediate | TAVR (255) | 81.3 | 6.7 | 41.6 |
| SAVR (255) | 70.3 | 11 | 61.6 | ||||
| TAVIK (2015) | Observant | Euroscore I score 6.0–11.6 | Low to intermediate | TAVR (216) | 78.3 | 5.2 | 46.3 |
| SAVR (216) | 78.2 | 4.6 | 51.4 | ||||
| US PIVOTAL (retrospective 2016) | RCT | STS 4.1–6.1 | Low to intermediate | TAVR (202) | 81.5 | 7.6 | 57.9 |
| SAVR (181) | 81.2 | 6.6 | 55.8 | ||||
2.4
Risk of bias assessment
Methodological quality was defined as the control of bias assessed through the reported methods in each individual study using the Cochrane risk of bias tool to assess the quality of randomized trials. Newcastle–Ottawa Scale (NOS) was used to assess the quality of observational studies. Two reviewers (JO, TE) independently assessed each study’s quality by examining risk of bias tool components. No evidence of publication bias was detected by noticing symmetry of the funnel plot ( Fig. 2 ) which provides a better display by spreading out the smaller studies on the bottom half of the plot, but have no bearing on the statistics. There was no performance bias due to non-blinded studies ( Tables 2–3 ). Disagreements between the reviewers were resolved by discussion or arbitrated with a third coauthor (ASA).
| Study ID | Study design | Selection | Comparability | Outcome | ||||
|---|---|---|---|---|---|---|---|---|
| Representativeness of exposed cohort | Non-exposed cohort | Ascertainment of exposure | Demonstration that outcome of interest was not present at start of study | Assessment of outcome | Enough follow-up length | |||
| Latib (2012) | Cohort | Truly representative | Same community | Secured records | Yes | Risk score | Independent assessment | Yes (30 day, 1 year) |
| ITALIAN OBSERVANT (2015) | Cohort | Truly representative | Multi-center | Secured records | Yes | Risk score | Independent assessment | Yes (30 day, 1 year) |
| TAVIK (2015) | Cohort | Truly representative | Single center | Secured records | Yes | Risk score | Independent assessment | Not all outcome |
| Piazza (2013) | Cohort | Truly representative | Multi-center | Secured records | Yes | Risk score | Independent assessment | No (prospective) |
| Study ID | Adequate randomization | Allocation concealment | Blinding | Baseline characteristics Balanced | Lost to follow-up (<20%) | Incomplete data |
|---|---|---|---|---|---|---|
| Notion (2015) | Yes | Yes | No | Yes | Yes | No |
| PARTNER 2A (2016) | Yes | Yes | No | Yes | Yes | No |
| STACCATO (2012) | Yes | Yes | NR | Yes | Yes | Terminated early |
| US PIVOTAL (STS < 7%) (2016) | Yes | NR | No | Yes | NR | Low–intermediate risk only |
2.5
Statistical analysis and data synthesis
From the abstracted data, we calculated the risk ratio (RR) using the inverse variance method for each study outcome to allow for pooling of similar outcomes. The average effects for the outcomes and 95% confidence intervals (CI) were obtained using a random effects model, as described by Higgins and Green . We chose the random effects method as primary analysis because of its conservative summary estimate and incorporation between and within study variance. When analysis was repeated with the fixed-effect method, the results were consistent as well.
To assess the heterogeneity of treatment effect among trials, we used the I 2 statistic. The I 2 statistic represents the proportion of heterogeneity of treatment effect across trials that were not attributable to chance or random error. Hence, a value of 50% reflects significant heterogeneity that is due to real differences in study populations, protocols, interventions, and outcomes .
The P -value threshold for statistical significance was set at 0.05 for effect sizes. Analyses were conducted using features on RevMan version 5.3.5 (The Nordic Cochrane Center, Copenhagen, Denmark).
2.6
Methods for including both-armed zero events
In the case of zero events for an outcome in both groups simultaneously, we utilized a continuity factor of 1. We added this to each arm including FFR events, FFR total, ANGIO events and ANGIO total in order to avoid computational errors. Studies without reported outcomes were not included in the analysis .
2
Methods
The aim of this meta-analysis was to examine the 1- and 12-month outcomes associated with TAVR versus SAVR in patients with AS who are classified as intermediate and low to intermediate surgical risk. Primary endpoints assessed at 1 and 12 months included: mortality, cerebrovascular accident (CVA) and myocardial infarction (MI). Secondary outcomes assessed at 1 month included: bleeding, acute kidney injury, need for new permanent pacemaker and vascular access site complications.
The study was performed following procedures recommended by the Cochrane collaboration and is reported in accordance with the recommendations set forth by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement .
2.1
Information sources and search methods
A comprehensive literature search was conducted through the electronic databases MEDLINE™, EMBASE and the Cochrane Central Register of Controlled Trials (CENTRAL) for abstracts using various combinations of the terms severe aortic stenosis, low and intermediate risk trancatheter aortic valve replacement, low and intermediate risk TAVR and non-high risk TAVR
Two reviewers (JO, TE) identified articles eligible for further review by performing a screen of abstracts and titles. If a study met the inclusion criteria, the manuscript was obtained and reviewed. Inclusion criteria were (1) studies comparing TAVR to SAVR for severe AS in non-high (i.e. low and intermediate) surgical risk and (2) double-armed studies (TAVR versus SAVR). Exclusion criteria were (1) high surgical risk or mixed data (high and non-high) severe AS and (2) single-armed studies (no SAVR group). In addition, bibliographic references of identified randomized clinical trials and review articles, in order to find randomized clinical trials not identified by the electronic searches, were reviewed.
2.2
Studies identification
The previously described data sources were searched for possible studies irrespective for dates of publications. The search was limited to English-language literature. We included original papers irrespective of randomization. The initial search identified 250 citations. 155 citations were excluded by identifying abstract/title. The final search identified 8 original papers that fulfilled the criteria for inclusion ( Fig. 1 ).
