This prespecified analysis of a randomized clinical trial assesses the long-term health status of surviving patients randomized to self-expanding transcatheter or surgical aortic valve replacement.
Key Points
Question
What are the long-term health status outcomes after transcatheter vs surgical aortic valve placement in high-risk patients?
Findings
In this prespecified analysis of a randomized clinical trial of 713 patients, high-risk patients randomized to transcatheter or surgical aortic valve replacement both had large improvements in health status that were stable through 2 years with gradual decline thereafter. Despite marked differences in the complication profiles of the 2 treatments, there were no differences in health status between treatment groups from 6 months through 5 years.
Meaning
While long-term mortality remains high in these high-risk patients, the majority of surviving patients treated with either transcatheter or surgical aortic valve replacement continue to report reasonable health status at 5 years.
Abstract
Importance
In the CoreValve High-Risk Trial, patients with severe symptomatic aortic stenosis had similar clinical outcomes with transcatheter aortic valve replacement (TAVR) vs surgical aortic valve replacement (SAVR) over 5 years of follow-up, with mortality rates of more than 50% in both groups.
Objective
To describe the long-term health status of surviving patients randomized to self-expanding TAVR vs SAVR.
Design, Setting, and Participants
This randomized clinical trial included patients at high surgical risk with severe aortic stenosis who completed a baseline Kansas City Cardiomyopathy Questionnaire (KCCQ) and were randomized to either self-expanding TAVR or SAVR from 45 US clinical sites. Patients were enrolled from February 2011 to September 2012. Analysis began May 2018 and ended June 2020.
Main Outcomes and Measures
Change in KCCQ and the 12-Item Short-Form Health Survey over 5 years, as assessed by repeated-measures analysis of covariance. Because there were significant interactions between access site and treatment for 1-month health status outcomes, all analyses were stratified by access site (iliofemoral or noniliofemoral).
Results
Of 713 patients, 377 (53%) were men, and the mean (SD) age was 83 (7) years. Prior to treatment, the mean (SD) KCCQ overall summary score (range, 0-100; higher score indicated better health status) was 47 (23), indicating substantial health status impairment. Among surviving patients, the KCCQ overall summary score increased significantly in both groups with greater early benefit with iliofemoral TAVR than SAVR (1-month difference, 16.8 points; 95% CI, 12.4-21.2). However, this early treatment difference between TAVR and SAVR was no longer apparent by 6 months, and there was no significant difference in health status between groups thereafter. At 5 years, 44% (134 of 305) of patients who underwent iliofemoral TAVR and 39% (105 of 266) who underwent SAVR were alive in this high-risk elderly cohort. Among surviving patients for whom health status data were available, 61% (48 of 79) in the TAVR group and 65% (46 of 71) in the SAVR group had KCCQ overall summary score more than 60 (P = .61). In the noniliofemoral cohort, there were no significant health status differences at any time between TAVR and SAVR. Results were similar for individual KCCQ domains and the Short-Form Health Survey.
Conclusions and Relevance
In high-risk patients with severe symptomatic aortic stenosis, there was an early health status benefit with self-expanding iliofemoral TAVR vs SAVR but no difference between groups in long-term health status. Although mortality at 5 years was high in this population, the majority of surviving patients continued to report reasonable health status.
Trial Registration
ClinicalTrials.gov Identifier: NCT01240902
Introduction
High-risk patients with severe, symptomatic aortic stenosis derive substantial survival and health status benefit from transcatheter aortic valve replacement (TAVR) or surgical aortic valve replacement (SAVR). However, long-term survival is limited even with valve replacement, with recent studies demonstrating 50% to 60% mortality within 5 years.1,2 While survival remains an important outcome, health status may be even more meaningful to older patients. Longer-term health status outcomes of TAVR vs SAVR are particularly relevant given concerns for differential rates of paravalvular leak, pacemakers, and valve degeneration, all of which may affect long-term health status.3,4 Therefore, we used data from the CoreValve US High Risk Pivotal Trial to compare the 5-year health status among surviving patients with severe, symptomatic aortic stenosis who were randomized to either self-expanding TAVR or SAVR.5
Methods
The trial protocol is available in Supplement 1. Disease-specific and generic health status were assessed using the Kansas City Cardiomyopathy Questionnaire (KCCQ)6 and the 12-Item Short-Form Health Survey, respectively.7 The KCCQ includes 5 domains (physical limitations, symptoms, quality of life, social limitations, and self-efficacy), the first 4 of which are combined into an overall summary (KCCQ-OS) score.8 Values range from 0 to 100; higher scores indicate better health status; and changes of 5, 10, and 20 points indicate small, moderate, and large improvements, respectively.9 The SF-12 provides physical component summary scores (PCS) and mental component summary scores (MCS), which are scaled to US norms of 50 (SD, 10); higher scores indicate better health status.7 The institutional review board at each site approved the study. All patients provided written informed consent. Patients were enrolled from February 2011 to September 2012.
The primary analysis compared the health status of patients randomized to TAVR vs SAVR on an intention-to-treat basis. Follow-up health status measures were compared within each treatment group to baseline measures using paired t tests. The relative association of TAVR vs SAVR with health status over time was assessed using repeated-measures analysis of covariance, which incorporated the available health status data from all follow-up times, including those for patients who subsequently died, withdrew from the study, or were lost to follow-up. We also conducted a sensitivity analysis using pattern-mixture models, which account for potential informative dropout by including missing data patterns over follow-up in a longitudinal random-effects model as main effects and as interactions with follow-up time. All analyses were performed using SAS version 9.4 (SAS Institute). All tests were 2-tailed with a P value of .05 used to determine statistical significance without adjustment for multiple comparisons. Analysis began May 2018 and ended June 2020.
Results
Among 795 patients randomized at 45 US sites, baseline health status data were available for 713 (89%), 84% (n = 599) of whom were eligible for iliofemoral access (Table 1). The baseline mean (SD) KCCQ-OS score was 46.7 (22.8) points, SF-12 PCS was 30.8 (8.9) points, and SF-12 MCS was 47.9 (11.8) points, indicating substantial impairment in disease-specific and physical health status. Treatment groups were well matched, other than slightly higher prevalence of diabetes in the SAVR arm (eTable 1 in Supplement 2).
Table 1. Baseline Characteristics.
| Characteristic | No. (%) | |
|---|---|---|
| TAVR (n = 377) | SAVR (n = 336) | |
| Age, mean (SD), y | 83.2 (7.1) | 83.5 (6.3) |
| Male | 199 (52.8) | 178 (53.0) |
| STS risk score, mean (SD), % | 7.3 (3.0) | 7.6 (3.4) |
| Prior myocardial infarction | 93 (24.7) | 82 (24.4) |
| Prior coronary artery bypass surgery | 110 (29.2) | 106 (31.5) |
| Prior percutaneous coronary intervention | 132 (35.0) | 128 (38.1) |
| Prior pacemaker | 86 (22.8) | 72 (21.4) |
| Prior stroke | 63 (16.7) | 71 (21.2) |
| Peripheral vascular disease | 153 (40.9) | 140 (41.9) |
| Chronic lung disease | 169 (44.8) | 150 (44.6) |
| Home oxygen | 48 (12.8) | 36 (10.7) |
| Diabetes mellitus | 128 (34.0) | 146 (43.5) |
| Atrial fibrillation | 155 (41.2) | 160 (47.6) |
| Chronic kidney disease | ||
| Stage III | 224 (59.4) | 202 (60.1) |
| Stage IV/V | 42 (11.1) | 37 (11.0) |
| Ejection fraction, mean (SD), % | 57.9 (11.5) | 57.5 (11.9) |
| Aortic gradient, mean (SD), mm Hg | 48.2 (15.5) | 47.7 (14.0) |
| Health status scores, mean (SD) | ||
| KCCQ | ||
| Overall summary score | 46.8 (23.4) | 46.7 (22.2) |
| Physical limitations | 46.5 (24.9) | 46.0 (25.0) |
| Total symptoms | 56.2 (24.6) | 55.5 (24.5) |
| Quality of life | 41.5 (24.8) | 42.4 (23.7) |
| Social limitations | 41.8 (30.5) | 41.7 (29.5) |
| SF-12 | ||
| Physical summary score | 30.7 (9.2) | 31.0 (8.5) |
| Mental summary score | 47.4 (12.0) | 48.4 (11.6) |
Abbreviations: KCCQ, Kansas City Cardiomyopathy Questionnaire; SAVR, surgical aortic valve replacement; SF-12, 12-Item Short-Form Health Survey; STS, Society of Thoracic Surgeons; TAVR, transcatheter aortic valve replacement.
Health status data were available for between 59% (eg, 429 of 720 at 1 month) and 75% (eg, 495 of 659 at 6 months) of eligible patients over follow-up, with similar rates between groups (Table 2). Both disease-specific and generic health status improved substantially after TAVR or SAVR, with improvement by 1 month with iliofemoral TAVR and by 6 months with noniliofemoral TAVR and SAVR (eTable 2 in Supplement 2). Maximal health status improvement was observed at approximately 6 months regardless of treatment group or access site; health status remained stable from 6 months to 2 years with gradual declines thereafter, particularly in the physical domains. At 5 years, surviving patients had approximately 20-point higher KCCQ-OS scores vs baseline. Symptoms, quality of life, and social limitations scores also demonstrated durable improvement, while the initial increases in the physical limitations domain and SF-12 PCS and MCS were no longer significant at 5 years.
Table 2. Compliance Summary for the Quality of Life Substudy.
| Visit | TAVR (n = 394) | SAVR (n = 401) | ||||
|---|---|---|---|---|---|---|
| No./ total No. (%) | Observed/eligible, No. (%) | No./ total No. (%) | Observed/eligible, No. (%) | |||
| Died | Withdrew | Died | Withdrew | |||
| Baseline | 0 | 0 | 377/394 (96) | 0 | 0 | 336/401 (84) |
| 1 mo | 17/393 (4) | 1/394 (0) | 249/376 (66) | 18/362 (5) | 37/401 (9) | 180/346 (52) |
| 6 mo | 38/393 (10) | 1/394 (0) | 276/355 (78) | 55/359 (15) | 40/401 (10) | 219/306 (72) |
| 1 y | 62/391 (16) | 3/394 (1) | 256/329 (78) | 73/355 (21) | 44/401 (11) | 201/284 (71) |
| 2 y | 90/388 (23) | 6/394 (2) | 222/298 (74) | 106/342 (31) | 59/401 (15) | 170/236 (72) |
| 3 y | 130/379 (34) | 15/394 (4) | 168/249 (67) | 139/334 (42) | 67/401 (17) | 137/195 (70) |
| 4 y | 170/374 (45) | 20/394 (5) | 136/204 (67) | 164/327 (50) | 74/401 (18) | 122/163 (75) |
| 5 y | 210/368 (57) | 23/394 (6) | 100/161 (62) | 190/320 (59) | 77/401 (19) | 88/134 (66) |
Abbreviations: SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement.
There was a significant interaction between treatment and access site at 1 month; thus, all between-group comparisons were stratified by access site. At 1 month, iliofemoral TAVR was associated with greater improvement in health status compared with SAVR (treatment difference in KCCQ-OS score, 16.8 points; 95% CI, 12.4-21.2; P < .001; Figure, A); similar benefits were seen across the KCCQ domains and SF-12 PCS and MCS (Figure, B and C). By 6 months and through 5 years, there were no differences between iliofemoral TAVR and SAVR in any of the health status measures. In a sensitivity analysis using pattern-mixture models to account for potential informative missing data, there remained no significant treatment difference in KCCQ-OS score at 5 years (TAVR-SAVR: 5.2 points; 95% CI −2.5 to 13.0; P = .19; eTable 4 in Supplement 2). At 5 years, 44% (134 of 305) of patients assigned to iliofemoral TAVR and 39% (105 of 266) of patients assigned to SAVR were alive. Among surviving patients for whom health status data were available, 61% (48 of 79) of patients who underwent TAVR and 65% (46 of 71) of patients who underwent SAVR had a KCCQ-OS score more than 60 (P = .61). Among noniliofemoral patients, there were no significant differences between TAVR and SAVR for any of the health status measures at any time, although confidence intervals for the differences were wide because of the small sample size (eTable 3 in Supplement 2).
Figure. Health Status After Iliofemoral TAVR vs SAVR, According to Repeated-Measures Analysis of Covariance Models.
Error bars indicate 95% CIs.
KCCQ-OS indicates Kansas City Cardiomyopathy Questionnaire Overall Summary score; SAVR, surgical aortic valve replacement; SF-12 MCS, 12-Item Short-Form Health Survey Mental Component Summary score; SF-12 PCS, 12-Item Short-Form Health Survey Physical Component Summary score; TAVR, transcatheter aortic valve replacement.
Discussion
In this multicenter trial of high-risk patients with severe aortic stenosis, surviving patients treated with either TAVR or SAVR had substantial improvements in disease-specific and generic health status. These improvements were maximal at approximately 6 to 12 months and maintained through 2 years with gradual decline in physical health status thereafter. While patients treated with iliofemoral TAVR had early health status benefit compared with SAVR across all disease-specific and generic health status domains, there were no significant between-group differences by 6 months through 5 years, and the pattern of gradual decline was consistent across treatments. Notwithstanding this gradual decline, more than 60% of surviving patients had KCCQ-OS scores higher than 60 at 5 years after valve replacement, a level consistent with New York Heart Association class I to II.9
This study represents the longest follow-up of patient-reported health status after aortic valve replacement. We found a gradual decline in health status from 2 to 5 years, particularly in the physical domains (disease-specific and generic), but even at 5 years, surviving patients had a KCCQ-OS score approximately 20 points higher than baseline (a large improvement9). It is unclear whether this gradual health status decline represents progression of heart failure or advanced aging combined with multiple comorbidities. Nonetheless, for patients with a 5-year survival less than 50% even after valve replacement, it is encouraging that most surviving patients continue to have reasonable quality of life. Of note, the survival rates in the intention-to-treat cohort differ slightly from those previously reported in the as-treated cohort.1
Another important insight from this study is the similar health status trajectories from 6 months through 5 years for those treated with TAVR and SAVR. In light of concerns about potential deleterious effects of TAVR-related complications, including paravalvular leak, need for permanent pacing, leaflet thrombosis, or structural valve deterioration (each of which could adversely affect health status),5 our findings are reassuring that there were no long-term differences in health status by treatment, despite differential rates of complications.
Limitations
Regarding limitations, importantly, health status can only be assessed among surviving patients, who represent the minority of patients at 5-year follow-up in this high-risk population. However, because long-term survival was similar between treatments, this should not bias treatment comparisons. Moreover, there was a fair amount of missing health status data over follow-up, although missing data rates were similar between groups. We conducted a sensitivity analysis using pattern-mixture models that account for missing data, and results were similar.
Conclusions
In conclusion, in high-risk patients with severe symptomatic aortic stenosis, treatment with either TAVR or SAVR resulted in substantial and sustained improvements in disease-specific and generic health status. Both disease-specific and generic health status were similar after 1 month in patients treated with TAVR or SAVR, despite differences in complication profiles. Although the 5-year mortality rate is high in these patients, most surviving patients treated with either TAVR or SAVR continue to report reasonable health status.
Trial protocol
eTable 1. Baseline characteristics stratified by access site
eTable 2. Unadjusted mean (± SD) health status scores over time and within group comparisons to baseline
eTable 3. Between group differences in health status, stratified by access site
eTable 4. Predicted mean values and between group comparisons of KCCQ-OS, according to pattern mixture models (Iliofemoral cohort)
Data sharing statement
References
- 1.Gleason TG, Reardon MJ, Popma JJ, et al. ; CoreValve U.S. Pivotal High Risk Trial Clinical Investigators . 5-Year outcomes of self-expanding transcatheter versus surgical aortic valve replacement in high-risk patients. J Am Coll Cardiol. 2018;72(22):2687-2696. doi: 10.1016/j.jacc.2018.08.2146 [DOI] [PubMed] [Google Scholar]
- 2.Mack MJ, Leon MB, Smith CR, et al. ; PARTNER 1 trial investigators . 5-year outcomes of transcatheter aortic valve replacement or surgical aortic valve replacement for high surgical risk patients with aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet. 2015;385(9986):2477-2484. doi: 10.1016/S0140-6736(15)60308-7 [DOI] [PubMed] [Google Scholar]
- 3.Arnold SV, Manandhar P, Vemulapalli S, et al. Impact of Short-term complications of TAVR on longer-term outcomes: results from the STS/ACC Transcatheter Valve Therapy Registry. Eur Heart J Qual Care Clin Outcomes. Published online January 11, 2020. doi: 10.1093/ehjqcco/qcaa001 [DOI] [PubMed] [Google Scholar]
- 4.Arnold SV, Zhang Y, Baron SJ, et al. Impact of short-term complications on mortality and quality of life after transcatheter aortic valve replacement. JACC Cardiovasc Interv. 2019;12(4):362-369. doi: 10.1016/j.jcin.2018.11.008 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Adams DH, Popma JJ, Reardon MJ, et al. ; U.S. CoreValve Clinical Investigators . Transcatheter aortic-valve replacement with a self-expanding prosthesis. N Engl J Med. 2014;370(19):1790-1798. doi: 10.1056/NEJMoa1400590 [DOI] [PubMed] [Google Scholar]
- 6.Green CP, Porter CB, Bresnahan DR, Spertus JA. Development and evaluation of the Kansas City Cardiomyopathy Questionnaire: a new health status measure for heart failure. J Am Coll Cardiol. 2000;35(5):1245-1255. doi: 10.1016/S0735-1097(00)00531-3 [DOI] [PubMed] [Google Scholar]
- 7.Ware J Jr, Kosinski M, Keller SDA. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220-233. doi: 10.1097/00005650-199603000-00003 [DOI] [PubMed] [Google Scholar]
- 8.Arnold SV, Spertus JA, Lei Y, et al. Use of the Kansas City Cardiomyopathy Questionnaire for monitoring health status in patients with aortic stenosis. Circ Heart Fail. 2013;6(1):61-67. doi: 10.1161/CIRCHEARTFAILURE.112.970053 [DOI] [PubMed] [Google Scholar]
- 9.Spertus J, Peterson E, Conard MW, et al. ; Cardiovascular Outcomes Research Consortium . Monitoring clinical changes in patients with heart failure: a comparison of methods. Am Heart J. 2005;150(4):707-715. doi: 10.1016/j.ahj.2004.12.010 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Trial protocol
eTable 1. Baseline characteristics stratified by access site
eTable 2. Unadjusted mean (± SD) health status scores over time and within group comparisons to baseline
eTable 3. Between group differences in health status, stratified by access site
eTable 4. Predicted mean values and between group comparisons of KCCQ-OS, according to pattern mixture models (Iliofemoral cohort)
Data sharing statement