Placement of Aortic Transcatheter Valves (Cohort A): TAVR vs. Surgical AVR - PARTNER Cohort A: TAVR vs. Surgical AVR
Contribution To Literature:
The PARTNER Cohort A trial showed that TAVR is noninferior to surgical AVR for clinical outcomes up to 5 years for the management of patients with high surgical risk.
As many as one-third of patients with severe aortic stenosis (AS) are high-risk surgical candidates and are conservatively managed. However, nonsurgical management of symptomatic AS is associated with a median survival of about 2 years. The results of the PARTNER trial (cohort B) comparing medical therapy to transfemoral transcatheter aortic valve replacement (TAVR) in inoperable patients demonstrated a significant mortality benefit with TAVR over medical therapy.
The current trial (cohort A) sought to compare outcomes between TAVR (either transfemoral or transapical) versus surgical aortic valve replacement (AVR) in patients who were high risk.
- Severe AS: Echo-derived aortic valve area (AVA) <0.8 cm2 (or AVA index <0.5 cm2/m2) and mean AV gradient >40 mm Hg or peak jet velocity >4.0 m/s
- Cardiac symptoms: NYHA functional class ≥II
- High surgical risk: Predicted risk of operative mortality ≥15% (determined by site surgeon and cardiologist); guideline = STS score ≥10
- Number of enrollees: 699
- Duration of follow-up: 2 years
- Mean patient age: 84 years
- Percentage female: 43
- Ejection fraction: 53%
- NYHA class: III/IV: 94%
- Bicuspid or noncalcified aortic valve
- Aortic annulus diameter (echo measurement) <18 mm or >25 mm
- Aortic dissection or iliac-femoral dimensions or disease precluding safe sheath insertion (especially calcification)
- Severe LV dysfunction (LV ejection fraction <20%)
- Untreated coronary artery disease requiring revascularization
- Severe AR or MR (>3+) or prosthetic valve (any location)
- Serum creatinine >3.0 mg/dl or dialysis dependent
- Acute myocardial infarction within 1 month
- Upper gastrointestinal bleed within 3 months
- Cerebrovascular accident or TIA within 6 months
- Any cardiac procedure, other than balloon aortic valvuloplasty, within 1 month or within 6 months for drug-eluting stent
- Hemodynamic instability (e.g., requiring inotropic support)
- All-cause mortality at 1 year
- Neurologic events
- Prospective: Stroke and stroke plus transient ischemic attack (TIA) (all neurologic events)
- Retrospective: Major stroke (modified Rankin score ≥2 at ≥30 days)
- Major vascular complications (Vascular Academic Research Consortium [VARC] definition)
- Major bleeding (modified VARC definition)
- Repeat hospitalization
- New pacemakers and new-onset atrial fibrillation (ECG core lab)
- Procedural events (assigned therapy aborted or converted to AVR, multiple valves, etc.)
- Surgical complications (reoperation for bleeding, sternal infection, etc.)
- NYHA symptoms
- Six-minute walk tests
- Quality-of-life measures and cost-effectiveness (core lab)
- Echocardiography assessment of valve performance
- Peak and mean gradients
- Effective orifice area
- Bioprosthetic valve regurgitation (especially paravalvular)
- Other: left ventricular (LV) ejection fraction, mitral regurgitation (MR), LV mass, evidence of structural valve deterioration
High-risk patients with adequate femoral/iliac vessel diameter (≥7 mm for #23 mm valve, and ≥8 mm for #26 mm valve) were randomized in a 1:1 fashion to either transfemoral TAVR or surgical AVR. Those with inadequate femoral/iliac vessel diameter were randomized in a 1:1 fashion to either transapical TAVR or surgical AVR.
A total of 699 patients were randomized, 492 to the transfemoral randomization (TAVR = 244, AVR = 248), and 207 to transapical (TAVR = 104, AVR = 103). The mean Society of Thoracic Surgeons (STS) score was 11.7%. Most patients had severely symptomatic AS, with a mean valve area of 0.7 cm2 and a mean gradient of 43 mm Hg. About 75% had coronary artery disease, 28% had cerebrovascular disease, 43% had undergone prior coronary artery bypass grafting, 42% had peripheral vascular disease, and 43% had chronic obstructive pulmonary disease. A porcelain aorta and chest wall radiation were reported in <1% of patients, and about 16% of patients were considered “frail.” The mean STS score mortality was 11.8%.
All-cause mortality was noninferior between TAVR and surgical AVR at 1 year (24.2% vs. 26.8%, hazard ratio [HR] 0.93, 95% confidence interval [CI] 0.71-1.22, p for noninferiority = 0.001, p for superiority = 0.62). When the two access routes were separately assessed, transfemoral TAVR was noninferior compared with AVR (22.2% vs. 26.4%, HR 0.83, 95% CI 0.60-0.15, p for noninferiority = 0.44). The comparison between transapical TAVR versus AVR was underpowered (29.0% vs. 27.9%, HR 1.22, 95% CI 0.75-1.98, p = 0.41).
There was one procedural death in the AVR arm, and three in the TAVR arm. The transcatheter procedure was either aborted or converted to open AVR in 4.6% (16/348) of patients. Thirty-day mortality was similar between TAVR and AVR (3.4% vs. 6.5%, p = 0.07). Vascular complications at 30 days (17.0% vs. 3.8%, p < 0.001) and at 1 year (18.0% vs. 4.8%, p < 0.001) were higher with TAVR, but major bleeding at 30 days (9.3% vs. 19.5%, p < 0.001) and at 1 year (14.7% vs. 25.7%, p < 0.001) was lower in the TAVR arm. The need for new permanent pacemaker was similar at 30 days (3.8% vs. 3.6%, p = 0.89) and at 1 year (5.7% vs. 5.0%, p = 0.68). All strokes were higher with TAVR at 30 days (5.5% vs. 2.4%, p = 0.04), and at 1 year (8.3% vs. 4.3%, p = 0.04), but not major strokes (p > 0.05 at both time points). The 6-minute walk test was superior for TAVR at 30 days (p = 0.002), but not at 1 year.
Mean echo gradients at 1 year were clinically similar, but statistically lower with TAVR (10.2 mm Hg vs. 11.5 mm Hg, p = 0.008). Moderate to severe paravalvular AR was greater with TAVR at all time points (p < 0.05).
Two-year follow-up: All-cause mortality (33.9% vs. 35.0%, HR 0.88, 95% CI 0.70-1.12, p = 0.31) was similar between the TAVR and surgical AVR arms. This was true on landmark analysis at 1 year as well. Mortality rates were numerically better in the transfemoral TAVR arm when compared with surgical AVR (30.9% vs. 34.6%, p = 0.38), and numerically worse in the transapical TAVR arm when compared with surgical AVR (41.1% vs. 35.8%, p = 0.44). Other outcomes such as cardiovascular mortality (20.5% vs. 21.4%, p = 0.48), mortality or repeat hospitalization (46.6% vs. 46.5%, p = 0.84), and strokes (7.7% vs. 4.9%, p = 0.52) were also similar at 2 years. Interestingly, strokes in the TAVR arm were more common up to 30 days post-implant and less common after 30 days when compared with surgical AVR. The incidence of permanent pacemaker requirement was similar (7.2% vs. 6.4%, p = 0.69). Most patients were still asymptomatic/minimally symptomatic 2 years out, with only 15% endorsing New York Heart Association (NYHA) class III/IV symptoms in both arms.
Echocardiography demonstrated that valve area and gradients were similar between the two arms at 2 years. Paravalvular AR was still significantly higher in the TAVR arm at 2 years (moderate: 6.9% vs. 0.9%, p < 0.001). Paravalvular AR remained unchanged in 46.2% of patients, improved in 31.5%, and worsened in 22.4% of patients. Even mild AR was significantly associated with a higher risk of all-cause mortality on long-term follow-up.
Three-year follow-up: All-cause mortality (44.2% vs. 44.8%, HR 0.93, 95% CI 0.74-1.15, p = 0.48) was similar between the TAVR and surgical AVR arms. This was true on landmark analysis at 1 year as well. Three-year mortality in the TAVR arm was independent of STS score, but higher in patients with comorbidities such as chronic kidney disease, liver disease, and atrial fibrillation. Strokes (8.2% vs. 9.3%, p = 0.76) were similar at 3 years. Interestingly, strokes in the surgical AVR arm were higher after year 2 (1 vs. 9). The incidence of permanent pacemaker requirement was stable (8.1% vs. 6.8%, p = 0.56). Endocarditis was also similar (1.5% vs. 2.6%, p = 0.37). Echocardiography demonstrated that valve area and gradients were similar between the two arms at 3 years. The majority of patients were still asymptomatic/minimally symptomatic 3 years out. Moderate to severe paravalvular AR was still significantly higher in the TAVR arm at 3 years, and even mild paravalvular AR portended increased mortality at 3 years.
Five-year follow-up: All-cause mortality (67.8% vs. 62.4%, p = 0.76), repeat hospitalizations (42.3% vs. 34.2%, p = 0.17), and all strokes (15.9% vs. 14.7%, p = 0.35) were similar between the TAVR and surgical AVR arms. Echocardiography demonstrated that valve area and gradients were similar between the two arms at 5 years. Outcomes were better in patients undergoing transfemoral TAVR compared with transapical TAVR (p = 0.05). The majority of patients were still asymptomatic/minimally symptomatic 5 years out. Moderate to severe paravalvular AR was still significantly higher in the TAVR arm at 5 years, and even mild paravalvular AR portended increased mortality at 5 years.
Cost-effectiveness analyses: These were conducted separately for transfemoral and transapical TAVR versus surgical AVR. For the transfemoral TAVR cohort, TAVR procedural costs were significantly higher ($36,652 vs. $14,475). However, overall admission costs were similar as compared with surgical AVR ($73,219 vs. $74,067) due to a reduction in hospital length of stay with transfemoral TAVR (10.2 vs. 16.4 days). Follow-up costs through 12 months were similar for the transfemoral TAVR and AVR groups (mean difference = $1,247). Quality-adjusted life-years (QALYs) were superior in transfemoral TAVR compared with AVR (0.66 vs. 0.59). Bootstrap simulation demonstrated that transfemoral TAVR was economically dominant compared with AVR in 55.7% of replicates, and economically attractive at an incremental cost-effectiveness ratio (ICER) of $50,000/QALY gained in 70.9%. Threshold analysis demonstrated that the ICER for transfemoral TAVR would remain <$50,000/QALY as long as the difference in acquisition cost between the TAVR device and a standard bioprosthetic valve remained <$29,390.
For the transapical TAVR cohort, total procedure costs ($40,368 vs. $15,076) and total admission costs ($90,919 vs. $79,024) were both higher, as compared with surgical AVR. Follow-up costs through 12 months were similar for the transapical TAVR and AVR groups (mean difference = -$1,102). QALYs were lower in the transapical TAVR cohort (0.57 vs. 0.64). Transapical TAVR resulted in higher 12-month costs and lower quality-adjusted life expectancy than AVR in the primary analysis, and was economically dominated by AVR in 86.6% of bootstrap replicates. At an ICER threshold of $50,000/QALY, transapical TAVR was economically attractive relative to AVR in just 7.1% of replicates. Transapical TAVR was likely to be economically attractive at the $50,000 ICER threshold only if the difference in valve acquisition costs was <$11,324.
Long-term hemodynamic evaluation: Data from the PARTNER 1 trial (cohorts A and B) as well the continued access registry were analyzed (n = 2,795; 2,482 TAVR, 313 surgical AVR). Complete 5-year echocardiographic data were available in 424 patients who received TAVR, while serial paired echocardiographic data were available in 399 patients. LV mass was significantly lower at 5 years (76.3 vs. 63.7 g post-implant vs. 5 years, p < 0.001). The average mean gradient was 6.4 mm Hg, and the average AVA was 1.57 cm2.
Among patients with ≥1 follow-up echocardiogram, at a medium follow-up of 3.1 years, moderate/severe transvalvular regurgitation was noted in 3.7% after TAVR and increased over time. Patients with surgical AVR showed no significant changes. Reintervention occurred in 20 patients (0.8%) after TAVR and in one (0.3%) after surgical AVR and became less frequent over time. However, reintervention was caused by structural deterioration of transcatheter heart valves in only five patients. Overall mortality was very high: 63.1%.
PARTNER is a landmark trial in the field of structural heart disease and in the management of patients with severe AS. The results of cohort A of the PARTNER trial presented here comparing TAVR (transfemoral or transapical) to AVR demonstrated noninferiority for all-cause mortality at 1 year, with transfemoral TAVR individually demonstrating noninferiority as well. Thirty-day mortality was also significantly lower with TAVR (3.4% in a cohort with an expected mortality of >11%). As noted in cohort B, vascular complications and all strokes were higher with TAVR. The incidence of permanent pacemaker implantation was similar between TAVR and AVR.
Cost-effectiveness analyses at 1 year indicate a wide disparity between transfemoral and transapical TAVR compared with surgical AVR. Transfemoral TAVR appears to be a cost-effective (and perhaps cost-dominant) strategy, whereas transapical TAVR seems to be inferior to TAVR on economic grounds. These differences in cost-effectiveness are driven by differences in both hospital resource use (shorter length of stay with transfemoral TAVR) and short-term clinical outcomes of TAVR according to access site. These results are impressive, given that these are early times in the transfemoral TAVR technology evolution.
Two- and three-year results indicate that hemodynamic and clinical outcomes remain similar between patients undergoing TAVR. The incidence of stroke actually decreased compared with surgical AVR beyond 30 days. Paravalvular AR, however, still remained high, and on post-hoc analysis, was a significant predictor of long-term mortality. Thus, although no difference in long-term mortality was noted at 3 years, longer-term follow-up is necessary. Overall though, these results are very encouraging, and highlight the emerging importance of TAVR in the management of high surgical-risk AS patients. Studies are ongoing to assess the utility of TAVR in lower risk patients (such as PARTNER II, where patients are being enrolled with an STS score of ≥4%). Future studies will also need to assess the relative utility of transfemoral versus transapical TAVR.
Five-year results indicate that hemodynamic and clinical outcomes remain similar between patients undergoing TAVR. The incidence of stroke actually decreased compared with surgical AVR beyond 30 days. Paravalvular AR, however, still remained high, and on post-hoc analysis, was a significant predictor of long-term mortality. Overall though, these results are very encouraging, and highlight the emerging importance of TAVR in the management of high surgical-risk AS patients. Studies are ongoing to assess the utility of TAVR in lower risk patients. Future studies will also need to assess the relative utility of transfemoral versus transapical TAVR.
Douglas PS, Leon MB, Mack MJ, et al., on behalf of the PARTNER Trial Investigators. Longitudinal Hemodynamics of Transcatheter and Surgical Aortic Valves in the PARTNER Trial. JAMA Cardiol 2017;Sep 27:[Epub ahead of print].
Presented by Dr. Vinod Thourani at ACC.13, San Francisco, March 11, 2013.
Reynolds MR, Magnuson EA, Lei Y, et al. Cost-effectiveness of transcatheter aortic valve replacement compared with surgical aortic valve replacement in high-risk patients with severe aortic stenosis: results of the PARTNER (Placement of Aortic Transcatheter Valves) trial (Cohort A). J Am Coll Cardiol 2012;60:2683-92.
Kodali SK, Williams MR, Smith CR, et al., on behalf of the PARTNER Trial Investigators. Two-year outcomes after transcatheter or surgical aortic-valve replacement. N Engl J Med 2012;366:1686-95.
Presented by Dr. Susheel Kodali at ACC.12 & ACC-i2 with TCT, Chicago, IL, March 26, 2012.
Smith CR, Leon MB, Mack MJ, et al., on behalf of the PARTNER Trial Investigators. Transcatheter Versus Surgical Aortic-Valve Replacement in High-Risk Patients. N Engl J Med 2011;364:2187-98.
Presented by Dr. Craig Smith at the ACC.11/i2 Summit, New Orleans, LA, April 3, 2011.
Mack M, Leon MB, Smith CR, et al. 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;Mar 15:[Epub ahead of print].
Kapadia SR, Leon MB, Makkar RR, et al. 5-year outcomes of transcatheter aortic valve replacement compared with standard treatment for patients with inoperable aortic stenosis (PARTNER 1): a randomised controlled trial. Lancet 2015;Mar 15:[Epub ahead of print].
Presented by Dr. Michael Mack at ACC.15, San Diego, CA, March 15, 2015.
< Back to Listings