Transcatheter vs Surgical Aortic Valve Replacement in High-Risk Patients: The PARTNER IA Results


Surgical aortic valve replacement (AVR) has been the standard of care in patients with severe aortic stenosis (AS) for the past half century. Nonetheless, it has become clear from multiple U.S. and European registries that, despite the dismal natural history of untreated AS, at least 30% of symptomatic patients do not receive surgical AVR.(1) The principle reasons cited for this lack of treatment are usually advanced age and/or other comorbidities. Over the course of the last 9 years, a less invasive alternative has been developed involving the transcatheter insertion of an aortic prosthesis that excludes the diseased calcific valve. The first such aortic prosthesis was successfully implanted on an emergency basis by Alain Cribier in 2002.(2) Since that time, transcatheter AVR (TAVR) has rapidly evolved and, with commercial approval in almost 30 countries, there have been over 30,000 implants worldwide. However, despite the widespread adoption of this technology, there has thus far been only one randomized clinical trial comparing TAVR to the current surgical standard: The PARTNER Trial.(3)

While there is an obvious appeal to less invasive treatment of AS without the need for sternotomy and cardiopulmonary bypass, it is vital for patients and the medical community to have a clear understanding of the potential advantages and hazards of this new therapy when compared to the current standard of care.


Partner Study Design

The Edwards SAPIEN valve used in the PARTNER trial evolved from a concept by Alain Cribier. The device consists of a standard bovine pericardial valve sewn onto a stainless steel “stent-like” support frame which is mounted on a balloon expandable delivery system. This is in contradistinction to the other widely employed percutaneously-implanted valve (CoreValve, Medtronic Vascular, Minn, MN) which is based on a different principal of a self-expanding nitinol platform.(4) The assembly of the valve and the catheter in the PARTNER Trial was an early generation device which required large size sheaths for percutaneous entry. There were both a 23- and a 26-mm valve sizes employed, which allowed inclusion of aortic annulus diameters from 18 to 25 mm. Given the limitations of the large sheath size (up to 26 French), there were many patients who did not have iliac-femoral arteries suitable for intra-arterial placement. Thus, a second parallel arm was incorporated in the trial using an identical shorter version of the device which was implanted antegrade via the LV apex (transapical) approach.

Trial Design

All-Cause Mortality at 30 Days and 1 Year: Patient Subgroups Slide

The PARTNER Trial was designed as two, separately powered trials; the first arm was dedicated to inoperable patients, as deemed by a multi-disciplinary Heart Valve Team with cardiac surgeons as the “gatekeepers” and eligibility further adjudicated by the Executive Committee on weekly conference calls. This trial enrolled 358 patients with 1:1 randomization between standard therapy (which frequently included balloon valvuloplasty) and trans-femoral TAVR. The primary endpoint was all-cause mortality over the duration of the trial. This portion of PARTNER revealed a dramatic absolute 20% reduction in mortality at 1 year and was reported at TCT and in The New England Journal of Medicine in September, 2010.(5) The focus of this report is to review the 699 patients who were enrolled in the high surgical risk arm; that is, high-risk AS patients who were still candidates for open surgical AVR.(3) This trial required the patients to be randomized (1:1) to either conventional AVR or to the catheter-based SAPIEN valve. The randomization was stratified based on planned transfemoral approach (492 patients) or the transapical approach (207 patients), when arterial access was problematic. The primary endpoint of this trial was all-cause mortality, powered to demonstrate “noninferiority” of the test TAVR arm vs. the control standard therapy arm at 1-year follow-up. Of note, the transfemoral arm was separately powered for this endpoint as well.

Patient Characteristics and Procedural Outcomes

Clinical Outcomes at 30 Days and 1 Year: All Patients (N=699)

These high-risk patients had a mean age of 84 years old and high-risk patient characteristics were well balanced between the two arms. The mean STS score was 11.8% and over 94% of patients were New York Heart Association Class III or IV. Three quarters of the patients had coronary artery disease, 40% had peripheral vascular disease, and 40% had COPD (8% oxygen dependent). Rhythm disturbances were common with over 40% of patients having a history of atrial fibrillation and over 20% having a permanent pacemaker. The average aortic valve area was less than 0.7 cm(2) and average ejection fraction was 53%.

Considering the early generation device and the limited operator experience, the procedural outcomes after TAVR were excellent; in only 16 of the 348 patients (4.6%) randomized to TAVR was the procedure aborted or converted to open surgical AVR. Among these 16 patients, 9 immediately underwent open surgery (1died), 2 underwent open surgery more than 30 days later, and 5 did not undergo valve repair (3 died). Multiple transcatheter valves (≥2) were implanted in 7 patients because of valve embolization (2 patients) or residual aortic regurgitation (5 patients).

Clinical Outcomes at 30 Days and 1 Year: All Patients (N=699)

Of note, the surgical outcomes were also excellent with an as-treated AVR mortality of 8%, which is 30% less than predicted by their STS scores. Thus the predicate surgical treatment was an exemplary comparator for this novel new therapy.


The study easily met its primary noninferiority endpoint; all-cause mortality at 1 year (intention-to-treat analysis) was 24.2% for TAVR and 26.8% for AVR (P=0.001). The separately powered transfemoral arm also was noninferior to surgical AVR (1-year mortality 22.2% with TAVR and 26.4% with AVR; P=0.002). The transapical outcomes revealed a 29.0% mortality at 1 year with TAVR and 27.9% with AVR.

Since 38 (10.8%) of the patients randomized to surgical AVR did not receive surgery due to (1) deaths or deterioration before scheduled surgery and (2) patient refusals or withdrawals from the trial, the “as treated” analyses have special significance and may be a more accurate reflection of procedural risks and benefits. As treated, the 30-day mortality for all patients (transfemoral and transapical cohorts) was 5.2% for TAVR and 8.0% for AVR (P=0.15) and for the transfemoral cohort was 3.7% for TAVR and 8.2% for AVR (P=0.045). These represent some of the lowest reported 30-day mortality results after TAVR ever reported. As treated, at 1 year, for all patients, mortality was 23.7% vs. 25.2% (P=0.64) and in transfemoral patients was 21.3% vs. 25.2%, for TAVR and AVR respectively (P=0.33).

Neurological Events at 30 Days and 1 Year: All Patients (N=699)Other important outcomes included a two-fold higher frequency of all neurologic events and major strokes, after TAVR compared with surgical AVR, both at 30-days and at 1 year (30 days: all neuro events 5.5% vs. 2.4%, P=0.04; and major strokes 3.8% vs. 2.1%, P=0.20; 1 year: all neuro events 8.35% vs. 4.3%, P=0.04; and major strokes 5.1% vs. 2.4%, P=0.07). However, the composite endpoint of death and major stroke was similar in TAVR and AVR patients at both 30 days and 1 year (30 days: 6.9% vs. 8.2%, P=0.52 and at 1 year: 26.5% vs.28%, P=0.68).

Other notable complications included a much greater frequency of 30-day major vascular complications in TAVR patients (11.0% vs. 3.2%, P<0.001) and a two-fold increase with surgical AVR in major bleeding (19.5% vs. 9.3, P<0.001) and new-onset atrial fibrillation (16.0% vs. 8.6%, P<0.001). Importantly, there was no difference in the rate of new permanent pacemakers for TAVR (3.8%) vs. AVR (3.6%), which is especially relevant given the higher pacemaker rates noted with the self-expanding TAVR systems.

New York Heart Association functional class and six-minute walk distances improved markedly with both therapies; more rapidly at 30 days for TAVR, but by 1 year they were equivalent in the surgical and the TAVR patients. More than 80% of survivors in both groups were either New York Heart Association Class I or II at 1 year.

Subgroup analyses revealed interesting trends, suggesting improved mortality outcomes in females with TAVR and improved outcomes in patients without prior CABG after AVR. This latter finding seems paradoxical and has not been replicated in many other studies.

The echocardiography findings at 1 year indicated a slightly lower mean gradient and an increased valve orifice area with TAVR vs. AVR (10.2 mmHg vs. 11.5mmHg, P=0.008; and 1.6 cm2 vs. 1.4cm2, P=0.002). Importantly, there was more paravalvular regurgitation at one year with TAVR (P<0.001), though moderate to severe aortic regurgitation was present in less than 7% of the TAVR patients. The clinical significance of these differences in valve performance, especially paravalvular aortic regurgitation after TAVR remains to be defined and will require longer term follow-up.


The landmark PARTNER IA Trial indicates that a less-invasive catheter-based technology has now been developed that is an alternative to open surgical AVR in patients who are at high risk for surgery. Obviously, there are many issues that remain controversial in interpreting the results. The implications of increased strokes that were observed with TAVR and their potential causes and prevention are important considerations in the advancement of this technology. New accessory devices for cerebral embolic protection are now being incorporated into these procedures and are being explored in early clinical trials.(6) Further reduction in major vascular complications can be anticipated with lower profile devices that are already available outside the U.S. and are being evaluated in clinical trials, including the PARTNER II Trial. With lower profile devices, an expected increase in transfemoral TAVR procedures will allow percutaneous closure in most patients and the likely reduction of general anesthesia requirements. The clinical importance of mild or moderate paravalvular aortic regurgitation after TAVR requires careful attention and must be addressed in long-term follow-up analyses.

A key question is transcatheter biologic valve durability compared with a surgically implanted valve. Concerns have been expressed about the trauma induced by valve crimping and the barotrauma of balloon expansion and deployment. While the valves used in clinical trials (including the PARTNER Trial) are prepared in a manner similar to surgically implanted valves, only long-term systematic echocardiography and clinical follow-up can satisfy the durability concerns. However, considering the elderly patient population being treated with TAVR and the few reports of important structural valve deterioration during 3 to 5 years follow-up from outside the U.S. registries, there is growing optimism that the current biologic valves will have acceptable durability.

Future Studies

The preliminary success of TAVR has naturally spawned interest in further refinements in these transcatheter valve systems for both transapical and transfemoral applications. Several of these next generation devices already have had first-human-use clinical experiences outside the U.S.

Importantly, the encouraging results from the PARTNER Trial and other overseas registries have raised the question of whether TAVR can be expanded to intermediate risk AS populations. There are two such trials in the advanced planning stages; SURTAVI with the Medtronic CoreValve, and the PARTNER IIA Trial with the new Sapien XT-Novaflex system. These randomized trials will most likely involve patients with STS scores as low as 3 or 4 and will require larger study cohorts and longer term follow-up than the current trials.

Another important area of exploration is the so-called “valve-in-valve” option. With the expansion of bioprosthetic valve use, there is an opportunity for treating deteriorated bioprostheses with the placement of a catheter-mounted valve within the failed surgical prosthesis.(7) There has been preliminary success in various centers with the valve-in-valve procedure and hopefully clinical trials for this important subset of patients will be underway in the near future.


  1. The PARTNER IA Trial has demonstrated non-inferiority of all-cause mortality with TAVR compared to surgical AVR in high-risk patients with severe aortic stenosis.
  2. While mortality outcomes are equivalent, there are important differences in certain complications; increased strokes and vascular complications after TAVR and more bleeding and arrhythmia complications after standard surgical AVR.
  3. Longer term clinical and echocardiography follow-up will be required to determine the durability and safety of these valves.
  4. Expansion of TAVR into intermediate-risk cohorts and valve-in-valve treatment for bioprosthetic valve failure will be tested in planned clinical trials.


  1. Bach DS, Siao D, Girard SE, Duvernoy C, McCallister BD, Jr., Gualano SK. Evaluation of patients with severe symptomatic aortic stenosis who do not undergo aortic valve replacement: The potential role of subjectively overestimated operative risk. Circulation. Cardiovascular quality and outcomes. 2009;2:533-539.
  2. Cribier A, Eltchaninoff H, Bash A, Borenstein N, Tron C, Bauer F, Derumeaux G, Anselme F, Laborde F, Leon MB. Percutaneous transcatheter implantation of an aortic valve prosthesis for calcific aortic stenosis: First human case description. Circulation. 2002;106:3006-3008.
  3. Smith CR, Leon MB, Mack MJ, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG, Fontana GP, Makkar RR, Williams M, Dewey T, Kapadia S, Babaliaros V, Thourani VH, Corso P, Pichard AD, Bavaria JE, Herrmann HC, Akin JJ, Anderson WN, Wang D, Pocock SJ. Transcatheter versus surgical aortic-valve replacement in high-risk patients. N Engl J Med. 2011; 364: 2187-2198.
  4. Tamburino C, Capodanno D, Ramondo A, Petronio AS, Ettori F, Santoro G, Klugmann S, Bedogni F, Maisano F, Marzocchi A, Poli A, Antoniucci D, Napodano M, De Carlo M, Fiorina C, Ussia GP. Incidence and predictors of early and late mortality after transcatheter aortic valve implantation in 663 patients with severe aortic stenosis. Circulation. 2011;123:299-308.
  5. Leon MB, Smith CR, Mack M, Miller DC, Moses JW, Svensson LG, Tuzcu EM, Webb JG, Fontana GP, Makkar RR, Brown DL, Block PC, Guyton RA, Pichard AD, Bavaria JE, Herrmann HC, Douglas PS, Petersen JL, Akin JJ, Anderson WN, Wang D, Pocock S. Transcatheter aortic-valve implantation for aortic stenosis in patients who cannot undergo surgery. N Eng J Med. 2010;363:1597-1607.
  6. Nietlispach F, Wijesinghe N, Gurvitch R, Tay E, Carpenter JP, Burns C, Wood DA, Webb JG. An embolic deflection device for aortic valve interventions. JACC Cardiovasc Interv. 2010;3:1133-1138.
  7. Azadani AN, Tseng EE. Transcatheter valve-in-valve implantation for failing bioprosthetic valves. Future Cardiol. 2010;6:811-831.

Keywords: Aortic Valve, Aortic Valve Stenosis, Comorbidity, Heart Valve Prosthesis, Standard of Care

< Back to Listings