TAVR changing the game again- Dr Anoop Agrawal

Two recent trials published in NEJM establishes Transcatheter Aortic Valve Replacement (TAVR) as an excellent alternative to surgery in patients with symptomatic severe aortic stenosis (AS) who are low risk for surgery (STS <4).

Since its first-in-human use in 2002 and FDA approval in 2011, TAVR has seen exponential growth. The therapy has been adopted worldwide with thousands of patients who are at an intermediate to extreme risk for surgical aortic valve replacement (SAVR) benefiting with this therapy annually. TAVR volumes have surpassed isolated SAVR in many developed countries. TAVR is currently recommended (class I) in patients with severe symptomatic AS who are at high or prohibitive risk for SAVR and is an effective alternative to SAVR in patients with intermediate surgical risk. SAVR remains the preferred therapy for patients who are at low surgical risk. We are about to witness a change very soon.

The two trials presented at the ACC 2019 (New Orleans, USA) were the showstoppers. The EVOLUT trial randomized 1468 patients with AS who met criteria for AVR and were low surgical risk for SAVR (STS risk ≤3) to TAVR (using self-expanding valve) vs SAVR in 1:1 fashion. Relevant exclusion criteria were patients who would qualify for mechanical prosthesis (typically young patients), bicuspid aortic valve, and patients with complex coronary artery disease. Primary endpoint at 24 months, composite of all-cause death or disabling stroke, was seen in 5.3% in TAVR arm vs 6.7% in SAVR arm. While TAVR was non-inferior to SAVR at 24 months, disabling strokes were lower in TAVR cohort. Only less than 2% of patients in the TAVR arm had cerebral embolic protection used. Interestingly, major vascular complications were similar in both groups. Clinical endpoints like new-onset atrial fibrillation, major bleeding, renal injury, and hospitalization for heart failure were all in favour of TAVR. Severe patient-prosthesis mismatch at 1 year was seen in 1.8% in TAVR group vs 8.2% in SAVR. Pacemaker rates were higher in TAVR at 17% vs 6%, and severe paravalvular leak in TAVR at 1 year was 0.2%. Observed to expected mortality in TAVR arm was 0.26 consistent with prior reports, while that in SAVR arm was 0.68 indicating excellent surgical outcomes.

PARTNER 3 trial randomized 1000 patients to TAVR using a balloon expandable valve vs SAVR in a 1:1 fashion. Primary endpoint at 1 year, which also included re-hospitalization, was seen in 8.5% in TAVR vs 15.1% in SAVR, a difference which was statistically superior favouring TAVR. Mortality and stroke rates in TAVR vs SAVR were 1% vs 2.5% and 1.2% vs 3.1% respectively. Combining hard clinical endpoints (mortality and stroke), one-year incidence of 2.2% vs 5.6% was significantly in favour of TAVR. A new pacemaker was needed in 6.6% of patients undergoing TAVR vs 4.1% with SAVR.

These two studies make a convincing statement that transfemoral TAVR may be as good as or better than SAVR in low risk severe AS patients. TAVR offered better hemodynamics compared to SAVR while having similar or lower rates of disabling stroke, mortality, and major vascular complications. Pacemaker requirement was expectedly higher in TAVR arm. Certain salient points must be noted from the above studies. Firstly, the mean age of study participants was 73 years. Patients who would be a candidate for mechanical prosthesis i.e. young individuals were excluded from the study. Also, patients with bicuspid aortic valve were excluded. While there is mounting evidence that new generation transcatheter valves can achieve similar procedural outcomes in bicuspid or tricuspid valve anatomy, whether clinical outcomes as seen in these trials can be replicated in patients with bicuspid aortic valve remains to be seen. Secondly, patients with complex coronary artery disease or multi-valvular pathology were also excluded. This outlines the fact that SAVR may offer them a theoretical advantage of single procedure vs multi-staged percutaneous approach which may add procedural complication. Thirdly, the need for a pacemaker is higher in TAVR. While pacemaker in itself has not been conclusively shown to affect mortality, it adds procedural cost and morbidity. Fourthly, while the EVOLUT data is reported for 2 years, most of the complete follow up was only up to one year. Even if we stretch it, we know TAVR performs well up to 2 years in these selected patients. Since these patients are expected to live much longer, they may outlast the valve’s longevity. The question of valve durability remains unanswered. Last, but not the least, the current trial data didn’t report sub-clinical valve thrombosis incidence. We know that transcatheter valves are associated with sub-clinical valve thrombosis in about 10% of cases. While these have not been associated with clinical neurological events, their effect on valve longevity is unknown. Most of such lesions resolve with anticoagulation though. Both the trials have parallel study groups which will address this in the future.

We are certainly heading towards a new era where TAVR may become the default treatment strategy for all patients with severe AS, with SAVR reserved for special situations. Across the risk spectrum, more than 50% of patients with severe AS fall into low surgical risk. If TAVR is approved in the low-risk subset, the global presence of TAVR can be expected to increase many folds. 53 patients randomized to SAVR in EVOLUT study and 43 patients randomized to SAVR in PARTNER 3 study didn’t undergo SAVR (vs 12 and 7 for TAVR respectively). This, to a certain degree, can be attributed to the patient’s unwillingness to undergo surgery. Going forward, the patient’s preference will certainly be taken into account which will invariably favour TAVR over SAVR.

As of this writing, SAVR remains the treatment of choice for patients who are otherwise a candidate for mechanical prosthesis in severe AS, has pure aortic regurgitation, has aortic valve endocarditis, or any concomitant pathology that requires open heart surgery.