The risk of stroke in patients undergoing TAVI


By Alexandra Lansky and Cody Pietras

Stroke is among the most feared complications of cardiac interventions because of its extreme morbidity and mortality burden,1 and many patients consider the prospect of a disabling stroke to be a fate worse than death.2 While transcatheter aortic valve implantation (TAVI) is the standard of care for inoperable patients with severe symptomatic aortic stenosis and a compelling approved alternative to surgical aortic valve replacement for patients at high operative risk, the early experience with the intervention raised concerns regarding the risk of stroke and other neurological events. Alexandra Lansky and Cody Pietras explore the risk of stroke in patients undergoing TAVI and review how this risk can be modified.

In the PARTNER (Placement of aortic transcatheter valve) trial, high-risk operative candidates randomised to TAVI had a substantially increased risk of stroke or transient ischaemic attack compared with surgery at both 30 days (5.5% vs. 2.4%) and one year (8.3% vs. 4.3%);3 among inoperable patients, the corresponding rates were 6.7% and 10.6%, the majority of which were disabling strokes.4 Approximately 50% of events were periprocedural, followed by a later constant hazard phase driven by traditional risk factors.5

In the six years since PARTNER trial enrolment, TAVI has evolved substantially. More recent comparisons generally find a similar or lower stroke risk in TAVI compared with surgery, probably due to increased operator experience and lower-profile devices. Nonetheless, neurological events continue to affect a substantial proportion of patients, with 30-day stroke rates generally in the range of 3–5% in randomised trials.6–8 While a report from the Society of Thoracic Surgeons (STS)/American College of Cardiology (ACC) Transcatheter Valve Therapies registry (n = 12,182) reported a 2.5% stroke rate at 30 days and 4.1% at one year,9 there is evidence that neurological events are substantially under-reported: Messe et al found a disturbing 17% early stroke rate after surgery when routine neurologist evaluation was performed.10 Likewise, routine neurological and neuroimaging follow-up in TAVI populations has revealed clinically apparent post-procedure neurologic deficits accompanied by cerebral ischaemia in 15–28% of patients.11,12

While initially limited to an elderly and highly comorbid population, the clinical application of TAVI will probably continue to broaden in coming years. Recent registries indicate that TAVI has achieved parity with surgery even in low-to-intermediate risk patients;13,14 if this finding is confirmed in ongoing randomised trials, TAVI will soon be extended to an intermediate-risk population with increased life expectancy—in whom reducing the risk of neurological events and the cumulative embolic burden to the brain are even more critical.

To minimise the risk of neurological injury following TAVI, a range of complementary approaches will be required, including improved patient selection and risk prediction, continued procedural and device improvements, and possibly the use of adjunctive embolic protection devices. Patient and procedural risk factors for early stroke include female sex, major vascular complications, acute kidney injury (which could be a surrogate for overall risk or an indication of high levels of embolisation), and pre-existing (but not new-onset) atrial fibrillation.15,16 The risks associated with female sex and major vascular complications reinforce the importance of lower-profile TAVI devices and minimal manipulation in atherosclerotic vessels.17 Transcranial Doppler monitoring indicates that the primary mechanism of periprocedural stroke is ischaemic and embolic, caused by the dislodgement of debris from the aortic arch, annulus, and native valve,18 and 68–100% of TAVI patients have new post-procedure cerebral ischaemic lesions on diffusion-weighted magnetic resonance imaging;19,20 in addition to serving as a surrogate for cerebral embolisation that could result in clinical stroke, such lesions have been linked to cognitive decline and an increased risk of future stroke. Although avoidance of pre- and post-implant balloon dilatation is feasible and may help prevent neurologic events,17,21,22 the embolisation of some debris appears to be an inevitable consequence of the TAVI procedure.

Several devices have been developed for placement in the aortic arch or supra-aortic vessels to prevent emboli from reaching the cerebral circulation, three of which are currently undergoing clinical evaluation. Of these, the first two are filters placed in the aortic arch to cover two (Embrella embolic deflector system, Edwards Lifesciences) or all three (TriGuard HDH embolic deflection device, Keystone Heart) cerebral vessel take-offs, while the third (Sentinel cerebral protection system, Claret Medical) is an embolic capture device consisting of dual basket-type filters placed above the aortic arch. Initial trials of these devices have been encouraging: in DEFLECT III (A prospective, randomized evaluation of the TriGuard HDH embolic deflection device during TAVI) (n=85), patients treated with the TriGuard HDH device had more freedom from new ischaemic lesions, fewer new neurologic deficits, and improved cognitive function.12 Similarly, the randomised CLEAN-TAVI (Claret embolic protection and TAVI) trial of the Sentinel device (N=100) reported a significant reduction in the number and volume of postprocedure cerebral ischaemic lesions as well as fewer neurologic deficits.11 The PROTAVI-C (Prospective randomized outcome study in patients undergoing TAVI to examine cerebral ischaemia and bleeding complications) pilot study of the Embrella device (n=52) reported a smaller average volume of new lesions, although all subjects had new post-procedure lesions.20 Before such devices are adopted, however, pivotal randomised trials are required to confirm the observed effects on clinical or subclinical neurologic and cognitive outcomes and to conclusively assess the overall impact of procedural neuroprotection. The REFLECT trial of the TriGuard HDH device and the US SENTINEL trial of the Sentinel device are currently underway.

Many other challenges highlighted by the early TAVI experience have since been overcome, and the intense scrutiny of TAVI-related stroke will only accelerate the development of methods to minimise the procedure’s neurological and cognitive impact. Doing so is essential to ensure that TAVI provides the maximum benefit to the growing number of patients with aortic stenosis who are appropriate candidates for valve replacement.


  1. Eggebrecht et al. Eurointervention 2012; 8: 129–38.
  2. Stolker et al. Circulation 2014; 130: 1254–61.
  3. Smith et al. N Engl J Med 2011; 364: 2187–98.
  4. Leon et al. N Engl J Med 2010; 363: 1597–1607.
  5. Miller et al. J Thorac Cardiovasc Surg 2012; 143: 832–43.
  6. Popma et al. J Am Coll Cardiol 2014; 63: 1972–81.
  7. Adams et al. N Engl J Med 2014; 370: 1790–08.
  8. Leon et al. Data presented at American College of Cardiology Scientific Sessions 2013.
  9. Holmes et al. JAMA 2015; 313: 1019–28.
  10. Messe et al. Circulation 2014; 129: 2253–61.
  11. Linke et al. Data presented at Transcatheter Cardiovascular Therapeutics 2014.
  12. Lansky et al. Eur Heart J 2015; 36: 2070–78.
  13. Tamburino et al. J Am Coll Cardiol 2015; 66: 804–12.
  14. Thyregod et al. J Am Coll Cardiol 2015; 65: 2184–94.
  15. Bosmans et al. J Am Coll Cardiol 2015; 66: 209-17.
  16. O’Connor et al. J Am Coll Cardiol 2015; 66: 221–28.
  17. Nombela-Franco et al. JACC Cardiovasc Interv 2012; 5: 499–512.
  18. Kahlert et al. Circulation 2012; 126: 1245-55.
  19. Arnold et al. JACC Cardiovasc Interv 2010; 3:1 126–32.
  20. Rodes-Cabau et al. JACC Cardiovasc Interv 2014; 7: 1146–55.
  21. Mollmann et al. Clin Cardiol 2014; 37: 667–71.
  22. Hahn et al. JACC Cardiovasc Interv 2014; 7: 781–89.

Alexandra Lansky, professor of Medicine (Cardiology) and director, Heart and Vascular Clinical Research Program, Yale School of Medicine, New Haven, USA; Cody Pietras, associate director, Yale Cardiovascular Research Group, Yale School