Post-TAVI Neurologic Problems an Ongoing Issue

Even though post-TAVI stroke has become less frequent over the years due to increased operator experience, advancements in technology, and rising numbers of lower-risk patients, it remains an important and dreaded complication that needs to be addressed, according to two recent studies providing additional insights into neurologic issues that occur following the procedure.

In one, lead author Alexandra Lansky, MD (Yale School of Medicine, New Haven, CT), and colleagues demonstrate that the number, size, and total volume of acute brain infarcts detected by diffusion-weight MRI after TAVI correlate with risk of clinical stroke and disability. They suggest that these measures may be suitable as surrogate endpoints in future trials of cerebral embolic protection (CEP) devices.

And in the other study, researchers led by Marisa Avvedimento, MD, and Pedro Cepas-Guillén, MD, PhD (both from the Quebec Heart and Lung Institute, Laval University, Quebec City, Canada), provide a clinical validation of the updated VARC-3 definitions of neurologic events, confirming that periprocedural stroke has a significant impact on risk of 1-year mortality after TAVI.

Speaking with TCTMD, Josep Rodés-Cabau, MD, PhD (Quebec Heart and Lung Institute, Laval University), senior author of the latter paper, said a message that can be taken away from both studies together is that “this problem has not disappeared and there is still room for improvement. . . . You see that this neurologic issue in TAVR candidates is still there.”

Although stroke rates have declined over time, there has not been a dramatic change, he said, noting that most patients undergoing TAVI are still elderly with a heavy burden of comorbidities that increase risks of cerebrovascular events. In that context, he added, it will be difficult to get the post-TAVI stroke rate below 1%.

CEP devices may have some benefit for reducing neurologic events, but their use remains controversial, Rodés-Cabau said: “Clinically, it’s difficult to see a strong benefit.”

The ideal solution, then, might be to come up with predictors that can identify patients at highest risk who would benefit from an intervention—whether a CEP device, aggressive antithrombotic therapy, or something else—to reduce these complications, he said.

This problem has not disappeared and there is still room for improvement. Josep Rodés-Cabau

Amar Krishnaswamy, MD (Cleveland Clinic, OH), told TCTMD that “this is yet another series of papers that demonstrates the frequency of stroke and the devastating implications of stroke in both the near term and the midterm.”

Highlighting the ongoing debate about use of CEP devices, he said there have been at least suggestions of benefit with deploying the Sentinel device (Boston Scientific) during TAVI.

“We know definitively that there is no safety concern or safety signal related to the use of that device,” he said. “And so at least for us at Cleveland Clinic, we apply the Sentinel to every single TAVR that we perform and recently published a paper demonstrating that it's feasible to use in about 97% of cases. So there seems to be little reason not to use it at this point in time.”

Role of Diffusion-Weighted MRI

The study by Lansky et al, published in the August 20, 2024, issue of the Journal of the American College of Cardiology, explored the impact of acute brain infarcts seen on diffusion-weighted MRI, which have been seen after TAVI in up to 100% of patients in some studies. The clinical relevance of these lesions, however, has been questioned.

For this analysis, the investigators pooled patient-level data from four prospective studies, including three RCTs evaluating the TriGuard CEP device (Keystone Heart)—DEFLECT III, REFLECT I, and REFLECT II—and the Neuro-TAVI registry of TAVI without CEP use; all included serial neurologic assessments. There was a total of 495 patients (mean age 81 years; 56% men) who underwent the procedure and completed a predischarge diffusion-weight MRI scan. Patients had an intermediate surgical risk, with a mean STS score of 5.2 and EuroSCORE II of 4.8.

All neurologic events were defined and adjudicated according to the NeuroARC recommendations. In the first 30 days, 6.9% of patients had a clinical ischemic stroke, 3.1% had a disabling stroke, and 0.8% died.

Imaging-detected acute ischemic brain injury was seen in 85% of patients, including all of those who had a stroke and 84% of those who didn’t. Median total lesion number was 3.0 per patient, median individual lesion volume was 35.9 mm³, and median total lesion volume was 227 mm³; all of these figures tended to be higher in patients with stroke.

The measure that best discriminated clinical ischemic stroke at 30 days was total lesion volume (C-statistic 0.84), followed by maximum individual lesion volume (0.82) and total lesion number (0.81). After adjustment for potential confounders, the C-statistic for total lesion volume increased to 0.87.

Using a threshold of 500 mm³, patients with a high total lesion volume (29% of the population) had higher rates of ischemic stroke (18.2% vs 2.3%) and disabling stroke (8.8% vs 0.9%) and a lower rate of complete stroke recovery (44.0% vs 62.5%; P < 0.0001 for all) at 30 days.

“Our study is the first to provide a robust validation of acute brain injury quantified by diffusion-weighted MRI as a strong discriminator of stroke after TAVR,” Lansky told TCTMD via email.

This is important because “until now we have been struggling with designing meaningful clinical trials to evaluate the efficacy of cerebral embolic protection devices,” she said, noting that low stroke rates after TAVI mean that 8,000 to 10,000 patients would be needed to power a trial capable of detecting a difference in that endpoint. “Our data provides a powerful surrogate of stroke that can be used in pivotal trials to demonstrate CEP device efficacy.”

Lansky estimated that using a total lesion volume greater than 500 mm³ as a surrogate endpoint would enable much smaller studies of 400 to 500 patients that could demonstrate a benefit of CEP devices. “Approval studies for adjunctive CEP devices must be reasonably sized, yet reliably demonstrate safety and efficacy,” she said. “Using diffusion-weighted MRI in designing CEP trials can accomplish this.”

Krishnaswamy, who co-wrote an editorial accompanying the paper with Samir Kapadia, MD (Cleveland Clinic), and Michael Mack, MD (Baylor Scott & White Health, Dallas, TX), said “these MRI-based studies are important because they remind us of the degree to which the transcatheter aortic valve replacements that we do can cause debris embolization to the brain.”

As for whether these imaging measures can serve as surrogates in future trials of CEP devices, Krishnaswamy pointed to some obstacles, including variability in MRI image acquisition protocols and the expense of performing the scans. “Oftentimes these newer CEP devices are coming from smaller companies that may not have the money necessary to fund a trial that has MRI-related endpoints,” he said. “It would be really nice to have, but I think that the logistics are going to be what makes it more difficult.”

Lansky acknowledged the difficulty of using MRI endpoints for clinical trials, noting that the proportion of patients who don’t undergo the mandated scans is currently about 15% to 20%. But, she noted, “portable MRI systems are rapidly being developed, where the MRI comes to the patient rather than the patient going to the MRI suite. The resolution of these systems still needs some refining, but ultimately point-of-care MRIs will change the diagnostic yield, detect more strokes in our postprocedural patient, and emphasize the need for effective CEP devices.”

Validating VARC-3 Definitions

The study by Avvedimento, Cepas-Guillén, Rodés-Cabau, and others, published in the August 12, 2024, issue of JACC: Cardiovascular Interventions, was designed to clinically validate the updated VARC-3 definitions of neurologic events post-TAVI that came out in 2021, broadening the scope of the VARC-2 criteria. The endpoint was harmonized with NeuroARC recommendations to include the following events:

• NeuroARC type 1 (overt central nervous system [CNS] injury that includes stroke and symptomatic hypoxic-ischemic injury)
• NeuroARC type 2 (covert CNS injury that includes neuroimaging or pathological evidence of focal or multifocal ischemia or hemorrhage without acute neurologic symptoms)
• NeuroARC type 3 (neurologic dysfunction without CNS injury, which includes TIA and delirium)

The analysis included 2,924 patients (mean age 80 years; 48% women) with severe aortic stenosis undergoing TAVI with newer-generation valves at three centers in Canada and Italy between 2014 and 2023. Median STS score was 3.6, and median EuroSCORE II was 3.7.

After a median follow-up of 13 months, 16.1% of patients had a neurologic event, including type 1 events in 37.4%, type 2 events in 4.7%, and type 3 events in 58.0%. More than half of the events (58.6%) were periprocedural, occurring within 30 days of TAVI.

These periprocedural events overall were independently associated with a greater risk of mortality at 1 year (18.2% vs 6.3%; adjusted HR 1.91; 95% CI 1.23-1.97). But that risk was driven by NeuroARC type 1 events—ischemic stroke (adjusted HR 4.42; 95% CI 2.72-7.18) and hemorrhagic stroke (HR 13.6; 95% CI 1.28-33.12)—particularly with increasing severity according to the NIHSS score.

NeuroARC type 2 and 3 events were not associated with mortality risk, although Rodés-Cabau said that doesn’t mean they’re not clinically relevant. He and his colleagues note that delirium, for instance, has been associated with impaired recovery and longer hospital stays after TAVI.

Regarding overall events, “given their negative impact on outcomes, every attempt should be made to reduce the risk of neurologic complications after TAVR,” the authors write. “Likewise, the implementation of protective measures needs to remain a major research priority in this field to advance the care and the prognosis of TAVR candidates.”

Implications for Use of CEP Devices

Krishnaswamy said “these are two nice papers to look at together because I think, ultimately, the thoughts go in the same direction” in underscoring that stroke still occurs after TAVI and, when it does, signals poorer functional and survival outcomes moving forward. This is, he added, “still a really important complication that patients suffer of from TAVR.”

As for use of CEP devices, he pointed out that the PROTECTED TAVR trial and an analysis of the Society of Thoracic Surgeons/American College of Cardiology TVT Registry reported earlier this year indicate that disabling strokes in particular are less frequent when protection is added to TAVI. “I think you can probably draw a straight line between the findings of lesion size and stroke severity in the [Lansky et al] analysis with studies that have shown smaller lesion size with CEP use and clinical outcome studies that have suggested lower disabling stroke rates with CEP use,” he said.

And the analysis validating the VARC-3 definitions of neurologic events showing an association between periprocedural strokes and mortality also “I think points to the need to reduce embolic material or debris material during the TAVR procedure,” Krishnaswamy said.

Currently, TAVI operators have firm opinions on whether CEP devices have a clinical benefit, possibly related to debate over the clinical data, to the financial impact on TAVI programs, or both, he said. Without definitive clinical data, there’s not likely to be much change in practice, he said.

Krishnaswamy noted that the BHF PROTECT-TAVI trial, with a target enrollment of 7,730 patients, is currently ongoing in the United Kingdom and will inform this issue, especially when combined with the PROTECTED TAVR results. “But barring that, I'm not sure what would change the current inertia,” he said.