Assessment of Neurologic Injury Subsequent to Transcatheter Aortic Valve Replacement: A Feasibility Study

Nearly 1 in 10 adults over 65 years have aortic valve stenosis (AS), defined as an obstruction of blood flow across the aortic valve.(Faggiano, Antonini-Canterin et al. 2006) AS is a life-threatening disease, and one whose incidence increases with age. Natural history studies suggest that the long-term survival among patients with severe AS is unfavorable, even among patients who are asymptomatic, with event-free survival for AS being 64% at 1-year, 36% at 2-years, 12% at 4-years, and 3% at 6-years.(Rosenhek, Zilberszac et al. 2010) Until recently surgery has been the gold standard approach for treatment for severe AS. Recently, a less invasive approach, transcathether aortic valve replacement (TAVR) has emerged as a viable treatment alternative, including among those previously not thought of as suitable candidates for surgery. Unlike its surgical counterpart that utilizes cardiopulmonary bypass and direct vision by a cardiothoracic surgeon, TAVR is performed (by a surgeon in conjunction with an interventional cardiologist) by threading a wire mesh valve through a catheter using fluoroscopy while the heart is still beating. Concern regarding broader adoption of TAVR often revolves around the higher stroke rate relative to surgery (5.5% vs. 2.4%, p = 0.04).(Leon, Smith et al. 2010) Much of the risk associated with neurologic injuries (whether stroke, neurocognitive deficits or silent infarcts) revolves around embolically-generated sources, including: threading a guidewire across diseased vessels, removal of the native valve, or insertion/expansion of the new valve.(Miller, Blackstone et al. 2012) Among 47 patients studied by Miller within a neurologic sub-study of the PARTNER Trial, there were 49 (n=31 TAVR, 16 AVR) neurologic events (defined as a transient ischemic attack or stroke).(Miller, Blackstone et al. 2012) In a recent review article, Daneault cited risk of post-procedural cerebral infarcts within 5-7 days (using Diffusion-weighted MRI) of 38-47% with standard aortic valve surgery vs. 68-84% with TAVR.(Daneault, Kirtane et al. 2011) Given the growing interest and anticipated broadening of indications for TAVR in and outside of the United States, it is increasingly important to develop a sound methodological approach for evaluating the safety and effectiveness of this emerging treatment modality. In the absence of such information, it is impossible for a patient or clinician to estimate the likelihood for developing a neurologic injury subsequent to TAVR. Additionally, linkage of processes of care with embolism detection (through transcranial Doppler) would provide evidence to support targeted quality improvement efforts. Such a strategy has been useful in prior studies applied to coronary artery bypass grafting (CABG) surgery.(Groom, Quinn et al. 2009) Indeed, early studies evaluating TAVR have found periods of the TAVR procedure which may be more prone to the generation of embolic debris, although they have used varied methodological approaches. Importantly, the relationship between these emboli and development of neurobehavioral or ischemic lesions has not been explored in this setting.

The overlap between sleep disorders and stroke is an emerging field. Sleep apnea is a serious medical condition that is very common after stroke, affecting over half of acute ischemic stroke patients. (Broadley, Jorgensen et al. 2007) Recently, sleep apnea has been recognized as an independent risk factor for stroke. (Munoz, Martinez-Vila et al. 2006; Redline, Gottlieb et al. 2010; Yaggi, Kernan et al. 2005) Furthermore, sleep apnea has been identified as an important predictor of both poor functional outcome and death following stroke. (Sahlin, Sandberg et al. 2008; Turkington, Allgar et al. 2004) There remains controversy over whether OSA predates stroke, whether stroke predates sleep apnea, and whether stroke exacerbates sleep apnea severity. To answer the questions definitely, sleep apnea testing would have to be performed just prior to and again after stroke. Because stroke is typically unpredictable, this has been logistically challenging to pursue. The current study however provides a rare opportunity to study patients for sleep-disordered breathing just prior to and after a type of procedure that has an association with acute cerebral infarction identified on MRI. (Kalert, Knipp et al. 2010) Within this context, we seek to determine the feasibility of assessing neurologic injuries subsequent to TAVR. Such a model has been applied previously by the principal investigator to assess and improve neurologic outcomes for other cardiac surgical procedures.(Groom, Quinn et al. 2009) We shall assess patients during the following intervals: pre-procedure, within 72-96 hours post-procedure, and 3 months post-procedure (see Appendix). Case videos will be established to assist in identifying and associating emboli (using transcranial Doppler) and processes of clinical care during the TAVR procedure. Neurologic injury will be assessed in the following ways: stroke (neurologic exam, NIH Stroke Scale), silent infarcts (diffusion-weighted MRI, diffusion-tensor imaging), and neurobehavioral deficits (a battery of neuropsychological tests). Secondly, we will investigate changes in the apnea-hypopnea index (AHI), a measure of sleep-disordered breathing, before vs after surgery between those subjects who develop post-operative acute brain infarction and those who do not. We hypothesize that subjects who develop acute brain infarction will have an increase in AHI between baseline and post-op measurements compared with those subjects who do not develop acute brain infarction. A research coordinator will coordinate the testing.