Abstract
INTRODUCTION: In China, evidence regarding cerebral embolic protection device (CEPD) use during transcatheter aortic valve replacement (TAVR) for severe aortic stenosis treatment is limited. This study evaluated the TriGUARD 3 (TG3) CEPD performance in patients undergoing TAVR. METHODS: Data from two studies were pooled: the CEPD group was derived from a multicenter TG3 trial in China, whereas the control group was obtained from a single-center registry. All participants underwent transfemoral TAVR and completed pre- and postoperative diffusion-weighted magnetic resonance imaging (DW-MRI). The primary outcome was total cerebral ischemic lesion volume on DW-MRI. RESULTS: No significant difference was observed between groups in total lesion volume {CEPD [n = 62] vs. control [n = 56]; 256.53 [interquartile range (IQR), 44.12-667.99] vs. 271.88 [IQR, 96.10-650.87]; p = 0.456}. Median regression analysis in the overall cohort showed no significant association between CEPD use and total lesion volume (p = 0.181). Nonetheless, among patients with bicuspid aortic valve (BAV) stenosis, the CEPD group demonstrated significantly lower total lesion volume [165.43 (IQR, 32.96-311.13) vs. 309.38 (IQR, 96.10-788.49); p = 0.025], average lesion volume [61.3 (IQR, 23.44-89.65) vs. 93.75 (IQR, 51.73-137.07); p = 0.019], and maximum single-lesion volume [89.65 (IQR, 28.13-174.02) vs. 164.14 (IQR, 75.00-365.08); p = 0.019]. Median regression revealed that CEPD use was significantly associated with reductions in total, average, and maximum single-lesion volumes (median differences: -406.1, -82.2, and -137.6; all p < 0.05), independent of age, sex, hypertension, diabetes, valve type, and pre-dilatation. CONCLUSION: In patients with severe aortic stenosis undergoing transfemoral TAVR, TG3 CEPD did not significantly reduce the total lesion volume on DW-MRI. In the BAV subgroup, an association was observed between device use and reductions in total, average, and maximum single-lesion volumes. This exploratory finding is hypothesis-generating and should be further elucidated in larger randomized studies.