Background: Coronary artery disease in patients with severe aortic valve stenosis treated with transcatheter aortic valve implantation (TAVI) is common. Self-expanding (SE) devices cover the aortic root and might make the coronary access more challenging compared to balloon-expandable (BE) devices.  

The aim of this study was to report the incidence of coronary angiography (CA) and percutaneous coronary intervention (PCI) after TAVI and to compare coronary access in SE and BE devices.

Methods: This is a retrospective analysis of 1346 patients undergoing TAVI between 09/2007 and 03/2020 included in a single registry of the Heart Center, Segeberger Kliniken. Patients undergoing CA or PCI at any time point after TAVI were identified. Clinical, angiographic and procedural characteristics were reviewed and analyzed. Data of patients treated with SE (CoreValve, Evolut R/Pro) and BE devices (Sapien XT/3) were compared. 

Results:  CA after TAVI was performed in 140 patients (10.6%). In 130 patients (92.9%) only one catheter was needed to engage each coronary artery. Successful engagement was mostly performed using JL4 for the left coronary artery (LCA) and JR4 for the right coronary artery (RCA) with no significant difference between SE and BE devices (p= 0.49 for LCA and p= 0.13 For RCA). A selective engagement was higher in the BE group (91.9% vs. 65.5% for the LCA and 91.9% vs. 55.3% for the RCA, each p<0.001).

Eighty patients (5.9%) (mean age 78.6 ± 5.8 years) underwent PCI over 87 procedures in 118 lesions at a median follow-up period of 19 month (IQR 4-38) after TAVI, of those 39 patients with BE, and 41 patients with SE devices. Overall, in 14 (16.1%) of the procedures PCI was performed in the setting of acute coronary syndrome and 47% of the treated lesions were of type B2/C, and 36% of the lesions involved bifurcations. In 30.2% (19/63) of PCIs in the LCA two or more catheters were used to engage the left main (LM) (range 1-10), with no difference between SE and BE groups (p = 0.9). For the RCA, more than one guide catheter was used in In 37.5% (9/24 cases) (p=0.42 for difference between SE and BE). The most used guiding to engage the LM was different between the two groups (SE: JL4.0 in 71% and in BE: extra back-up (EBU) 4.0 in 48.3%, p <0.001). The most used guiding to engage the RCA was similar (JR 4) in both groups (54.5% in SE and 55.5% in BE; p = 0.77).  There was not significant difference in terms of the engagement selectivity during PCI (94.9% vs. 82.9%; p = 0.09). The use of guide extension was not different between the groups (4.9% in SE vs. 5.1% in BE group; p = 0.5). Engagement across the valve was the main access in SE (95.1%), whereas BE devices were mostly accessed above the valve (84.6%)(p value <0.001). Procedural success was similar between the two groups (97.5% in SE vs. 100% in BE; p = 0.34). Of interest, the time of the procedure and the used contrast amount were significant lower in the SE vs. the BE group (p = 0.035 and p = 0.049, respectively).

Conclusion: Although more unselective coronary engagement during CA was the case in SE devices, catheterization of the coronary ostia after transfemoral TAVI with self-expanding or balloon-expandable devices was successfull in almost all cases. Use of special technic such as guide extension and dedicated guide catheter facilitates selective CA in cases with difficult access.