In this randomised prospective study, the main finding was the increased SCI rate demonstrated by new NSE elevations and diffusion MR imaging in patients undergoing routine pre-dilation before valve implantation when compared to the direct TAVI method. Besides other factors including presence of DM, calcification at arcus aorta, cusp calcification volume effected the development of SCI in patients undergoing TAVI with a self-expandable valve. Another important finding was the increased risk of SCI in case of failure at the first try of self-expandable valve implantation and prosthetic valve retrieval and re-implantation.
In this study, we used both a bio-marker, NSE and diffusion MR imaging in eligible patients to show SCI development. New NSE elevations were significantly correlated with diffusion MR imaging results (r = 0.908,p < 0.001) in eligible patients. Due to this significant correlation, we used only new NSE elevations (studied in whole population) as the marker of SCI in our study. Besides, NSE was also studied as a good marker for SCI development in previous studies in several patient populations (18–22).
SCIs are common after many cardiac interventions including coronary angiography, percutaneous coronary interventions, pulmonary vein isolation and CABG operation (23–25). TAVI is the leading procedure among these with an incidence of > 70% in previous reports (11, 26–32). Herein, we have encountered with a rate of 70.5% in the whole population and 62% in the pre-dilation group. This decrease in SCI rates in our study population was thought to be due to the presence of a direct TAVI group with lower SCI rates.
Direct TAVI implantation without pre-dilation has become popular in the recent years. It has some advantages due to by-passing the risk of several complications which might be a result of balloon valvuloplasty including aortic regurgitation, annular rupture and hemodynamic instability due to rapid pacing. Two randomised clinical studies also demonstrated the safety and efficacy of direct TAVI method (5, 6). However, balloon pre-dilation may be necessary in some cases. In this study, we used pre-dilation in the direct TAVI group in two patients. In one of these patients, the valve could not be improved into the LV and in the other one, AL2 catheter improvement was impossible, instead a 6F JR4 catheter was used to get into the LV. These two patients who were included in the direct TAVI group had elevated NSE levels and SCI was also detected with diffusion MR imaging. As a result, despite this cross-over, the results were not expected to change significantly and were similar with the main results of the study.
The decrease of SCI with avoidance of pre-dilation might be explained by the fact that less manipulation with catheters decreases the embolization risk. Besides, balloon valvotomy might result in more separation of the calcium debris from the native valve. In the literature, no randomised study is found comparing SCI risk with vs. without pre-dilation, but some studies show that during pre-dilation, the risk of SCI could increase due to fragmentation of calcific debris present on a heavily calcified degenerative valve (33) as demonstrated by transcranial Doppler signals (34, 35) but the same Doppler studies also showed that positioning of the new valve and fast pacing during deployment were associated with the highest embolic load. Besides, fast pacing during pre-dilation might have caused transient ischemia. Some controversies also exist in the literature so that avoidance of pre-dilation increased cerebral emboli risk in a study by Bijuklic et al(36). This might have been speculated to be due to difficulty while crossing the native valve and therefore additional manipulations which might have increased embolization risk (36, 37). However, in this study, we have found for the first time in a randomised study that avoidance of pre-dilation decreases the likelihood of SCI development. This result may be due to the new generation Evolut-R valve use which has a better profile and easier passage through the calcified valve which decreases the additional manipulation need. Besides, in this study we have also found that SCI rates were higher in case of failure at the first try of implantation and retrieval and reimplantation. This finding also supports the theory that additional manipulations during TAVI increase SCI rate.
Other factors were also found to be related with the development of SCI in patients undergoing TAVI. DM was an important but a non-surprising factor, because diabetic patients have a lower threshold for development of an ischemic cerebrovascular event (38). Calcification at arcus aorta was also related with new SCI development. As expected, passage of the guidewires, catheters and the bulky valve from the atherosclerotic arcus aorta should have increased the likelihood of plaque embolization. Total calcium debris detected with cusp calcification volume was also an important risk factor for SCI development in this patient group. This has been also proved in a previous study by Bron et al (32).
Similar to the literature, we have also found that direct TAVI is an effective method without increasing the complication rates. The post-dilation rate was also higher in the direct TAVI group when compared to the pre-dilation group similar to the literature. However, post-dilation did not affect SCI risk in these patients. New pacemaker rate was low in our study group when compared to the similar studies. This was thought to be due to high implantation with cusp overlap technique.
There were several limitations of the current study. First of all, we could not get an MR imaging in the whole population because of some contraindications of MRI including presence of a pacemaker or any other prosthesis. However, NSE was significantly correlated with the MRI results and in all patients, NSE was measured to find out new SCIs. Besides, in this study, we did not use embolic protection devices during the TAVI procedure, however these devices have not been proven clinically effective till now.