Response to Letters by Settaci et al, Cremonesi et al, and Gröschel et al
We appreciate the comments on our publication “Does carotid stent cell design matter” and would like to take the opportunity to address some of the critical points.
Regarding the comments of Drs Cremonesi et al:
First, we disagree that a reanalysis of events according to the timing of their occurrence is of any value. Microembolism occurring during the procedure under cerebral protection may cause delayed symptoms after removal of the sheath and still has to be attributed to the procedure itself rather than to postimplantation stent properties. Continuous monitoring for microembolic signals would be the only way to differentiate intraprocedure versus postprocedure events. Furthermore, the vast majority of events at day 0 occurred during the procedure with an even higher incidence in the group of patients who received a closed-cell design stent. The numbers would therefore hardly change if such a reanalysis is done. Finally, the timing of stroke seems mostly irrelevant from the patients’ perspective, and closed-cell design stents had a nonsignificant 2.0% higher incidence of overall events and a significant 2.6% higher incidence of events at day 0, suggesting that these devices did not add to the safety of the procedure.
Second, the purpose of any retrospective analysis is reporting personal experience and hypothesis generation. We did not claim that the study was adequately powered to demonstrate or exclude a difference between the stent designs. However, the authors have to acknowledge that even combining the findings of the formerly reported BIC registry and our study, there is currently no scientific evidence to support their hypothesis.
Third, in light of the limitations of the retrospective study design and length restrictions for the publication, we believed that the information on the propensity score modeling strategy would not be essential for the reader. The Hosmer-Lemeshow test was not statistically significant indicating an acceptable model fit (c=9.2, df=8, P=0.33), and the area under the receiver-operating characteristic curve for the final model was 0.70.
In summary, we feel that the formerly reported benefit of closed-cell stent design over the open-cell design lacks even more good evidence and supports the call for properly designed randomized studies.
Regarding the comments of Drs Setacci et al:
First, the number of late events was small and not statistically significantly different in our series; these findings therefore do not support the conclusion that closed-cell design stent is superior to open-cell stents in the late postprocedure phase. This is why statistics rather than “wishful thinking” are used to assess data. The intraprocedure event rate was even higher (statistically significant) with closed-cell design stents. This of course may be attributed to other factors than the stent design, such as delivery system or patient selection, but still questions the superiority of these devices.
Second, as stated above we disagree that a reanalysis of subacute events is helpful to clarify the situation.
Third, as outlined in the Discussion section of the original publication, we agree that the underlying hypothesis of the BIC registry on stent-scaffolding properties is intriguing and the ongoing discussion hopefully will trigger development of improved devices. Nevertheless, currently available data do not provide scientific evidence to support this hypothesis.
In summary, we are not saying the same thing. The take-home message from our study was that in 10 European centers no significant differences of neurological outcome according to currently available stent designs in symptomatic and asymptomatic patients were observed.
Regarding the comments of Drs Gröschel et al:
First, we agree that length of the stent might have an important impact on the occurrence of microembolic complications during and after CAS. Stroke rates in patients receiving stents of 20, 30, 40 and 50 mm in length were 0%, 2.9%, 2.7% and 2.2%, respectively (P=0.41); rates of all neurological adverse events including transient ischemic attack, stroke and death were 0%, 5.2%, 5.3% and 7.7%, respectively (P=0.34). However, it has to be acknowledged that the length of the stent is mainly determined by the length of the stenosis and thus the plaque volume. In this context, length of the stent may be just a surrogate marker for the severity of the disease (eg, length of the stenosis and plaque volume) rather than an independent risk factor for complications.