Carotid artery stenting in a single center, single operator, single type of device and 15 years of follow-up

Cerebrovascular disease is an increasing global health problem, responsible for 10% of deaths, worldwide (Dorn et al., 2012). Carotid artery stenosis due to atherosclerotic disease is liable for approximately 20–30% of these strokes (Chaturvedi et al., 2005; Roger et al., 2011), with this stroke subtype associated with the highest rate of recurrence (Coutts et al., 2008).

The American Heart Association/American Stroke Association (AHA/ASA) (Kernan et al., 2014) considers revascularization with stent to be a therapeutic alternative to surgery to prevent stroke, in select patients with symptomatic carotid artery stenosis, although in many centers surgery is still considered as the gold standard. While this recommendation is supported by multiple large randomized clinical trials, including CREST (Bonati et al., 2015; Brott et al., 2016), we currently lack long-term outcome data for CAS procedures. Further, critics of CAS cite that this approach fails to remove the plaque, which can result in restenosis and potentially new stroke events. With an aging population and an increasing life expectancy, long-term outcomes following CAS are now needed to guide treatment appropriately.

The objective of this study was to report the efficacy and durability of CAS in terms of stroke prevention. The primary outcome of the trial was stroke, myocardial infarction (MI), or death during a periprocedural period (30 days after treatment), or any stroke event over a 15-year follow-up. The secondary aim was to identify risk factors for 30-day complications, long-term neurological complications, and intra-stent restenosis (ISR).

This paper has been approved by the Ethics Committee in our Hospital.

The carotid stent deployed was the Acculink with no procedures aborted before insertion of the stent. Protection devices were used for 60.2% of patients. Post-stent balloon dilatation was performed in 73.8% of patients, with predilatation necessary for 20/344 arteries (5.8%).

Technical success was achieved in 335 of the 344 arteries (97.4%) treated. Although it was possible to pass a wire through the stenosis in 9 patients, complete stent expansion was not achieved for this group.

For long-term analysis, patients were followed for a median of 50 months (range 0-155 months). A total of 50 patients were lost to follow-up (Table 4), with long-term follow-up achieved in 294 patients (85.5%). Fifty-six (16.3%) patients died within this period, including 4 stroke-related deaths, and 52 non-lesion related deaths due to comorbidity. Two patients died from intraparenchymal hemorrhage and two from ischemic stroke, one ipsilateral to the treated carotid artery and the other, contralateral to the treated carotid artery. Among the non-neurological deaths, there were 5 cases of vascular death: 3 cases due to chronic ischemia of the lower extremities, and 2 due to acute myocardial infarction (Fig 1).

Follow-up (months)	n	%	Average	SD	Median	(IQR)
Keep in follow-up	238	69,19%	56,40	36,30	48,90	(59,50)
Loss to follow-up	50	14,53%	34,08	29,28	30,04	(41,50)
Neurological death	4	1,16%	10,36	10,69	8,14	(19,25)
No neurological death	52	15,12%	38,33	31,28	35,53	(44,50)

During the follow-up period, 8 strokes and 3 TIAs were diagnosed (3.2%). The majority of the strokes (75%) were contralateral to the treated carotid artery and most were ischemic. We did not identify any predictive risk factors or any procedural characteristics for ipsilateral neurological complication.

Stroke is a major cause of mortality and morbidity in industrialized countries. These data, combined with an ever-increasing life expectancy, necessitates that we conduct more detailed analyses of whether carotid revascularization is a safe and effective treatment with which to prevent stroke.

In our series, the 30-day outcomes after CAS show that carotid stenting is an effective treatment in preventing future vascular events, with a low incidence of periprocedural complications: 2.3% for disabling stroke, death, or MI. Previous systematic reviews of nonrandomized cases series (Kastrup et al., 2003), as well as several studies (Castriota et al., 2002; Cremonesi et al., 2003; Garg et al., 2009; Giri et al., 2015), have shown that EPD-use appears to reduce the incidence of new ischemic lesions. However, various reports have also criticized the use of EPDs in CAS as these devices must pass through the arterial stenosis, which might itself provoke complications (Reimers et al., 2004; Wu et al., 2011) and a greater incidence of microemboli (El-Koussy et al., 2007). Nevertheless, the results of our series show a decreased rate of major complications when EPDs are used (p = 0.005), which therefore classifies these devices as protective against stroke.

To our knowledge, our study comprises the largest patient cohort, with the longest national follow-up to compare stroke prevention in patients treated by the same surgical team. In our series, stenting was performed electively as an alternative to endarterectomy. This is because the committee composed by neurologist, radiologist, and interventional radiologist (vascular surgeons refused to participate) decided, more than 15 years ago to send patients directly to CAS, depending on the poor results of the surgery department. Thus, our indications were not limited to patients at a high surgical risk. Nevertheless, the risk of periprocedural complication was low, and similar to data from surgical registries with more stringent inclusion criteria (North American Symptomatic Carotid Endoarterectomy Trial Collaboration, 1991; Executive Committee for the Asymptomatic Carotid Atherosclerosis Study, 1995).

The CREST study showed no significant differences between stent and endarterectomy in terms of the risk of stroke or death over its 10-year follow-up. Indeed, the higher risk of stroke reported for CAS can be attributed to the periprocedural differences between the two groups (Brott et al., 2016; Brott et al., 2010), and the infrequent use of EPDs (El-Koussy et al., 2007). Other randomized trials comparing both techniques have been reported, and showed no significant differences after long-term follow-up (Bonati et al., 2015; Gurm et al., 2008; Mas et al., 2008).

Our data appears to indicate that CAS is preventive against ipsilateral stroke with a low long-term risk of severe stroke The recurrence rate in our series of 3.2% primarily involved strokes contralateral (75%) to the treated carotid artery, and most were ischemic. This finding has also been reported in previous related studies in which no significant difference in cumulative rates of fatal or disabling stroke were found for stenting vs. endarterectomy (Bonati et al., 2015; Mas et al., 2008). It is interesting to note that our observed stroke rate of 3.2% is (fairly) consistent with, and even lower than the 8.5% reported in the ECST at 3 years (European Carotid Surgery Trial, 1998), and the 13% rate of the NASCET (North American Symptomatic Carotid Endoarterectomy Trial Collaboration, 1991). However, our patient population was closely monitored in terms of risk factors, which may have lowered our rate of complication. Regarding risk factors, Donato et al. (de Donato et al., 2008) and Brooks et al. (Brooks et al., 2014) found that being symptomatic at the time of enrollment and intervention was a good predictor of early and late neurological complication. We failed to find any risk factor associated with late neurological complications, although this may be related to our low recurrence rate.

Prevention of ipsilateral stroke is the ultimate goal of any treatment for carotid stenosis. However, durability, defined as patency determined by Doppler ultrasound, may also serve as a useful definition of therapeutic utility. Our analysis of long-term outcomes showed that the long-term rate of restenosis after stenting was low (4.4%), and comparable, or even lower, than those achieved with CEA. However, direct comparisons are complicated by the criteria used to make a diagnosis in Doppler sonography. Of the restenosis cases, only 2 (0.6%) were symptomatic. We found no association between restenosis and recurrent events, but pre-stent angioplasty was significantly associated with the incidence of ISR. This is difficult to interpret given the small number of recurrent events, pre-stent angioplasty, and restenoses seen in our study. A certain amount of controversy surrounds the issue of long-term rates of restenosis after stenting vs. endarterectomy. Some trials (Bonati et al., 2015; Brott et al., 2016; Mas et al., 2008) have demonstrated no differences when comparing both techniques. For example, 5-year risks of 10.8% vs. 8.6%, 3-year risks of 3.3% vs. 2.8%, and 2-year risks of 6.0% vs. 6.3%, for stenting vs. endarterectomy, have been reported, respectively. In contrast, the CAVATAS (Bonati et al., 2009) and SPACE trials (Eckstein et al., 2008) reported higher rates of restenosis 2 years after treatment in the stenting group vs. endarterectomy group (10.7% vs. 4.6%). Consequently, further studies are needed to compare the rate of restenosis and to investigate the association between restenosis and recurrent stroke.

The current study has several limitations. The main limitation was its small sample size and retrospective design. Moreover, results obtained in this study were compared with the literature, as it was not possible to undertake a randomized comparison of CAS versus CEA. Further, the efficacy of the EPDs could not be evaluated throughout the entirety of this study as these devices were not available until 2005.

The long-term follow-up results of our study validate CAS as a safe and durable procedure with which to prevent ipsilateral stroke. Our data also suggest that CAS is effective in terms of long-term functional outcome and risk of fatal or disabling stroke. Post CAS restenosis was infrequent, and, in the majority of cases, asymptomatic.