Background and purpose In-stent restenosis (ISR) is a potentially preventable cause of stroke in carotid artery stent (CAS) patients. Understanding the frequency and timing of ISR would be useful in developing optimal protocols for carotid stent surveillance. The time course and frequency of moderate and severe ISR in our single institution prospective registry of CAS procedures is reported here.
Methods Data were collected prospectively from 296 consecutive elective CAS procedures. Doppler surveillance was performed at 1, 6 and 12 months and annually thereafter in some cases. Moderate ISR (>50%) was defined as a peak systolic velocity (PSV) > 200 cm/s. Severe ISR (>70%) was defined as PSV > 200 cm/s and end diastolic velocity > 125 cm/s or internal carotid artery/common carotid artery ratio >4. Patients with severe ISR underwent digital subtraction angiography for confirmation and possible retreatment.
Results Clinical follow-up at 1 year was 98%. Clinical and ultrasound follow-up at 1 month was 100%, at 6 months 96% and at 1 year 91%. The incidence of all ISR at 6 months was 8%; 5% moderate and 3% severe. Two patients had asymptomatic occlusions at 6 months (0.8%). Patients with moderate ISR at 6 months did not progress to severe ISR. There were two strokes caused by stent thrombosis, one acute and one delayed.
Conclusions Doppler surveillance is important for detecting ISR after CAS procedures. Severe ISR (>70%) should be retreated in most cases as ISR may progress to occlusion. Asymptomatic patients with moderate ISR (50–69%) at 6 months can be safely followed conservatively. Progression of ISR after 6–12 months is uncommon over a 2–3 year period.
Statistics from Altmetric.com
Carotid artery stenting (CAS) has been approved by the Centers for Medicare Services as an alternative treatment for severe symptomatic carotid artery stenosis in patients with medical comorbidities or anatomic factors that increase the risk of perioperative complications of carotid endarterectomy. Utilization of CAS may expand, pending the results of pivotal randomized trials such as CREST and ACT I. While the scope of future applications for CAS is uncertain, there is no argument that CAS will continue to have a role in many patients, and that physicians need to understand the incidence of in-stent restenosis (ISR), and the best management and treatment strategies for patients with moderate and severe ISR.
Two hundred and ninety-six consecutive elective carotid artery stenting procedures were performed by the author between December 2002 and December 2008 (procedures performed for acute carotid artery occlusion during emergent stroke intervention were excluded). There was one technical failure (inability to wire the lesion). That patient had an endarterectomy and is not included in the follow-up. Hence there were 295 elective carotid stents for analysis. Forty per cent (118) of the stents were placed in women. Forty-two per cent (124) were symptomatic.
The follow-up protocol included carotid ultrasound at 1, 6 and 12 months, and then annually in some cases. The vast majority of carotid ultrasound examinations were performed at one facility (Maplewood Vascular Clinic), which was an American College of Radiology accredited facility from 2002 to 2006 and Intersocietal Commission for Accreditation of Vascular Laboratories accredited from 2007 to 2009. In rare cases, follow-up was obtained from other laboratories when the patient had moved or was unable to travel to our site for follow-up. Moderate ISR (>50%) was defined as a peak systolic velocity (PSV) > 200 s. Severe ISR (>70%) was defined as a PSV > 200 and end diastolic velocity > 125 or internal carotid artery/common carotid artery (ICA/CCA) ratio >4. These criteria, utilized empirically since 2002, are very similar to those recently reported by Setacci et al in Stroke based on a prospective study of 814 carotid stents.1 They suggested a PSV > 175 for >50% stenosis and an end diastolic velocity > 140 or ICA/CCA ratio >3.8 for >70% stenosis. Others have reported very similar criteria.2 3 4
Several different carotid stents were used. Post-dilation was performed in the majority of cases with a 4.5 or 5 mm balloon, slightly undersizing relative to the cervical internal carotid. All patients were pretreated with clopidogrel for at least 5 days and with aspirin 325 mg for at least 2 days, or alternatively 450–600 mg of clopidogrel and 325 mg of aspirin at least 4 h prior to the procedure. Patients intolerant of clopidogrel received ticlopidine 250 mg twice daily for 5 days. Patients were maintained on aspirin and clopidogrel or ticlopidine for at least a month post-procedure, and aspirin either 325 or 81 mg daily thereafter unless clopidogrel was continued for other reasons. Assays for platelet inhibition were not performed. Use of cholesterol lowering medications, including statins, was left to the discretion of the patient's internist.
Patients who met the criteria for moderate restenosis, but not severe restenosis, at 6 months were followed, with clinical evaluations and Doppler at 6 months and then annually. Six patients who developed severe restenosis by ultrasound were taken for angiography and repeat intervention. Severe restenosis (>70%) was confirmed in five of six patients who were retreated with cutting balloon angioplasty. Results and follow-up of those procedures are reported separately.
All 295 patients had 1 month clinical and ultrasound follow-up. Two patients were lost to follow-up between the 1 month and 6 month follow-up visits. Seven patients died between the 1 and 6 month follow-up visits due to pneumonia in two cases, malignancy in three cases (known pre-existing lung cancer in one symptomatic patient, and two new diagnoses of pancreatic cancer and lymphoma), and undocumented cause in two cases, both of whom were octogenarians. One patient ceased follow-up because of general debilitation requiring full time nursing home care from the stroke which led to the index procedure. Thirty-three patients were not eligible for the 6 month follow-up at the time of writing. Hence 252 of 262 patients (96%) eligible for the 6 month follow-up completed the 6 month clinical and ultrasound follow-up. An additional 10 patients who completed the 6 month follow-up and were eligible for the 1 year follow-up did not complete the 1 year follow-up. Two were lost to follow-up. Six patients died of malignancy, which was known before the index stent procedure but thought to be in remission in two symptomatic patients (laryngeal cancer, lung cancer) and was previously unknown in four patients (acute myeloid leukemia, pancreatic carcinoma, bladder carcinoma, lung carcinoma). One patient died of a myocardial infarction and one died of an unknown cause. Sixty-three patients were not yet eligible for the 1 year follow-up at the time of writing. Hence 212/232 (91%) patients eligible for the 1 year follow-up completed a 1 year ultrasound and clinical follow-up. Duration of follow-up for the entire 295 patient cohort is presented in table 1. Overall, only four patients were lost to follow-up at 1 year, 15 had died (6.4%) and one was in full time nursing care (overall follow-up rate 98%).
All 295 stents were patent at 1 month. One patient developed acute thrombus in the stent on post-procedure day 1. One month earlier, this 59 year-old man had suffered a myocardial infarction and minor left hemispheric stroke due to a >80% left carotid stenosis which was stented uneventfully. He was readmitted for elective treatment of an asymptomatic right carotid stenosis 1 month later. The morning after the right carotid stent procedure, the patient developed a partially occlusive thrombus within the right carotid stent (fig 1) which embolized to the middle cerebral artery, requiring emergent intra-arterial thrombolysis. Assay for clopidogrel activity was not available at that time. At 30 days, the National Institutes of Health Stroke Scale (NIHSS) score was 4 and the modified Rankin scale (MRS) was 3 due to severe neglect. At 6 months, the NIHSS was 1 and the MRS was 1, and both stents were patent with no stenosis. One additional 64-year-old man, who underwent stenting for asymptomatic nearly occlusive left common carotid stenosis in the setting of previous neck dissection, radiation and a contralateral ICA occlusion, had acute stent thrombosis and a left hemispheric stroke at 3 months, 1 month after stopping clopidogrel (fig 2). He was treated with intravenous tissue plasminogen activator on arrival in the emergency department and improved. Angiography demonstrated acute partially obstructive thrombus throughout the stent. No intervention was attempted. Repeat angiography at 1 week showed complete resolution of the thrombus (fig 2). The NIHSS score was 4 and the MRS was 2 at 30 days after the stroke. The stent was patent without ISR at 1 year, with an NIHSS score of 2 and MRS of 1. Two additional patients developed asymptomatic stent occlusion. One of these was occluded at the 6 month visit. The other patient had severe ISR, and progressed to complete occlusion by the time of angiography 1 week later. Hence the 6 month primary patency was 248/252 (98%) and the secondary patency was 250/252 (99%). There have been no additional occlusions after the 6 month follow-up.
Of the 231 stents that did not have restenosis at 6 months, 128 had follow-up of at least 1 year, 52 of at least 2 years and 22 of 3 or more years. None subsequently developed restenosis.
Moderate or severe restenosis (defined by a PSV > 200 or occlusion in two patients) developed in 21/252 (8%) stents at 6 months (table 2). Nine were women (42%). Seven of these patients (five with confirmed restenosis at angiography and the two with occlusion) developed severe restenosis for a rate of 3% (fig 3). None of these patients were symptomatic.
Of the 14 patients with moderate restenosis, two have not yet reached the 1 year follow-up and 12 have had additional follow-up of 1 year in six patients, 2 years in three patients or at least 3 years in three patients. Importantly, no patient in this group has developed a severe restenosis or occlusion. Hence at least in the 12 patients and time period observed here, progression of moderate restenosis after 6 months did not occur.
Acute thrombosis of electively placed carotid stents, utilizing dual antiplatelet therapy, is a rare event, occurring in only 1/295 cases in this series. Suboptimal response to clopidogrel, as detected with point of care testing with the Verify Now assay (Accumetrics, San Diego, California, USA), may be present in as many as 40% of neurointerventional patients.5 While a subtherapeutic assay may indicate a higher risk of acute carotid stent thrombosis, the risk must be relatively low. The effect of clopidogrel resistance on ISR of carotid stents is not known.
Delayed acute thrombosis in the absence of ISR is also rare, occurring at 3 months in one patient with a history of neck irradiation. It is plausible that the radiation inhibited the neointimal response. Eskandari et al reported 26 cases of CAS in patients with a history of radiation, and there was just one case of ISR and no cases of acute thrombosis.6 Dual antiplatelet therapy was used for just 30 days. The risk of acute thrombosis in this group would thus seem to be low. Nonetheless, a longer period of dual antiplatelet therapy in the previously irradiated patient, of the order of 6–12 months, seems reasonable. The optimal length of dual antiplatelet therapy in carotid stenting in general has not been defined. The use of high dose atorvastatin (80 mg/day) has been shown to reduce the need for revascularization procedures (coronary, carotid or peripheral) in patients who have had a previous stroke7 but the effect on carotid in-stent restenosis has not been investigated.
The 8% overall incidence of restenosis and the 3% incidence of severe restenosis is similar to previous reports. Undersizing the post-dilatation balloon did not result in a higher incidence of ISR in this series compared with other published series.1 6 8 9 10 11 12 13 14 15 Female sex has been suggested as a risk factor for restenosis but the proportion of female patients in the ISR group was 43%, nearly equal to their 40% representation in the population studied. Initial symptomatic status also did not correlate with ISR. ISR occurred with all stents used (table 2). While there was a statistically non-significant (p = 0.15) slightly higher incidence of moderate restenosis observed with the Acculink stent compared with the other three stents, this was not reflected in a higher incidence of severe restenosis (table 2). In a multi-institutional retrospective review of over 3000 carotid stent cases, De Donato et al reported that stent design did not correlate with restenosis or subsequent neurological symptoms.8
Delayed complete occlusion of carotid stents (after a normal 1 month ultrasound) occurred in 2/252 patients (0.75%). Severe ISR should be retreated as it may progress to occlusion as in the two cases reported here, and may also present with transient ischemic attack or stroke.8 15
Some investigators have reported late (after 1 year) restenosis.8 15 While some patients in the present series may yet develop restenosis, no patient has yet developed a new restenosis beyond the initial 6 month period. Hence the 6 month restenosis rate is a good indicator of the durability of the procedure, at least in the first few years. It is also noteworthy that the 12 patients with a moderate restenosis by ultrasound at 6 months who had additional follow-up have not progressed or become symptomatic. Based on these data, one can recommend clinical follow-up and ultrasound surveillance for this group.
1 month, 6 month and 1 year clinical and ultrasound follow-up has been the standard for monitoring patients with CAS. As the restenosis process occurs mostly in the first 6 months and does not seem to progress rapidly thereafter, closer monitoring of this period might be beneficial. In the series presented here, there were two asymptomatic occlusions at 6 months, which possibly could have been prevented with earlier detection of severe ISR. A follow-up protocol of 1 month, 3–4 months, 6 months and 1 year might be more likely to detect severe ISR before occlusion or ischemic symptoms occur. For cost and patient convenience purposes, the 1 month ultrasound could probably be omitted as it added little value to the management of patients in this series.
Competing interests Local PI CREST, ACT I, SAPPHIRE WW, CHOICE, CABANA.
Ethics approval Ethics approval was obtained.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.