In-stent restenosis after carotid artery stenting as diagnosed during routine duplex sonography follow-up. Duplex sonography (B-mode) of a carotid artery after stenting showing a narrowing in the middle part of the stent due to a calcified plaque (A). During the routine follow-up investigation after six months there was a typical aliasing phenomenon indicating focal flow acceleration (B). Peak systolic velocity reached up to 520 cm/s (C) with a markedly disturbed poststenotic frequency pattern (D). The reconstructed contrast enhanced computer tomography confirmed the high-grade in-stent restenosis (E). doi:10.1371/journal.pone.0022683.g001 

Contexts in source publication

Context 1

... be endangered by a pathological neointimal proliferation and subsequent in-stent restenosis (ISR). ISR is frequently asymptomatic at diagnosis, but may nevertheless adversely affect the long-term safety and efficacy of CAS because it could necessitate a second intervention which might again be associated with a periprocedural complication. We know from coronary artery stenting that inflammation plays a pivotal role in the pathogenesis of ISR, causing neointimal proliferation through the stent meshes [8–10]. In this scenario, technical factors such as stent dimensions or pre- and postdilation during CAS may result in a vascular burden due to vessel injury. This could cause inflammation and thus contribute to the development of an ISR after CAS due to neointimal proliferation. In vivo, an inflammatory process can easily be monitored with inflammatory serum biomarker for instance C-reactive protein (CRP), which has shown its predictive value in the clinical and imaging outcome of patients undergoing coronary and carotid artery stenting [11–13]. The aim of the current study was to investigate the influence of periprocedural serum inflammatory biomarkers and procedural technical characteristics on the incidence of ISR in patients undergoing CAS during long-term follow-up. The current study is in accordance with the Declaration of Helsinki and ICH/GCP guidelines. The analysis was approved by the Ethics Committee of the University of G ̈ttingen, Germany. 197 patients (215 arteries) with a symptomatic carotid artery stenosis $ 70% or an asymptomatic carotid artery stenosis $ 90% (degree of stenosis was measured according to the European guidelines (ECST) [14]) who underwent carotid artery stenting between May 2003 and June 2010 and were prospectively examined at our institution were consecutively included in our study. Exclusion criteria for this current analysis were unstable neurological conditions, progressive stroke, stenosis caused by dissection or markedly elevated CRP levels or leukocyte counts preinterventionally due to manifest infections such as pneumonia or urinary tract infections. Patients who had experienced a transient or permanent ipsilateral ocular or cerebral ischemic event within the past six months due to carotid artery stenosis were considered symptomatic. The degree of stenosis was determined by carotid duplex ultrasound imaging according to the ECST guidelines and angiographically confirmed during the stenting procedure. All patients received detailed information about the three different treatment strategies (CEA, CAS, and best medical treatment) and their specific advantages and potential complications. With respect to the CAS procedure, all patients were informed about the investigational nature of CAS and gave their written informed consent. An experienced stroke neurologist (K.W., S.S. or K.G.) performed a complete neurological examination before the procedure, immediately after CAS and at every visit during long-term follow-up, and documented all clinical data. The following cerebrovascular risk factors were recorded using history or direct measurements: hypertension (blood pressure $ 140/ 90 mmHg measured on repeated occasions or presence of antihypertensive drugs), hyperlipidemia (fasting serum cholesterol levels $ 200 mg/dl or statin therapy), smoking (current or within the previous year), diabetes mellitus (HbA1c $ 6.5%, fasting blood glucose $ 120 mg/dl, or presence of antidiabetic drugs), coronary artery disease (history of angina, myocardial infarction, percuta- neous transluminal angioplasty or surgery), peripheral occlusive arterial disease (history of typical clinical presentation, percutane- ous transluminal angioplasty or surgery) and the presence of contralateral carotid artery stenosis or occlusion (as assessed with ultrasound). Carotid duplex ultrasound imaging using a combination of direct and indirect criteria as well as the presence and extent of intrastenotic and poststenotic flow yielded the degree of stenosis before CAS and the diagnosis of an ISR. In detail, as direct criteria for the local degree of stenosis, the peak systolic flow velocities (PSV) within and distal of the carotid artery stenosis, the peak enddiastolic flow velocity in the stenosis, the internal carotid artery-to-common carotid artery PSV ratio, and the prestenotic and poststenotic frequency patterns were documented. Whenever possible, the residual vessel lumen in the B image and the colour- coded residual vessel area were determined. As indirect criteria, a reversal flow of the supratrochlear artery and the anterior cerebral artery as an indicator of an inadequate intracerebral crossflow and the pulsatility of the ipsilateral common carotid artery were taken into account. As there are no common criteria available for defining an ISR and current literature suppose different criteria [15–17], we used locally adopted criteria with a PSV $ 300 cm/s as a key feature representing an ISR of $ 70% (Figure 1). All examinations were performed in a standardized form in the same vascular laboratory with the same ultrasound equipment (Acuson Sequoia TM 512, Siemens, San Jos ́, CA) under the supervision of an experienced, board certified vascular neurologist (K.G.). Contrast-enhanced reference imaging (computer tomography angiography and/or conventional angiography) was done at the discretion of the treating physician, in most of the cases if the collateral flow pattern changed, in case of the occurrence of clinical stroke-symptoms attributable to the treated vessel or if a re- intervention was anticipated. Peripheral blood samples for routine measurement of CRP and leukocyte count were taken 24 hours before and after CAS. Furthermore, a serum lipid profile containing high density lipoprotein (HDL), low density lipoprotein (LDL) and total cholesterol as well as triglycerides were tested during hospitalization after overnight fasting. All parameters were determined by standard laboratory methods at the Department of Clinical Chemistry of the University Medical Center of the Georg-August-University G ̈ttingen, Germany. In detail, C-reactive protein was assayed by an automated immuno- precipitation technique with photometric analysis. The reference range of our local laboratory is # 8.0 mg/l, the lower detection limit 0.2 mg/dl. The leukocyte count was also measured sample dependent by automated techniques with one of the following haematology analyzers: Beckman Coulter AcT 5 diff AL (Beck- mann Coulter, Brea, CA, USA), ADVIA H 120 (Bayer Diagnostics, M ̈nchen, Germany) or Abbot Cell-Dyn Sapphire (Diamond Diagnostics, Holliston, MA, USA) with a local reference range of 4000–11000/ m l. A following manual count was carried out if pathological results occurred. Serum triglycerides, total cholesterol and HDL cholesterol were measured in fasted blood samples using an autoanalyzer. LDL cholesterol was calculated using the Friedwald’s formula [18]. In the majority of cases, the intervention was done under anaesthesiology stand-by via a femoral approach by an experienced interventional neuroradiologist. Stent-type and the use of filter-based protection devices were chosen at the discretion of the interventionalist. All patients received orally administered aspirin (100 mg/d) and clopidogrel (75 mg/d) at least 3 days before the procedure. Aspirin was administered indefinitely and clopidogrel was withdrawn 6 weeks after CAS. All patients were routinely monitored in our intensive care or stroke unit overnight and were discharged to normal ward or home thereafter. A complete neurological status of the patient and a duplex sonography was performed before discharge. A neurologist experienced in neurovascular diseases examined each patient and recorded clinical complications at the hospital’s outpatient clinic at 3, 6, and 12 months after the CAS-procedure and every 6 months thereafter. During these postinterventional visits serial duplex sonography was routinely performed according to a standardized protocol. Continuous values were expressed as mean 6 SD and nominal variables as count and percentages. Median values with the corresponding interquartile range (IQR) were computed for non-normally distributed variables. A two-sided T-test was used for comparison of normally distributed variables and the non- parametric Kruskal Wallis test for not normally distributed values. For comparisons of categorical data we used two-tailed Chi-square statistics with Yates’ correction or the Fisher’s exact test when the predicted contingency table cell values were less than five. A two-sided p value of less than 0.05 was considered to indicate a statistically significant difference. Multiple binominal regression analysis was conducted for those variables with a p , 0.1 on univariate level to estimate a potential effect on the development of an ISR (p to enter = 0.05, p to leave = 0.1) using additive and multiple interaction terms. All statistical analyses were performed using SPSS (Version 17, SPSS Inc., Chicago, Ill). We retrospectively screened 215 patients (237 arteries) consecutively undergoing elective carotid artery stenting between May 2003 and June 2010 from our prospectively achieved database. The data of 21 patients (22 arteries) had to be excluded due to missing follow-up data, additionally 5 patients (5 arteries) with clinical signs of pulmonary or urinary tract infections and/or markedly elevated CRP levels or leukocyte counts were excluded. Complete clinical follow-up data were available for the remaining 210 CAS procedures (mean age of 67.9 6 9.7 years, 71.9% male), with a median duration of follow-up of 33.4 months (IQR: 14.9– 53.7 months). A PSV $ 300 cm/s corresponding to an ISR $ 70% was detected in 12/210 (5.7%) arteries after a median of 8.6 months (IQR: 3.4–17.3). In 9 of the 12 ISR a new crossfilling via the anterior communicating artery or supraorbital artery could be identified as additional supporting ...

Context 2

... same ultrasound equipment (Acuson Sequoia TM 512, Siemens, San Jos ́, CA) under the supervision of an experienced, board certified vascular neurologist (K.G.). Contrast-enhanced reference imaging (computer tomography angiography and/or conventional angiography) was done at the discretion of the treating physician, in most of the cases if the collateral flow pattern changed, in case of the occurrence of clinical stroke-symptoms attributable to the treated vessel or if a re- intervention was anticipated. Peripheral blood samples for routine measurement of CRP and leukocyte count were taken 24 hours before and after CAS. Furthermore, a serum lipid profile containing high density lipoprotein (HDL), low density lipoprotein (LDL) and total cholesterol as well as triglycerides were tested during hospitalization after overnight fasting. All parameters were determined by standard laboratory methods at the Department of Clinical Chemistry of the University Medical Center of the Georg-August-University G ̈ttingen, Germany. In detail, C-reactive protein was assayed by an automated immuno- precipitation technique with photometric analysis. The reference range of our local laboratory is # 8.0 mg/l, the lower detection limit 0.2 mg/dl. The leukocyte count was also measured sample dependent by automated techniques with one of the following haematology analyzers: Beckman Coulter AcT 5 diff AL (Beck- mann Coulter, Brea, CA, USA), ADVIA H 120 (Bayer Diagnostics, M ̈nchen, Germany) or Abbot Cell-Dyn Sapphire (Diamond Diagnostics, Holliston, MA, USA) with a local reference range of 4000–11000/ m l. A following manual count was carried out if pathological results occurred. Serum triglycerides, total cholesterol and HDL cholesterol were measured in fasted blood samples using an autoanalyzer. LDL cholesterol was calculated using the Friedwald’s formula [18]. In the majority of cases, the intervention was done under anaesthesiology stand-by via a femoral approach by an experienced interventional neuroradiologist. Stent-type and the use of filter-based protection devices were chosen at the discretion of the interventionalist. All patients received orally administered aspirin (100 mg/d) and clopidogrel (75 mg/d) at least 3 days before the procedure. Aspirin was administered indefinitely and clopidogrel was withdrawn 6 weeks after CAS. All patients were routinely monitored in our intensive care or stroke unit overnight and were discharged to normal ward or home thereafter. A complete neurological status of the patient and a duplex sonography was performed before discharge. A neurologist experienced in neurovascular diseases examined each patient and recorded clinical complications at the hospital’s outpatient clinic at 3, 6, and 12 months after the CAS-procedure and every 6 months thereafter. During these postinterventional visits serial duplex sonography was routinely performed according to a standardized protocol. Continuous values were expressed as mean 6 SD and nominal variables as count and percentages. Median values with the corresponding interquartile range (IQR) were computed for non-normally distributed variables. A two-sided T-test was used for comparison of normally distributed variables and the non- parametric Kruskal Wallis test for not normally distributed values. For comparisons of categorical data we used two-tailed Chi-square statistics with Yates’ correction or the Fisher’s exact test when the predicted contingency table cell values were less than five. A two-sided p value of less than 0.05 was considered to indicate a statistically significant difference. Multiple binominal regression analysis was conducted for those variables with a p , 0.1 on univariate level to estimate a potential effect on the development of an ISR (p to enter = 0.05, p to leave = 0.1) using additive and multiple interaction terms. All statistical analyses were performed using SPSS (Version 17, SPSS Inc., Chicago, Ill). We retrospectively screened 215 patients (237 arteries) consecutively undergoing elective carotid artery stenting between May 2003 and June 2010 from our prospectively achieved database. The data of 21 patients (22 arteries) had to be excluded due to missing follow-up data, additionally 5 patients (5 arteries) with clinical signs of pulmonary or urinary tract infections and/or markedly elevated CRP levels or leukocyte counts were excluded. Complete clinical follow-up data were available for the remaining 210 CAS procedures (mean age of 67.9 6 9.7 years, 71.9% male), with a median duration of follow-up of 33.4 months (IQR: 14.9– 53.7 months). A PSV $ 300 cm/s corresponding to an ISR $ 70% was detected in 12/210 (5.7%) arteries after a median of 8.6 months (IQR: 3.4–17.3). In 9 of the 12 ISR a new crossfilling via the anterior communicating artery or supraorbital artery could be identified as additional supporting sonographic criteria for a high grade restenosis. A contrast-enhanced reference imaging confirmed the ISR (see Figure 1) in nine cases (seven by conventional angiography, two by CT angiography), failed in one case (CT angiography) and was not done in two cases. The detailed clinical baseline patient characteristics are given in table 1. Preinterventional CRP (CRP pre ) values were not normally distributed and had a median value of 2.1 mg/dl (IQR: 0.0– 6.8 mg/dl). According to our local laboratory reference values pathological CRP pre values of . 8.0 mg/dl occurred in 37 patients (21.1%). There was a statistically significantly higher median CRP value of 10.4 (IQR: 1.5–23.7 mg/dl) in the group of patients with an ISR in comparison to the group of patients without an ISR (2.0 mg/dl, IQR: 0.0–6.6 mg/dl) during follow up (p = 0.022, see table 2). Postinterventional CRP (CRP post ) did not differ between the groups (p = 0.314) on univariate level. Preinterventional leukocyte counts (Leuko pre ) were normally distributed and reached 7626/ m l ( 6 2040/ m l) in the group without and 8300/ m l ( 6 3013/ m l) with ISR (p = 0.283). The postinterventional leukocyte count (Leuko post ) showed a significant increase in patients with the occurrence of an ISR during follow-up in the univariate analysis (8526/ m l vs. 10433/ m l; p = 0.035). The lipid profile did not differ significantly between both groups. In most of the cases (79%) a closed-cell design Carotid WALLSTENT (Boston Scientific, Natick, Massachusetts) was used. The stent cell design did not differ between the groups with and without ISR during follow-up (see Table 2, p = 0.695). There was yet a significantly narrower stent width in the group with the occurrence of an ISR (mean: p = 0.019), as well as a trend towards ISR formation during follow-up in those patients with a longer stent length (p = 0.068). There was no significant association between the use of multiple stents, pre- and postdilation and ISR on the univariate level (Table 2). After applying binominal multivariate analysis with variables which were imbalanced on univariate analysis (CRP pre , Leuko post , stent length and width) and correcting for possible influencing variables such as age, gender and symptomatic status of the carotid artery (symptomatic vs. asymptomatic), the variables Leuko post , stent length and stent width remained significant for predicting the occurrence of an ISR (Odds ratio (OR) and 95% confidence interval (CI): OR 1.31, 95% CI 1.02–1.69, p = 0.036; OR 1.25, 95% CI 1.05–1.65, p = 0.022 and OR 0.28, 95% CI 0.09–0.84, p = 0.01, respectively, see table 3). The aim of our study was to analyze the effect of periinterventional serum inflammation markers and procedure related technical factors during CAS on the development of an in-stent restenosis during long-term follow-up. Our results suggest that post-procedurally elevated leukocyte count as an inflammatory serum marker is an independent predictor for subsequent ISR. Moreover, the occurrence of ISR is markedly regulated by stent dimensions, i.e. the thinner and longer the deployed stent the more frequently a restenosis could be detected. The pivotal role of vascular inflammation in the development of ISR has been well documented in the context of coronary artery stenting [11]. ISR is mostly triggered by an endothelial disruption and abrasion which is caused by balloon inflation and stent placement. This vascular injury initiates the release of several mediators leading to adhesion of thrombocytes, neutrophiles and monocytes. These cells, for their part, release vasoactive, thrombogenic, lymphocytic, and mitogen substances, which lead to vasoconstriction, vascular remodelling, neointimal proliferation, thrombosis and inflammation finally resulting in ISR [8,19]. Assuming a similar pathophysiology of ISR within the carotid arteries, the periinterventional monitoring of such inflammatory markers may be useful for the prediction of ISR. The rate of ISR $ 70% within our study population was 5.7% and occurred early after a median duration of 8.6 months follow-up, suggesting more an intimal hyperplasia than an atherosclerotic burden as the main cause of stenosis. To identify patients with a higher risk for ISR is of high relevance for clinical routine practice because a tight follow-up in these patients is warranted. The inflammatory biomarker CRP has proven its predictive value of clinical outcome and ISR in patients undergoing coronary and carotid artery stenting [11,13,20]. Within our patient cohort, the predictive value of preinterventionally elevated CRP could only be shown on a univariate level whereas it did not remain significant after applying multivariate regression analysis due to interaction effects. Moreover, our data could not support an association between CRP post and the formation of ISR as has been demonstrated previously [21]. This may be attributable to the plasmakinetics of CRP showing a delayed increase after the triggering event [22]. In fact, the most predictive increase of postprocedural CRP for ISR could be ...