|
|
||||||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Vascular and Interventional Radiology |
1 From the Departments of Angiology (M.S., W.M., M.H., R.A., E.M.) and Laboratory Medicine (M.E., H.R., O.W.), University of Vienna Medical School, Währinger Gürtel 18-20/6J, 1090 Vienna, Austria. Received July 23, 2001; revision requested September 12; revision received October 12; accepted December 10. Address correspondence to M.S. (e-mail: martin.schillinger@akh-wien .ac.at).
| ABSTRACT |
|---|
|
|
|---|
MATERIALS AND METHODS: This prospective cohort study included 274 consecutive patients who underwent stent implantation in the femoropopliteal (n = 95), iliac (n = 70), and carotid (n = 109) arteries. C-reactive protein (CRP), serum amyloid A (SAA), and fibrinogen levels were measured at baseline and at 48 hours after intervention. Polynomial logistic regression analysis was applied to assess the independent association of the course of acute-phase reactants and the site of stent implantation.
RESULTS: Stent implantation in the femoropopliteal artery was associated with a higher postintervention increase in CRP (P = .01), SAA (P = .04), and fibrinogen (P = .01) values compared with values with iliac artery stent implantation, with adjustment for age, sex, fluoroscopy duration, contrast agent dose, complication occurrence, stenosis grade, total vessel occlusion, and stent cumulative length. No significant difference in the postintervention course of CRP (P = .9) and SAA (P = .1) levels was determined for stents implanted in the carotid artery compared with those implanted in the iliac artery; however, a higher increase in fibrinogen levels (P = .04) was noted.
CONCLUSION: Stent implantation in the muscular femoropopliteal artery was associated with a more extensive vascular inflammatory response than was stent implantation in the elastic iliac and carotid arteries, independent of lesion morphology and interventional factors.
© RSNA, 2002
Index terms: Arteries, stenosis or obstruction, 172.721, 9*.7212 Stents and prostheses, 172.2079, 172.458, 172.629, 92.1268, 92.24, 98.1268, 98.24
| INTRODUCTION |
|---|
|
|
|---|
The severity of arterial injury during stent implantation correlates with increased inflammation (4). This vascular inflammatory process plays an important role in vascular smooth muscle cell proliferation and late neointimal growth (5,6). Vascular smooth muscle cell proliferation and subsequent hypertrophic neointima formation at the treated artery lead to in-stent restenosis.
Patency rates after percutaneous transluminal angioplasty (PTA) and stent implantation widely depend on the location of the treated lesion. Endovascular treatment of large elastic arteries, such as the internal carotid artery and the iliac arteries, is associated with a relatively low rate of recurrence (710). In contrast, in muscular arteries, such as the femoropopliteal artery, restenosis after PTA occurs in up to 50% of the patients within the 1st year (7,1117). Hemodynamic and wall characteristics at each vessel segment are believed to influence the susceptibility for restenosis. However, the extent of the inflammatory response might also contribute to the specific pattern of in-stent restenosis in femoropopliteal, iliac, and carotid arteries. Recently, baseline levels of C-reactive protein (CRP) were demonstrated to be an independent predictor of restenosis after balloon angioplasty of the femoropopliteal artery (18).
The perivascular inflammatory process after stent implantation can be adequately quantified by means of measurement of serum acute-phase reactants (19,20). We hypothesized that the postintervention course of serum acute-phase reactants varies according to the arterial site of stent implantation in large elastic or muscular arteries. The aim of this study was to investigate the postintervention course of serum acute-phase reactants after stent implantation in the femoropopliteal, iliac, and carotid arteries.
| MATERIALS AND METHODS |
|---|
|
|
|---|
Definitions
The Fontaine classification was used to categorize the clinical stage of peripheral arterial disease (PAD) (7). Patients with PAD in Fontaine II, III, and IV classifications and lesions with at least 50% reduction in diameter in the iliac (common and external iliac arteries) or the femoropopliteal artery were scheduled for endovascular interventions of the lower limbs. Assessment of data in patients with PAD was in accordance with the TransAtlantic Inter-Society Concensus for treatment of PAD (7). Carotid artery interventions were performed in asymptomatic patients with internal carotid artery stenosis of 90% or greater and in symptomatic patients with internal carotid artery stenosis of 70% or greater (8). Cerebral vascular occlusive disease was categorized as stage I if it was asymptomatic, stage II if it was associated with a recent transitory ischemic attack, stage III if it was associated with a recent prolonged reversible ischemic neurologic deficit, and stage IV if it was associated with a recent stroke.
At the time of hospital admission, patients age, sex, body mass index, medical history, cardiovascular risk factors, and comorbidities were recorded, according to a standardized questionnaire, by two independent observers (M.S., W.M.). Data sets of both observers were then compared to identify any inconsistency in data collection. Hypercholesterolemia was defined as a baseline serum cholesterol level greater than 200 mg/dL (5.17 mmol/L) or serum low-density lipoprotein value greater than 130 mg/dL (3.36 mmol/L). Diabetes mellitus was classified in patients with a history of diabetes, in patients with fasting blood glucose levels greater than 110 mg/dL (6.1 mmol/L) measured on three occasions, in patients with oral glucose tolerance test results suggestive of disease, and in patients with a glycosylated hemoglobin level greater than 0.06 (6.5%).
Laboratory Investigations
A complete series of routine laboratory investigations, including glycosylated hemoglobin level, low- and high-density lipoprotein cholesterol values, complete blood cell count, and serum creatinine value, was performed at baseline before PTA. Antecubital venous blood samples were obtained for determination of CRP, serum amyloid A (SAA), and fibrinogen values at baseline before the intervention and at 48 hours after the intervention. For measurement of serum CRP values, the high-sensitivity assay (N Latex CRP Mono; Dade Behring, Vienna, Austria) was used. SAA level was measured with a high-sensitivity assay (N Latex SAA; Dade Behring). Fibrinogen level also was measured with a test (Fibrinogen Clauss; Stago, Roche, Vienna, Austria). The detection levels of CRP, SAA, and fibrinogen assays, respectively, were 0.03 mg/dL (0.3 mg/L), 3.8 mg/L (3.8 U/L), and 20 mg/dL (2.0 g/L), respectively. The coefficients of variation of CRP, SAA, and fibrinogen values were 4.6%, 6.4%, and 5.2%, respectively.
Interventions
PTA and stent implantation were performed according to a standardized protocol. All interventions were performed by using the transfemoral access. Length and grade of lesion were calculated from the initial angiogram in at least two planes. Predilation was routinely performed in all vessel areas with the balloon diameter corresponding to the proximal nondiseased vessel diameter. Indication for interventional treatment and diameter, length, and number of stents implanted were routinely documented. The diameter of the stent was chosen according to the proximal nondiseased vessel area. For all carotid artery interventions, an over-the-wire carotid stent (Carotid Wallstent OTW; Boston Scientific, Natick, Mass) was used. All carotid artery interventions were performed by one experienced interventionist (R.A.) with experience in more than 300 procedures before initiation of the study. Implantation of stents in iliac arteries was performed by using one of three self-expandable stents (Easy Wallstent and Symphony Stent, Boston Scientific; Dynalink Stent, Guidant, Santa Clara, Calif).
In the femoropopliteal vessel area, self-expandable stents, including one made by another manufacturer (SmartStent; Cordis, Miami, Fla), were implanted. Peripheral interventions were performed by two authors (R.A. or E.M.) with more than 15 years of interventional experience. Duration of fluoroscopy and dose of contrast agent and amount of intraarterially administered heparin were recorded. The nonionic low-osmolality contrast agent ioversol (Optiray 320; Mallinckrodt, St Louis, Mo) in a dilution of 50% was used for all interventions. Peri- and postintervention complications at the site of arterial puncture and at the dilated vessel were documented up to 48 hours after the intervention. All complications that could be managed conservatively were classified as minor complications and included hematoma or pseudoaneurysm at the site of puncture and a significant dissecting membrane at the dilated vessel segment or peripheral emboli. Major bleeding (hemoglobin decrease > 2 mg/dL [0.02 g/L]) and all complications that necessitated emergency surgery within 48 hours were classified as major complications.
Standardized medication administered after stent implantation, irrespective of the site of intervention, included 100 mg of acetylsalicylic acid and 75 mg of clopidogrel bisulfate (Sanofi Pharma Bristol-Myers Squibb, Paris, France) daily. Pharmacologic treatment was initiated at least 1 day before the intervention.
Statistical Analysis
Continuous data were presented as the median and the interquartile range (range, 25th to the 75th percentile) or, if adequate, as the mean and the SD. Percentages were calculated for dichotomous variables. For univariate comparison of continuous data, the Kruskal-Wallis test or, if adequate, the Mann Whitney U test was used. Categorical data were compared by means of the
2 test. Spearman rank correlation coefficient was calculated for correlation of continuous variables. The course of laboratory values before intervention to 48 hours after intervention was measured by means of the relative change (ie, relative change in CRP [
CRP], relative change in SAA [
SAA], and relative change in fibrinogen [
fibrinogen]).
Multivariate polynomial logistic regression analysis was performed to assess the independent association of the course of acute-phase reactants and the location of stent implantation and to adjust for confounding factors and preselection bias. Variables that differed between the two groups at a level of significance with a P value less than .2 in univariate analysis were entered into the models as possible confounders. The results of the logistic regression model were presented as the odds ratio and the 95% CI. All P values were two sided; a P value less than .05 was considered statistically significant. Calculations were performed with statistical software (SPSS, version 10.0, for Microsoft Windows; SPSS, Chicago, Ill).
| RESULTS |
|---|
|
|
|---|
|
Carotid artery interventions were performed in 76 (70%) of 109 asymptomatic patients with internal carotid artery stenosis of 90% or greater. Thirty-three (30%) of these patients had a symptomatic internal carotid artery stenosis and recent neurologic events: 24 (22%) patients had a transitory ischemic attack, two (2%) patients had a prolonged reversible ischemic neurologic deficit, and seven (6%) patients had a major stroke and were examined 6 weeks afterward.
Baseline characteristics and interventional data in the 274 patients according to the arterial site of stent implantation are shown in Table 2. Patients who underwent revascularization of the carotid artery were substantially older compared with patients who underwent interventions in the iliac and femoropopliteal arteries. Patients with femoropopliteal artery stent implantation more frequently had complete vessel occlusion and an overall higher grade of stenosis compared with the grade of stenosis in other vessel areas. Furthermore, femoropopliteal artery lesions were longer compared with carotid and iliac artery lesions, and longer stents were implanted in this vessel area. Internal carotid artery stenoses were the shortest treated lesions, with a maximum length of 30 mm. Duration of carotid and iliac artery interventions was shorter compared with duration of femoropopliteal artery endovascular treatment (Table 2). The rate of complications was similar after stent implantation in the carotid, iliac, and femoropopliteal arteries; in particular, no differences in vessel dissection before stent implantation were observed in the three groups. No significant differences were observed between male and female patients within the three intervention groups.
|
|
|
|
|
CRP,
SAA, and
fibrinogen and length of the lesion, grade of stenosis, and length and diameter of the stent; no association was observed between total vessel occlusion and
CRP (P = .9),
SAA (P = .9), and
fibrinogen (P = .1). Furthermore, no significant differences in
CRP,
SAA, and
fibrinogen were determined for self-expandable stents implanted in the femoropopliteal artery and self-expandable stents implanted in the iliac artery. Multivariate polynomial logistic regression analysis was performed to independently assess differences in the postintervention course of acute-phase parameters with regard to the arterial site of stent implantation. The models were adjusted for possible confounding factors, as indicated by univariate comparison (Table 2). Patients with iliac stent implantation served as the reference group. Because of colinearity between the acute-phase reactants, separate models were applied for the courses of CRP, SAA, and fibrinogen values. The final models were adjusted for age, sex, fluoroscopy duration, contrast agent dose, complications, stenosis grade, total vessel occlusion, and cumulative stent length.
Femoropopliteal artery stent implantation was independently associated with a higher increase in CRP (Table 4), SAA (Table 5), and fibrinogen (Table 6) values compared with the values in the reference group of patients with iliac artery stent implantation. Carotid artery stent implantation was associated with a similar course of postintervention CRP (Table 4) and SAA (Table 5) values compared with the course of these values with iliac stent implantation; however, a more pronounced increase in fibrinogen (Table 6) level was noted. The final models had an acceptable fit (P < .001, log likelihood ratio).
|
|
|
| DISCUSSION |
|---|
|
|
|---|
Intimal and medial injuries after balloon dilation and stent implantation induced a perivascular inflammatory response (1,4,6,21). Differences in the extent of the inflammatory process depended on the type of stent used (2,22). In particular, the material of the stent mesh, such as stainless steel or nitinol, and the presence of a covered fabric, such as polytetrafluoroethylene or synthetic polyester fiber (ie, Dacron), influenced the local vascular reaction at the site of implantation (2,22). Furthermore, the distribution of shear stress after arterial stent implantation correlated with in-stent neointimal hyperplasia (23), which was a result of the inflammatory reaction.
However, the postimplantation reaction after stent deployment may also depend on the arterial site. So far, to our knowledge, large elastic arteries and muscular vessels have not been investigated with regard to their inflammatory potential in response to stent implantation. A marked increase in CRP, SAA, and fibrinogen values was described 48 hours after peripheral and coronary PTA and stent implantation (19,20). These acute-phase parameters, therefore, seemed adequate to quantify inflammation after stent implantation in muscular or elastic arteries.
Elastic and geometric properties of the vascular wall and local hemodynamics significantly vary in the carotid, iliac, and femoropopliteal arteries (24). In particular, maintenance of brisk flow through the carotid and iliac arteries in diastole as opposed to the diastolic flow characteristic in the femoropopliteal arteries may contribute to the patency of stents. Furthermore, atherosclerotic lesions in these three vessel areas were shown to differ with regard to plaque morphology and remodeling to luminal narrowing (25). Similarly, results in studies with animal models used for interventions to prevent coronary and peripheral artery restenosis are not uniformly predictive, depending on the arterial site of stent implantation (26).
Local vessel wall properties of muscular and elastic arteries probably cause the observed differences in the postimplantation inflammatory reaction. The distribution and density of vascular smooth muscle cells significantly vary between elastic and muscular arteries. Vascular smooth muscle cells may be the key factors in the chain of vascular inflammatory processes. However, the specific mechanisms of the more extensive inflammatory response to stent implantation in the femoropopliteal artery compared with those of the response to stent implantation in the carotid and iliac arteries remain unclear so far.
Balloon angioplasty and stent implantation apply a transient strain to the vessel wall that activates smooth muscle cells and modulates a proliferative and inflammatory phase of vessel repair (27). Migration of smooth muscle cells through the meshes of the stent and lumen-narrowing proliferation are the hallmarks of in-stent restenosis. Initial lumen diameter, flow characteristics, and endogenous antiinflammatory and antiproliferative mechanisms influence the occurrence of restenosis (23,28). However, the specific differences in the inflammatory response at various arterial sites are also likely to contribute to different patterns of restenosis after stent implantation in femoropopliteal, iliac, and carotid arteries.
Baseline CRP and SAA values have been reported to be independent predictors of restenosis after coronary angioplasty (29,30). Recently, the baseline CRP level before intervention was shown to be an independent predictor of restenosis after angioplasty in the femoropopliteal artery as well (18). The finding that stent implantation in the femoropopliteal artery induces a more extensive inflammatory response than do interventions in the carotid or iliac artery may have a potential effect on the understanding of mechanisms of restenosis. Further studies may reveal whether the postintervention increase in these acute-phase reactants is also associated with the occurrence of restenosis in the individual patient.
Limitations
These findings are derived from a between-patient comparison and the use of different types of stents. Therefore, we adjusted for several possible confounding factors, such as lesion morphology and length of the stent. Lesion morphology and stent characteristics were not associated with the postintervention inflammatory reaction. Furthermore, there was no relationship between a given stent in a given vessel and
CRP,
SAA, and
fibrinogen. However, it cannot be entirely excluded that different stent materials caused an altered inflammatory response and influenced the course of acute-phase parameters. However, all stent meshes were of comparable fabricates and geometries, no covered stents were implanted, and experimental studies suggest that a similar degree of inflammatory response can be expected from each of the stents used in this patient series (2,3).
In conclusion, stent implantation in the muscular femoropopliteal artery was associated with a more extensive vascular inflammatory response than was stent implantation in the elastic iliac and carotid arteries, independent of lesion morphology and interventional factors. The enhanced inflammatory response after stent implantation in the femoropopliteal artery might contribute to the higher rates of restenosis in this vessel area.
| FOOTNOTES |
|---|
Abbreviations: CRP = C-reactive protein,
CRP = relative change in CRP,
fibrinogen = relative change in fibrinogen,
SAA = relative change in serum amyloid A,
PAD = peripheral arterial disease,
PTA = percutaneous transluminal angioplasty,
SAA = serum amyloid A
Author contributions: Guarantors of integrity of entire study, E.M., M.S.; study concepts, M.S., E.M.; study design, M.E., W.M., M.H.; literature research, M.S., W.M.; clinical studies, R.A., E.M.; data acquisition, M.S., W.M.; data analysis/interpretation, E.M., R.A., H.R., O.W.; statistical analysis, M.S.; manuscript preparation, all authors; manuscript definition of intellectual content, E.M., M.S.; manuscript editing, revision/review, and final version approval, all authors.
| REFERENCES |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
G. Tepe, P. Muschick, M. Laule, F. Reddig, C. D. Claussen, L. M. Dinkelborg, H. Tielemans, M. Wehrmann, and S. H. Duda Prevention of Carotid Artery Restenosis After Sirolimus-Coated Stent Implantation in Pigs Stroke, February 1, 2006; 37(2): 492 - 494. [Abstract] [Full Text] [PDF] |
||||
![]() |
S. P. Lownie, D. M. Pelz, D. H. Lee, S. Men, I. Gulka, and P. Kalapos Efficacy of Treatment of Severe Carotid Bifurcation Stenosis By Using Self-Expanding Stents without Deliberate Use of Angioplasty Balloons AJNR Am. J. Neuroradiol., May 1, 2005; 26(5): 1241 - 1248. [Abstract] [Full Text] [PDF] |
||||
![]() |
B. J. Pearce and J. F. McKinsey Current Status of Intravascular Stents as Delivery Devices to Prevent Restenosis Vascular and Endovascular Surgery, July 1, 2003; 37(4): 231 - 237. [Abstract] [PDF] |
||||
![]() |
D. F. Denny Jr Prediction of Restenosis after Carotid Artery Stent Implantation Radiology, May 1, 2003; 227(2): 316 - 318. [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |