Elevated Levels of Pre-Procedural High-Sensitivity C-ReactiveProtein Is Associated with Midterm Restenosis after Extra- andIntracranial Stenting

Rishi Gupta, MD, Archit Bhatt, MD, Mounzer Kassab, MD, Arshad Majid, MDFrom the Department of Neurology, Division of Cerebrovascular Diseases, Michigan State University/Sparrow Health System, East Lansing, MI (RG, AB, MK, AM);and Cerebrovascular Center, the Cleveland Clinic Foundation, Cleveland, OH (RG).

[Correction added after online publication 15-December-2009: Received date has been corrected.]

Keywords: Re-stenosis, stents, C-reactive protein.

Acceptance: Received January 5, 2008,and in revised form February 11, 2008.Accepted for publication February 27,2008.

Correspondence: Address corre-spondence to Rishi Gupta, MD, Cere-brovascular Center, the Cleveland ClinicFoundation, 9500 Euclid Avenue, S80,Cleveland, OH 44195. E-mail: guptar@ccf.org.

J Neuroimaging 2010;20:74-77.DOI: 10.1111/j.1552-6569.2008.00313.x

A B S T R A C T

BACKGROUND AND PURPOSEHigh-sensitivity C-reactive protein (hsCRP) is an inflammatory marker associated withsubsequent coronary events and neointimal hyperplasia after coronary artery stent place-ment. We sought to determine if elevated levels of hsCRP are associated with restenosisafter placement of extra- and intracranial stents.METHODSWe retrospectively reviewed 73 consecutive patients at Michigan State University fromJuly 2006 until June 2007 who underwent treatment with carotid artery stent place-ment or intracranial stent placement. Data were collected in regards to demographics,pre-procedural hsCRP, and LDL levels, and angiographic variables characterizing the le-sion before and after treatment. A binary logistic regression model was constructed todetermine independent predictors of restenosis.RESULTSA total of 73 patients with a mean age of 69 ± 11 years were studied. A total of 57 patientswere treated with extracranial carotid stenting, 22 (38%) of whom were symptomatic,while 16 patients underwent intracranial stenting (all were symptomatic). There were9 patients (4 intracranial stents [25%] and 5 carotid stents [8.8%]) who developed arestenosis of >50%. In binary logistic regression modeling, the following variables werefound to be independently predictive of developing restenosis: smaller vessel diameter(OR .49, 95% CI .23-.98, P-value .046) and elevated hsCRP (OR 2.2, 95% CI 1.29-6.66,P-value .018).CONCLUSIONSElevated levels of pre-procedural hsCRP may be predictive of the development of neointi-mal hyperplasia in patients treated with extra- or intracranial stenting procedures. Futureprospective multicenter studies will be required to confirm these findings.

IntroductionRecent evidence suggests that high-sensitivity C-reactive pro-tein (hsCRP) plays an important role in the prediction of car-diovascular events in seemingly healthy patients.1 Elevation ofhsCRP is also associated with progression of carotid atheroscle-rosis2 and in patients at higher risk for stroke or death aftercarotid artery stent placement.3 Revascularization of extracra-nial and intracranial atherosclerotic lesion is being increasinglyperformed with stenting and angioplasty therapy. The durabil-ity of stent placement has been limited by restenosis as a resultof neointimal hyperplasia. This appears to be most significantin intracranial stent placement with short-term restenosis ratesreported as high as 30%.4 There have been few reports associ-ating elevated levels of hsCRP in patients undergoing coronaryartery stent placement with a higher risk of developing in stentrestenosis.5 We hypothesized that patients with elevated lev-els of hsCRP were at a significantly higher risk of developingin stent restenosis after placement of an extra- or intracranialstent.

MethodsAfter approval from our Institutional Review Board, we retro-spectively reviewed all patients at our institution that underwentstenting and angioplasty for extracranial atherosclerotic carotidartery disease or intracranial atherosclerotic disease betweenJuly 2006 and June 2007. A total of 78 consecutive patientswere identified. Patients were included in this study if theywere treated with a stent, an hsCRP was assessed prior to theprocedure, and a follow-up study was available to assess forrestenosis. Of these patients, 5 were excluded: 2 patients didnot have an hsCRP level drawn pre-procedure, 2 patients havenot had a follow-up study to assess for restenosis, and 1 patientwas treated with angioplasty without stent placement.

Demographic data and laboratory data (hsCRP and low-density lipoprotein [LDL] levels) were extracted from the med-ical records of each patient by one author blinded to the an-giography studies (AB). Data were also collected with regardsto if a patient was on a statin medication prior to hospitaliza-tion and if the patient presented with symptoms of a stroke

74 Copyright ◦C 2008 by the American Society of Neuroimaging

or transient ischemic attack (TIA) referable to the vessel thatwas treated with a stent. The angiograms were reviewed byone of the authors blinded to demographic and laboratory data(RG). Angiograms were reviewed for measurement of the ves-sel diameter pre-treatment, the percentage of stenosis of thelesion pre-treatment, the length of the lesion, and the percent-age of stenosis after stent deployment. All patients treated withintracranial stents had failed medical therapy with an anti-platelet agent or warfarin and exhibited neurological symp-toms referable to the ipsilateral stenosis. Extracranial carotidartery lesions were treated with stents in 22 (38%) patients withclinical signs and symptoms of stroke or TIA referable to theipsilateral vessel and a stenosis greater than 50%, while 35 pa-tients (62%) were treated for an asymptomatic stenosis greaterthan 80%.

All patients treated with extracranial carotid stents were fol-lowed with carotid duplex ultrasonography at 3 and 6 monthspostprocedure. If there was evidence of a restenosis >50% thena 64-channel CT angiogram was performed to verify the carotidultrasound findings. If both concurred then these patients wereconsidered to have a restenosis of >50%. In 1 patient with adiscordance between CT angiography and carotid duplex ul-trasonography, a catheter angiogram was performed to confirmthe restenosis. Only 1 patient required retreatment with carotidangioplasty for a restenosis >80%. No patient with a carotidrestenosis developed neurological symptoms.

All patients with an intracranial stent underwent follow-up catheter angiography at 3 months or greater to assess forrestenosis. These images were assessed in several different viewsto assess for restenosis. A >50% luminal narrowing was re-ported as a restenosis. Only 1 patient with a symptomatic MCArestenosis was retreated successfully with angioplasty.

Laboratory Analysis

Each patient in this study had a serum sample collected within24 hours prior to placement of a stent. Blood samples werecentrifuged for 10 minutes and stored at 4 ◦C until the assaywas run within 24 hours. A chemiluminescent immunometricassay (Immulite 2500 assay, Diagnostics Products Corporation,Los Angeles, CA) was used. Levels of hsCRP are expressed inmilligrams per deciliter.

Interventional Procedure

All patients were pre-treated with an oral loading dose of clopi-dogrel 300-450 mg within 24 hours of the stent procedure andaspirin 325 mg. All extracranial stent procedures were per-formed with the patient awake, while intracranial interventionswere performed under general anesthesia. A 6-French guidecatheter was used for intracranial interventions and a 6-Frenchshuttle sheath for extracranial carotid stent procedures. All pa-tients were fully heparinized with a target activated clotting timeof 200 to 250 seconds prior to placement of the guide catheter.For intracranial stenting procedures, an .014-inch wire was usedto traverse the lesion. A Wingspan self-expanding stent (BostonScientific Corp., Natick, MA) was utilized in 2 patients withlesions in the middle cerebral artery (MCA) territory. Each le-

sion was pre-dilated with a Gateway Balloon (Boston Scientific,Natick, MA) that was undersized by 10-15%. Balloon-mountedstents were used for the other 14 lesions (7 MCA and 7 verte-brobasilar). The Mini-Vision or Vision Cobalt chromium stent(Abbott-Guidant Corp., Indianapolis, IN) was utilized in 12of these lesions, while the Taxus drug-eluting stent (BostonScientific Corp., Natick, MA) was placed in 2 patients. Nopre-dilation was used for lesions treated with balloon-mountedstents.

In patients treated with extracranial carotid disease, the Ac-cunet distal protection device was navigated across the stenosisin all cases. An Acculink self-expanding stent (Guidant Corp.,Indianapolis, IN) was sized to the vessel diameter and lesionlength and deployed successfully in all patients. The stent waspost-dilated using a Viatrac Balloon (Guidant Corporation, In-dianapolis, IN) or Maverick balloon (Boston Scientific, Natick,MA). The balloon was inflated to nominal pressure in order toexpand the stent to the diameter of the target vessel.

All patients were treated with clopidogrel 75 mg daily andaspirin 325 mg daily for at least 8 weeks after extracranial stentplacement and at least 12 weeks after intracranial stent place-ment. After these time points, patients were placed on aspirin 81mg daily as monotherapy. Each patient was placed on a statinmedication after stent placement with the intent to reduce theLDL cholesterol to <70 mg/dL. Each patient with hyperten-sion was placed on anti-hypertensive medications with target ofreducing the systolic blood pressure to less then 140 mmHg.

Statistical Analysis

Statistical analysis was performed using SPSS Version 15. Aunivariate analysis was performed to determine predictors ofrestenosis and also predictors of patients that presented withsymptomatic lesions. Categorical variables were assessed usingthe Fisher’s exact test and continuous variables with the stu-dent’s t-test. Variables with a P -value <.10 were entered intoa binary logistic regression model using the enter method todetermine the independent predictors of the development of arestenosis of >50% and symptomatic lesions.

ResultsA total of 73 patients with a mean age of 69 ± 11 years werestudied. A total of 57 patients were treated with carotid stent-ing, 22 (38%) of whom were symptomatic, while 16 patientsunderwent intracranial stenting (all were symptomatic). Therewere 9 patients (4 intracranial stents [25%] and 5 carotid stents[8.8%]) who developed a restenosis of >50%. Patients who pre-sented with symptomatic lesion were more likely to have anelevated LDL at admission (99.4 ± 26.9 mg/dL vs. 84.7 ± 21.1,P < .0001), elevated levels of hsCRP (1.72 ± 1.65 vs. .30 ± .24,P < .0001), and a higher percentage of stenosis of the vessel(90.9 ± 9.5% vs. 81.8 ± 9.1%, P < .02) in univariate analysis.In binary logistic regression modeling, patients with elevatedLDL cholesterol (OR 1.04, 95% CI 1.005-1.068, P < .02) andelevated hsCRP (OR 22.9, 95% CI 3.9-95.2, P < .0001) wereat a significantly higher risk of presenting with a symptomaticlesion.

Gupta et al: hsCRP and Midterm Restenosis 75

Table 1. Univariate Analysis of the Potential Predictors ofRestenosis in Patients Treated with Extra- or Intracra-nial Stent Placement

NoRestenosis Restenosis

(N = 9), (N = 64), P-N (%) N (%) Value

DemographicsAge (years ± SD) 70 ± 9 69 ± 12 .94Female 5 (56) 26 (41) .20Hypertension 8 (89) 56 (88) .79Diabetes 4 (44) 21 (33) .48Hyperlipidemia 6 (67) 45 (70) .55Smoking 6 (67) 27 (42) .28Symptomatic at 4 (44) 33 (52) .41

presentationOn statin at admission 3 (33) 46 (72) .05

Angiographic variablesIntracranial lesion 4 (44) 12 (19) .14Lesion length,(mm ± SD) 17 ± 9 14 ± 7 .20Stenosis pre, (% ± SD) 91 ± 9 86 ± 10 .15Stenosis post, (% ± SD) 5.2 ± 4.9 2.9 ± 5.7 .18Vessel diameter, 3.8 ± 0.8 4.5 ± 1.1 .035

(mm ±SD)Laboratory values

LDL, (mg/dL ± SD) 95.7 ± 28.4 91.7 ± 24.9 .66hsCRP, (mg/dL ± SD) 2.29 ± 2.24 .85 ± 1.13 .003

Of the 16 intracranial stent procedures, 1 patient treatedwith a balloon expandable stent in the basilar artery developeda stroke secondary to perforator occlusion to the pons. Thepatient recovered after 3 months to independent function, buthad a significant hemiparesis post-procedure. The overall peri-procedural rate of stroke was thus 1 of 16 procedures (6.3%) forintracranial stenting. Of the 57 patients treated with extracranialcarotid stent placement, 1 patient developed a transient apha-sia that resolved after 30 minutes. The peri-procedural strokeor TIA rate during carotid stent placement was thus 1 of 57procedures (1.7%).

A univariate analysis was performed to identify potentialpredictors of the development of restenosis (Table 1). Pa-tients with a smaller pre-treatment vessel diameter and elevatedhsCRP were at a higher risk for the development of restenosisat follow-up. Additionally, patients on statin medications priorto hospitalization were at a lower risk of developing restenosis.Of the 4 patients with intracranial stent restenosis, 3 were inthe middle cerebral artery (one of whom had a Wingspan stentplaced, and two with a balloon-mounted stent), while 1 patienthad a basilar artery restenosis in a balloon-mounted stent. Onepatient with a MCA restenosis had symptoms of TIA referableto the restenosis and underwent successful angioplasty to treatthe restenosis. All 5 patients with restenosis of the extracranialcarotid artery were asymptomatic, with only 1 patient requir-ing retreatment with angioplasty for a restenosis of greater than80%.

In binary logistic regression modeling the following vari-ables were found to be independently predictive of developingrestenosis: smaller vessel diameter (OR .49, 95% CI.23-.98, P -value .046) and elevated hsCRP (OR 2.2, 95% CI 1.29-6.66,P -value .018).

DiscussionThis study shows that patients who undergo treatment withextra- or intracranial stents appear to be at a higher risk ofdeveloping restenosis if their pre-procedural hsCRP levels areelevated. Additionally, patients who presented with a symp-tomatic stenosis in the extra- or intracranial location were foundto have elevated LDL cholesterol levels and elevated hsCRPlevels. This would suggest that inflammatory markers may beof use in determining vulnerable atherosclerotic lesions in theextra- and intracranial vasculature and that hsCRP may be amarker of future restenosis.

Inflammatory markers are felt to play a pivotal role in theprogression of atherosclerosis. Elevations of hsCRP has beenshown to be an independent predictor of future coronary eventsin asymptomatic patients particularly with levels greater than.3 mg/dL.6 This marker has been shown to be additive to tradi-tional risk factor assessments for prediction of vascular eventsincluding LDL cholesterol.7 High doses of statin medicationscan reduce the levels of hsCRP and also reduce progression ofcoronary plaque and subsequent coronary events. Patients withan LDL cholesterol less than 70 mg/dL and hsCRP less than.2 mg/dL were least likely to develop a recurrent myocardialinfarction.8 In our cohort of patients, those who presented witha symptomatic lesion were found to have significantly higherlevels of hsCRP and LDL cholesterol. This may signify the pres-ence of a recently ruptured plaque that has led to a stroke orTIA.

Although several inflammatory markers have been linkedto predicting vascular events in patients, hsCRP appears tobe the most sensitive marker.9 C-reactive protein appears tohave a role in recruitment of moncytes into the arterial walland reducing the production of nitric oxide.10,11 It is felt thatthis may create instability of the plaque architecture and thusplace patients at a higher risk of plaque rupture or cardiovas-cular events.12 There is also evidence that hsCRP is secretedby coronary arteries and also by smooth muscle cells.13 Au-topsy studies of the coronary vessels after stent placement haveshown that neointimal hyperplasia occurs in varying degrees af-ter stent placement. The vessels that developed restenosis dueto the neointimal hyperplasia may be a result of inflammatorymacrophages and T cells with redifferentiated smooth musclecells (α-actin-positive).14,15 The inflammatory cascade may beresponsible for restenosis and elevations of hsCRP may be amarker of patients at risk.

The following variables have been found to be predictive offuture coronary artery restenosis after stent placement: longerlesion length,16 if more then one stent was placed,17 a greaterpost-procedural percentage of stenosis, smaller vessel diame-ters,18 and the use of drug-eluting stents.19 Coronary vesselsless then 2.6 mm appear to be at a higher risk of restenosis com-pared with larger vessels.18 Patients in our series with smaller

76 Journal of Neuroimaging Vol 20 No 1 January 2010

vessel diameters pre-treatment were at a significantly higherrisk of developing restenosis. In univariate modeling there wasa trend toward restenosis in patients with a higher percentage ofstenosis after stent placement. We noted that both patients witha Wingspan stent in the MCA developed restenosis. This maybe partly due to leaving a higher percentage of residual stenosiswith the Wingspan stent in the intracranial vasculature. A recentmulticenter registry revealed that the average stenosis poststentplacement with the Wingspan stent system is 27.2 ± 16.7%.20

This is in contrast to the experience from placing balloon-mounted stents intracranially where the mean post-stent place-ment stenosis was 5 ± 16%.21 We did not see a statistical differ-ence in poststent percentage of stenosis, but this may be due tothe small number of patients in this series. Additionally, we uti-lized balloon-mounted stents in a majority (88%) of intracraniallesions and the post-stent percentage of stenosis for this groupwas 3.5% versus 20% for the two Wingspan patients. Only 2 pa-tients were treated with drug-eluting stents intracranially in thisseries and neither developed restenosis. The short-term resteno-sis rate for these stents placed intracranially has been reportedto be 5%.22

There are limitations to this study given its retrospectivenature. Additionally, this is a single-center experience and re-sults may vary with a more heterogeneous population. Third,different stent types were used in the intracranial vasculature, al-though a majority (14 of 16 [88%]) were balloon-mounted stents.Fourth, we followed patients with carotid duplex ultrasound toassess for restenosis. This may have potentially underestimatedthe degree of restenosis as we did not follow each patient withcatheter angiography. Carotid duplex ultrasonography tends tooverestimate the degree of stenosis after stent placement23 andthus it is less likely that this study underestimates the percent-age of patients with restenosis. Last, we do not have follow-upinformation in regards to hsCRP and LDL values at a later timepoint and patient compliance with medications. Nonetheless,these results are of importance as they reveal that inflamma-tory markers may play a crucial role in the development ofrestenosis in patients treated with extra- or intracranial stentplacement.

In conclusion, patients with elevations of hsCRP pre-procedure and smaller reference vessel diameters are at a higherrisk of developing in stent restenosis in short-term follow-up.Additionally, patients presenting with symptoms of stroke andTIA have significantly higher hsCRP and LDL levels at admis-sion in comparison to patients that are asymptomatic from thestenosis. Further study is required on a more heterogeneouspopulation in a prospective manner. Additional study shouldbe considered to determine if lowering of hsCRP with aggres-sive statin use will help to reduce the number of patients thatdevelop restenosis.

References1. Schlager O, Exnerr M, Mlekusch W, et al. C-reactive protein pre-

dicts future cardiovascular events in patients with carotid stenosis.Stroke 2007;38:1263-1268.

2. Ridker PM, Buring JE, Shih J, et al. Prospective study of C-reactiveprotein and the risk of future cardiovascular events among appar-ently healthy women. Circulation 1998;98:731-733.

3. Groschel K, Ernemann U, Larsen J, et al. Preprocedural C-reactiveprotein levels predict stroke and death in patients undergoingcarotid stenting. Am J Neuroradiol 2007;28:1743-1746.

4. Levy EI, Turk AS, Albuquerque FC, et al. Wingspan in-stentrestenosis and thrombosis: incidence, clinical presentation andmanagement. Neurosurgery 2007;61:644-651.

5. Hong YJ, Jeong MH, Lim SY, et al. Elevated preprocedural high-sensitivity C-reactive protein levels are associated with neointimalhyperplasia and restenosis development after successful coronaryartery stenting. Circ J 2005;69:1477-1483.

6. Pearson TA, Mensah GA, Alexander RW, et al. Markers for in-flammation and the cardiovascular disease: application to clinicaland public health practice: a statement for healthcare profession-als from the Centers for Disease Control and Prevention and theAmerican Heart Association. Circulation 2003;107:499-511.

7. Ridker PM, Rifai N, Rose L, et al. Comparison of C-reactive pro-tein and low-density lipoprotein cholesterol levels in the predictionof first cardiovascular events. N Engl J Med 2002;347:1557-1565.

8. Ridker PM, Cannon CP, Morrow D, et al. C-reactive protein levelsand outcomes after statin therapy. N Engl J Med 2005;352:20-28.

9. Smith SC Jr, Anderson JL, Cannon RO III, et al. CDC/AHAworkshop on markers of inflammation and cardiovascular disease:application to clinical and public health practice: report from theclinical practice discussion group. Circulation 2004;110:e550-e553.

10. Verma S, Wang CH, Li SH, et al. A self-fulfilling prophecy: C-reactive protein attenuates nitric oxide production and inhibitsangiogenesis. Circulation 2002;106:913-919.

11. Pasceri V, Willerson JT, Yeh ET. Direct proinflammatory ef-fect of C-reactive protein on human endothelial cells. Circulation2000;102:2165-2168.

12. Ross R. Atherosclerosis: an inflammatory disease. N Engl J Med1999;340:115-126.

13. Calabro P, Willerson JT, Yeh ET. Inflammatory cytokines stim-ulated C-reactive protein production by human coronary arterysmooth muscle cells. Circulation 2003;108:1930-1932.

14. Komatsu R, Ueda M, Naruko T, et al. Neointimal tissue responseat sites of coronary stenting in humans. Macroscopic, histological,and immunohistochemical analyses. Circulation 1998;98:224-233.

15. Grewe PH, Deneke T, Machraoui A, et al. Acute and chronic tissueresponse to coronary stent implantation: pathological findings inhuman specimen. J Am Coll Cardiol 2000;35:157-163.

16. Lee CW, Park CB, Kim YH, et al. Incidence and predictors ofrecurrent restenosis following implantation of drug-eluting stentsfor in-stent restenosis. Catheter Cardiovasc Interv 2007;69:104-108.

17. Roy P, Okabe T, Pinto Slottow TL, et al. Correlates of clini-cal restenosis following intracoronary implantation of drug-elutingstents. Am J Cardiol 2007;100:965-969.

18. Kastrati A, Dibra A, Mehilli J, et al. Predictive factors of resteno-sis after coronary implantation of sirolimus- or paclitaxel-elutingstents. Circulation 2006;113:2293-2300.

19. Gershlick A, De Scheerder I, Chevalier B, et al. Inhibition ofrestenosis with a paclitaxel-eluting, polymer-free coronary stent.The European evaluation of the palitaxel eluting stent (ELUTES)trial. Circulation 2004;109:487-493.

20. Fiorella D, Levy EI, Turk AS, et al. US multicenter experiencewith the wingspan stent system for the treatment of intracranialatheromatous disease: periprocedural results. Stroke 2007;38:881-887.

21. Freitas JM, Zenteno M, Alburto-Murrieta Y, et al. Intracranial ar-terial stenting for symptomatic stenoses: a Latin American experi-ence. Surg Neurol 2007;68:378-386.

22. Gupta R, Al-Ali F, Thomas AJ, et al. Safety, feasibility, and short-term follow-up of drug-eluting stent placement in the intracranialand extracranial circulation. Stroke 2006;37:2562-2566.

23. AbuRahma AF, Maxwell D, Eads K, et al. Carotid duplex velocitycriteria revisited for the diagnosis of carotid in-stent restenosis.Vascular 2007;15(3):119-125.

Gupta et al: hsCRP and Midterm Restenosis 77