Carotid angioplasty and stenting in high-risk patients

Carotid Angioplasty and Stenting in High-Risk Patients George P. Teitelbaum, M.D., Michael A. Lefkowitz, M.D., and Steven L. Giannotta, M.D. Department of Neurological Surgery, USC School of Medicine, Los Angeles, California

Teitelbaum GP, Lefkowitz MA, Giannotta SL. Carotid angioplasty and stenting in high-risk patients. Surg Neurol 1998;50:300 –12. BACKGROUND

To examine our initial experience in carotid stenting (CS) for the prevention of stroke in patients with high-grade carotid stenoses. METHODS

The authors performed 26 CS procedures in 25 carotid vessels in 22 patients over a 15-month period. All carotid stenoses treated, except one, were 70% or greater. Of all CS procedures, 84% were performed for obstructing atherosclerotic plaques. CS was performed in one patient each for carotid dissection and pseudoaneurysms caused by a gunshot wound, post-radiation stenosis, post-carotid endarterectomy (CEA) restenosis, and a flow-obstructing post-CEA intimal flap. Of all patients, 68.2% were symptomatic, with a history of stroke or transient ischemic attacks ipsilateral to the treated carotid artery. In addition, 36.4% of our patients were either hospitalized or required skilled nursing care before CS because of severe neurologic deficits. Using the Sundt CEA-risk classification system, 59.1% of our patients were classified as Grade III and 40.9% were Grade IV pre-CS. All but one patient had either a compelling medical or anatomic reason for endovascular treatment of their carotid disease. We used both Wallstents and Palmaz stents, and all procedures were performed via the transfemoral route. Three procedures were performed in conjunction with detachable platinum coil embolization for multiple carotid pseudoaneurysms, a residual carotid “stump” after previous ICA thrombosis, and an ipsilateral MCA saccular aneurysm. RESULTS

We had a 96.2% procedural technical success rate. There was one death in our series 3 weeks post-CS attributable to myocardial infarction. Despite a high 30-day combined death, stroke, and ipsilateral blindness rate of 27.3% (6/22 patients), only two ipsilateral strokes directly related to CS occurred (7.7% per procedures performed) from which one patient recovered fully within 5 days. The average follow-up post-CS was 5.9 months (range, 3 weeks–15 months). Of successfully treated vessels, 58.3% have undergone 6-month follow-up vascular imaging, which has revealed a 14.3% rate of occlusion or restenoPresented at SCVIR 1997, Washington, D.C. Address reprint requests to: Dr. George P. Teitelbaum, USC University Hospital, 1500 San Pablo Street, Los Angeles, CA 90033. Received June 9, 1997; accepted November 21, 1997. 0090-3019/98/$19.00 PII S0090-3019(98)00038-X

sis greater than 50%. At or beyond 1 month post-CS, 19 of 21 surviving patients (90.5%) were ambulatory, fluent of speech, and independent, and none has thus far suffered a delayed stroke or TIA. CONCLUSION

CS seems to be a reasonable alternative to medical management for the treatment of carotid disease in patients deemed to be poor candidates for standard carotid surgery. Longer term follow-up is necessary to assess the durability of carotid revascularization using CS. KEY WORDS

Angioplasty, carotid, stroke, stenting, carotid, high-risk carotid surgery.

revious work by Sundt and associates [24,25] examined the risk of carotid endarterectomy (CEA) in patients who were neurologically unstable and/or had medical conditions or angiographic findings that would increase the risk of mortality and morbidity from CEA. Medical risk factors identified by Sundt included angina, myocardial infarction within 6 months, congestive heart failure (CHF), severe hypertension (greater than 180/110 mm Hg), chronic obstructive pulmonary disease (COPD), age greater than 70 years, and severe obesity. Neurological risks were defined as progressing deficit, resolved neuro deficit less than 24 hours pre-CEA, global cerebral ischemia, recent stroke less than 7 days preoperative, deficits from multiple previous strokes, and crescendo (frequent and uncontrolled by anticoagulation) transient ischemic attacks (TIAs). Contralateral internal carotid artery (ICA) occlusion, ipsilateral ICA siphon stenosis, ICA plaque extending more than 3 cm distal to the carotid bifurcation, common carotid artery (CCA) plaque extending more than 5 cm proximal to the carotid bifurcation, high carotid bifurcation, and intraluminal thrombus propagating from a carotid ulcerative lesion were all described as angiographic risk factors. Using these risk categories, Sundt [25] defined four classes of CEA patients with regard to mortal-

P

© 1998 by Elsevier Science Inc. 655 Avenue of the Americas, New York, NY 10010

Carotid Stenting

1

Carotid Endarterectomy Morbidity and Mortality [25]

GRADE I II III IV

Surg Neurol 301 1998;50:300 –12

RISK

FACTORS*

RISK (IN 1,935

Neurologically stable; no medical or angiographic risks; with unilateral or bilateral ulcerative-stenotic carotid disease Neurologically stable; no medical risks, but with angiographic risks (contralateral ICA occlusion most common) Neurologically stable; medical risks, with or without angiographic risks Neurologically unstable, with or without medical or angiographic risks

ity and morbidity risk (Table 1). The highest risk classification was Grade IV, in which patients were neurologically unstable with or without medical or angiographic risk factors. These patients had an 8.5% risk of mortality or morbidity resulting from CEA. Other studies examining the risks of CEA have reported an overall mortality rate as high as 2.3% [3], a risk of stroke (intra- and postoperative) of 5% [3], intracranial postoperative hemorrhage of less than 0.6% [20], and a long-term restenosis (greater than 50% by 1 year) rate as high as 25% [1]. Some authors have reported a higher incidence of complications (including stroke, TIAs, cranial nerve injuries, and restenosis) associated with surgery to treat post-CEA restenosis [2,4,6,29]. The NASCET study [16], which demonstrated the superiority of CEA over medical management for symptomatic high-grade carotid stenosis, reported CEA mortality and morbidity rates of 0.6% and 4.9%, respectively. Recently published series of patients undergoing carotid artery percutaneous transluminal angioplasty (PTA) with and without stenting have reported procedural technical success rates of 93.2– 100%, combined mortality and major morbidity rates of 1.4 –10.2%, and post-PTA carotid restenosis/occlusion rates of 0 –16% [7,10 –12,14,28]. However, Theron et al [28] reported a 0% cerebral embolization rate in 136 patients undergoing carotid PTA with cerebral protection (an ICA occlusion balloon inflated during PTA). Yadav et al [31] reported a stroke rate of 4% (per vessels treated) in a series of 25 carotid arteries treated with PTA and stenting for restenosis post-CEA. Although these early carotid PTA and stenting results are encouraging, more long-term data need to be accumulated before carotid PTA can be measured accurately against the results of CEA. Presently in the planning stages is a major North American multicenter protocol prospectively comparing carotid PTA and stenting with CEA.

OPERATIONS)

,1% risk (5 RIND, 6 CVA) 1.8% risk (6 RIND, 7 CVA) 4% risk (9 fatal MIs, 10 RIND, 10 CVAs (1 fatal) 8.5% risk (27 CVAs (8 fatal), 14 RIND,

We now report our initial experience with carotid stenting (CS) for the prevention of stroke in patients considered at increased risk for conventional CEA.

Materials and Methods Patient data are summarized in Table 2. Twenty-six CS procedures were performed in 25 vessels in 22 patients. There were 13 males and 9 females, with an age range of 32– 83 years (average, 63.3 years). Of these 22 patients, 15 (68.2%) were symptomatic with either stroke or TIAs associated with the treated carotid artery. Of our 22 patients, 8 (36.4%) [1,4,6,17–21] were either hospitalized or required skilled nursing care before CS because of severe neurologic deficits. One of these patients (Patient No. 17) had recent right ICA occlusion (with an ICA cul-de-sac or “stump”) and was suffering right hemispheric crescendo TIAs. Right CCA/external carotid artery (ECA) CS was performed to provide increased collateral blood flow to the right hemisphere via the ipsilateral ophthalmic artery. In addition, trans-stent Guglielmi detachable coil (GDC) (Target Therapeutics, Fremont, CA) embolization of the ICA remnant was performed to avoid the occurrence of “stump” syndrome [5] (Figure 1). This same technique of trans-stent coil embolization, described previously by Perez-Cruet et al [18], was performed in a case of ICA dissection and multiple pseudoaneurysms after a point-blank gunshot wound to the neck and base of skull (Patient No. 20). Here, the arterial stent and GDC coils were used to “reconstruct” a smoother ICA lumen and avoid the risk of embolic stroke attributable to arterial dissection and pseudoaneurysms (Figure 2). We now generally wait a minimum of 4 weeks after a cerebral infarction before attempting to treat an ipsilateral carotid lesion. In the seven patients where there had been no

55/M

69/F

78/F

77/M

32/M

48/M

67/M

59/F

66/M

82/M

60/F

71/M

69/F

59/F

82/M

1

2

3

4

5

6

7

8

9

10

11

12

13

14

15

TYPE

Atherosclerosis

LESION

Post-radiation stenosis Atherosclerosis

Atherosclerosis

Atherosclerosis

Atherosclerosis

Atherosclerosis

Atherosclerosis

Atherosclerosis

Lt. ICA

Lt. CCA

Restenosis post-CEA

Atherosclerosis

Rt. carotid Atherosclerosis bifurcation Lt. ICA

Rt. ICA

Rt. ICA

Rt. ICA

Lt. ICA

Rt. carotid Atherosclerosis bifurcation Lt. ICA Atherosclerosis

Lt. ICA

Lt. ICA

Rt. ICA Lt. ICA

Rt. ICA

Rt. ICA Atherosclerosis Lt. ICA & carotid bifurcation

Rt. ICA

LOCATION

LESION

Summary of Patients’ Data

PATIENT NO. AGE/SEX

2

III

III

III

III

III

III

III

III

III

IV

III

IV

III

IV

IV

P

1

W W

P W1P W W

P W

1 1

1 1 1 1

1 1

1

1

—

1. Lt. occipital and bilat. cerebellar infarcts causing Rt. homonymous hemianopsia 2. Uremia not requiring hemodialysis 2

2

Contralateral (Lt) hemispheric stroke due to A-fib. 1 day post-CS. 2

TIA 3 days post-CS.

2

2

2

COMPLICATIONS

AT

6

2 No change in Lt. ICA stenosis

*

*

*(3 mos. F/U carotid Doppler sonogram revealed no restenosis)

2

2

2

2

Asymptomatic occlusion at 6 mos. P-S1W Fatal M.I. 3 wks. post- N/A CS W-CCA 1 ICA 2 2 P-S-ECA W 2 2

W 2

1

30-DAY

RESTENOSIS >50% MOS.

Lost to follow-up after 1 month W 2 1 (,70%) W-CCA & ICA Lt. hemispheric stroke 2 P-S-ECA (recovered function completely within 5 days) W 2 2

W

1 2

1

1 1

1

SUNDT TECHNICAL STENT/ GRADE† SUCCESS DEVICES USED

90

90

90

90

90

80

100

90

90

20

80

30

80

90

30

90

100

90

90

100

80

100

90

100

0

80

60

80

100

30

STATUS AT LATEST FOLLOW-UP

No change from baseline. No new stroke or TIA. Ambulatory, fluent of speech, and independent without recurrent TIAs or stroke.

NEUROLOGIC

4

5

Continued

Ambulatory, fluent of speech, and independent without recurrent TIA. No change from baseline. No new stroke or TIA.

8.5 Ambulatory, fluent of speech, and independent. No recurrent TIA or new stroke. 7 No change from baseline. No new stroke or TIA. 5.5 Rt. homonomous hemianopsia. Ambulatory, fluent of speech, and independent without recurrent stroke of TIA.

No change from baseline. No new stroke or TIA. 9 No change from baseline. No new stroke or TIA. 8.5 No change from baseline. No new stroke or TIA. 8.5 No change from baseline. No new stroke or TIA.

10

No change from baseline. No new stroke or TIA. 11.5 Ambulatory with walker, fluent of speech, and independent. No history of recurrent stroke. Possible candidate for Lt. CEA. 11 No change from baseline. No new stroke or TIA. 3 wks Died.

14

15

1

F/U PRE-CS AT F/U (MOS)

KARNOFSKY SCORE††

302 Surg Neurol 1998;50:300 –12 Teitelbaum et al

57/F

83/F

71/M

27/M

82/M

59/F

17

18

19

20

21

22

Atherosclerosis

TYPE

Obstructing intimal flap post-CEA. Recurrent intimal flap

Multiple pseudoaneurysms from gunshot wound

Atherosclerosis

Lt. ICA & Atherosclerosis carotid bifurcation

Lt. CCA

Lt. CCA

Lt. ICA

Lt. ICA

Rt. carotid Atherosclerosis bifurcation & Rt. MCA aneurysm

Rt. CCA/ECA Atherosclerosis (Rt. ICA “stump”)

Lt. CCA

LESION

III

IV

IV

IV

IV

IV

III

P

W 1 GDC coils W 1 GDC coils

W

W1P1 GDC coils

P P

W1P

1

1

1

1

1

1 1

1

SUNDT TECHNICAL STENT/ GRADE† SUCCESS DEVICES USED COMPLICATIONS

Recurrence of intimal flap 2 days later. Need for Rt. femoral cutdown to retrieve Palmaz stent dislodged from balloon catheter. Asymptomatic Lt. ECA occlusion at time of CS.

2

*

*

*

1. OS visual loss due *(3 mo. F/U angio to central retinal revealed stent artery embolus 3 patency without hrs. post-CS (not restenosis) responsive to IV tPA). 2. OD amaurosis fugax 5 days post-CS, cardiac in origin (treated with Coumadin). Recurrent TIAs * resolved using volume expansion and Florinef. Rt. frontal stroke * causing Lt. hemiparesis and increased dysphagia and confusion 6 hrs. post-CS. 2 *

30-DAY

RESTENOSIS >50% MOS. AT

6

90

30

20

40

40

40

90

90

60

100

60

30

90

90

1

1

1

1

1

3

4

F/U PRE-CS AT F/U (MOS)

KARNOFSKY SCORE††

STATUS AT LATEST FOLLOW-UP

No change from baseline. No new stroke or TIA.

No new stroke or TIA. Ambulatory, and fluent of speech, with improved memory and orientation. No new stroke or TIA.

Mild residual Rt.-sided weakness requiring physical therapy. Ambulatory with walker, fluent of speech, and independent. No recurrent stroke or TIA. Ambulatory, fluent of speech, and independent. No new stroke or TIA.

Increased confusion and dysphagia with left arm paralysis. Requires skilled nursing care. No new stroke or SAH.

Ambulatory, fluent of speech, and independent without recurrence of TIAs. No new stroke.

OS blindness. Ambulatory, fluent of speech, and independent without recurrence of OD amaurosis fugax. No new stroke or TIA.

NEUROLOGIC

CCA, common carotid artery; CEA, carotid endarterectomy; CS, carotid stenting; ECA, external carotid artery; F/U, follow-up; GDC, Guglielmi detachable coil; ICA, internal carotid artery; MCA, middle cerebral artery; MI, myocardial infarction; OD, right eye; OS, left eye; P, Palmaz stent; P-S, Palmaz-Schatz stent; SAH 5 Subarachnoid hemorrhage; TIA 5 Transient ischemic attack; and W, Wallstent. †, See Table 1. ††, See Table 3.

47/M

16

LOCATION

LESION

Summary of Patients’ Data (continued)

PATIENT NO. AGE/SEX

2

Carotid Stenting Surg Neurol 303 1998;50:300 –12

304 Surg Neurol 1998;50:300 –12

Teitelbaum et al

Patient No. 17, a 57-year-old female was admitted to the hospital for crescendo TIAs causing left-sided weakness. She also had a history of hypertension, coronary vessel disease, and severe systemic atherosclerosis. (A) Right common carotid lateral digital arteriogram demonstrating proximal occlusion of the right ICA (arrow) together with a high-grade stenosis of the proximal ECA (curved arrow) and an eccentric plaque at the carotid bifurcation (open arrow). (B) After deployment of an 8 3 20-mm Wallstent device within the right ECA, extending into the right CCA and after balloon angioplasty of the stented ECA to 5 mm, we advanced a Tracker 10 microcatheter (arrowheads) through the interstices of the Wallstent into the ICA cul-de-sac, where multiple GDC platinum embolization coils were detached (arrows). (C) Common carotid lateral digital arteriogram demonstrating successful stent deployment and dilatation of distal right CCA and proximal ECA and occlusion of the ICA remnant with GDC coils (arrows). After this procedure, the patient continued to have multiple transient ischemic attacks, which were managed successfully with fluid expansion and oral Florinef therapy. At 3 months post-procedure, she is ambulatory, fluent of speech, and independent without recurrent TIAs.

1

stroke or TIA referable to the treated carotid artery, CS was performed for a variety of reasons: prophylaxis before spinal surgery (Patient No. 16); in highrisk patients with carotid stenoses greater than 70% (Patients No. 8, 12, 15, 16, 22); to provide increased collateral blood flow to the contralateral cerebral hemisphere (Patient No. 1); to allow ICA guiding catheter access for an intracranial endovascular procedure (Patient No. 18, see below) (Figure 3). Patient No. 22 did have a history of contralateral (right eye) amaurosis fugax that we treated successfully 5 months before her carotid procedure by PTA and stenting of an innominate artery stenosis. Each patient was classified into Sundt risk grades I–IV according to criteria summarized in Table 1. Of 22 patients, 13 (59.1%) were classified as Grade III; 9 of 22 (40.9%) as Grade IV. All but one patient had either a compelling medical or anatomic reason for endovascular treatment of their carotid disease and after evaluation by a neurovascular surgeon (SLG), were thought to be poor candidates for CEA. The one patient not meeting these criteria (Patient No. 8) was a 59-year-old female with severe diabetes and obesity who self-referred herself for endovas-

cular treatment of a 90% ICA origin stenosis because she adamantly refused surgery. All carotid stenoses treated, except one, measured 70% or greater by NASCET criteria [16]. A 60% carotid bifurcational stenosis was treated (Patient No. 18) to allow passage of a 6F guiding catheter into the ICA to perform GDC coil embolization of a middle cerebral artery (MCA) aneurysm. Of 25 total vessels treated, 21 (84%) were narrowed by atherosclerotic plaque. CS was performed in Patient No. 20 for the treatment of ICA dissection and multiple pseudoaneurysms. One vessel each was stented for treatment of post-CEA restenosis, post-radiation stenosis, and a flow-obstructing post-CEA intimal flap. The only absolute contraindication to CS with reference to lesion morphology was the presence of intraluminal thrombus. The presence of plaque ulceration was not considered a contraindication to CS. Of 26 procedures, 21 (80.1%) involved stenting of the ICA and/or carotid bifurcation, 4/26 (15.4%) the CCA, and 1/26 (3.8%) ECA/carotid bifurcation. Two of the above ICA/carotid bifurcation cases also included stenting of an ipsilateral ECA origin stenosis. Bilateral carotid procedures were attempted in

Carotid Stenting

Surg Neurol 305 1998;50:300 –12

Patient No. 20, a 27-year-old male, the victim of a point-blank gunshot wound to the base of the left skull, presented with decreased mental status, requiring intubation. (A) This cut film left common carotid frontal arteriogram demonstrates the bullet fragment (curved arrow) together with irregular stenoses and several fusiform pseudoaneurysms of the cervical ICA (arrows). He was considered to be a poor candidate for carotid surgery and was referred for either left internal carotid revascularization versus permanent coil or detachable balloon ICA occlusion for prevention of stroke. After deployment of two 8 3 20 Wallstent devices and a 1-cm Palmaz stent (at the junction between the cervical and petrous portions of the ICA). (B) GDC (arrows) coils were detached within the pseudoaneurysms. (C, D), Frontal (C) and lateral (D) left common carotid digital arteriograms of the neck and base of skull demonstrate a much smoother patent ICA lumen without pseudoaneurysm filling. The patient was extubated 1 day later and made an uneventful recovery. He suffered no TIA or stroke during either the peri-procedural or 30-day post-procedural period. Carotid Doppler sonography during the week post-CS demonstrated no flow obstruction within the left ICA.

2

three patients (Nos. 2, 4, and 13). However, in only one patient (No. 13) did we treat both carotid lesions during the same procedure. Local femoral puncture site lidocaine anesthesia was used in all cases with or without IV propofol anesthesia. All procedures were performed through

long transfemoral 9 or 10F sheaths initially using 10F and later in our experience, 9F braided guiding catheters (Cordis, Miami Lakes, FL) under full systemic heparinization. Initial access to the carotid artery was achieved with a Simmons I–III catheter. Through this catheter, nonionic contrast injections

306 Surg Neurol 1998;50:300 –12

Teitelbaum et al

Patient No. 18, an 83-year-old female who had suffered two recent devastating intracranial hemorrhages, leaving her with confusion, dysphagia, and left-sided weakness. She required constant skilled nursing care in a rehabilitation facility. She had two unprotected aneurysms: a wide neck saccular aneurysm arising from an azygous anterior cerebral artery as well as a 10 3 4-mm wide-neck saccular aneurysm arising from the trifurcation of the right MCA. The MCA aneurysm was thought to be the cause of her subarachnoid hemorrhages. She also had severe systemic atherosclerosis. (A) Right common carotid lateral cervical digital arteriography demonstrates an approximately 60% eccentric stenosis involving the carotid bifurcation (arrow). Also note the extreme tortuousity of the cervical ICA (curved arrows). (B), Right common carotid frontal digital arteriographic imaging of the skull demonstrates the wide-neck saccular aneurysm of the right MCA M1 segment trifurcation (arrow). After evaluation by two neurosurgeons, she was thought to be an extremely poor surgical risk for both clip ligation of her MCA aneurysm and CEA. (C), In an effort to advance a 6F guiding catheter into the ICA to allow intracranial catheterization of her MCA aneurysm, we performed placement of an 8 320 mm Wallstent and balloon angioplasty (to 8 mm) within the right carotid bifurcation as seen on this post-stenting right common carotid lateral digital arteriogram. (D), We were then able to easily advance a 6F envoy guiding catheter (Cordis) through the Wallstent and into the proximal right ICA up to the first cervical “loop” of this vessel. This frontal digital roadmap image demonstrates the presence of a 2-tip marker Rapid Transit microcatheter (Cordis) coaxially advanced through the 6F guiding catheter through the right M1 segment, and finally, into the right MCA aneurysm where GDC coils were deployed (arrow). After the aneurysm embolization, it was thought that the patient could not be heparinized because of her recent multiple intracranial hemorrhages and the continued presence of an unprotected intracranial aneurysm. Unfortunately, she suffered a right frontal stroke approximately 6 hours post-procedure. The patient’s Karnofsky score decreased from a pre-CS level of 40 down to 30 at 1 month post-CS. She has not, however, experienced any recurrent stroke or intracranial hemorrhage.

3

Carotid Stenting

permitted pre-CS cervical and intracranial digital arteriography in frontal and lateral planes. This catheter was then exchanged for the 9-10F guiding catheter over a 0.0350 Amplatz 3-mm J 260-cm guidewire (Cook, Bloomington, IN) with a custom-made 7-tapering-to-5F 135-cm central catheter (Cook) inserted within the Cordis guiding catheter. Using digital roadmap imaging, the carotid stenosis was crossed with a steerable 0.014 – 0.0180 platinum-tip 300-cm guidewire. Over this small gauge wire, a low-profile (less than 4F) balloon angioplasty catheter was used to predilate the carotid stenosis to 3 or 4 mm. All balloon inflations, both before and after stent deployment, were kept to less than 10 seconds. All stenting was performed using 8-mm 3 20- or 40-mm Wallstents (Schneider, Minneapolis, MN), loaded within a 7– 8F introduction device, or 1–2 cm-long Palmaz stents (Johnson and Johnson, Warren, NJ) crimped onto a 5F balloon angioplasty catheter. The aforementioned ECA origin stenoses were stented using 1.5-cm Palmaz-Schatz stents premounted onto a 4-mm coronary balloon dilatation catheter. The stent devices were advanced to the appropriate deployment site over the above 0.014 – 0.0180 exchange wire. Wallstents, which were used in the treatment of most of the ICA/carotid bifurcation stenoses, were intentionally deployed to cover the origin of the ECA. Any residual stenosis visualized within the deployed Wallstent was further dilated with a 5- to 8-mm 5F balloon angioplasty catheter. After CS, repeat cervical and intracranial digital arteriography was performed in frontal and lateral planes with injection of nonionic contrast through the CCA guiding catheter. These images were assessed for residual carotid stenosis and intracranial branch occlusions. Technical success of a procedure was defined as the ability to deploy the stent with a residual carotid stenosis of less than 20%. At the conclusion of the procedures, systemic heparinization was allowed to wear off gradually (no IV protamine sulfate was used). Several hours later, in the intensive care unit (where all patients were monitored overnight post-CS), the groin sheath was removed and hemostasis was achieved with both manual compression and the application of a Femostop device (USCI, Billerica, MA) for 5– 6 hours. Barring complications, most patients were discharged within 24 hours post-CS. All patients were treated with oral ticlopidine (250 mg po BID) for 4 weeks and daily aspirin (325 mg po qd) for life after CS. A complete blood count was performed 2 weeks post-CS to assess for any signs of aplastic anemia. All patients were to undergo either carotid Doppler sonography or contrast angiography at 6 months

Surg Neurol 307 1998;50:300 –12

3

Karnofksy Scale (modified)20

SCORE

MEANING

100 90

Normal: no complaints, no evidence of disease Able to carry on normal activity: minor symptoms Normal activity with effort: some symptoms Cares for self: unable to carry on normal activity Requires occasional assistance: cares for most needs Requires considerable assistance and frequent care Disabled: requires special care and assistance Severely disabled: hospitalized, death not imminent Very sick: active supportive care needed Moribund: fatal processes are progressing rapidly Dead

80 70 60 50 40 30 20 10 0

post-stenting to assess for carotid restenosis of greater than 50%. Patients were assessed for both neurologic and non-neurologic complications occurring within and after 30 days post procedure. Patients’ neurologic functional status pre- and at least 30 days post-CS was compared. We evaluated patients for specific deficits, ambulatory status, fluency of speech, and level of independence. Patients were considered independent if they were able to bathe, dress, and feed themselves. Patients were also scored using the modified Karnofsky Scale (Table 3) [13] before CS and at last follow-up.

Results Please refer to the summary of patient data in Table 2. Technical success was achieved in 25/26 procedures (96.2%). However, there was recurrence of a post-CEA intimal flap 2 days post-stenting in one patient (Patient No. 21) requiring placement of a second Palmaz CCA stent to adequately treat this problem. In none of the attempted stenting procedures was there an ipsilateral post-procedural angiographic branch occlusion within the distributions of the anterior cerebral artery or MCA or within the distal ICA branches. All patients undergoing balloon dilatation of the carotid bifurcation experienced transient bradycardia and/or asystole (lasting less than 30 sec), which in all cases was adequately managed with angioplasty balloon deflation and/or IV glycopyrrolate or atropine. In no case was it necessary to place a temporary cardiac pacemaker for the management of bradycardia or asystole.

308 Surg Neurol 1998;50:300 –12

A total of 4/22 patients (18.2%) suffered strokes within 30 days of their carotid procedures: one case of aphasia and right hemiparesis that resolved completely by 5 days post-PTA with volume expansion, IV dobutamine, and anticoagulation (Patient No. 2); one case of right homonymous hemianopsia caused by vertebrobasilar embolization, which we treated with IV t-PA and volume expansion (Patient No. 13); one case of contralateral MCA stroke 1 day after PTA because of atrial fibrillation, which was treated with anticoagulation (Patient No. 10); and a right frontal stroke (resulting in left arm paralysis, increased dysphagia, and decreased mental status) 6 hours after combined right carotid bifurcation stenting and right MCA aneurysm coil embolization (Patient No. 18). All but one of these patients (No. 18) are presently ambulatory, fluent of speech, and independent at .30 days post PTA. Patient No. 18, who was dependent on skilled nursing care pre-CS (because of two recent devastating intracranial hemorrhages from her right MCA trifurcation aneurysm), remained so post-CS. It is not completely clear whether the right frontal stroke suffered by this patient was attributable to CS or aneurysm coil embolization. The average Karnofsky score of our 22 patients increased from 67.7 pre-CS to 77.7 at last follow-up post-CS. Patient No. 16 suffered ipsilateral central retinal artery occlusion (and left eye blindness) 3 hours post-CS. This was unresponsive to IV t-PA. Of 22 patients, 3 (13.6%) experienced reversible ischemic neurologic deficits within 30 days of CS: Patient No. 16 suffered contralateral eye amaurosis fugax 5 days post-CS (thought to be cardiac in origin and treated with oral Coumadin); one case of mild transient left hand hypesthesia 3 days post-CS (Patient No. 9); one case of episodic right hemispheric ischemia for 1 week post-CS associated with systemic hypotension responsive to IV volume expansion and oral Florinef (Apothecon) therapy (Patient No. 17). Other than these episodes, no other surviving patient has thus far reported a new recurrent stroke or TIA during follow-up periods ranging from 3 weeks to 15 months (average follow-up period 5 5.9 months). Of 22 patients, 7 (31.8%) were treated with anticoagulation for variable lengths of time after PTA. However, as a result of CS, Patients No. 3 and 22 had their pre-CS Coumadin therapy discontinued. One patient was lost to follow-up after 1 month (Patient No. 1), and one patient (4.5%) died 3 weeks post-PTA because of a myocardial infarction (Patient No. 6). To summarize, despite a high 30-day combined death, stroke, and ipsilateral blindness rate of 6/22 (27.3%), only two ipsilateral strokes directly related to CS occurred (7.7% per proce-

Teitelbaum et al

dures performed), from which one patient fully recovered. At or beyond 1 month post-CS, 19 of 21 surviving patients (90.5%) were ambulatory, fluent of speech, and independent, and none has thus far suffered a delayed stroke or TIA. Other procedural complications have included one case of post-CS renal failure in a 69-year-old female whose serum creatinine level increased from 1.5 to 3.5 mg/dl in the weeks after PTA (Patient No. 13). She has thus far not required hemodialysis. This complication, as well as her vertebrobasilar stroke (see above), were thought to be attributable to cholesterol emboli to the renal and left subclavian arteries as a result of her carotid procedures. Although there have been no significant groin hematomas associated with our procedures, one patient has required a right femoral artery cut-down to aid in the retrieval of a Palmaz stent that became dislodged from its balloon catheter (Patient No. 21). However, after retrieval of the dislodged stent, a 9F vascular sheath was reintroduced via the arteriotomy, and the carotid stenting procedure was successfully completed. Patient No. 22 suffered an asymptomatic ipsilateral ECA occlusion as a result of her CS procedure. However, antegrade flow at the origin of this ECA was observed the following day using Doppler sonography. Thus far, 14 of the 24 successfully treated carotid vessels (58.3%) have undergone 6-month follow-up vascular imaging studies (10 carotid Doppler sonograms and one carotid arteriogram), which have revealed one case each of ipsilateral asymptomatic carotid occlusion (Patient No. 5) and restenosis (Patient No. 2) (50 –70%) post-CS (14.3% rate of restenosis/occlusion at 6 months). No patient, except for Patient No. 21 (who underwent a second stent placement 2 days after the first because of recurrence of a post-CEA intimal flap), has thus far required re-intervention.

Discussion In this limited series of patients, all considered to be at high risk for standard CEA, we actually experienced only two strokes (from which one patient recovered completely within 5 days) directly related to dilatation and stenting of an affected carotid artery. The other two strokes experienced in our series were caused by atrial fibrillation and aortic arch catheterization (where the strokes were not in the vascular distribution being treated with CS). Obviously, the two major hurdles to be crossed before there can be general acceptance of CS as an

Carotid Stenting

Surg Neurol 309 1998;50:300 –12

Patient No. 12, a 71-year-old male with a history of hypertension and coronary vessel disease, underwent previous left carotid surgery for a ligation of a carotid stump. He suffered left vocal cord paralysis due to cranial nerve injury from this surgery. Because standard CEA for treatment of this right-sided high-grade carotid stenosis was contraindicated, he was referred for balloon angioplasty and stenting. (A) Right common carotid lateral digital arteriogram demonstrating a 75% stenosis at the origin of the right ICA (arrow). (B) After balloon angioplasty and stenting of the right carotid bifurcation with an 8 3 20-mm Wallstent device that extends into both the ICA and CCA, there is noted to be lack of apposition of the distal Wallstent to the lumen of the right ICA (arrow). Despite multiple balloon dilatations of the distal portion of this stent, the same distal stent configuration persisted. (C) Right common carotid lateral digital arteriogram after deployment of a 1-cm Palmaz stent (expanded to 6 mm) within the distal Wallstent, “tacking” the distal margins of the Wallstent down to the intimal surface of the right ICA (arrows). Also note the excellent flow into the right ECA through the interstices of the Wallstent (curved arrow). The patient suffered no neurologic deficit as a result of this procedure and was discharged home the following day.

4

alternative to CEA are reduction in the rates of procedure-related strokes and post-stenting restenosis. Theron [28] showed a 0% stroke rate for carotid angioplasty and stenting performed with balloon protection of the cerebral circulation. Other work has demonstrated that PVA particle embolization can be performed relatively safely within the cavernous ICA with balloon protection of the distal intracranial circulation [27]. As more experience is gained with this technique, perhaps it will prove to be a reliable method to reduce the rate of CS-related strokes. The rate of post-CS restenosis is central to the issue of durability of endovascular management of carotid disease, because stenting of the carotid artery may obviate surgical treatment options should restenosis occur. Patch angioplasty or bypass grafting may be made impossible by the presence of a long carotid stent. In such circumstances, the only option left for a symptomatic patient would be extracranial-to-intracranial bypass surgery which, even in the best of hands, has not provided effective long-term stroke prevention [8]. For the above reasons, it is imperative to use the shortest stent possible in the endovascular management of carotid disease. We, as well as other investigators, have chosen to use ticlopidine therapy in the post-stenting period,

because this antiplatelet agent is associated with a significant reduction in restenosis compared with Coumadin therapy after coronary artery stenting [22]. However, this agent may be associated with adverse gastrointestinal symptoms and aplastic anemia. Newer methods being investigated to prevent restenosis include endovascular brachytherapy [19,23,26]. No significant work has yet been published reporting the use of endovascular radiation therapy post-CS for the prevention of restenosis. Theron [28] demonstrated a significantly lower restenosis rate with the use of a stent in association with carotid angioplasty compared with balloon angioplasty alone. Much discussion has centered on the ideal stent design for endovascular treatment of carotid disease. We feel that any interventionalist involved with such work should be thoroughly familiar with the placement of both balloon-expanded Palmaz stents and self-expanding Wallstents. Both Wallstents [15,21] and Palmaz stents [17,30] have been used successfully in the treatment of iliac obstructive lesions. However, there is evidence that the rigid Palmaz stent, although maintaining larger vessel diameters over the long term, may also promote a thicker neointima formation, at least in the swine model [9]. Also, there is some concern that the inflexible Palmaz stent could be externally com-

310 Surg Neurol 1998;50:300 –12

pressed and kinked in the neck, thus causing acute flow compromise in the stented carotid artery. To its advantage, the Palmaz stent has greater hoop strength and creates a lower metal burden within stented vessels compared with the Wallstent. Both stents permit effective follow-up color Doppler sonography of the stented carotid lumen. Although the Wallstent has excellent flexibility and conforms well to curves normally found at the carotid bifurcation, it can sometimes fail to fully appose the intima at its proximal or distal ends (Figure 4). In two of our patients, we needed to use a 1-cm Palmaz stent to fully deploy the distal portion of the Wallstent against the luminal surface of the ICA. In addition, we have observed several cases in which the proximal portion of the Wallstent did not make the necessary proximal turn into the CCA, but was positioned at an angle to the direction of blood flow. This necessitated the placement of a second Wallstent within the first to make the stents conform to the curved path between the CCA and ICA. In these cases, it is imperative to maintain guidewire access across the Wallstent during its placement because of the possibility of suboptimal stent deployment. There have been concerns over blood flow to the ECA through the struts of an endovascular stent (“jailing” of the ECA). Except for one asymptomatic, and apparently transient, CS-induced ECA occlusion, we have not yet encountered any difficulty with this condition. However, only further experience with these procedures will reveal whether this is merely a benign situation or actually creates delayed ECA occlusion or embolization. In two cases, we chose to prophylactically stent a stenotic ECA origin before stenting of the ICA/carotid bifurcation in an effort to avoid acute occlusion of the ECA (presumably because of plaque deformation). In conclusion, our preliminary experience with CS for the prevention of stroke indicates that this technique is a reasonable alternative to medical management for the treatment of carotid disease in patients who have compelling anatomic and/or medical reasons why standard carotid surgery cannot be performed. However, longer term follow-up is necessary to assess the durability of carotid revascularization using CS.

Teitelbaum et al

2. 3. 4. 5. 6. 7. 8. 9.

10.

11. 12. 13. 14.

15.

16.

17.

18. The authors acknowledge the assistance of Drs. Hal Damuth, Anibal Gauto, Dinesh Patel, and Stuart May. 19.

REFERENCES 1. Bernstein EF, Humber PB, Collins GM, Dilley RB, Devin JB, Stuart SH. Life expectancy and late stroke

20.

following carotid endarterectomy. Ann Surg 1983;198: 80 – 6. Bernstein EF, Torem S, Dilley RB. Does carotid restenosis predict an increased risk of late symptoms, stroke or death? Ann Surg 1990;212:629 –36. Brott TG, Labutta RJ, Kempczinski RF. Changing patterns in the practice of carotid endarterectomy in a large metropolitan area. JAMA 1986;255:2609 –12. Callow AD. Recurrent stenosis after carotid endarterectomy. Arch Surg 1982;117:1082–5. Cassidy L, Grace A, Bouchier-Hayes DJ. The carotid stump syndrome. Eur J Vasc Surg 1992;6:368 –70. Das MB, Hertzer NR, Ratliff NB, Ohara PJ, Beven EG. Recurrent carotid stenosis: a five-year series of 65 operations. Ann Surg 1985;202:28 –35. Diethrich EB, Mouhamadou N, Reid DB. Stenting in the carotid artery: initial experience in 110 patients. J Endovasc Surg 1996;3:42– 62. EC/IC Study Group. Failure of EC-IC arterial bypass to reduce the risk of ischemic stroke. N Engl J Med 1985;313:1191–200. Fontaine AB, Spigos DG, Eaton G, Das-Passos S, Christoforidis G, Khabiri H, Jung S. Stent-induced intimal hyperplasia: are there fundamental differences between flexible and rigid stent designs? JVIR 1994;5: 739 – 44. Gil-Peralta A, Mayol A, Marcos JR, Gonzalez A, Ruano J, Boza F, Duran F. Percutaneous transluminal angioplasty of the symptomatic atherosclerotic carotid arteries: results, complications, and follow-up. Stroke 1996;27:2271–3. Iyer SS, Roubin GS, Yadav S, Vitek J, Parks JM, Wadlington V, Doblar D, Jordan W. Elective carotid stenting (abstr). J Endovasc Surg 1996;3:105– 6. Kachel R. Results of balloon angioplasty in the carotid arteries. J Endovasc Surg 1996;3:22–30. Karnofsky DA, Burchenal JH. In: Macleod CM, ed. Evaluation of chemotherapy agents. New York: Columbia University Press, 1949:191–205. Mathias KD, Jaeger HJ, Mau C, Goetz F. Problems and complications of internal carotid artery percutaneous transluminal angioplasty and stent placement (abstr). Radiology 1995;197:649. Murphy TP, Webb MS, Lambiase RE, Haas RA, Dorfman GS, Carney WI Jr, Morin CJ. Percutaneous revascularization of complex iliac artery stenoses and occlusion with use of Wallstents: three-year experience. JVIR 1996;7:21–7. North American Symptomatic Carotid Endarterectomy Trial Collaborators. Beneficial effect of carotid endarterectomy in symptomatic patients with highgrade carotid stenosis. N Engl J Med 1991;15:445–53. Palmaz JC, Laborde JC, Rivera FJ, Encarnacion CE, Lutz JD, Moss JG. Stenting of the iliac arteries with the Palmaz stent: experience from a multicenter trial. Cardiovas Intervent Radiol 1992;15:291–7. Perez-Cruet M, Patwasdhan R, Mawad M. Treatment of dissecting pseudoaneurysm of the cervical ICA using Wallstent and detachable coils: case report. Neurosurgery 1997;40:622– 6. Popowski Y, Verin V, Urban P. Endovascular b-irradiation after percutaneous transluminal coronary balloon angioplasty. Int J Radiation Oncology Biol Phys 1996;36:841–5. Reigel MM, Hollier LH, Sundt TM Jr, Piepgras DG,

Carotid Stenting

21.

22.

23.

24. 25. 26.

27. 28.

29.

30.

31.

Sharbrough FW, Cherry KJ. Cerebral hyperperfusion syndrome: a cause of neurologic dysfunction after carotid endarterectomy. J Vasc Surg 1987;5:628 –34. Sapoval MR, Chatellier G, Long AL, Ravani C, Pagny JY, Raynaud AC, Beyssen BM, Gaux JC. Selfexpandable stents for the treatment of iliac artery obstructive lesions: long-term success and prognostic factors. AJR 1996;166:1173–9. Schomig A, Neumann FJ, Kastrati A, Schuhlen H, Blasini R, Hadamitzky M, Walter H, Zitzmann-Roth EM, Richardt G, Alt E, Schmitt C, Ulm K. A randomized comparison of antiplatelet and anticoagulant therapy after the placement of coronary artery stents. N Engl J Med 1996;334:1084 –9. Schopohl B, Liermann D, Pohlit LJ, Heyd R, Strassmann G, Bauersachs R, Schulte-Huermann D, Rahl CG, Manegold KH, Kollath J, Bottcher HD. 192IR endovascular brachytherapy for avoidance of intimal hyperplasia after percutaneous transluminal angioplasty and stent implantation in peripheral vessels: 6 years of experience. Int J Radiation Oncology Biol Phys 1996;36:835– 40. Sundt TM, Sandok BA, Whisnant JP. Carotid endarterectomy: complications and preoperative assessment of risk. Mayo Clin Proced 1975;50:301– 6. Sundt TM. Occlusive cerebrovascular disease. Philadelphia: W. B. Saunders, 1987:226 –30. Teirstein PS, Massullo V, Jani S, Popma JJ, Mintz GS, Russo RJ, Schatz RA, Guarneri EM, Steutermans, Morris NB, Leon MB, Tripuraneni P. Catheter-based radiotherapy to inhibit restenosis after coronary stenting. N Engl J Med 1997;336:1697–703. Teitelbaum GP. Head and Neck Therapeutic Embolizations with Assistance of Balloon Occlusion. JVIR Supplement, 1996;7:220. Theron JG, Payelle GG, Coskun O, Huet HF, Guimaraens L. Carotid artery stenosis: treatment with protected balloon angioplasty and stent placement. Radiology 1996;201:627–36. Treiman GS, Jenkins JM, Edwards WH Sr, Barlow W, Edwards WH Jr, Martin RS 3d, Mulherin JL Jr. The evolving surgical management of recurrent carotid stenosis. J Vasc Surg 1992;16:354 – 63. Williams JB, Watts PW, Nguyen VA, Peterson CL. Balloon angioplasty with intraluminal stenting as the initial treatment modality in aorto-iliac occlusive disease. Am J Surg 1994;168:202– 4. Yadav JS, Roubin GS, King P, Iyer S, Vitek J. Angioplasty and stenting for restenosis after carotid endarterectomy: initial experience. Stroke 1996;27: 2075–9.

COMMENTARY

The authors present their experience with carotid angioplasty and stenting in a series of patients who were not suitable for standard endarterectomy procedures. As their patients represent a riskier subgroup, we would not expect their complication record to match that of a large endarterectomy series. The therapies provided in this presentation are appropriate and warranted. I do believe that endarterectomy in the hands of an experienced sur-

Surg Neurol 311 1998;50:300 –12

geon represents the current “gold standard” for garden-variety carotid bifurcation disease, but endovascular procedures (also in the hands of knowledgeable neurointerventionalists) can provide definitive therapy in selected patients with unusual anatomy or a medical condition that precludes endarterectomy. Christopher F. Dowd, M.D. Department of Radiology University of California San Francisco, California

Teitelbaum et al present an honest and frank paper on their experience with carotid angioplasty and stenting. Twenty-two patients underwent stenting for various indications, including atherosclerosis (18 procedures), trauma (1), radiation-induced intimal fibrosis (1), and recurrent stenosis or postsurgical intimal flap (2). The reported complication rates are: seven neurologic complications (31%; 4 permanent and 3 reversible), one mortality (4.5%), and two non-neurologic complications (9%). Of the 11 patients available for 6-month follow-up, 7 (64%) had no recurrence of stenosis, 1 had no change (9%), 1 had recurrent stenosis (9%), and 1 artery had occluded silently (9%). We agree with the authors that carotid endarterectomy (CEA) remains the mainstay of therapy for carotid artery disease. We are not as enthusiastic as members of the interventional community that percutaneous carotid artery angioplasty with stenting (PTAS) will imminently unseat CEA as the primary mode of therapy for carotid artery disease [1]. Proven superiority in at least these three areas will be needed before angioplasty supplants CEA: (1) safety (Is angioplasty as safe or safer than CEA?); (2) cost (Does angioplasty shorten hospital stay, and is it less expensive?); and (3) durability (Is angioplasty durable? Does it last?). Teitelbaum et al have addressed these issues except for cost in their paper. The combined morbidity and mortality of this procedure in their hands is too high. The safety issues have been underscored in our studies as well, where the risk of minor strokes during PTAS was significantly greater than for CEA [2,3]. We feel that this is probably because the number of emboli shed during PTAS (documented by transcranial doppler) far exceeds that shed during CEA (unpublished data). Also of considerable concern to us in our report was the trend (but not significance) of nonneurologic complications, including asystole, requiring permanent pacemaker; retroperitoneal he-

312 Surg Neurol 1998;50:300 –12

matoma; prolonged hospitalization because of arrythmias; and lower extremity ischemia [3]. These findings coupled with mortality and morbidity data have convinced us that, as yet, PTAS is too risky to justify involvement in a cooperative study comparing PTAS and CEA. Finally, we share the authors’ concern that these stents may require removal in the future. We have removed a stent, and have realized that length of the stent and involvement of the external carotid artery will make any stent removal in the future difficult. Stents need to be kept short, and engineering efforts are addressing this issue, as well as the observed problem of stent deformation caused by kinking (which we have also observed). We approach carotid stenosis very much like we do intracranial aneurysmal disease, in that we are not convinced that anything short of surgical intervention is superior, but there may be cases in which interventional procedures are justified. Identifying such cases remains difficult because safety issues have not been fully assessed. We applaud the authors for their honest report, but we remain con-

Teitelbaum et al

vinced that the best treatment for carotid artery disease involves the shortest route—through the neck. Wink S. Fisher, III, M.D. Division of Neurosurgery William D. Jordan, M.D. Section of Vascular Surgery University of Alabama Birmingham, Alabama

REFERENCES 1. Hopkins LN. Vascular neurosurgery: past-presentfuture. The Richard C. Schneider Lecture. Presented at the American Association of Neurological Surgeons Annual Meeting, Denver, Colorado; April 1997. 2. Jordan WD, Schroeder PT, Fisher WS, McDowell HA. A comparison of angioplasty with stenting versus endarterectomy for the treatment of carotid artery stenosis. Ann Vasc Surg 1997;11:2– 8. 3. Yadav JS, Roubin GS, Iyer S, Vitek J, King P, Jordan WD, Fisher WS. Elective stenting of the extracranial carotid arteries. Circulation 1997;95:376 – 81.