Intracranial Percutaneous Transluminal Arterial Angioplasty (PTA) and Stenting

A number of investigators have reported the management of intracranial (subarachnoid location) arterial atherosclerotic stenosis using percutaneous transluminal angioplasty (PTA). In 1990 Purdy, et al described a case of middle cerebral artery stenosis managed with PTA and demonstrated post-procedure improvement in cerebral perfusion. In 1992 Ahuja, et al. reported PTA for symptomatic basilar artery stenosis. A six month follow-up arteriogram demonstrated basilar occlusion. In 1996 McKenzie, et al. published their series of 17 PTA cases for skull-base and intracranial cases of atherosclerotic and vasculitic stenoses. Three procedures resulted in dissections one of which resulted in severe narrowing. Sixteen of the seventeen cases improved angiographically although 35% of the cases had moderate or severe residual stenoses after dilatation. In 1996 Nakatsuka, et al. reported two cases of basilar artery angioplasty for atherosclerotic stenoses both patients had clinical and blood flow improvements. A 1997 paper by Takis, et al. described experiences with 10 patients undergoing PTA. Four patients suffered peri-procedural strokes secondary to vasospasm or perforator occlusion. Kellogg et al's topic review in 1998 cited their groups results with 34 intracranial angioplasties. Thirty day neurologic morbidity rate was 8.8% per treated vessel. Jimenez, et al published one case of successful basilar artery angioplasty and provided an excellent review of 36 published cases relating to basilar stenosis and PTA. Of those 36 cases, 13 demonstrated improvement in lumen diameter demonstrated improvement in lumen diameter following treatment. Nineteen of the cases did not have sufficient information reported to determine the therapeutic results.

The risks of intracranial PTA were best described by Mori, et al. These authors reported their results in 42 patients with 42 lesions having greater than a 70% stenosis. The series was divided into three groups based upon lesion morphology. Type A lesions were 5mm or less in length, concentric or moderately eccentric, and less than totally occlusive. Type B lesions were 5-10 mm in length, extremely eccentric or totally occlusive, and less than three months old. Type C lesions were more than 10 mm in length, angulated greater than 90 degrees, excessively tortuous, or totally occluded and three months or older. Clinical success rates were 92%, 86%, and 33% for Types A, B, and C respectively while the risk of fatal or nonfatal stroke, or bypass surgery in Types A. B, and C were 8%, 26%, and 87% respectively. Restenosis at one year for type A was 0% (92% follow-up), type B 33% (86% follow-up), and type C 100% (33% follow-up). From their data the authors recommended PTA for Type A lesions only.

Intracranial Arterial Stenting

The concept of using stents along with PTA for intracranial arterial stenoses comes from the cardiac intervention experience. Coronary balloon angioplasty is accompanied by abrupt closure secondary to elastic recoil and vessel dissection in approximately 5% of patients. In the presence of thrombus the risk of abrupt closure increases to 7.2 to 27.8%. While such closure may be tolerable in the cardiac system it is not tolerable in a basilar artery without significant posterior communicating artery collateral flow. The use of stents for the management of intracranial arterial disease has been limited by the interventionalist's inability to negotiate the relatively stiff balloon/stent assembly through the tortuous carotid and vertebral artery systems without damaging the vessel in the process. Recent advances in coronary stents, however, have made delivery of more flexible systems possible. In 1999 Horowitz reported three cases of basilar stenosis managed successfully with angioplasty and stenting.

We have found the coronary stent to be quite flexible with excellent trackability. There was no difficulty negotiating vertebral artery turns and the stent demonstrated no appreciable shortening upon expansion thus making final stent positioning and placement quite accurate and consistent with our pre-angioplasty positioning.

Pre and Post Angioplasty Management

We feel patient preparation prior to and after procedure performance is integral to achieving a good immediate outcome with stenting procedures especially when carried out in the intradural compartment. Our rationale for a number of interventions will be discussed below.

The use of clopidogrel is based upon the CAPRIE stud which demonstrated a 8-10% relative risk reduction of ischemic stroke, myocardial infarction, or vascular death in patients taking 75mg/day clopidogrel compared to those ingesting 325 mg/day of aspirin. The Stent Anticoagulation Regimen Study (STARS) also supported the use of thienopyridine derivative antiplatelet substances like ticlopidine and clopidogrel which inhibit platelet fibrinogen binding. The STARS study demonstrated a reduction in stent thombosis in patients treated with 350 mg aspirin/day and 250 mg ticlopidine following stent placement as compared to those managed with coumadin alone or coumadin and aspirin. We chose clopidogrel as opposed to ticlopidine because of its greater anti-thrombotic properties as well as its better patient tolerance.

As alluded to above, the major etiology of acute and subacute stent thrombosis is platelet deposition on stent struts and damaged intima/media following PTA and stent placement. Exuberant platelet deposition takes place within one hour of deployment. By one week stents are covered by a thin layer of thrombus, fibrin, and neointimal cells and by four to eight weeks stents are fully covered by proliferating smooth muscle cells and fibroblasts, making subsequent thrombosis unlikely. Acute stent thrombosis within the first 24 hours is rare. Subacute thrombosis occurs at a median of 5-6 days but is rarely observed after 21 days. Multiple series have found coronary stent thrombosis rates of 0.3 to 18%.

In order to reduce the incidence of stent thrombosis a variety of pharmaceutical manipulations have been tried. The use of clopidogrel and aspirin is one strategy that was outlined previously. The use of abciximab is another. Abciximab (ReoPro, Eli Lilly and Company, Indianapolis, IN) binds to the human platelet glycoprotein IIb/IIIa receptor and inhibits platelet aggregation. It also binds to the vitronectin receptor found on platelets and vessel wall endothelial and smooth muscle cells thus inhibiting wall deposition. The EPIC study demonstrated a significantly lower rate of 30 day coronary stent thrombosis, death, nonfatal myocardial infarction, unplanned surgical revascularization, unplanned repeat PTA, and unplanned stent implantation in patients treated with abciximab bolus (0.25mg/kg over 10-60 minutes) followed by a twelve hour infusion (0.125 micrograms/kilogram/minute; maximum 10 micrograms/minute) vs. those treated with placebo (12.8% vs. 8.3%).

Urokinase (UK) is a protein that promotes the dissolution of thrombi by stimulating the activation of endogenous plasminogen to plasmin which in turn hydrolyzes fibrin. Because of these properties, UK was infused below the lesion prior to crossing it with a wire based upon reading and personal conjecture. Prior to cervical internal carotid angioplasty Lanzino, et al suggest a 150,000 unit UK infusion at 50,000 units/minute. The aim of this infusion is to both dissolve any debris from the plaque face which might have a propensity to break free when the wire and the balloon/stent system are advanced across the stenosis and to possibly increase the available working vessel lumen. A UK infusion, however, runs the risk of creating emboli from a plaque. It also may increase the risk of a hemorrhagic event especially when a thrombolytic agent is combined with a glycoprotein IIb/IIIa receptor blocker (Lange, personal communication). Finally, and most importantly, because of its effects on plasminogen UK also has procoagulant properties which may be mediated through release of clot bound thrombin and/or through platelet activation. The latter may occur through platelet activation. The latter may occur through UK's conversion of plasminogen to plasmin which in turn activates factor V to Va which in turn helps to accelerate the generation of thrombin which activates platelets, converts fibrinogen to fibrin, and converts factor XIII to XIIIa which cross links fibrin monomers. The Thrombolysis and Angioplasty in Unstable Angina (TAUSA) Trial found that abrupt closure of lesions prior to angioplasty was higher in the group receiving prophylactic UK than in the group receiving prophylactic UK thank in the group given placebo (15% vs 5.9%). In view of the finding from the TAUSA trial UK administration is controversial. Though tolerated in this patient, our procedure has subsequently been modified to exclude this step in routine practice.

Heparin infusion was maintained throughout the procedure with ACT kept between 200 and 250 seconds. Acts greater than 300 seconds are associated with increased risk of non-intracranial hemorrhage when patients are also being infused with abciximab. Following completion of the procedure the heparin infusion was stopped and the drug's effects were allowed to wear off gradually. Most acute and subacute stent failures are secondary to platelet aggregation and as such, post procedure heparinization is not necessary.

Patient Follow-up

Assuming the stent remains patent for the initial 30 days following implantation, we must consider the risks of delayed stenosis. There is inadequate literature concerning intracranial stent patency simply because of the paucity of cases. In the cardiology literature, intimal hyperplasia has been shown to be the major component of late lumen loss after stent implantation. Risk factors for restenosis include stent implantation at a site that stenosed following a previous angioplasty, use of multiple stents regardless of whether or not they overlap, extent of residual stenosis following initial stenting, stent placement in a totally occluded vessel, diabetes mellitus, and stent placement in a vessel with a normal diameter less than 3.0mm.

Intimal hyperplasia follows a fairly predictable time course. Serial angiographic studies in humans demonstrate that the greatest smooth muscle proliferation occurs between one and three months with only a small number of stents exhibiting further narrowing between six and twelve months. These findings are supported by experimental work that has shown late replacement of proliferating smooth muscle cells by relatively inactive ground matrix and fibrosis.