Abstract
Background. Non-atherosclerotic internal carotid artery (ICA) lesions are uncommon. The incidence of ICA dissections is estimated to be 2.5 to 3 per 100,000, with a 10% annual recurrence rate of thromboembolic. The incidence of fibromuscular dysplasia (FMD), the most common non-traumatic, non-atherosclerotic, non-inflammatory lesion of the ICA is estimated to be 0.02%, with a more benign natural history. For both pathologies, antithrombotic therapy and blood pressure control are initial therapeutic modalities, surgery only being indicated after ischemic events or aneurysmal degeneration. Herein, we report our experience of a combined conventional and endovascular covered stent treatment of these patients under reversed flow and their respective long-term results.

Patients and methods. From February 1999 to June 2005, 13 patients and 16 ICAs were revascularized using a hybrid conventional/endovascular technique. The patients’ age ranged from 30 to 78 years (mean 48 years). All patients have had neurological symptoms. All patients were treated with a covered stent introduced via a cervical approach.

Results. Two patients suffered from a perioperative transient neurological deficit. Duplex revealed ideally patent prosthesis, both patients fully recovered, with no changes visible on repeat CT scans. One patient developed a recurrent laryngeal nerve weakness perioperatively. However, no perioperative strokes or deaths were observed. The follow-up ranged from 1 to 68 months (mean = 43 months). No thromboembolic neurological events, graft occlusions or hemodynamically significant stenoses were seen during the observation period.

Conclusion. Stent grafting of non-atherosclerotic ICA lesions is safe with low perioperative morbidity and good long-term patency.

Introduction
The current incidence of stroke in Europe and the United States is about 200 per 100,000 population per annum.^1,2 Of these, about 20% are due to atherosclerotic internal carotid artery (ICA) lesions.³ For patients with symptomatic high-grade ICA lesions, it has been proven that surgery significantly reduces stroke recurrence, compared to best medical therapy alone.^4,5
Non-atherosclerotic ICA lesions are much more uncommon. The incidence of ICA dissections is estimated to be 2.5 to 3 per 100,000 with a 10% annual recurrence rate of thromboembolic events.^6–8 The incidence of fibromuscular dysplasia (FMD), the most common non-traumatic, non-atherosclerotic, non-inflammatory lesion of the ICA is estimated to be 0.02%⁹ with a more benign natural history.10 For both pathologies, antithrombotic therapy and blood pressure control are initial therapeutic modalities. Only patients with complications (recurrent ischemic events with high grade stenosis, aneurysmal degeneration of the ICA) after at least 6 months of follow-up are considered surgical candidates.¹¹ The reason for this hesitant surgical approach to these ICA pathologies may well be the unfavorable results. Perioperative stroke rates of 10%, peripheral cranial nerve injury rates of 58%, and mortality rates of 2% in surgically managed patients are reported.¹¹

Recently, we reported on endovascular treatment strategies with covered stents in patients with non-atherosclerotic ICA lesions (FMD and primary dissections) with excellent long-term results and low perioperative morbidity and no mortality.^12,13 Herein, we report our experience using a combined conventional and endovascular covered stent treatment in these patients under reversed flow and their respective long-term results.

Patients and Methods
Patient selection. Symptomatic and asymptomatic patients referred to our department with primary dissection or FMD of the ICA were initially screened with duplex sonography of the carotid and vertebral arteries, accompanied by a magnetic resonance angiography. In inconclusive cases, a digital subtraction angiography was performed. In cases of aneurysm formation, a CT scan (computed tomography) was initiated to evaluate maximum diameter and length of the aneurysm. After assessing all patients neurologically, as well as the morphology of the lesions, antiplatelet therapy was commenced or continued.

The indication for operation was given in patients under antiplatelet therapy with recurrently symptomatic high-grade stenosis measured by flow velocities and DSA as well as aneurysm formation and enlargement during follow up.

Patients
From February 1999 to June 2005, 13 patients and 16 ICA were revascularized using a hybrid conventional/endovascular technique (Table 1).

All patients with FMD were female; two patients with dissection were female, four male ages 30–78, with a mean of 48 years). All patients reported had a history of neurological symptoms.

Operative procedure
After transverse skin incision of the neck, and administration of 5000 international units of heparin, the common carotid artery (CCA), external carotid artery (ECA) and the ICA were dissected free with minimal manipulation. Subsequently, a reverse blood flow in the ICA was induced by clamping the CCA and ECA. After arteriotomy, an introducer sheath was inserted into the CCA. Under fluoroscopy, a guidewire was advanced through the pathologically altered ICA up to the siphon. A 5 cm x 6 mm PTFE-HEMOBAHN (WL Gore, Flagstaff, Arizona, US) endoprosthesis was deployed into the ICA, overlapping the diseased portion of the artery. Subsequently, a 6 mm dilatation balloon was inserted and the graft was molded to the ICA. At the end of this procedure, blood was drawn from the side port of the introducer in order to remove intravascular debris. Finally, a digital subtraction angiography with the guidewire in place was performed. Before removing the introducer system, it was flushed retrogradely and the arteriotomy of the CCA was closed with interrupted sutures (Figure 1).

Results
Perioperative results. Twelve patients and 15 ICAs were operated on using general anesthesia, one patient and ICA were operated on in loco-regional anesthesia. Two patients suffered from a perioperative transient neurological deficit. After skin closure, weakness of the upper extremity, corresponding with the operated ICA was noted in both patients. Duplex revealed ideally patent prosthesis, both patients fully recovered, not showing any changes on repeat CT scans.

One patient perioperatively developed a recurrent laryngeal nerve weakness. The symptoms gradually dissolved within 1 month of follow up. No perioperative strokes or deaths were observed.

Follow-up
Patients were followed up 1, 3, 6 and 12 months postoperatively. Thereafter, routine follow-up with neurological evaluation and duplex sonography was performed once every 12 months from 1–68 months, with a mean of 43 months.

No thromboembolic neurological events, graft occlusions or hemodynamically significant stenoses were seen during the observation period. One patient was lost to follow-up after 6 months (left the country), and one patient died 3 years postoperatively from an unrelated cause.

Discussion
The prevention of stroke in young patients is the aim of surgical therapy of non-atherosclerotic ICA pathologies. This is an especially challenging task, as experience with patients with these pathologies is quite limited and surgical techniques are much more demanding compared to endarterectomy of atherosclerotic ICA lesions.^11,14

Surgical treatment options of symptomatic patients include resection and interposition of the diseased artery, endarterectomy with patch grafting, gradual internal dilatation, open angioplasty and stent grafting.^14–17

Surgical treatment of symptomatic carotid artery lesions due to atherosclerosis is a well-established method and confers significant reduction of stroke and death compared to medical therapy alone in patients with ICA lesions. To date, ICA stenting of high-risk patients with atherosclerotic stenosis, however, has only been shown to be non-inferior to surgery.¹⁸ Yet, patients in both treatment groups had a worse outcome compared to the natural history of symptomatic ICA lesions under best medical therapy. For these patients, the benefit of intervention has not been shown unequivocally. The main reasons are low morbidity and mortality rates caused by surgery. In experienced hands, surgical morbidity is even much lower¹⁹ than described in the two landmark studies for symptomatic atherosclerotic ICA lesions.^4,5 Non-atherosclerotic ICA lesions, on the other hand, do have similar or superior morbidity and mortality rates when treated medically compared to surgery. Therefore, there appears to be no compelling indication for intervention. However, patients in this series treated by endografting of the diseased artery have excellent perioperative morbidity rates and long-term results. These results are superior to the natural history of these pathologies and could possibly change indications for intervention for symptomatic dissections and FMD.

ICA dissections and FMD usually affect the middle and distal portion of the ICA at the level of the first and second cervical vertebrae.²⁰ Therefore, revascularization with covered stents is an appealing option, as all morphological presentations of ICA dissections and FMD can be treated with one surgical technique.

For our patients, a carotid approach was chosen, as we believe that the device has to be implanted with a retrograde flow of the ICA in order to avoid periprocedural embolization. This strategy is especially important, as mixed thrombi are frequently adherent in dissections and aneurysms, potentially embolizing during passage of a catheter in a prograde flow. It has been described that during all steps of carotid artery stenting, microembolizations occur^,21,22 potentially causing stroke. Also, at the time of treating our first patients, we did not have access to a Parodi balloon-occluding sheath or similar devices.

For non-atherosclerotic ICA lesions, a polytetrafluorethylene (PTFE) covered stent is an ideal therapeutic option, as the diseased portion of the vessel is in the transition zone of the ICA, where the elastic type of the artery changes into the muscular type,²³ usually a few centimeters distal to the carotid bulb. Therefore, a cylindrically-shaped graft can be used as no caliber changes are noted in this particular area. Therefore, aneurysm exclusion can be performed equally as safe as stenting stenosis.

Two patients suffered from an intraoperative neurological deficit, affecting the upper extremity corresponding to the operated carotid artery. The symptoms only lasted a few hours; patients did not have any CT changes. One patient was the first one operated on with this technique; the other had a long clamping time with multiple catheter passages and intraoperative angiograms. One patient suffered from transient recurrent laryngeal nerve palsy, but no residual symptoms one month postoperatively. This is a rare complication, and in our own series of carotid endarterectomy, applying a similar skin incision and more surgical exposure of the carotid arteries, morbidity occurred only once in 180 operated arteries.²⁴

Long-term durability of reconstructions of non-atherosclerotic lesions is imperative, especially considering that the mean age of these patients is below 50 years of age with no relevant comorbidities. The long-term results in our patients treated with a covered stent are excellent with no symptomatic recurrences or hemodynamically relevant stenosis during follow up.