Carotid thromboendarterectomy for recent total occlusion of the internal carotid artery☆☆☆

Article Outline

• Abstract
• Methods
• Results
• Discussion
• Discussion
• References
• Copyright

Abstract 

Background: The efficacy of emergency carotid thromboendarterectomy (CTEA) for acute internal carotid artery (ICA) thrombosis has been questioned. We evaluated the use of CTEA in patients with recent ICA occlusion. Methods: From August 1989 to December 1999 patients who underwent urgent CTEA for recent ICA thrombosis were retrospectively evaluated. Patient data analyzed included age, sex, comorbid risk factors, diagnostic evaluation, operative procedure, and long-term follow-up with clinical assessment and carotid duplex scan. Neurologic status was evaluated with the Modified Rankin Scale (MRS) before the operation, immediately after the operation, and at 3- to 6-months' follow-up. Results: Twenty-nine patients underwent emergency ipsilateral CTEA for acute ICA thrombosis over the last 10 years. The average age of the patients was 69.9 ± 1.7 years, and 66% were men. Patient risk factors included diabetes (7 [24%]), hypertension (21 [72%]), coronary artery disease (8 [29%]), and history of tobacco abuse (20 [69%]). Presenting symptoms included cerebrovascular accident (7 [24%]), transient ischemic attack (nonamaurosis) (10 [34%]), crescendo transient ischemic attack (7 [24%]), stroke in evolution (2 [7%]), and amaurosis fugax (3 [10%]). Diagnostic evaluation included computed tomographic scan (29 [100%]), magnetic resonance imaging/magnetic resonance angiography (4 [14%]), duplex scan evaluation of the carotid arteries (23 [79%]), and cerebral angiography (18 [64%]). Antegrade flow in the ICA was successfully established in 24 (83%) of 29 patients and confirmed with intraoperative angiography or duplex sonography. Postoperative morbidity included 2 hematomas (7%), 4 transient cranial nerve deficits (14%), and 1 conversion to hemorrhagic stroke (3.6%), which resulted in the only death (3.6%). MRS scores averaged 3.4 ± 0.2 preoperatively. Follow-up averaging 74.1 ± 21 months (range, 3-140 months) was obtained in 27 (93%) patients. Improvement or deterioration was defined as a change in MRS ± 1. Immediately postoperatively, 14 (48%) patients were improved, 2 (7%) deteriorated, and 13 (45%) had no change. At 3 to 6 months, 20 (74%) of 27 patients were improved, seven (26%) had no change, and none deteriorated. Of patients with successful CTEA, 23 (96%) of 24 had a patent ICA on follow-up duplex scan evaluation, and there was no evidence of recurrent ipsilateral neurologic events at an average of 49 months. Conclusion: These data support an aggressive early surgical intervention for acute ICA thrombosis in carefully selected patients. In the previous decade we reported a 46% success rate for establishing antegrade flow in the ICA long term. Data from this decade show a 79% (P = .0114) success rate for establishing antegrade flow long term in all patients undergoing emergency CTEA. New and improved imaging modalities have allowed better patient selection, resulting in improved outcomes. (J Vasc Surg 2001;33:242-50.)

The efficacy of carotid thromboendarterectomy (CTEA) for acute carotid occlusion has not been established. Although several randomized studies regarding the management of symptomatic and asymptomatic patients with hemodynamically significant internal carotid artery (ICA) stenosis have validated the use of operative endarterectomy, the same conclusion has not been attained for patients with ICA occlusions.¹, ², ³

ICA occlusions usually occur at a point of critical stenosis in the proximal ICA. Often the thrombus will propagate cephalad to where patency of the ICA can be maintained by collateral flow, commonly occurring at the level of the ophthalmic artery. Reviews of the surgical treatment for ICA occlusions have been historic and have shown mostly dismal results.⁴, ⁵, ⁶, ⁷

Patients with acute ICA occlusions can have a variable presentation ranging from asymptomatic to profound neurologic impairment and death. In addition, long-term patient neurologic outcome is unpredictable. The retrospective reviews of operative intervention in patients with acute ICA occlusions and neurologic events have often shown worse outcomes when compared with medical management of matched patients.⁴, ⁵, ⁶, ⁷ However, a few reports, including our previously published data, have shown a benefit to urgent CTEA in appropriately selected patients.⁸, ⁹, ¹⁰, ¹¹, ¹² In this article we review our past decade of experience in the management of patients with acute ICA occlusion.

Back to Article Outline

Methods 

Twenty-nine patients presenting with acute neurologic symptoms were thought to have recent total occlusion of the extracranial ICA from August 1989 to December 1999. Urgent CTEA was defined as operative intervention within 8 days of presentation. Total occlusions that occurred after carotid bifurcation endarterectomy, trauma, dissection, or angiographic procedures were excluded, as were isolated occlusions of the common carotid artery (CCA). Patients exhibiting a decreased level of consciousness or lethargy were not considered for CTEA.

Patient data analyzed included age, sex, presenting symptoms, neurologic status, comorbid risk factors, diagnostic evaluation, operative procedure, intraoperative findings, and long-term follow-up. Comorbid risk factors evaluated included coronary artery disease (CAD), peripheral vascular disease, prior cerebral vascular events (transient ischemic attack [TIA] or cerebrovascular accident [CVA]), hypertension, diabetes, renal insufficiency (creatinine level > 1.5 mg/dL), chronic obstructive pulmonary disease (COPD), and tobacco use.

Neurologic status was based on the Modified Rankin Scale (MRS). This scale consists of six grades, from 0 to 5, with 0 corresponding to no symptoms and 5 corresponding to severe disability (Table I).

Table I. Modified Rankin Scale

Neurologic status was scored preoperatively, immediately postoperatively, 3 to 6 months postoperatively, and then yearly. Improvement or deterioration was defined as a change in the MRS ± 1. Neurologic status at the time of this report was confirmed with each patient through a telephone interview.

Diagnostic imaging studies were performed with computed tomography (CT) (Siemens Somatom Plus/Plus S; Siemens, Iselin, NJ); duplex ultrasonography (DU) (Acuson; Acuson, Mountainview, Calif; or ATL HDI 3000; Advanced Technology Laboratores, Bothell, Wash); magnetic resonance imaging/magnetic resonance angiography (MRI/MRA) (Picker Vista HPQ 1.oT; Picker International, Cleveland, Ohio, or GE Signa Horizon LX1.5T; GE Signa Horizon, Milwaukee, Wis); and catheter angiography (CA) (Siemens Multistar TOP; Siemens, or Philips Maximus C1250; Philips, Shelton, Conn). CT scans of the head were performed with a standard protocol to evaluate for cerebral infarction or hemorrhage. The protocol consisted of 4-mm images through the posterior fossa, then 8 × 8-mm images through the vertex. DU studies were performed in a laboratory accredited by the Intersocietal Commission for the Accreditation of Vascular Laboratories, by the use of color duplex scanners (Acuson or ATL HDI 3000; Aucson or Advanced Technology Laboratories) with standard techniques. The CCA, ICA, and external carotid artery (ECA) were each evaluated. Strandness criteria, which have been internally corroborated at this institution, were used to determine stenosis or occlusion.

MRI and MRA were used to evaluate the carotid system, the brain, or both. Standard protocol to evaluate for cerebral infarction included sagittal and axial T1 images, axial T2 images, axial proton density or fluid attenuated inversion recovery. In more recent cases axial diffusion–weighted images have been used. The protocol to evaluate the carotid and intracranial arterial system included two-dimensional and three-dimensional time-of-flight MRA through the neck, three-dimensional time-of-flight through the head, and more recently, gadolinium bolus fast spoiled gradient recalled echo MRA through the arch and neck.

CA was done with the digital subtraction technique. This consisted of four-vessel angiography and evaluation of the intracranial arterial system. The subclavian or vertebral arteries were injected only in select cases depending on the results of the arch injection. All of these studies were performed in the interventional radiology suite with multiplanar imaging.

The anesthetic technique included the use of regional cervical block or general anesthesia. Stringent control of perioperative hypertension was used for patients during their hospitalization. Cerebral perfusion was evaluated for patients who had general anesthesia with computer-assisted electroencephalogram (Neurotrac) monitoring. Standard positioning and operative technique for carotid endarterectomy (CEA) was used. All patients were heparinized with an 80 U/kg bolus. Control of the CCA and ECA only was obtained initially. Next, an arteriotomy was performed, and if the thrombus could be removed locally or backbleeding expelled the thrombus, then control of the ICA was obtained. If backbleeding was not achieved or deemed to be inadequate, then a Fogarty thrombectomy catheter was used on a selective basis with 2F or 3F balloons. Shunting was performed on a selective basis. Arteriotomies were repaired primarily or with polyester fiber (Dacron) patch angioplasty. Intraoperative evaluation of the endarterectomized ICA was performed with either duplex sonography or angiography. All patients were observed in an intensive care unit setting for at least the first 24 hours postoperatively. All patients were given an infusion of dextran (20%/500 mL) at a rate of 20 cc/h for 24 hours postoperatively. Blood pressure was monitored continuously with arterial catheterization and maintained at 130 to 160 mm Hg systolic. Mannitol at 0.5 to 1 g/kg every 6 hours for 24 to 48 hours and dexamethasone at 10 mg intraoperatively and 4 mg every 6 hours for 24 to 48 hours became standard neuroprotective therapies over the study period.

Kaplan-Meier tables were created in accordance to the revised recommendations of the Ad Hoc Committee on Reporting Standards for the Society for Vascular Surgery/International Society for Cardiovascular Surgery. Intergroup comparisons of proportional data were made with χ² analysis. Differences were considered significant if the P value was less than .05.

Back to Article Outline

Results 

Twenty-nine patients were identified as having undergone urgent CTEA for acutely symptomatic ICA occlusion. The average age of these patients at presentation was 69.9 ± 1.7 years (range, 48-84 years). There were 10 women (34%) and 19 men (66%). Comorbid risk factors in these patients included CAD in 8 (28%), prior cerebral vascular event in 1 (3.4%), hypertension in 21 (72%), diabetes mellitus in 7 (24%), renal insufficiency in 1 (3.4%), COPD in 2 (7%), and tobacco use in 20 (69%) (Table II).

Table II. Perioperative risk factors

Presenting symptoms included CVA in 7 (24%), single TIA in 10 (34%), crescendo TIA in 7 (24%), stroke in evolution in 2 (7%), and amaurosis fugax in 3 (10%) (Table III).

Table III. Clinical presentation

TIA was defined as a single ipsilateral hemispheric neurologic event in which the deficit resolved within 24 hours. Crescendo TIA was defined as multiple similar ipsilateral hemispheric neurologic events with the deficit continuing to resolve within 24 hours. Any ipsilateral hemispheric neurologic event that lasted more than 24 hours was termed a CVA. The preoperative MRS score was 3.4 ± 0.2 (range, 1-5). The average time from onset of symptoms to endarterectomy was 39 hours (range, 5-192 hours; median, 22.5 hours). Neurologic exclusion criteria for urgent CTEA included depressed level of consciousness, profound neurologic deficit, and significant lethargy.

Diagnostic imaging varied over the 10-year study period. In the studies CT scan, MRI, or both of the head and imaging of the cervical carotid circulation were used in all 29 patients. Preoperative CT scans were obtained in all 29 patients. CT scan exclusion criteria for urgent CTEA include evidence of intracranial hemorrhage, significant cerebral edema, hemispheric asymmetry, or cerebral infarct of 1 cm or more. Only five patients (17%) had evidence of acute infarction on CT. MRI was obtained preoperatively in only four patients. Exclusion criteria for CTEA with MRI included intracranial hemorrhage, an infarction size of 2 cm or more, significant cerebral edema, or hemispheric asymmetry. Carotid imaging was obtained by means of MRA in four patients (14%), DU in 23 patients (79%), and CA in 18 patients (64%).

Fogarty embolectomy catheters (2F and 3F) were needed in 14 cases (48%) to help reestablish backbleeding and allow for antegrade flow. This procedure was performed under fluoroscopic guidance to avoid injury to the carotid siphon.

Intraoperative findings included the identification of intraplaque hemorrhage in 25 cases (86%). A thrombus in the cervical ICA was noted in 25 cases (86%). The arteriotomy was repaired primarily in 25 cases (86%) and with a woven polyester fiber (Dacron) patch in four (14%). Backbleeding was established in 24 patients (83%).

Antegrade flow in the ICA was confirmed intraoperatively with the use of DU in 11 cases (46%) and with angiography in 13 cases (54%). The ICA was ligated in five patients (17%). Two of these patients had small atretic ICAs that appeared to be occluded for a long time. In three patients we could not establish backflow. This subgroup of patients had an external CEA with primary closure performed in conjunction with ICA ligation. Intra-arterial thrombolysis was not attempted. These five patients presented with TIA (2 [40%]), reversible ischemic neurologic deficit (2 [40%]), and CVA (1 [20%]).

Immediate postoperative morbidity included 1 case of tachyarrhythmia (3.4%), 2 neck hematomas (7%), 2 transient cranial nerve injuries (7%), 1 CVA (3.4%), and 1 conversion to hemorrhagic stroke (3.4%), which resulted in the only death (3.4%). There were two neck hematomas (7%); one required operative evacuation (3.4%) (Table IV).

Table IV. Perioperative morbidity and mortality

The one postoperative CVA was in a patient who had the ICA ligated for what appeared to be an atretic, chronically occluded vessel. This patient had a more profound upper extremity hemiplegia. This was resolved by a 3-month follow-up evaluation. The cranial nerve injuries were all transient and included one to the marginal mandibular nerve and one to the hypoglossal nerve. Eighteen patients (62%) received mannitol and dexamethasone as neuroprotective adjuncts. This practice was instituted over the second half of the study period and is now used routinely. The average length of postoperative stay was 3.2 days (range, 1-8 days). Twenty-four patients were discharged home, two were discharged to a skilled nursing unit, and two were discharged to an extended care facility.

Patient follow-up, which averaged 74.1 ± 21 months (range, 3-140 months), was obtained in 27 patients (93%). One patient was lost to follow-up after 1 month, and the other patient died 5 days postoperatively. The immediate postoperative MRS showed improvement in 14 patients (48%), no change in 13 (45%), and deterioration in 2 (7%). At 3 to 6 months postoperatively, 20 (74%) of 27 patients were improved, and seven (26%) of 27 were unchanged. No patient followed up beyond 1 month showed any deterioration (Table V).

Table V. Neurologic events

In the 24 patients who had undergone successful CTEAs, the extracranial ICA remained patent with antegrade flow in 23 (96%) patients on follow-up DU at an average of 49 months (range, 1-168 months) (Table VI).

Table VI. Patency

In addition, none of the 24 patients had evidence of ipsilateral neurologic events during that time. There were four deaths (14%) in the follow-up period. Two deaths were due to myocardial infarctions, and two were unexplained. These latter patients died in their sleep, and the request for autopsies was denied. One of the unexplained deaths occurred in a patient whose attempt at revascularization was unsuccessful. The ICA had been ligated at the time of surgery, and this patient died 46 months postoperatively. In the 29 patients undergoing CTEA, long-term patency was achieved in 23 (79%). This was an improvement compared with our previous series in which antegrade flow on follow-up was achieved in 11 (46%) of 24 patients (P = .0114).

Back to Article Outline

Discussion 

CEA has been validated as efficacious for stroke prophylaxis in symptomatic and asymptomatic carotid artery stenosis.¹, ², ³ CEA reduces the mortality rates and increases stroke-free survival in symptomatic patients if it is performed in a patient population with a perioperative morbidity rate below 5.8%.¹ Over the past 35 years the role of CEA in the setting of acute stroke has not been as clear. The intracranial risks of reperfusion and hemorrhagic conversion temper the application of CEA in the setting of an acute cerebrovascular accident.⁴, ⁵, ⁶, ⁷, ¹³

Although some patients with occlusion of the ICA can have a catastrophic brain injury, others have only a TIA or a minor stroke. In other cases there are no immediate symptoms, and the occluded artery is found incidentally. This wide range of clinical consequences confirms that multiple factors influence the outcome and natural history of an occluded ICA. Nicholls et al¹⁴ followed 24 patients over a 7-year period with unilateral carotid occlusion. The occlusions were associated with a neurologic event at presentation in 46% of patients. At a mean follow-up of 39 months after identification of the occlusion, the neurologic event rate was 20% per year. Cote et al¹⁵ prospectively followed 47 patients with unilateral carotid artery occlusion for an average of 34 months. These patients presented with neurologic symptoms from none to mild. The neurologic event rate over 34 months was 23.5%. Two thirds of these events were ipsilateral to the occluded artery. The stroke rate distal to the occluded artery was 5% per year, whereas 51% of patients experienced ipsilateral TIAs during the study period.

Strokes and TIAs after total occlusion of the ICA have been attributed to four main pathogenic mechanisms: (1) emboli into the circle of Willis from the distal end of the occlusive thrombus; (2) emboli originating from a residual internal artery stump through the ECA; (3) emboli from ipsilateral CCA, ECA, or aorta through collateral channels; or most likely, (4) hypoperfusion distal to the occlusion.¹³, ¹⁶, ¹⁷, ¹⁸, ¹⁹ The capacity of the cerebral circulation to develop collateral flow depends on autoregulation, which is dependent on native cerebral collateral anatomy. Impairment of autoregulation leads to chronic cerebral hypoperfusion, and therefore, neurologic function becomes susceptible to changes in blood pressure and cardiac output.¹⁶ Grubb et al¹⁸ have used positron emission tomography to delineate a subgroup of patients at risk for this type of cerebral hemodynamic failure. Their data showed a 15.8% incidence of ipsilateral stroke at 2 years for all patients. In patients identified as having cerebral hemodynamic failure as diagnosed with increased oxygen extraction measured with positron emission tomography scan, the incidence of ipsilateral neurologic event was 26.5%. This study further defined a subgroup of patients who would benefit from revascularization.

The role of surgical therapy to improve the natural history of ICA occlusion has been visited previously.⁴, ⁵, ⁶, ⁷, ⁸, ⁹, ¹⁰, ¹¹, ¹² In the 1960s and 1970s, before CT, in multiple series there were reports of surgical morbidity and mortality rates of 16% to 51% for thromboendarterectomy of the occluded ICA.⁴, ⁵, ⁶, ⁷ These studies also identified a subgroup of patients who fared better than the group as a whole, albeit without the sophistication of Grubb et al.⁴, ⁵, ⁶, ⁷, ²⁰, ²¹, ²², ²³, ²⁴ This subgroup included patients who presented early in their course and without profound neurologic deficits. The selection of this subgroup has been further delineated with the newer imaging modalities made available over the last decade. The routine use of CT and MRI have helped identify patients with dense infarcts versus those with a potentially recoverable ischemic penumbra. Improvements in contrast angiography, DU, and MRA have helped differentiate chronic from acute carotid occlusive disease. The improvements in contrast angiography, especially digital subtraction angiography, have provided a better assessment of the carotid bifurcation, intracranial arterial anatomy, and patency of the carotid siphon. The B-mode imaging definition of current ultrasound scan devices has significantly improved over the last 15 to 20 years. Thrombus that appears retracted and brightly echogenic is undoubtedly chronic. Also, patients with a small atretic-appearing ICA, in which the ICA is disproportionately smaller than the CCA or ECA, are likely to have a chronically occluded ICA. These factors have allowed for better preoperative patient selection. A patent intracranial ICA in the supraophthalmic region is predictive for the successful establishment of patent antegrade flow. MRA is helpful in demonstrating patency in this portion of the distal ICA. Intraoperative angiography should be performed after the establishment of antegrade ICA flow. This allows for the identification of intracranial atherosclerotic disease, which may affect long-term patency. This was predictive for the one postoperative occlusion in this series. When compared with our previous series, our long-term patency improved from 46% (11/24) to 79% (23/29; P = .0114). We think this is a result of improved patient selection. Experience and progress in imaging enhancement have provided improved patient selection over the last 20 years.

The Rankin scale was the first comprehensive functional assessment scale published for use on stroke survivors. It is an ordinal scale that provides an overall estimate of the level of functional dependence in a given stroke survivor. The MRS measures independence rather than performance of specific tasks.²⁵ Newer, more sophisticated indexes, such as the National Institutes of Health stroke scale, may be more appropriate for prospective studies. However, because our study was retrospective, we used the MRS as an objective assessment.

These results support the concept that carotid artery surgery may be performed safely in properly selected patients with acute ICA occlusion. It is thought that ongoing hypertension and cerebral edema are the primary contributing factors resulting in the conversion of a bland infarct to a hemorrhagic infarct.⁶, ⁷, ¹³, ¹⁹, ²⁶ Intensive patient management with precise control of hypertension, anticoagulation, and adjunctive means of reducing intracranial pressure is a means of diminishing the risk of hemorrhagic conversion. We have used mannitol and steroids (dexamethasone) to decrease intracranial pressure, to act as oxygen-free radical scavengers, and to provide membrane stabilization. We think that these therapies have been useful neuroprotective adjuncts in the postoperative care of these patients.

We previously reported a series of 24 patients undergoing surgical exploration for suspected recent total occlusion of the ICA. Antegrade flow was established in 15 (63%). There was no aggravation of preoperative neurologic deficits, there were no hemorrhagic conversions, and there were no operative deaths.⁸ McCormick et al⁹ reviewed 42 patients with surgically managed symptomatic ICA occlusion. Antegrade flow was established in 24 (56%), there was one postoperative neurologic event (2.3%), and one patient had worsening of a preoperative neurologic deficit (2.3%). There were no hemorrhagic conversions and no postoperative deaths.⁹ The one hemorrhagic conversion in this series resulted in the only death. This patient had a dense neurologic deficit and significant cerebral edema on CT. The patient did not undergo CTEA until 8 days after presentation, at the recommendation of the referring neurologist. This represented an aggressive approach to a high-risk patient, which was a violation of our protocol. This supports the need for careful patient selection. Urgent CTEA should not be attempted in patients with profound neurologic deficits or a depressed level of consciousness. Table VII summarizes the outcomes of the few modern series in which surgical restoration of ICA flow after recent ICA occlusion has been reported.

Table VII. Outcome of patients for surgical restoration of the ICA flow in six series*

ICA, Internal carotid artery; ICH, intracerebral hemorrhage; NA, not available.

These data support an aggressive approach to a select subgroup of symptomatic patients presenting with suspected recent total occlusion of the ICA. Proper identification should include patients presenting with mild or improving neurologic deficits, no evidence of bland or hemorrhagic infarct on CT, and a suggestion of recent ICA thrombosis with a normal-appearing patent distal ICA. The preoperative use of modern cerebral imaging techniques and the stringent perioperative control of hypertension, anticoagulation, and intracerebral pressure have virtually eliminated the risk of hemorrhagic conversion in appropriately selected patients (Fig 1).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 1. 

  Preoperative flowchart. ICA, Internal carotid artery; MRA, magnetic resonance angiography.

Urgent CTEA is technically easier when performed early and before thrombus organization and adherence (Fig 2).

• 
  • View Large Image
  • Download to PowerPoint

• Fig. 2. 

  Intraoperative flowchart. CEA, Carotid endarterectomy; ICA, internal carotid artery.

With proper selection and technique, urgent surgical treatment for ICA occlusion has a better long-term neurologic outcome than as shown in present natural history studies. These data show that urgent CTEA for suspected recent total occlusion of the ICA is safe and improves stroke-free survival in a selected group of patients.

Back to Article Outline

Discussion 

Dr Jerry Goldstone (Cleveland, Ohio). At the 31st annual meeting of this society in 1983, Welling and his colleagues presented a similar series of carotid explorations in 24 asymptomatic patients with internal carotid artery occlusion. These operations were performed between 5 and 21 days after the onset of the neurologic deficit. Successful restoration of flow within the internal carotid artery was accomplished in 63% of the cases with no perioperative new strokes or mortality. So, what, if anything, is different about the current series and what has been learned in the last 20 years? Certainly the current series of patients had better documentation not only of their neurologic status but also of the status of their arteries and brains because of the availability and use of contemporary imaging techniques. Reestablishment of flow in the internal carotid artery was even more successful, being achieved in 83% in this series compared with 63% in the earlier one with only one stroke and that occurred in a patient in whom flow could not be reestablished. This patient had a dense neurologic deficit that was 8 days old at the time of the operation. I suspect that a similar patient would not be operated on. The authors believe, as I do, that acute profound neurologic deficits are a contraindication to carotid surgery. Clearly, the most important feature in attaining results of this caliber is the careful selection of patients and the proper timing of the operation. The authors emphasize this in their manuscript. This leads to my first question: The interval from onset of symptoms to operation ranged from 5 to 192 hours. Frankly, I don't consider an operation performed 8 days into an illness as being urgent, but nevertheless, what was the distribution of preoperative intervals according to patient categories, and when is the optimum time to proceed?

I have two additional comments and questions:

Several technical points about the operations deserve emphasis. The authors were careful not to clamp the internal carotid artery until after patency was reestablished. This avoids the potentially disastrous complication of fracturing the thrombus and permitting distal embolization. They also used small embolectomy catheters to extract thrombus from the distal extracranial internal carotid artery. Although this is often said to be an ill-advised thing to do, they did it under fluoroscopic control so that they knew where the tips of the catheters were at all times. Parenthetically, this is another example of the benefits of using intraoperative digital fluoroscopy. Lastly, they used intraoperative imaging, duplex ultrasound, or arteriography to ensure an adequate technical results in each of these cases.

Perhaps more important was the careful attention to control of blood pressure in the perioperative period. This series and the previous one clearly debunk the myth that reperfusion alone converts anemic into hemorrhagic cerebral infarction. This is no doubt, however, what reperfusion plus hypertension does.

Finally, I believe the most important lesson from this paper is that there have been no late strokes in this series of patients after an average follow-up of 74 months. This stands in stark contrast to the natural history of internal carotid occlusion for which the subsequent ipsilateral stroke rate ranges from 2% to 6% per year. I have long considered carotid occlusion to be a another contraindication to carotid surgery, but occlusion of the internal carotid artery is clearly not the end of the story for a substantial number of patients. In a prospective, blinded longitudinal cohort study of 81 symptomatic patients with internal carotid artery occlusion studied with PET scans, increased cerebral oxygen extraction or what is known as stage two hemodynamic failure, was identified as an independent risk factor for subsequent ischemic strokes that occurred at an annual rate of greater than 10%. Although the current study did not include PET scan data, I suspect that many of the patients would meet these criteria and that this explains the long-identifiable subgroup of patients for whom early, if not urgent, carotid thromboendarterectomy could have highly beneficial results.

I appreciate the opportunity to discuss this stimulating work and congratulate Dr Kasper for an excellent presentation.

Dr Gregory C. Kasper. Thank you for your kind comments, Dr Goldstone.

With regard to preoperative intervals, there was quite a variation. Three of 29 patients were operated on at 6, 7, and 8 days postpresentation. This was the result of referral patterns. If you eliminate those three outliers, all patients were operated on within 24 hours of presentation.

The average time to CEA in patients with compelling neurologic events or unstable symptomatology was 38 hours. There was one patient evaluated 7 days postpresentation who had become neurologically unstable after initial discharge. When you eliminate this outlier, the subgroup underwent CEA prior to our involvement at an average 20 hours postpresentation.

There were five patients in whom we were unable to reestablish antegrade carotid flow; two had small, chronically atretic vessels. No attempt was made to open these vessels. In the remaining three patients we were unable to establish carotid backbleeding intraoperatively. We believe this was the result of propagated thrombus to the carotid siphon. It is important to be cautious with thrombectomy catheters in that area to prevent carotid cavernous fistulas.

Regarding your question about the NASCET subgroup, which showed a significant increased stroke rate in patients, if patients present with a tight stenosis and antegrade carotid flow, manipulation of the internal carotid artery can dislodge material and result in an embolic stroke. In an occluded internal carotid artery back pressure supports retrograde flow, which theoretically decreases the risk of emboli. In 52% of our patients, after the carotid arteriotomy, the thrombus was easily expelled into the operative field because of back pressure.

Lastly, with regard to imaging, we attempt to image the intracranial portion of the internal carotid artery below the level of the ophthalmic artery. We use gadolinium-enhanced MRA for this. A few tiny branches of the petrous portion of the internal carotid artery have been described, and these may maintain flow in this portion of the ICA. If patency is demonstrated, there is an excellent chance of establishing antegrade flow in the ICA. I believe our data support this approach.

Dr John M. Porter (Portland, Ore). Let me ask a quick question in the meantime. What size was the balloon catheter, and did you inject contrast into the balloon so you could see it?

Dr Kasper. Yes, this is an excellent point. We diluted the contrast 20% with saline. We perform balloon thrombectomy with a 2F or 3F Fogarty catheter. We only used Fogarty thrombectomy in 48% of the patients in the series.

Dr Porter. And I assume you did not recognize any creation of a carotid cavernous fistula?

Dr Kasper. We did not. In the previous series from our institution, there was one patient with a carotid-cavernous fistula who was treated with a detachable balloon.

Dr M. Ashraf Mansour (Maywood, Ill). My question is, what was the status of the opposite side? How many patients had a contralateral occlusion?

Also, did you have any patients who initially got worse, the stroke got worse, and then got better as they recovered?

Dr Kasper. With regard to status of the contralateral internal carotid artery, there were no patients with contralateral occlusion.

With regard to postoperative neurologic events, there was only one CVA in a patient whose ICA was ligated. This patient had resolution of this neurologic deficit at 3 months. We believe these patients are at risk for watershed infarcts due to decreased or altered cerebral hemodynamics. Early revascularization, if possible, lessens this risk.

Dr Stanley O. Snyder, Jr (Nashville, Tenn). I applaud your efforts here, and I agree with a lot of the things you've actually done.

My question is, once you identify these patients, do you then treat them as an emergency and do them as quickly as you have the tests completed, or do you give the patient heparin and do them electively the next day?

Dr Kasper. When we evaluate these patients, they're already on heparin. It is important to put the diagnostic measures into high gear and, if possible, perform CEA as soon as possible.

It's been shown in previous studies that success is greatest if the thrombectomy and CEA are performed early.

Back to Article Outline

References 

1. North American Symptomatic Carotid Endarterectomy Trial Collaborators . Beneficial effect of carotid endarterectomy in symptomatic patients with high-grade carotid stenosis. N Engl J Med. 1991;325:445–453
   • View In Article
   • MEDLINE
   • CrossRef
2. Executive Committee for the Asymptomatic Carotid Atherosclerosis Study . Endarterectomy for asymptomatic carotid artery stenosis. JAMA. 1995;273:1421–1428
   • View In Article
   • MEDLINE
3. Moore WS, Mohr JP, Najafi H, Robertson JT, Stoney RJ, Toole JF. Carotid endarterectomy: practice guideline: report of the ad hoc committee to the joint council of the Society for Vascular Surgery and the North American Chapter of the International Society for Cardiovascular Surgery. J Vasc Surg. 1992;15:469–479
   • View In Article
   • Full Text
   • Full-Text PDF (1221 KB)
   • CrossRef
4. Blaisdell WF, Clauss RH, Galbraith JG, Imparato AM, Wylie EJ. Joint study of extracranial arterial occlusion. JAMA. 1969;209:1889–1895
   • View In Article
   • MEDLINE
5. Rob CG. Operation for acute completed stroke due to thrombosis of the internal carotid artery. Surgery. 1969;65:862–865
   • View In Article
   • MEDLINE
6. Wylie EJ, Hein MF, Adams JE. Intracranial hemorrhage following surgical revascularization for treatment of acute strokes. J Neurosurg. 1964;21:212–218
   • View In Article
   • MEDLINE
   • CrossRef
7. Bruetman ME, Fields WS, Crawford ES, DeBakey ME. Cerebral hemorrhage in carotid artery surgery. Arch Neurol. 1963;9:458–467
   • View In Article
   • MEDLINE
8. Welling RE, Cranley JJ, Krause RJ, Hafner CD, Arbaugh JJ, Roedersheimer LR. Surgical therapy for recent total occlusion of the internal carotid artery. J Vasc Surg. 1984;1:57–61
   • View In Article
   • Abstract
   • Full Text
   • Full-Text PDF (469 KB)
   • CrossRef
9. McCormick PW, Spetzler RF, Bailes JE, Zabramski JM, Frey JL. Thromboendarterectomy of the symptomatic occluded internal carotid artery. J Neurosurg. 1992;76:752–758
   • View In Article
   • MEDLINE
   • CrossRef
10. Kusunoki T, Rowed DW, Tator CH, Loughweed WM. Thromboendarterectomy for total occlusion of the internal carotid artery: a reappraisal of risks, success rate and potential benefits. Stroke. 1978;9:34–38
    • View In Article
    • MEDLINE
    • CrossRef
11. Hugenholtz H, Elgie RG. Carotid thromboendarterectomy: a reappraisal. Criteria for patient selection. J Neurosurg. 1980;53:776–783
    • View In Article
    • MEDLINE
    • CrossRef
12. Hafner CD, Tew JM. Surgical management of the totally occluded internal carotid artery: a ten-year study. Surgery. 1981;89:710–717
    • View In Article
    • MEDLINE
13. Hardin C, Hendren TH, Faris AA, Poser CM. Pathogenesis of hemorrhagic infarction of the brain, II: protective effect of low molecular weight dextran. Arch Neurol. 1963;9:473–476
    • View In Article
    • MEDLINE
14. Nicholls SC, Bergelin R, Strandness DE. Neurologic sequelae of unilateral carotid artery occlusion: immediate and late. J Vasc Surg. 1989;10:542–548
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (649 KB)
    • CrossRef
15. Cote R, Barnett HJM, Taylor DW. Internal carotid occlusion: a prospective study. Stroke. 1983;14:898–902
    • View In Article
    • MEDLINE
    • CrossRef
16. Klijn CJM, Kappelle LJ, Tulleken CAF, van Gijn J. Symptomatic carotid artery occlusion: a reappraisal of hemodynamic factors. Stroke. 1997;28:2084–2093
    • View In Article
    • MEDLINE
    • CrossRef
17. McIntyre KE, Ely RL, Malone JM, Bernhard VM, Goldstone J. External carotid artery reconstruction: its role in the treatment of cerebral ischemia. Am J Surg. 1985;150:58–62
    • View In Article
    • MEDLINE
    • CrossRef
18. Grubb RL, Derdeyn CP, Fritsch SM, Carpenter DA, Yundt KD, Videen TO, et al.  Importance of hemodynamic factors in the prognosis of symptomatic carotid occlusion. JAMA. 1998;280:1055–1060
    • View In Article
    • MEDLINE
    • CrossRef
19. Faris AA, Hardin CA, Poser CM. Pathogenesis of hemorrhagic infarction of the brain, I: experimental investigations of role of hypertension and of collateral circulation. Arch Neurol. 1963;9:468–472
    • View In Article
    • MEDLINE
20. Giordano JM, Trout HH, Kozloff L, DePalma RG. Timing of carotid artery endarterectomy after stroke. J Vasc Surg. 1985;2:250–254
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (802 KB)
    • CrossRef
21. Goldstone J, Effeney DJ. The role of carotid endarterectomy in the treatment of acute neurologic deficits. Prog Cardiovasc Dis. 1980;22:415–422
    • View In Article
    • MEDLINE
    • CrossRef
22. Goldstone J, Moore WS. Emergency carotid artery surgery in neurologically unstable patients. Arch Surg. 1976;111:1284–1291
    • View In Article
    • MEDLINE
23. Kahn MB, Patterson HK, Seltzer J, Fitzpatrick M, Smullens S, Bell R, et al.  Early carotid endarterectomy in selected stroke patients. Ann Vasc Surg. 1999;13:463–467
    • View In Article
    • Abstract
    • Full-Text PDF (496 KB)
    • CrossRef
24. Ricco JB, Illuminati G, Bouin-Pineau MH, Demarque C, Camiade C, Blecha L, et al.  Early carotid endarterectomy after a nondisabling stroke: a prospective study. Ann Vasc Surg. 2000;14:89–94
    • View In Article
    • Abstract
    • Full-Text PDF (529 KB)
    • CrossRef
25. Sulter G, Steen C, De Keyser J. Use of the Barthel Index and Modified Rankin Scale in acute stroke trials. Stroke. 1999;30:1538–1541
    • View In Article
    • MEDLINE
    • CrossRef
26. Eckstein HH, Schumacher H, Dörfler A, Forsting M, Jansen O, Ringleb P, et al.  Carotid endarterectomy and intracranial thrombolysis: simultaneous and staged procedures in ischemic stroke. J Vasc Surg. 1999;29:459–471
    • View In Article
    • Abstract
    • Full Text
    • Full-Text PDF (360 KB)
    • CrossRef