Symptomatic acute occlusion of the internalcarotid artery: Reappraisal of urgent vascularreconstruction based on current stroke imagingBarbara Theresia Weis-Müller, MD,a Rita Huber, MD,a Asya Spivak-Dats, MD,a Bernd Turowski, MD,b
Mario Siebler, MD,c and Wilhelm Sandmann, MD,a Düsseldorf, Germany
Objective: We hypothesized that a subgroup of patients with frank stroke due to sudden occlusion of the internal carotidartery could safely undergo surgery to restore carotid patency and to rescue brain tissue not yet irreversibly damaged ifcurrent stroke diagnostic methods were applied.Methods: From November 1997 to March 2007, 1810 patients underwent carotid endarterectomy of the internal carotidartery for occlusive disease at our department. Within the same period, 5369 patients were examined at our stroke unit,and 502 from this cohort underwent internal carotid artery reconstruction. A subgroup of 35 patients (28 men, 7 women;mean age, 61 � 10 years) underwent urgent surgical revascularization due to an acute internal carotid artery occlusion<72 hours (mean 25 � 17 hours) after the onset of stroke symptoms and <36 hours (mean 16 � 10 hours) afteradmission to our stroke unit. Our diagnostic workup consisted of extracranial intracranial duplex sonography, cerebralcomputed tomography, digital subtraction angiography, magnetic resonance imaging, and angiography, includingdiffusion- and perfusion-weighted imaging, to discriminate between viable and irreversibly damaged brain tissue. Thestudy excluded patients who presented an impaired level of consciousness, occlusion of the intracranial internal carotidartery, occlusion of the ipsilateral middle cerebral artery, or infarction more than one-third of the territory perfused bythe middle cerebral artery. Imaging showed signs of recent ischemic infarction in all 35 cases. On admission, eight patients(23%) scored 0 to 2 points and 27 (77%) scored 3 to 5 points in Rankin scale.Results: Confirmed by postoperative Doppler and duplex sonography at discharge, internal carotid artery patency couldbe achieved in 30 of 35 cases (86%). Intracranial hemorrhage occurred in two patients (6%) and reinfarction in anothertwo (6%). Two patients died during their hospital stay (30-day mortality, 6%). Compared with the preoperativeneurologic status, rates of clinical improvement (>1 point in Rankin scale), stability, and deterioration were 57%, 31%,and 6%, respectively.Conclusions: Restoration of blood flow in an acutely occluded internal carotid artery can only be achieved in the acutestage. Our pilot study demonstrated that a thorough diagnostic workup allows selection of patients who may benefit fromurgent revascularization of acute internal carotid artery occlusion in the stage of an acute stroke. A prospectiverandomized multicenter trial comparing surgery with conservative medical treatment is needed. (J Vasc Surg 2008;47:
752-9.)
The early attempts of the pioneers of vascular surgery torestore blood flow for acute occlusion of the internal carotidartery (ICA) often ended with catastrophic results, most likelydue to postoperative intracranial hemorrhage.1-5 As a conse-quence, carotid surgery was more or less abandoned, althoughfew experienced authors insisted that urgent surgery in se-lected patients could have good results.
Today, however, the diagnostic modalities available forevaluation of the vascular system and the brain are substan-tially different from those available at the time of the jointstudy.5 Ultrasound technology, angiography, and neuroim-
From the Clinic for Vascular Surgery and Kidney Transplantation,a Instituteof Diagnostic Radiology,b and the Department of Neurology,c UniversityHospital of Düsseldorf, Heinrich-Heine-University of Düsseldorf.
Competition of interest: none.Presented at the 2007 Vascular Annual Meeting, Baltimore, Md, Jun 6-10,
2007.Reprint requests: Barbara Theresia Weis-Müller, University Hospital of
aging such as cerebral computed tomography (CT)or magnetic resonance imaging (MRI) with diffusion-weighted (DWI) and perfusion-weighted imaging (PWI) nowallow for rapid and accurate diagnosis of extracranial andintracranial vascular occlusion and its associated acute isch-emic stroke. Systemic thrombolytic therapy �3 hours aftersymptoms onset has been shown to be effective in reducingthe severity of the neurologic sequelae of intracranial vascularocclusion.6-9 Here we report a single-center experience withsurgical revascularization of acute extracranial ICA occlusionin the acute stage. Our aim was to evaluate inclusion andexclusion criteria based on current diagnostic techniques tomake emergency carotid surgery safe and effective.
METHODS
Patients. We prospectively followed up 35 consecu-tive patients from November 1997 to March 2007 whounderwent urgent surgery for acute ICA occlusion �72hours after the onset of neurologic symptoms and �36hours after admission to our stroke unit. The study wasapproved by the local ethics committee, and patients or
JOURNAL OF VASCULAR SURGERYVolume 47, Number 4 Weis-Müller et al 753
During the study period, 1810 patients underwent ca-rotid surgery for occlusive disease in our department. Duringthis time, 5369 patients were examined at our stroke unit, and502 symptomatic patients from this cohort were transferred toour department for ICA reconstruction. A subgroup of 35patients (28 men, 7 women) with a mean age of 60.7 � 10.6years (range, 32-76 years) underwent urgent reconstructionof acute ICA occlusion within a mean of 24.1 � 17.1 hours
Fig 1. (a) Perfusion-weighted image 4 hours after onsedisplays a perfusion deficit (red and green) of the righinfarction. (b) Perfusion-weighted image taken 1 weekartery demonstrates a perfusion deficit in the area of threcovered (blue). The patient remained clinically stable.
Fig 2. (a) Diffusion-weighted image 4 hours after onseinternal carotid artery displays white spots in the middle carotitissue. (b) Diffusion-weighted image 1 week after successwhich approximately has the same size as the zone of severezone of impaired perfusion in the preoperative perfusion-w
(median, 18; range, 6-66 hours) after symptoms onset and
within a mean of 15.7 � 9.5 hours (median, 11; range, 3-34hours) after admission to our stroke unit.
Preoperative diagnostic work up. All patients under-went a standardized protocol of clinical neurologic andultrasound examination of the extracranial ICA by a staffneurologist, and all 35 patients underwent preoperativeneuroimaging (cerebral CT or MRI, or both. includingPWI and DWI assessment; Fig 1 and 2). When extracranial
roke symptoms (Rankin 4) and before revascularizationisphere, which represents “tissue at risk” for further
successful revascularization of the right internal carotidnitive infarction (green), while the surrounding tissue
roke symptoms and before revascularization of the rightry perfusion area region as sign for definitely damaged brainascularization displays an enlargement of the stroke area,sion deficit (red color Fig 1, a), but is smaller than the totaled image (red and green in Fig 1, a).
t of stt hemaftere defi
t of std arteful revperfu
and intracranial ultrasound imaging was suggestive of ICA
JOURNAL OF VASCULAR SURGERYApril 2008754 Weis-Müller et al
occlusion, additional MRA or digital subtraction angiogra-phy (DSA), or both, were performed to evaluate moreprecisely the intracranial vascular perfusion. Indications forurgent revascularization were fluctuating or progressingneurologic deficit, stroke in evolution, but also fixed, smallneurologic deficits in patients with a large PWI/DWI mis-match. Patients with an impaired level of consciousnesswere excluded from surgery (Table). Rankin scores at thetime of admission were 0 to 2 in eight patients (23%) and 3to 5 in the other 27 (77%) patients (Fig 3).
Inclusion criteria based on imaging were a fresh isch-emic brain infarction limited to not more than one-third ofthe territory perfused by the middle cerebral artery (MCA).Exclusion criteria were intracranial hemorrhage or infarc-tion volume that exceeded more than one-third of theMCA perfusion area (Table). In particular, preoperativecerebral CT was performed in nine patients (26%), MRIcombined with DWI and PWI in 15 (43%), and 11 (31%)
Table. Inclusion and exclusion criteria for emergencysurgery of acute internal carotid artery occlusion
Fig 3. Rankin scale scores before (gray) and after (black) carotidartery surgery.
were examined by both neuroimaging methods.
It should be noted that there is frequently an additionalocclusion of the distal ICA, in particular in the supraclinoidsegment, resulting from embolization or thrombus propa-gation. Surgical revascularization can only be achieved ifthe occlusion is limited to the extracranial ICA; therefore,visualization to confirm patency of the distal vessel is essen-tial before performing surgical revascularization (Fig 4).Accordingly, 26 patients (74%) were evaluated by usingMRA and 22 patients (63%) by a DSA alone or by bothangiographic methods. Apart from ipsilateral ICA occlu-sion, angiography demonstrated contralateral ICA chronicocclusion in seven patients (20%). Our inclusion criterionfor urgent revascularization was a fresh occlusion of theextracranial ICA. Patients with a suspected ICA occlusionolder than 3 days, an occlusion of the intracranial ICA or ofthe ipsilateral MCA, or both, were excluded from urgentrevascularization (Table).
Surgery. Before surgery, six patients (17%) had beentreated by systemic recombinant tissue plasminogen activa-tor lysis. Immediately after the diagnostic workup wascompleted, the patients were transferred to the operatingtheater. General anesthesia was initiated, and after exposureof the carotid bifurcation, heparin (2500 IU) was adminis-tered intravenously. The common carotid artery and theexternal carotid artery were cross-clamped, and a longitu-dinal arteriotomy of the common carotid artery extendinginto the ICA was performed. Chronic atherosclerotic ste-nosis as underlying cause for acute occlusion was treated bycarotid endarterectomy (CEA). The fresh thrombus, whichwas present usually above the chronic bifurcation process,was either flushed out by elevating the systemic systolicblood pressure to increase the retrograde blood flow, or incase of insufficient backflow, a 2F balloon catheter wasadvanced with caution towards the carotid siphon andthrombectomy a was performed.
Still insufficient backflow was an indication for an intra-operative arteriography. A 4F arterial double-lumen irrigationembolectomy catheter was inserted into the distal ICA, theballoon catheter was blocked, and 20 mL of contrast mediumwas applied with smooth pressure to visualize the siphonregion and the intracranial vessels. After adequate backflowwas achieved, the arterial incision was closed with an autoge-nous patch graft from the great saphenous vein harvested fromthe ankle. Restoration of ICA blood flow was evaluated intra-operatively by pulsed Doppler sonography. In four patients,patency could not be achieved and the operation was termi-nated by ligation of the ICA and a thromboendarterectomy ofthe external carotid artery.
In two patients, ICA occlusion was due to acute dissec-tion. In one patient, the dissecting membrane was limitedto the carotid bulb and the C2 segment and was treated byremoving the dissecting membrane together with thethrombus by a Fogarty maneuver and a typical carotidthromboendarterectomy. In the other patient, the dis-sected segment was excised and replaced using a venousinterposition graft from the groin, and the distal anastomo-
sis had to be performed at the skull base.
usion
JOURNAL OF VASCULAR SURGERYVolume 47, Number 4 Weis-Müller et al 755
Postoperative treatment and follow-up. After sur-gery, all patients were monitored in the intermediate orintensive care unit for 24 hours under continuous arterialblood pressure control with a systolic blood pressure be-tween 100 and 140 mm Hg before being transferred backto the stroke unit.
During the postoperative stay and before discharge, allpatients underwent a standardized clinical neurologic eval-uation and duplex ultrasound (DUS) examination of thesupra-aortic vessels to confirm the patency of the ICA.
Fig 4. Duplex subtraction angiography shows (a) an octhe right middle carotid artery, and (c) a retrograde perf
Fig 5. Magnetic resonance angiography taken 1 week after thesurgical revascularization demonstrates restoration of blood flowin the right internal carotid artery.
Cerebral CT or MRI (Fig 5) were used to re-evaluate 31
(89%) patients during their postoperative hospitalization,including especially those who had shown deterioration oftheir clinical status.
Follow-up. All surviving patients were contacted bytelephone. The patients were encouraged to come to ourdepartment of neurology for complete neurologic check-upand DUS examination of the neck arteries and a Dopplersonography of the intracranial circulation. Recorded wasthe patient’s present social and medical history, includinggeneral information on current state of health, recurrentischemic events, hospital stays, change in medication, andnewly acquired diseases. Current Rankin scores were deter-mined by using a standardized questionnaire. If patients werenot able to come to the hospital, general practitioners, prac-ticing neurologists, and local hospitals and rehabilitation clin-ics were also contacted for the latest medical records such asDoppler and DUS findings of the cerebral circulation, radio-logic reports, and letters of discharge. In case of death, date,the place and cause of death was documented.
RESULTS
Findings during surgery. After surgical exposure andopening of the carotid bifurcation, complete ICA occlusionwas confirmed in all 35 cases. The cause of acute carotidocclusion was atherosclerotic high-grade bifurcation steno-sis in 31 patients (89%), acute carotid dissection in 2 (6%),1 (3%) with embolism from cardiac source, and 1 (3%) withheparin-induced thrombocytopenia with fresh ICA throm-bus. We were able to restore blood flow in 31 of the 35 ICAocclusions (89%) but were not able to reopen the occludedICA in the other four. The reason for failure was an old fibroticICA occlusion in three cases and an ICA occlusion at thesiphon level with descending thrombus in another case.
Postoperative results. The DUS check at dischargeconfirmed that revascularization had been successful in 30 of
on of the right internal carotid artery, (b) a perfusion ofof the “siphon.”
clusi
35 ICA occlusions (86%). The failures comprised reocclusion
JOURNAL OF VASCULAR SURGERYApril 2008756 Weis-Müller et al
in one patient during the hospital stay, and restoration ofblood flow could not be achieved during surgery in 4 patients.
Neuroimaging displayed parenchymal hematoma intwo patients (6%), one with mass effect and the otherwithout clinical consequences. Reinfarction developed intwo further patients (6%); of these, one patient showed asignificant enlargement of his pre-existing infarction anddeteriorated, and the other sustained a small embolic rein-farction without clinical consequences. Hemorrhagic trans-formation of the pre-existing infarction developed in four(11%) patients, but without clinical consequences.
Compared with the preoperative clinical examination,20 patients (57%) had improved at least 1 point in Rankinscale, 11 patients (31%) remained unchanged, and twopatients (6%) had deteriorated. One patient, who had beentreated by systemic thrombolysis before surgery, deterio-rated owing to an intracranial hemorrhage. Reinfarctiondeveloped in another patient. Two patients died in thehospital for a 30-day mortality rate of 6% (Fig 3 and 6). Wewere not able to reopen an occluded ICA in one patientwho had already sustained a stroke due to brain infarctionof the contralateral hemisphere. Postoperative enlargementof this ipsilateral stroke area with malignant brain edemaoccurred, and the patient died. Another patient died after adeep vein thrombosis with fulminant pulmonary embolismdeveloped during the hospital stay.
In the subgroup of six patients who had been treated bysystemic thrombolysis before surgery, three patients im-proved clinically after surgery, two patients remained sta-ble, and one patient deteriorated due to postoperativeintracranial hemorrhage.
Follow-up data. We were able to get clinical informa-tion for 32 of 33 surviving patients (94%) with a meanfollow up of 34.4 � 23.3 months (range, 1-106 months).Five patients died during follow-up. One patient died 6weeks after carotid revascularization of the effect of hisprior stroke, and 4 patients died of causes unrelated to theircerebrovascular disease: 1 died during aortocoronary by-pass surgery, 2 died of lung cancer, and 1 of an unknowncause.
Of 27 surviving patients, 13 (48%) were re-examined inour department of neurology and 11 patients (41%) in an-other hospital or neurologic institution. Information for three
Fig 6. Postoperative clinical course at discharge improved ordeteriorated means change of 1 point or more in Rankin scale.
patients (11%) was obtained by a telephone call exclusively.
One of the patients in whom we had not been able torestore ICA patency presented with an ipsilateral transientischemic attack during follow-up. One patient sustained astroke of the contralateral hemisphere due to a hypertensivecrisis. The remaining 25 patients (93%) did not have anyneurologic disorders during follow-up. An asymptomaticrestenosis developed in one patient, which was treated bydilatation and stenting.
Five of the 13 patients (38%) who were re-examined inour department of neurology had completely recoveredfrom their stroke and were at Rankin stage 0. Comparedwith discharge, two further patients (15%) had clinicallyimproved at least 1 point in the Rankin scale, five patients(38%) had remained stable, and one patient (8%) haddeteriorated, despite CT scans that did not reveal reinfarc-tion.
DISCUSSION
Since systemic thrombolysis has proven to be effectivein reducing the severity of neurologic deficit and morbidityin acute stroke, there has been increased interest in acuteinterventions for stroke patients.6-9 Fibrinolysis alone canbe effective for treatment of strokes caused by middlecerebral artery occlusion but has been shown to be ineffec-tive in ICA occlusion, revealing a complete recanalizationrate of only 10%.10 The occluded segments of intracranialarteries, such as the MCA, usually contain an embolus or ashort length of thrombus in an otherwise patent healthyvessel,11-12 whereas as shown in our study, acute occlusionof the extracranial ICA is usually due to a tight stenosis ofatherosclerotic etiology at the carotid bifurcation and asuperimposed thrombus. Therefore, CEA and simulta-neous removal of the appositional thrombus have to becombined for restoration of blood flow. A prerequisite forsuccessful surgery is the limitation of the ICA thrombosis toits extracranial portion, because only then can the throm-bus be retrieved either by sufficient backflow or by a bal-loon catheter. Thus, all our patients were examined beforesurgery by MRA or DSA, or by both methods in selectedcases, to confirm patency of the distal ICA (Fig 4).
An additional reason for preoperative angiography isthe suspicion of an ICA occlusion, which develops from atight stenosis of the intracranial portion of the ICA with athrombus propagating from the skull base towards thebifurcation. In such cases surgery, cannot be successful andeven endovascular methods, as recently shown in the Me-chanical Embolus Removal in Cerebral Ischemia (MERCI)trial,13 have problems recanalizing these distal ICA occlu-sions. The participating study groups were able to achievedistal ICA revascularization in only 53% to 63%, with agood clinical outcome in 39% of the revascularized patientsand a 90-day mortality of 30%; in the group that was notrevascularized, 79% of the patients died.
The success of surgery in acute carotid occlusion partlydepends on the duration of the occlusion. The occludingthrombus usually originates from the carotid bifurcationand propagates towards the skull base. The thrombus
causes inflammation of the arterial wall, and organization of
JOURNAL OF VASCULAR SURGERYVolume 47, Number 4 Weis-Müller et al 757
the thrombus begins within days after arterial occlusion.Paty et al14 reviewed their data from 87 patients whounderwent surgical treatment for 90 symptomatic carotidocclusions �14 days of symptom onset. They were onlyable to restore blood flow in 31 of 90 occluded ICAs(34%).14 In contrast, we limited our time for urgent revas-cularization to 3 days after the very first onset of symptomsand were able to reopen 86% of the ICA occlusions. Thetime interval for successful recanalization may vary frompatient to patient, but in general the success rate will behigher with short intervals.
Early revascularization after acute stroke is importantnot only to preserve perfusion of viable brain but also tosalvage ischemic but not yet infarcted brain. In acute ICAocclusion, neurologic disorder occurs either due to throm-boembolism while the occluding thrombus is propagating,or from hypoperfusion of the dependent hemisphere inpatients with insufficient collateral blood flow.11 Magneticresonance imaging with DWI and PWI may help to differ-entiate between ischemic brain tissue that is definitively andirreversibly damaged (DWI), and “tissue at risk,” hypoper-fused but still viable, which may be saved by early revascu-larization (PWI; Fig 1, 2).15-20
Neumann-Haefelin et al18 demonstrated that PWI/DWI volume mismatch was significantly larger in strokepatients with severe ICA disease than in patients withoutextracranial carotid stenosis.18 The authors showed thatabout 80% of the patients presented with a PWI/DWImismatch during the first 24 hours after onset of strokesymptoms. Control PWI/DWI scans 1 week later revealedan enlargement of the infarction zone in 75% of the patientsif the perfusion time delay compared with the unaffectedhemisphere, as demonstrated in the first scans, had been �6seconds.18 The PWI/DWI volume mismatch on MRI wasone of our most important criterion for selecting patientsfor urgent carotid endarterectomy.
Neurologic examination at discharge showed thatmore than one-half of our patients with successful revascu-larization had clinically improved. A prospective multi-center trial randomizing the patients into a conservativemedical and a surgical group is now needed to investigatethe influence of early ICA revascularization on clinicaloutcome and its corresponding MRI changes in a largergroup of cases.
Summarizing our diagnostic algorithm, we now per-form first combined MRI/MRA if carotid occlusion issuspected by Doppler/DUS imaging. An additional angio-gram using DSA only becomes necessary if the supraclinoidregion cannot be evaluated by MRA alone. Patients withinfarctions of more than one-third of the MCA perfusionarea, with occlusion of the intracranial ICA or MCA, will beexcluded from urgent carotid revascularization. Patientswith impaired level of consciousness will be excluded with-out further invasive vascular imaging studies.
Apart from a possible benefit on clinical outcome in theacute stroke situation, reopening of the ICA occlusion mayprevent recurrence of stroke later in life. Chronic carotid
occlusion is associated with an increased risk of 18.5% to 26%
for later ipsilateral cerebral ischemia.21-24 Other groups re-cently focused on hemodynamic failure distal to the ICAocclusion and used positron emission tomography scanningto describe the importance of hemodynamic factors in predict-ing outcome in these patients.23,24 The necessity of restoringICA blood flow is emphasized by our observation that 20% ofour patients had a chronic occlusion contralateral to theacutely occluded ICA. After successful reopening of ICAocclusion, we did not observe any ipsilateral cerebral ischemiaduring a mean follow-up of 34 months.
Acute carotid surgery was abandoned for years becauseof the risk of postoperative intracranial hemorrhage. How-ever intracerebral hemorrhage after ischemic stroke is com-monly observed without invasive treatment if patients arestudied with serial CT.25,26 Okada et al25 found hemor-rhagic infarction during the first month after the embolicevent in 41% of the patients. The percentage of hemor-rhagic infarction in their study was higher in patients aged�70 years compared with younger patients and was alsohigher in moderate or large infarcts compared with smallerinfarcts. Of interest was that no correlation was foundbetween hemorrhage and a previous history of hyperten-sion or blood pressure elevation during the acute stage ofstroke.25 Hornig et al26 made a similar observation, detect-ing hemorrhagic transformation in 43% of the cases. Therisk of hemorrhage in their study was correlated with asevere neurologic deficit, disturbances of consciousness,and large infarctions with a mass effect.26 Consideringthese results, we excluded patients with impaired level ofconsciousness from consideration for emergency revascu-larization.
Following the example of others, we differentiatedbleeding between hemorrhagic infarction and parenchymalhematoma. Hemorrhagic infarction is petechiae or conflu-ent hemorrhage in the area of ischemic injury; parenchymalbleeding is homogeneous bleeding with circumscribed clotformation or mass effect.28,29
For many years we routinely perform urgent carotidsurgery in patients with transient ischemic attack or amau-rosis fugax, but in the present study, we only includedpatients with frank stroke so that we could evaluate the riskof intracranial bleeding and neurologic impairment. Thepostoperative neuroimaging of our patients displayed pa-renchymal hematoma in two patients (6%), one of whomhad a mass effect, and a hemorrhagic transformation of theischemic infarction without clinical impairment in fourpatients (11%). The frequency of postoperative intracranialhemorrhage in our patients is comparable with the naturalcourse of hemorrhagic transformation of ischemic cerebralinfarction.25,26
Apart from postoperative intracranial hemorrhage, re-infarction developed in two further patients (6%). Onepatient showed a significant enlargement of his pre-existinginfarction and deteriorated, and the other one sustainedfrom a small embolic reinfarction without clinical conse-quences. A postoperative stroke recurrence rate of 6%seems high compared with results of elective carotid sur-
gery; but in general, acute carotid occlusion seems to be
JOURNAL OF VASCULAR SURGERYApril 2008758 Weis-Müller et al
associated with a bad outcome. The German Stroke StudyCollaboration recruited 4157 patients prospectively whohad experienced an acute ischemic stroke and monitoredthem to assess recurrent stroke and death up to 1 year afterthe event. Stroke was caused by acute carotid occlusion in366 patients. Early recurrent cerebral ischemia �72 hoursafter hospital admission was observed in 7.4% of the pa-tients with carotid occlusion. After 100 days, already 21.2%of these patients had died.27
In the last decade, several surgical work groups pub-lished their results of urgent carotid surgery.14,30-34 Kasperet al30 reported the outcome of 29 patients with revascu-larization for acute ICA occlusion �8 days of clinical pre-sentation. In contrast to our study in which all patients hadexperienced a preoperative stroke, preoperative stroke hadoccurred in only nine (31%) patients, whereas most of thepatients complained about transient ischemic attack oramaurosis fugax. Their results were similar to ours, with asuccessful restoration of ICA blood flow in 83%, 48% of thepatients had clinically improved, 45% remained unchanged,7% had deteriorated, and 1 patient (3%) had died.30
Gay et al32 reported 22 emergency carotid revascular-izations �24 hours with a 30-day mortality rate of 9.5%.Berthet et al34 tried a surgical revascularization of 12 acuteICA occlusions �6 hours of symptoms; one patient died ofstroke and another died of reasons unrelated to stroke.34
All of these work groups performed careful preoperativeworkup and had well-defined inclusion and exclusioncriteria.
Nowadays interventional teams are also dealing withacute ICA occlusion, using new devices for mechanicalthrombectomy, intra-arterial lysis, dilatation, and stent-ing.35-38 The largest series was actually presented by aTaiwanese group, who treated 30 patients with acute cer-vical ICA occlusion by using endovascular techniques. Incontrast with our study, intervention was not performed inacute stroke; the most recent cerebral infarction had to be�2 weeks before the intervention. The group was able torestore ICA blood flow in 73%. They observed one fatalbrainstem infarction for a 30-day mortality rate of 3%.During a mean follow-up of 16 months, ICA restenosisdeveloped in one patient and two others sustained reocclu-sion, for a reocclusion or restenosis rate of 14%. We areconvinced that it is better to treat cervical ICA occlusionsurgically: after clamping of the common carotid artery, theentire thrombus and atherosclerotic material can be re-moved without flooding blood over the embolic source,which effectively prevents embolism. Endovascular maneu-vers could be considered if CEA is not an option, especiallyin intracranial ICA occlusions, where surgery fails andmedical treatment alone has a poor outcome.14,38-41
CONCLUSION
Our preliminary study has shown that urgent revascu-larization in selected stroke patients with acute carotidocclusion is both technically feasible and associated withacceptable morbidity and mortality rates when preceded by
careful preoperative evaluation. A prospective randomized
multicenter trial is needed to confirm the validity of ourfindings and to assess the relative merits of aggressive vsconservative management in these patients.
AUTHOR CONTRIBUTIONS
Conception and design: BW, MS, WSAnalysis and interpretation: BW, AS, MS, WSData collection: AS, RH, BT, BWWriting the article: BWCritical revision of the article: RH, AS, BT, MS, WSFinal approval of the article: RH, AS, BT, MS, WSStatistical analysis: BW, ASObtained funding: Not applicableOverall responsibility: BW
REFERENCES
1. Wylie EJ, Hein MF, Adams JE. Intracranial hemorrhage followingsurgical revascularization for treatment of acute strokes. Neurosurg1963;21:212-5.
2. Thompson JE, Dale JA, Patman RD. Endarteriectomy of the totallyoccluded carotid artery for stroke. Arch Surg 1967;95:791-801.
3. Rob GB. Operation for acute completed stroke due to thrombosis of theinternal carotid artery. Surgery 1969;65:862-5.
4. Bruetman ME, Fields WS, Crawford ES, de Bakey ME. Cerebralhemorrhage in carotid artery surgery. Arch Neurol 1963;9:458-67.
6. NINDS rt-PA Stroke Group. Tissue plasminogen activator for acuteischemic stroke. N Engl J Med 1995;333:1581-7.
7. The National Institute of Neurological Disorders and stroke rt-PAStroke Study Group. Tissue plasminogen activator for acute ischemicstroke at one year. N Engl J Med 1999;340:1781-7.
8. The ATLANTIS, ECASS, and NINDS rt-PA Study Group Investiga-tors. Association of outcome with early stroke treatment: pooled anal-ysis of ATLANTIS, ECASS, and NINDS rt-PA stroke trials. Lancet2004;363:768-74.
9. Straub S, Junghans U, Jovanovic V, Wittsack HJ, Seitz RJ, Siebler M.Systemic thrombolysis with recombinant plasminogen activator andtirofiban in acute middle cerebral artery occlusion. Stroke 2004;35:705-9.
10. Christou I, Felberg RA, Demchuk AM, Burgin WS, Malkoff M, GrottaJC, et al. Intravenous tissue plasminogen activator and flow improve-ment in acute ischemic stroke patients with internal carotid arteryocclusion. J Neuroimaging 2002;12:119-23.
12. Qureshi AI, Luft AR, Sharma M, Guterman LR, Hopkins LN. Preven-tion and treatment of thromboembolic and ischemic complicationsassociated with endovascular procedures: part I: pathophysiological andpharmacological features. Neurosurgery 2001;46:1344-59.
13. Flint AC, Duckwiler GR, Budzik RF, Liebeskind DS, Smith WS, for theMERCI and Multi Merci Writing Committee. Mechanical thrombec-tomy of intracranial internal carotid occlusion. Pooled results of theMERCI and Multi MERCI part I trials. Stroke 2007;38:1274-80.
14. Paty PSK, Adeniyi JA, Mehta M, Darling C 3rd, Chang BB, KreienbergPB, et al. Surgical treatment of internal artery occlusion. J Vasc Surg2003;37:785-8.
15. Schwamm LH, Koroshetz WJ, Sorensen G, Wang BS, Copen WA,Budzik R, et al. Time course of lesion development in patients withacute stroke. Serial diffusion- and hemodynamic-weighted magneticresonance imaging. Stroke 1998;29:2268-76.
16. Staroselskaya IA, Chaves C, Siver B, Linfante I, Edelmann RR, CaplanL, et al. Relationship between magnetic resonance arterial patency andperfusion-diffusion mismatch in acute ischemic stroke and its potential
clinical use. Arch Neurol 2001;58:1069-74.
JOURNAL OF VASCULAR SURGERYVolume 47, Number 4 Mackey 759
17. Neumann-Haefelin T, Wittsack HJ, Wenserski F, Siebler M, Seitz RJ,Mödder U, et al. Diffusion- and perfusion-weighted MRI. The DWI/PWI mismatch region in acute stroke. Stroke 1999;30:1591-7.
18. Neumann-Haefelin T, Wittsack HJ, Fink GR, Wenserski F, Li T-Q,Seitz RJ, et al. Diffusion- and perfusion-weighed MRI. Influence ofsevere carotid artery stenosis on the DW/PWI Mismatch in acutestroke. Stroke 2000;31:1311-7.
19. Wittsack HJ, Ritzl A, Fink GR, Wenserski F, Siebler M, Seitz RJ, et al.MR imaging in acute stroke: diffusion weighted and perfusion imagingparameters for predicting infarct size. Radiology 2002;222:397-403.
20. Seitz RJ, Meisel S, Weller P, Junghans U, Wittsach HJ, Siebler M. Initialischemic event: perfusion weighted MR-imaging and apparant diffusioncoefficient for stroke evolution. Radiology 2005;237:1020-8.
21. Nicholls SC, Bergelin R, Strandness DE. Neurologic sequelae of uni-lateral carotid artery occlusion: Immediate and late. J Vasc Surg 1989;10:542-8.
22. Klijn CJ, Van Buren PA, Kappelle LJ, Tulleken CA, Eikelboom BC,Alagara A, et al. Outcome in patients with symptomatic occlusion of theinternal carotid artery. Eur J Vasc Endovasc Surg 2000;19:579-86.
23. Vernieri F, Pasqualetti P, Passarelli F, Rossini PM, Silvestrini M. Out-come of carotid occlusion is predicted by cerebrovascular reactivity.Stroke 1999;30:2765-6.
24. Grupp RL, Powers WJ. Risk of stroke and current indications forcerebral revascularization in patients with carotid occlusion. NeurosurgClin North Am 2001;473-87.
25. Okada Y, Yamaguchi T, Minematsu K, Miyashita T, Sawada T, Sa-doshima S, et al. Hemorrhagic transformation in cerebral embolism.Stroke 1989;20:598-603.
27. Weimar C, Goerler M, Harmas L, Diener HC, for the German StrokeCollaboration. Distribution and outcome of symptomatic strenosis andocclusions in patients with acute cerebral ischemia. Arch Neuol 2006;63:1287-91.
28. Pessin MS, del Zoppo GJ, Estol C. Thrombolytic agents in the treat-ment of stroke. Clin Neuropharmacol 1990;13:271-89.
29. Jansen O, von Kummer R, Forsting M, Hacke W, Sartor K. Thrombo-lytic therapy in acute occlusion of the intracranial interna carotid artery
bifurcation. AJNR Am J Neuroradiol 1995;16:1977-86.
detailed late follow-up assessment, 5 had no residual deficit, 2 had
30. Kasper GC, Wladis AR, Lohr JM, Roedersheimer LR, Reed RL, MillerTJ, et al. Carotid thromboendarterectomy for recent occlusion of theinternal carotid artery. J Vasc Surg 2002;33:242-50.
31. Huber R, Muller BT, Seitz RJ, Siebler M, Modder U, Sandmann W.Carotid surgery in acute symptomatic patients. Eur J Vasc EndovascSurg 2003;25:60-7.
32. Gay Jl, Curtil A, Buffiere S, Favre JP, Barral X. Urgent carotid arteryrepair: retrospective study of 21 cases. Ann Vasc Surg 2002;16:401-6.
33. Sbarigia E, Toni D, Speziale F, Falcou A, Sacchetti ML, Panico MA, etal. Emergency and early carotid endarterectomy in patients with acuteischemic stroke selected with a predefined protocol. A prospective pilotstudy. Int Angiol 2003;22:426-30.
34. Berthet JP, Marty-Ane CH, Picard E, Branchereau P, Mary H, Veera-pen R, et al. Acute carotid artery thrombosis: description of 12 surgi-cally treated cases. Ann Vasc Surg 2005;19:11-8.
35. Kao HL, Lin M-S, Wang C-S, Lin Y-H, Lin L-C, Chao C-L, et al.Endovascular recanalization for cervical ICA occlusion is feasible withacceptable midterm clinical results. J Am Coll Cardiol 2007;49:765-71.
36. Dabitz R, Triebe S, Leppmeier U, Ochs G, Vorwerk D. Percutaneousrecanalization of acute internal artery occlusions in patients with severestroke. Cardiovasc Intervent Radiol 2007;30:34.
37. Srinivasan A, Goyal M, Stys P, Sharma M, Lum C. Microcatheternavigation and thrombolyis in acute symptomatic cervical internal ca-rotid occlusion. AJNR Am J Neuroradiol 2006;27:774-9.
39. Rubiera M, Ribo M, Delgado-Mederos R, Santamina E, Delgado P,Montaner J, et al. Tandem internal carotid artery/middle cerebralartery occlusion. Stroke 2006;37:2301-5.
40. Smith WS. Technology insight: recanalization with drugs and devicesduring acute ischemic stroke. Neurology 2007;3:45-53.
41. Arnold M, Kappeler L, Nedeltchev K, Brekenfeld C, Fischer U, KeserueB, et al. Recanalization and outcome after intra-arterial thrombolysis inmiddle cerebral artery and internal carotid artery occlusion: does sexmatter? Stroke 2007;38:1281-5.
Submitted May 28, 2007; accepted Nov 12, 2007.
INVITED COMMENTARY
William C. Mackey, MD, Boston, Mass
When faced with acute stroke patients who may benefit fromprompt revascularization, we face a dilemma. There is high risk butin many cases high reward. Should we try to restore carotid flow inthe setting of acute stroke? Dr Weis-Muller and colleagues reviewtheir experience with urgent endarterectomy for 35 patients withacute carotid occlusion related strokes to determine if currentimaging technology might refine patient selection and therebyimprove outcome. Each study patient underwent neurologic ex-amination, carotid duplex, and contrast enhanced computed to-mography (CT) scanning and/or magnetic resonance imaging(MRI). Patients with carotid occlusion underwent MRA and/ordigital subtraction angiography to visualize intracranial vessels.Patients with altered mental status, infarct volume exceeding onethird of the middle cerebral artery (MCA) territory, hemorrhagicinfarction, internal carotid artery (ICA) occlusion more than 3 daysold, nonvisualized intracranial internal carotid, or middle cerebralartery occlusion were excluded from urgent endarterectomy. Us-ing this algorithm, the authors restored flow in 31/35 cases withone reocclusion. Twenty of 35 (57%) patients improved in theirRankin Stroke Score, 11 (31%) remained unchanged, two (6%)worsened, and two (6%) died. On late follow-up (mean 34months), five patients had died (one from stroke), and one patientsuffered a contralateral stroke. Of 13 patients who underwent
improved further after discharge, 5 were stable, and 1 was worse.Of note is that early postoperative imaging studies revealed brainhematomas in 6%, reinfarction in 6%, and hemorrhagic infarcttransformation in 11%.
While encouraging, these results do not prove the value ofurgent endarterectomy. First, there is no medically managed con-trol group. If we are to subject patients to an intervention with a 6%risk of death, a 6% risk of neurological deterioration, and a 23% riskof hematoma, recurrent infarction or hemorrhagic transformation,we must be sure that nonsurgical treatment is associated witheither greater risk of adverse outcome, lesser likelihood of clinicalimprovement, or both. Second, inclusion and exclusion criteria arenot yet fully refined. Of particular interest is the potential forperfusion weighted image (PWI) and diffusion weighted image(DWI) mismatches to refine patient selection for urgent interven-tion by distinguishing infarcted from ischemic but viable braintissue. Valid guidelines for the optimal management of acutestroke patients await more detailed understanding of the ischemicborder zone and a properly conducted randomized trial of urgentsurgery vs medical management. Until we have such guidelines,the decision to intervene should be individualized based on thesurgeon’s careful assessment of the relevant risks and benefits
weighed in the context of the new information from this study.
