Results of Carotid Endarterectomy for Vertebrobasilar Insufficiency An Evaluation Over Ten Years David Rosenthal, MD; David

Cossman, MD; C.

Bruce

Ledig, MD; Allan

\s=b\ A review was performed of 114 patients with symptoms of vertebrobasilar insufficiency (VBI) alone, or in combination with carotid territory transient ischemic attacks or carotid territory completed stroke (cCS) with follow-up extending to ten years. The most frequent symptoms of VBI were visual changes (50%), dizziness (31%), and syncope (30%). Patients with symptoms of VBI and arteriographic evidence of intracranial disease, regardless of stump pressure, are at high risk for cerebral ischemia during endarterectomy. At late follow-up, ranging from one to ten years, 63% of the patients were alive; 88% were asymptomatic. Causes of death were mainly cardiac (44%)and stroke (36%), but patients with symptoms of VBI and cCS died earlier and from a second cerebrovascular accident. When a correct preoperative diagnosis was established, carotid endarterectomy produced relief of symptoms in 90% of the patients.

(Arch Surg 113:1361-1364, 1978) cerebral blood flow due to emboli from the carotid bifurcation or stenosis of the internal carotid may result in transient characteristic symptoms of the CNS. These symptoms include ipsilateral visual defects or contralateral sensory and motor changes of the face and extremities. If hypoxia occurs in the dominant hemisphere, expressive or sensory aphasia may result.1 Symptoms of vertebrobasilar insufficiency (VBI) are less clearly defined. Symptoms of vertebrobasilar insuffi¬ ciency may occur singly or in combination and may be unilateral, bilateral, or occur on alternate sides.- The patient may experience dizziness, vertigo, visual changes, bilateral sensory or motor changes, tinnitus, dysarthria, headache, or syncope.14 When disease is present in both the carotid and /or vertebrobasilar systems, any combination, of episodic symptoms may be present, such as symptoms of VBI alone, VBI and carotid territory transient ischémie attacks, or VBI and carotid territory completed stroke (cCS). The variation of symptoms may be explained by the anatomic variations in the circle of Willis and other collateral channels.5 Since the anterior (carotid) and poste¬ rior (vertebrobasilar) circulation communicate in more than 90% of the individuals studied angiographically,6 correction of carotid stenoses should improve total cerebral blood flow and thereby indirectly relieve the symptoms of VBI.

Reduced artery i

Accepted for publication July 28,

1978. From the Vascular Service, Department of Surgery, Tufts-New England Medical Center, and the Department of Surgery, Tufts University School of Medicine, Boston. Presented at the 26th scientific meeting of the International Cardiovascular Society, Los Angeles, June 24, 1978. Reprint requests to Department of Surgery, Tufts-New England Medical Center, 171 Harrison Ave, Boston, MA 02111 (Dr Callow).

D.

Callow,

MD

MATERIALS AND METHODS A computer-assisted retrospective review of 114 patients between 1965 and 1975 was performed. All of the patients had signs or symptoms of VBI alone, or in conjunction with transient ischémie attacks or cCS. Of the 114 patients, 75 were males. The mean age was 64 years, range 42 to 83 years. A history consistent with coronary heart disease and documented by ECG was present in 84 patients. Hypertension was present in 70 patients, whereas 41 patients had diabetes mellitus proved by glucose tolerance test. Forty-six patients had abnormal lipid profiles and 44 patients had a history of cigarette smoking. All patients underwent four vessel arch arteriography. One hundred and four had visualization of the extracranial and intracerebral circulation; ten, early in the series, had only extracranial arteriography. One hundred and thirty-nine carotid endarterectomies were performed with continuous ten-lead EEG monitoring to evaluate the adequacy of cerebral blood flow during carotid occlusion. All endarterectomies were performed under general anesthesia. All patients were clinically evaluated in the immediate postoperative period (30 days). Long-term follow-up ranged from one to ten years (mean, 59 months). Six patients were lost to long-term follow-up. If follow-up evaluation showed no further symptoms of VBI, the patient was classified as improved or asymptomatic. Patients with continued symptoms of VBI were classified as unimproved or unchanged.

RESULTS

Symptoms of VBI occurred singly or in combination, and transient, lasting minutes to hours and completely resolved. The most frequent symptom of VBI was visual changes occurring in 57 (50%) patients (Table 1). Visual changes consisted of polyopia, diplopia, and blurred vision.

were

Vertebrobasilar visual symptoms occurred with or without amaurosis fugax. However, amaurosis fugax was consid¬ ered a symptom of carotid insufficiency. Dizziness or lightheadedness was the presenting symptom in 35 (31%) patients, whereas syncope occurred in 34 (30%). Vertigo (eight), dysarthria (four), and ataxia (one) occurred less

frequently.

Based on symptoms, 114 patients were divided into the following three clinical groups: symptoms of VBI alone occurred in 24 patients, symptoms of VBI plus transient ischémie attacks occurred in 61 patients, and symptoms of VBI in patients with cCS occurred in 29 patients. In the 114 patients, 139 carotid endarterectomies were performed. Twenty-five patients had staged bilateral endarterecto¬ mies. Prophylactic "second side" endarterectomies were performed in 12 patients for contralateral high-grade stenoses, and eight for contralateral ulcerative plaque disease. Only five endarterectomies were necessary to

attain relief of persistent VBI symptoms. Cerebral ischem¬ ia during occlusion, sufficient to require placement of a temporary shunt, was observed in 26 patients. The 139 endarterectomies resulted in four patients suffering tran¬ sient postoperative CNS deficits: an incidence of 2.8%. Permanent CNS deficits developed in two patients in the postoperative period from which there was a slow improve¬ ment during several months. Transient peripheral nerve deficits developed in 12 patients: an incidence of 8%. These consisted of weakness of the tongue, ipsilateral vocal cord, ipsilateral half of the lower lip, and sensory deprivation of the larynx. Twenty-one of 24 (88%) patients with symptoms of VBI alone improved after carotid endarterectomy (Table 2). Fifty of 61 (82%) patients with symptoms of VBI plus transient ischémie attacks were relieved of both their VBI and carotid insufficiency symptoms. Twenty of 29 (69%) of patients with VBI plus cCS were improved after endarter¬ ectomy. Thus, 91 (80%) of 114 patients were improved by endarterectomy, whereas 17 (15%) were unchanged. Six (5%) were lost to follow-up. Of the 17 (15%) patients whose symptoms were unchanged by carotid endarterectomy (Table 2), 13 "preoperative diagnostic errors" were documented. These errors consisted of the following symptoms that mimicked those of VBI: three patients had subclavian steal syndrome; three, cardiac arrhythmias; two, seizure disorders; one, angiographically incomplete Circle of Willis; two, diabetic or hypertensive retinopathies (mimicking VBI visual symptoms); one, idiopathic hypertrophie subaortic stenosis; and one, labyrinthitis. Intraoperative carotid bifurcation pressures were obtained in 90 patients. Pressures were expressed as a gradient percentage: common carotid artery pressure minus internal carotid artery pressure divided by common carotid artery pressure. Eighty patients showed sympto¬ matic improvement. Fifty-nine of the 80 patients who showed symptomatic improvement had no gradient or a gradient percentage of less than 25% across the carotid bifurcation (Table 3). At late follow-up, ranging from one to ten years after endarterectomy, 63% (Fig 1) or 69 patients were alive. Seventy-seven percent (43/57) of the patients who originally had symptoms of VBI plus carotid transient ischémie attacks were alive. Of patients who originally had VBI alone, 62% (13/22) were alive, whereas only 44% (13/29) of patients who originally had symptoms of VBI plus cCS were living. Of the patients alive at ten years who initially had VBI alone, 85% (11/13) were asymptomatic. Eighty-eight Table

1.—Symptoms

percent (38/43) of the patients with symptoms of VBI plus

carotid transient ischémie attacks, and 92% (12/13) patients with symptoms of VBI plus cCS were symptomfree. Overall, 88% of the patients alive at follow-up extend¬ ing to ten years were asymptomatic. Causes of death (more than one year postoperative) included 17 (44%) cardiac, 14 (36%) cerebrovascular accidents (CVA), four cancer, and four unknown Fig 2. COMMENT Prior to suffering actual cerebral infarction, more than 60% of the patients will give a history of vertebral basilar or carotid territory transient ischémie attacks." If infarc¬ tion occurs in the vertebral basilar territory, the mortality is 2J/2 times greater than that seen with carotid system infarction." This "stroke prone" population may benefit from carotid endarterectomy by alleviating symptoms of VBI with or without carotid transient ischémie attacks, and by preventing stroke. The goal of this retrospective analysis was to increase our understanding of VBI and evaluate the results of carotid endarterectomy for symp¬ toms of VBI alone, or VBI in conjunction with carotid transient ischémie attacks or cCS. Visual changes, dizziness, and syncopal episodes were the most common presenting symptoms (Table 1). Of 29 patients with symptoms of VBI plus cCS, 19 experienced visual changes as a principal symptom. Of 61 patients with symptoms of VBI plus carotid transient ischémie attacks, visual changes in 30 and dizziness in 24 were the main presenting symptoms. Of 24 patients with symptoms of VBI alone, 12 had syncope. During carotid endarterectomy cerebral ischemia docu¬ mented by EEG changes necessitated the use of a tempo¬ rary shunt in 26 patients. To determine why ischemia occurred during endarterectomy, the four vessel arch arteriograms and operative records of these 26 patients were reviewed. In 25 of 26 patients, the posterior commu¬ nicating artery was patent. Evidence of intracranial disease (stenoses at the siphon, or stenoses in the circle of Willis) was present in 14 of 26 patients. Despite a stump pressure of greater than 50 mm Hg in 12 of 26 patients, ischémie changes in the EEG necessitated shunting. From this analysis, it seems that patients with symptoms of VBI and artériographie evidence of intracranial disease are at a high risk for cerebral ischemia during cross clamp. These patients, regardless of the stump pressure, should have placement of a temporary indwelling shunt during endar¬

terectomy. The

intraoperative bifurcation

of Vertebrobasilar

pressure

gradients

Insufficiency

Visual

_Changes_Dizziness_Syncope_Vertigo_Dysarthria alone Vertebrobasilar insufficiency Vertebrobasilar insufficiency ischémie attacks (N Vertebrobasilar insufficiency completed stroke (N =

=

Total

(%)(N

=

114)

(N

plus

=

24)

8

7

12

2

carotid transient

Ataxia 1

...

61)_30_24_14_5_2_L___ plus carotid 29)_19_4_8_1_2_L__ 34 1 57(50) 35(31)

(30)

8~(7)

4(3.5)

(.8)

obtained in 90 endarterectomies showed that 59 of 80 patient symptoms improved when there was a gradient percentage of 25% or less (Table 3). This indicates that a bifurcation pressure gradient is apparently not necessary for relief of VBI symptoms. On the basis of these pressure studies, it is difficult to state that symptomatic improve¬ ment after endarterectomy was due totally to elimination of a stenotic focus and increased cerebral blood flow. Instead, it may be postulated that emboli from the carotid bifurcation, causing transient decreases in cerebral perfu¬ sion (especially in the watershed region) may initiate a regional cerebral dysautoregulation'" phenomenon, there¬ by causing symptoms of VBI. This, however, remains to be further evaluated. Carotid endarterectomy produced relief of VBI symp¬ toms in 80% (91/114) of the patients (Table 2). Seventeen of 114 (15%) patient symptoms were unchanged after endar¬ terectomy. Preoperative diagnostic errors were noted in 13 patients, all of whose condition was unimproved by endar¬ terectomy. These errors consisted of symptoms that mimicked those of VBI. If these 13 patients were excluded from the operative series of 114, then 101 patients would Table

2.—Operative Results of Insufficiency Improved

Vertebrobasilar

Unchanged Unknown

_(%)_(%)_(%) Vertebrobasilar insufficiency alone (N 24)_21 (88)_2J8)_1 Vertebrobasilar insufficien¬ cy plus transient ischemic attacks (N 61) 50(82)_8(13)_3 =

=

Vertebrobasilar insufficien¬ cy plus completed stroke

(N

Total

=

(N

29)_20 (69)_7 (24) =

114)

91(80)

17(15)

2

(4)

have had endarterectomies, with 91 demonstrating relief of symptoms—an overall improvement of 90%. Most of these 13 diagnostic errors occurred early in the series. The present use of noninvasive carotid testing techniques, superior cardiovascular evaluation, improved arteriog¬ raphy, and extensive ear, nose, and throat, and neurologic evaluations have greatly reduced this incidence.

1 VBI r cTIA

(n ' 43)

20VERALL

(n

--

69J

ZVBI ALONE ln= 131

(n-13)

WBIrcCS

90-

Results of carotid endarterectomy for vertebrobasilar insufficiency: an evaluation over ten years.

Results of Carotid Endarterectomy for Vertebrobasilar Insufficiency An Evaluation Over Ten Years David Rosenthal, MD; David Cossman, MD; C. Bruce L...
1MB Sizes 0 Downloads 0 Views