Carotid Artery Endarterectomy in Patients With Contralateral Carotid Artery Occlusion: Perioperative Hazards and Late Results

Carotid Artery Endarterectomy in Patients With Contralateral Carotid Artery Occlusion: Perioperative Hazards and Late Results Giovanni P. Deriu, MD, Lorenza FranceschL MD, Domenico Milite, MD, Alessio Calabro, MD, Aldo Saia, MD, Franco Grego, MD, Diego Cognolato, MD, Paalo Frigatti, MD, and Mario Diana, MD, Padova, Italy

The aim of this study was to analyze and compare the perioperative hazards and late results of internal carotid endarterectomy (CEA) in patients with and without contralateral internal carotid artery occlusion. From March 1980 to April 1990, 375 consecutive patients underwent 439 CEAs at the First Department of Vascular Surgery of Padova Medical School. Patients were divided into two groups; group 1 (61 patients) had contralateral internal carotid artery occlusion and group 2 (314 patients) did not (378 CEAs, 64 bilateral). Indications for CEA were similar in both groups. The only significant difference in patient characteristics was a higher rate of previous stroke in group 1 (11% vs. 3%, p < 0.001). General anesthesia, continuous EEG monitoring, selective intraluminal shunt, and arteriotomy closure with a polytetrafluoroethylene patch (PTFE) were used routinely in both groups. An intraluminal shunt was inserted more frequently in group 1 than in group 2 (69% vs. 17%, p <0.001). Major perioperative stroke occurred in one patient in each group (1.7% vs. 0.31%, respectively; NS). Early fatal stroke rates were 0% and 0.95% in groups 1 and 2, respectively (NS). All patients had neurologic examinations and duplex scans every 6 months (range 6 to 118 months; mean 42 months). Kaplan-Meier survival curves were virtually identical in the two groups; the majority of deaths were caused by myocardial infarction and cancer. There were no stroke-related deaths in group 1 as compared with 8.2% in group 2 (NS). New neurologic symptoms appeared in 4.7% of patients in group 1 and 6% in group 2 (NS) whereas the late stroke rates were 0% and 3.1%, respectively (NS). Restenosis was observed in two and three patients in groups 1 and 2, respectively (NS). In conclusion, CEA for ulcerated or stenotic lesions of the internal carotid artery in patients with contralateral carotid occlusion is associated with very low early and long-term neurologic morbidity and mortality, similar to findings in patients who undergo CEA with a patent contralateral carotid artery. (Ann Vasc Surg 1994;8:337-342.)

Carotid endarterectomy (CEA) is widely used for the prevention of stroke. The immediate and long-term results are excellent if the operation is performed by skilled specialists; otherwise the risk of neurologic damage related to the operation is so high that it is preferable to let the disease From the Departments of Vascular Surgery, Internal Medicine, and Neurology, University of Padova, Padova, Italy. Reprint requests: G. Deriu, MD, Cattedra de Chirurgia Vascalare, Universita di Padova, Via Giustiniani, 2, Padova 35100, Italy.

follow its natural course. Recent attention has been focused on patients with carotid artery stenosis and occlusion of the contralateral carotid artery. As reported by several authors, 1-3 these patients are at risk (8% to 10%) for intraoperative and postoperative complications. Other surgical reports, however, have shown an acceptable risk and long-term benefits in the prevention of stroke for these same patients. 4s The aim of this study was to compare the perioperative hazards and late results of CEA in patients with and without contralateral carotid artery occlusion. The tactics 337

338

Annals of Vascular Sttrgery

Deriu et al.

and technical details of cerebral protection during carotid clamping will be discussed.

PATIENTS AND METHODS From March 1980 to April 1990, 375 patients underwent a total of 439 CEAs at the First Department of Vascular Surgery of Padova Medical School. Patients undergoing simultaneous reconstruction of one or more other cerebral arteries during the same operation were not included in this study. Patients were divided into two groups. Group 1 included 61 patients with occlusion of the contralateral carotid artery and group 2 consisted of 314 patients undergoing 348 CEAs who had a patent contralateral carotid artery. All patients were evaluated preoperatively by noninvasive techniques (continuous-wave Doppler velocimetry and duplex scanning) and arteriography. Preoperative CT scans were obtained in only 115 cases (31 in group 1, 84 in group 2). General anesthesia was used in all cases. All patients had systemic heparin therapy (5000 IU) without neutralization at the end of the operation. The arteriotomy was routinely extended to the internal carotid artery beyond the end of the plaque and was closed with a polytetrafluoroethylene patch (PTFE). Continuous EEG monitoring was routinely used to detect brain intolerance to carotid cross-clamping. EEG recordings during carotid cross-clamping were classified as follows: (1) no abnormalities, (2) bilateral delta activity, (3) focal ipsilateral delta activity, (4) focal contralateral delta activity, (5) decreased fast activity or flattened EEG signal, and (6) symmetric bursts of theta-delta waves occurring independent of carotid clamping. EEG categories 1 and 6 were considered "normal" whereas categories 2, 3, 4, and 5 were considered "abnormal." Whenever EEG alterations appeared and were not related to

arrhythmia or hypotension, a Brener (from 1980 to 1982) or a Pruitt-Inahara (from 1983 to 1990) shunt was inserted and normalization of EEG recordings was achieved in all cases. Patients were usually discharged 7 days postoperatively on antiplatelet therapy (dipyridamole, acetylsalicylic acid). All patients were examined clinically in the immediate postoperative period and with ultrasound hnaging at 1 m o n t h and every 6 m o n t h s thereafter. Whenever clinical or ultrasound signs of substantial hemodynamic progression of atherosclerosis were found at the 6-month checkup, new arteriograms were obtained. Statistical comparisons between the two groups were made with the chi-square test or Fisher's exact test as appropriate. Statistical significance was inferred for p < 0.5. Life-table comparisons were carried out with the log-rank test.

RESULTS Age, sex, and risk factors were comparable in both groups of patients (Table I). Preoperatively the two groups were relatively similar with regard to the absence of symptoms as well as the presence of transient ischemic attacks and nonhemispheric symptoms. The only significant difference was a higher rate of previous stroke in group 1 (11% vs. 3%, p <0.001) (Table II). Preoperative CT scans showed ischemic lesions in 66 patients (57.3%). Indwelling shunts were required in 42 of 61 patients in group 1 (69%) and 64 of 314 in group 2 (17%) (p <0.001). With regard to immediate results (during the first 30 days after operation), there were no significant differences in neurologic or general morbidity or in mortality rates between the two groups. Only one patient in each group ( 1.7% and 0.31% in groups 1 and 2, respectively) had per-

Table I. Age, sex, and risk factors in patients with (group 1) and without (group 2) contralateral carotid artery occlusion Patient characteristics Mean age (yr) Sex (M/F) CAD Smoking Diabetes Hypertension CAD = coronaryartery disease. *By Fisher's exact test.

Group 1 (N = 61) 64 38 43 9 40

57/4 (62.3%) (70.5%) (14.7%) (65.5%)

Group 2 (N = 314)

p value*

66 254/60 179 (57%) 211 (67%) 43 (13.7%) 193 (61.4%)

NS < 0.5 NS NS NS NS

Vol. 8, No. 4 1994

CEA in patients with contralateral occlusion 339

manent neurologic damage (NS). Fatal strokes occurred in 0% and 0.95% of patients in groups 1 and 2, respectively. The three fatal strokes were due to basilar artery thrombosis, cerebral hemorrhage, and thrombosis of the operated carotid artery, occurring 6, 10, and 15 days after operation, respectively. The immediate neurologic relevant complication rate (combined neurologic morbidity plus mortality) was 1.7% and 1.3% in groups 1 and 2, respectively (NS). The six minor neurologic deficits (two reversible ischemic neurologic deficits and four transient ischemic attacks), all with negative postoperative CT scans,

as well as the two nonneurologic deaths, involved patients in group 2 only (NS) (Table III). By the last clinical and investigational follow-up visit in the spring of 1990 (range 6 to 118 months; m e a n 42 months), 64 patients (19.5%) had died. The Kaplan-Meier survival plots for the two groups are illustrated in Fig. 1. No statistically significant difference was observed in the m e a n survival rates as the two curves were virtually superimposable. The majority of late deaths were due to myocardial infarction and cancer in both groups (Table IV). Of the 13 deaths in group 1, none was caused by stroke. This contrasts with

Table II. Preoperative symptoms in patients with (group 1) and without (group 2) contralateral carotid artery occlusion Group 2 (N = 314)

Group 1 (N = 61) Preoperative s y m p t o m s TIA Nonhemispheric Previous stroke Asymptomatic

symptoms

n

%

n

%

p value*

35 7 7 12

58 11 11 20

186 34 8 86

58 12 3 27

NS NS < 0.001 NS

TIA = transient ischemic attack. *By chi-square analysis.

Table IlL Early postoperative (30 days) results in patients with (group 1) and without (gr6up 2) contralateral carotid occlusion Group 2 (N = 314)

Group I ( N = 61) Postoperative r e s u l t s

n

%

n

%

p value*

Major strokes Neurologic deaths R e l e v e n t complications RINDs TIAs Nonneurologic deaths

1 0 1 0 0 0

1.7 -1.7 ----

1 3 4 2 4 2

0.31 0.95 1.26 0.63 1.27 0.63

NS NS NS NS NS NS

RINDs = reversible ischemic neurologic deficits; TIAs = transient ischemic attacks. *By Fisher's exact test.

Table IV. Causes of late death in patients with (group 1) and without (group 2) contralateral carotid occlusion Group 1 (N = 61)

Group 2 (N = 314)

C a u s e s of d e a t h

n

%

n

%

p value*

Cardiac Cancer Stroke Other

8 5 0 0 13

61.6 38.4 -

29 8 4 10 51

56.8 15.6 7.8 19.6

NS < 0.0001 NS NS

*By Fisher's exact test.

Annals of Vascular Surgery

340 Deriu et al.

ated whereas another patient became symptomatic because of lesions at the origin of the common carotid artery (as noted previously). Eight patients in group 2 had new atherosderotic involvement including three restenoses of the carotid bifurcation and five occlusions.

100%

75%

DISCUSSION

""V,,. V

50%

25%

0%

I

0

I

I

I

I

I

I

500 I000 1500 2000 2500 3000 3500 4000 Days !

[3 Group 1

--V-- Group 2 / J

Fig. 1. Kaplan-Meier life-table analysis of survival.

4 of the 51 deaths in group 2 that were due to stroke (8.2%) but the difference was not statistically significant. Overall mortality was 13 (21.3%) and 56 (17.8%) in groups 1 and 2, respectively. No significant difference was found in regard to late neurologic symptoms. Two patients in group 1 had repeated episodes of nonhemispheric symptoms. In one this was due to progression of atherosclerosis at the origin of the common carotid artery, which was corrected with a repeat operation. In the second patient, neurologic symptoms were secondary to a marked increase in the dimensions of an old ischemic area in the hemisphere contralateral to the operated carotid artery. Thirteen patients in group 2 had new neurologic symptoms (seven strokes, five transient ischemic attacks, and one nonhemispheric event; NS). No significant difference was noted between the two groups regarding the course of arterial disease in the operated carotid artery. One patient in group 1 had an asymptomatic restenosis at the carotid bifurcation that was successfully reoper-

The neurologic outcome of acute occlusion of the internal carotid artery is unpredictable, ranging from death due to massive cerebral infarction to complete asymptomaticity.9"~~Asymptomatic carotid occlusion does not, however, preclude subsequent stroke, which has been reported to occur at a rate of 2% to 5% per year in the ipsilateral hemisphere. H " Grillo and Patterson, 14 on the other hand, have indicated that the risk of stroke is higher in the contralateral hemisphere. Controversy still exists as to the proper treatment of patients with carotid artery stenosis associated with occlusion of the contralateral artery. CEA of the stenotic carotid artery can be considered effective in the prevention of stroke only if perioperative morbidity and mortality are lower t h a n the incidence of neurologic events in patients treated medically. Data in the literature on this subject are somewhat confusing. 15'~6 Most authors have reported an increased intraoperative risk associated with temporary carotid clamping in this setting and advocate the use of intraoperative cerebral monitoring and protective measures. Of the techniques currently available for control of cerebral function during carotid clamping, surgery under regional anesthesia 5 and continuous EEG monitoring 7 are both consistent with selective use of the intraluminal shunt. In our experience, continuous intraoperative EEG monitoring and selective use of an intraluminal shunt have been very effective in detecting patients who could not tolerate clamping and in guaranteeing adequate cerebral perfusion, respectively. EEG abnormalities were reversed in all cases. This and the very low rate of neurologic sequelae on awakening attest to the reliability of the intraluminal shunt. It is of note that the five relevant neurologic complications in our series ( t w o strokes and three neurologic deaths) occurred 36 hours and 2, 6, 10, and 15 days after surgery. Of interest in our series is that the intraluminal shunt was used significantly more frequently in group 1 as compared with group 2 (69% vs. 17%, p <0.001). This most likely reflects the inadequacy of collateral flow. However, although oc-

Vol, 8, No. 4 1994

clusion of the contralateral carotid artery increases the risk of cerebral ischemia during carotid clamping, this hazard is still present even w h e n the contralateral internal carotid artery is patent. The poor results achieved by surgeons w h o routinely perform CEA without the use of an intraluminal shunt 3"1719 or use it on the basis of subjective criteria such as arteriographic analysis or objective but debatable criteria such as stump pressure 2~ are therefore easily understandable. On the other hand, surgeons w h o routinely use an intraluminal shunt 4 have reported complications associated with the use of this shunt (extensive neck dissection, malpositioning) 21 or difficulty in performing the endarterectomy because the shunt impinges on the operative field. To avoid these problems we routinely perform the major part of the CEA on the segment most involved before inserting the shunt. This short period of ischernia (5 to 6 minutes at most) has never proved to be dangerous and EEG recordings return to normal within a few minutes. Intraoperative or immediate postoperative stroke may also be related to surgical technique. Limiting arteriotomy to the carotid bulb, without extending it to the internal carotid artery beyond the end of the plaque, can cause the formation of intimal flaps with an attendant risk of thrombosis. Moreover, primary suturing of the arteriotomy (without a patch) can lead to an immediate or progressive restenosis capable of causing thrombosis, especially if the arteriotomy was extended to the internal carotid artery. In our series, the carotid arteriotomy was roufinely extended to the internal carotid artery and closed with a PTFE patch. Although these technical details were time consuming, continuous EEG lnonitoring was an obvious security measure. Of note is that 25 (23%) of the EEG abnormalities appeared more than 5 minutes after carotid clamping. Other unexpected benefits of EEG monitoring were the detection of ischemia due to faulty head positioning (five cases, all of which resolved with correction of hyperextension of the head) and a poorly functioning intraluminal shunt (six cases, all of which resolved w h e n the shunt was changed). The real hazard of CEA in patients with contralateral carotid occlusion is the possibility of stroke occurring not only in the ipsilateral but also in the contralateral hemisphere) Even though this risk can be attributed to general factors such as an unstable hemodynamic compensatory state of the hemisphere ipsilateral to the occluded carotid artery, hypertension, or increased

CEA in patients with contralateral occlusion 341

blood viscosity, these complications must be considered intraoperative or early postoperative incidents directly related to the surgical procedure. The need to pay close attention to the treatment of the external as well as the internal carotid artery has already been underscored, 22 and this is especially true in the case of occlusion of the contralateral carotid artery. One of our patients in group ! had a postoperative complication consisting of occlusion of the external carotid artery on the operated side with cerebral infarction 36 hours after the operation in the hemisphere contralateral to the operated carotid artery. Endarterectomy had been performed blindly in the ipsiateral external carotid artery, which in this specific case provided the collateral ckculation in compensation for the occluded contralateral carotid artery. Presently there are several surgical techniques available for performing CEA, all of which have thek advantages and disadvantages. Nevertheless, it is important to stress that a correct, wellarticulated sequence of individual technical tasks should ensure an excellent final outcome. An essential factor in improving the long-term results is a very accurate routine clinical and ultrasound follow-up protocol capable of detecting restenoses, which in patients with an occluded contralateral carotid artery could be particularly critical. 23 In our series the rate of late neurologic symptoms was very low in both groups. Follow-up investigations showed that the operated carotid artery was normal in 97% of patients. Only 10 patients (two in group 1 and eight in group 2; NS) experienced ongoing evolution of arterial disease in the area of the CEA (myointimal or neointimal hyperplasia, new atherosclerotic plaque). Another reason for these excellent results, as confirmed recently by other researchers, 2. is undoubtedly the routine use of the patch after CEA, which in our opinion 25 not only improves the immediate results but also prevents long-term restenosis. Our life-table analysis of patients undergoing CEA with or without occlusion of the contralateral carotid artery demonstrates the absence of statistically significant differences in survival rates. The main causes of late death were myocardial infarction and cancer. The only four deaths related to stroke occurred in patients with patent contralateral carotid arteries. These findings are consistent with those of Sachs et al. 4 and Mackey et al.6 who noted no significant differences in early and late neurologic complications

342

Deriu et al.

or in survival w h e n groups with and without occlusion of the contralateral carotid artery were compared. CONCLUSION Our data demonstrate that patients with carotid stenosis and contralateral carotid artery occlusion may undergo CEA with morbidity and mortality rates similar to those in patients without contralateral carotid artery occlusion. Moreover, as shown in our series there was no difference concerning the risk during carotid clamping; this risk can be safely identified through EEG monitoring and minimized by selective use of an intraluminal shunt, allowing the neurologic complications to be virtually reduced to zero irrespective of the patency of the contralateral carotid artery.

REFERENCES 1. Moore DJ, Modi JR, Finch WJ, et al. Influence of the contralateral carotid artery on neurological complications following carotid endarterectomy. J Vasc Surg 1984;1:409414. 2. Andreassian B, Bouttier S, Nussattme O, et al. Evaluation preoperatoire du risque neurologique en chirurgie carotidieune. In Kieffer E, Natali J, eds. Aspects Techniques de la Chirurgie Carotidienne. Paris: AERCV, 1987, pp 57-59. 3. Lesage R, Paris E, Koskas F, et al. Surgical reconstruction of the internal carotid artery with contralateral occlusion without use of shunt. Ann Vasc Surg 1991;5:55-60. 4. Sachs SM, Fulenwider JT, Smith RB, et al. Does contralateral carotid occlusion influence neurologic fate of carotid endarterectomy? Surgery 1984;96:839-844. 5. Friedman SG, Riles TS, Lamparello P3, et al. Surgical therapy for the patient with internal carotid artery occlusion and contralateral stenosis. J Vasc Surg 1987;5:856-861. 6. Mackey WC, O'Donnell TF Jr, Callow AD. Carotid endarterectomy contralateral to an occluded carotid artery: Perioperative risk and late results. J Vasc Surg 1990;11:778-785. 7. Hammacher ER, Eikelboom DC, Bast TJ, et al. Surgical treatment of patients with a carotid artery occlusion and a contralateral stenosis. J Cardiovasc Surg 1984;25:513-517.

Annals of Vascular Surgery

8. La Muraglia GM, Darling RC, Brewster DC, et al. Neurologic sequelae with internal carotid artery occlusion. Arch Surg 1987;122:432-435. 9. Fritz VU, Voil CL, Levien Id. Internal carotid occlusion: Clinical and therapeutical implications. Stroke 1985;16:940944. 10. Nicholls SC, Bergelin R, Strandness DE. Neurological sequelae of unilateral carotid artery occlusion: Immediate and late results. J Vasc Surg 1989;10:542-548. 11. Nicholls SC, Kohler TR, Bergelin RO, et al. Carotid artery occlusion: Natural history. J Vasc Surg 1986;4:479-485. 12. Furlan AJ, Whisnant JP, Baker HL et al. Long-term prognosis and carotid artery occlusion. Neurology 1980;30:986-988. 13. Cote R, Barnett HJM, Taylor DW, et al. Internal carotid occlusion: A prospective study. Stroke 1983;14:898-902. 14. Grillo P, Patterson RH Jr. Occlusion of the carotid artery: Prognosis (natural history) and the possibilities of surgical revascularization. Stroke 1975;6:17-20. 15. Hertzer NR, Flanagan RA, O'Hara PJ, et al. Surgical versus nonoperative treatment of symptomatic carotid stenosis. Ann Surg 1986;204:154-162. 16. Fields WS, Lemak NA. Joint study of extracranial arterial occlusion: Internal carotid artery Occlusion. JAMA 1976;235: 2734-2738. 17. Ott DA, Cooley DA, Chapa L, et al. Carotid endarterectomy without temporary intraluminal shunt: Study of 308 consecutive operations. Ann Surg 1980;191:708-714. 18. Baker WH, Littooy FN, Hayes AC, et al. Carotid endarterectomy without a shunt: The control series. J Vasc Surg 1984; 1:50-56. 19. Littooy FN, Halstuk KS, Mamdani M, et al. Factors influencing morbidity of carotid endarterectomy without a shunt. Am Surg 1984;50:350-353. 20. Sillesen H, Schroeder T, Rasmussen L, et al. Carotid endarterectomy in patients with carotid artery occlusion and a contralateral carotid artery stenosis: Perioperative risk a n d late results. Eur J Vasc Surg 1987;1:85-89. 21. Calhorm TR, Kitten CM. Proximal shunt dissection: A potential problem in carotid endarterectomy. Texas Heart Inst J 1985;12:359-361. 22. McIntyre KE, Ely RL III, Maline JM, et al. External carotid artery reconstruction: Its role in the treatment of cerebral ischemia. Am d Surg 1985;150:58-64. 23. Dos M, Hertzer NR, Ratliff NB, et al. Recurrent carotid stenosis: A five-year series of 65 reoperations. Ann Surg 1985;202:28-35. 24. Piepgras DG, Sundt TM, Malsh WR, et al. Recurrent carotid stenosis: Results and complication of 57 operations. Ann Surg 1986;203:205-213. 25. Deriu GP, Ballotta E, Bonavina L, et al. The rationale for patch graft angioplasty after carotid endarterectomy: Early and long-term follow-up. Stroke 1986; 15:972-979.