A rational approach to recurrent carotid stenosis R o b e r t G. Atnip, M D , M a r k Wengrovitz, M D , R o b e r t R. M. Gifford, M D , Marsha M. Neumyer, BS, RVT, and Brian L. Thiele, M D , Hershey, Pa. To further characterize the incidence and morbidity of recurrent carotid stenosis, we reviewed 184 consecutive carotid endarterectomies performed in a university hospital between August 1983 and January 1988, in patients followed after operation with serial duplex ultrasonography. Recurrent stenosis of greater than 50% diameter reduction developed in eleven arteries (6.0%) at a mean interval of 10.2 4" 7.8 months. Three of the eleven (1.6% of the total) had associated transient ischemic attack, and none had strokes. Restenosis was significantlymore frequent in diabetic patients than in nondiabetic patients (13.3% vs 4.5%; p < 0.05); and among patients whose primary stenoses had been symptomatic compared to asymptomatic (11.0% vs 1.5%;p < 0.02). No statistically significant association with restenosis could be established for gender, hypertension, or smoking. Completion angiography and/or Doppler spectral analysis had been performed, and resuits were normal at the primary operation in 10 of the 11 patients. Only six of 184 arteries (3.3%) had vein patch closure, but none of these restenosed. Uneventful reoperation with patch closure was performed in three patients with transient ischemic attacks and two with preocclusive restenoses. Lesions were myointimal hyperplasia in four and atheroma in one. Three of the unoperated restenoses have shown regression on duplex scanning, but a fourth progressed to asymptomatic occlusion. Carotid restenosis is uncommon, even without routine use of vein patch angioplasty. Reoperation should be reserved for patients with associated symptoms or >80% restenosis. (J VASe SuRG 1990;11:511-6.)
No component of the surgical treatment of carotid atherosclerosis has escaped controversy. As the number of carotid endarterectomies performed in this country has increased, so has the diversity of opinion regarding all elements of this operation, from minute technical details to fundamental philosophical and ethical issues. The proponents of ca,tid endarterectomy have sought to identify means, of reducing both early and late surgical morbidity to improve surgical outcome. Among the surgical complications targeted by these efforts is the problem of recurrent carotid stenosis. At least from a morphologic Standpoint, recurrent stenosis represents a failure of arterial reconstruction and thus merits investigation into its pathogenesis and clinical behavior. The findings of other investigators in this field have kindled our own interest in recurrent carotid stenosis
From the Departmentof Surgery,The Milton S. HersheyMedical Center,The PennsylvaniaStateUniversityCollegeof Medicine. Presented at the Third Annual Meeting of the Eastern Vascular Society, Bermuda,May 4-7, 1989. Reprint requests: Robert G. Amip, MD, Vascular Surgery Section, The Milton S. Hershey Medical Center, P.O. Box 850, Hershey, PA 17033. 24/6/18089
and stimulated us to study its incidence and significance in our own patient population. MATERIAL AND METHODS Between August 1983 and January 1988, 184 consecutive carotid endarterectomies were performed in 175 patients at The Milton S. Hershey Medical Center of the Pennsylvania State University by or under the direct supervision of fellowshiptrained vascular surgeons. A uniform surgical technique was used as described herein. All patients were followed with duplex scanning of the extracranial vessels every 3 months for the first postoperative year and every 6 months thereafter. Records were reviewed retrospectively to determine the incidence and morbidity of recurrent stenosis of the operated carotid arteries and to identify possible predisposing factors. Statistical significance o f possible risk factors was tested with Fisher's exact test for two degrees of freedom. SURGICAL TECHNIQUE All operations were performed under general anesthesia, with continuous electroencephalographic monitoring by means of a compressed spectral array 511
Journal of VASCULAR SURGERY
512 Atnip et al.
device in most instances. Carotid shunts were used selectively. Carotid bifurcation endarterectomy was carried out with loupe magnification in the plane between the media and adventitia, removing all visible media that were not tightly adherent. The arteriotomy was extended onto the internal carotid artery beyond the distal termination of the plaque as far as necessary to achieve an easily visualized smooth endpoint. Tacking sutures were rarely required. Vein patches were used selectively for arteriotomy closure, usually in patients with carotid vessels judged to be abnormally small or in those requiring intraoperative revision of the endarterectomy for any reason. Vein was variably obtained from either groin or ankle. All closures were done with loupe magnification and 60 and 7-0 polypropylene sutures. Intraoperative assessment of the endarterectomy was performed by use of either contrast arteriography or high-frequency (20 MHz) Doppler velocity spectral analysis. Arteriography was always performed if the velocity spectra were abnormal. All patients were systemically heparinized during operation and given low-molecular weight dextran at 25 ml/hr for 48 hours after operation. RESULTS Among all patients three deaths (1.6%) and two permanent neurologic deficits (1.1%) occurred within the perioperative period. No immediate carotid occlusions and no stroke deaths occurred. All surviving patients were followed at least 1 year after surgery (range 12 to 56 months, mean 35 months). Recurrent stenosis of the operated carotid resulting in at least 50% diameter reduction by duplex criteria occurred in 11 patients (6.0%) at a mean of 10.2 _+ 7.8 months after surgery. Three of the 11 patients had ipsilateral transient ischemic attacks (two amaurosis, one arm/face paresis), yielding a symptomatic recurrent stenosis rate of only 1.6%. No patient with recurrent stenosis suffered a stroke, and the remaining eight have been free of symptoms. Reoperation was performed in the three patients with symptoms and in two other patients with preocelusive lesions. Duplex findings were confirmed angiographically in all of these patients before reoperation. Disease was myointimal hyperplasia in four patients and atheroma in one. In all cases an endarterectomy plane could be developed to excise the bulk of the recurrent lesion. Saphenous vein patches were used for closure in all instances. There were no complications of reoperation, and all five reoperated vessels remain patent from 3 to 23 months of follow-up.
Among the six patients not undergoing reoperation one stenosis has remained stable, and three have regressed at a mean of 34 months of follow-up. One patient progressed to asymptomatic internal carotid occlusion after refusing reoperation for a severe stenosis. A final patient with a preocclusive restenosis has remained free of symptoms but has declined any further intervention. Restenosis was more frequent in men (10 of 128; 7.8%) than in women (1 of 56; 1.8%), but this difference did not achieve statistical significance. Ten of the 11 patients with restenosis had originally undergone operation for carotid territory ischemic symptoms, whereas only one had operation for asymptomatic stenosis. The rate of restenosis was thus significantly higher in patients with symptomatic primary carotid stenosis than in those with asymptomatic primary stenosis (11.0% vs 1.5%; p < 0.02). Vein patch closure had been used in only six,~f the 184 primary procedures, and none of these arteries restenosed. The small number of patched arteries rendered statistical comparison invalid. Intraoperative high-frequency Doppler or arteriography had been performed at the primary operation in 10 of 11 patients with restenosis, but residual stenosis or technical error had not been shown in any. With regard to other possible factors predisposing to recurrent carotid stenosis, comparative data were available in 154 patients (Table I). No statistically significant association could be demonstrated between the risk ofrestenosis and either hypertension or cigarette smoking. Blood lipid values were not available for enough patients to make a comparison. It is surprising to note that there was a strong st• tistical association between diabetes mellitus and carotid restenosis. Six of 11 patients (54%) with restenosis were diabetic, compared to only 27% of patients without restenosis. Expressed differently, the risk of restenosis was 6/45 (13.3%) among diabetic patients, compared to 5/109 (4.5%) among nondiabetic patients (p < 0.05). DISCUSSION
Recurrent carotid stenosis is a proven entity whose existence has been recognized for at least a decade. Although first reported in the carotid artery by Stoney and String in 1976,1 the phenomenon of arterial restenosis had been recognized years earlier as contributing to vascular reconstructive failures, z,3 Unfortunately, surgeons have not always taken full advantage of the methods available for objective
Volume 11 Number 4 April 1990
Recurrent carotid stenosis 513
Table I. Relationship of clinical variables to carotid restenosis Number ofpatients with restenosis
Incidence of restenosis (%)
Factor
Number ofpatients
Significance~
Hypertension Normotension Diabetic Nondiabetic Smoker Nonsmoker Original stenosis Symptomatic Asymptomatic Gender Male Female
101 53 45 109 53 101
6 5 6 5 5 6
5.9 9.4 13.3 4.5 9.4 5.9
91 66
10 1
11.0 1.5
No component of the surgical treatment of carotid atherosclerosis has escaped controversy. As the number of carotid endarterectomies performed in this country has increased, so has the diversity of opinion regarding all elements of this operation, from minute technical details to fundamental philosophical and ethical issues. The proponents of ca,tid endarterectomy have sought to identify means, of reducing both early and late surgical morbidity to improve surgical outcome. Among the surgical complications targeted by these efforts is the problem of recurrent carotid stenosis. At least from a morphologic Standpoint, recurrent stenosis represents a failure of arterial reconstruction and thus merits investigation into its pathogenesis and clinical behavior. The findings of other investigators in this field have kindled our own interest in recurrent carotid stenosis
From the Departmentof Surgery,The Milton S. HersheyMedical Center,The PennsylvaniaStateUniversityCollegeof Medicine. Presented at the Third Annual Meeting of the Eastern Vascular Society, Bermuda,May 4-7, 1989. Reprint requests: Robert G. Amip, MD, Vascular Surgery Section, The Milton S. Hershey Medical Center, P.O. Box 850, Hershey, PA 17033. 24/6/18089
and stimulated us to study its incidence and significance in our own patient population. MATERIAL AND METHODS Between August 1983 and January 1988, 184 consecutive carotid endarterectomies were performed in 175 patients at The Milton S. Hershey Medical Center of the Pennsylvania State University by or under the direct supervision of fellowshiptrained vascular surgeons. A uniform surgical technique was used as described herein. All patients were followed with duplex scanning of the extracranial vessels every 3 months for the first postoperative year and every 6 months thereafter. Records were reviewed retrospectively to determine the incidence and morbidity of recurrent stenosis of the operated carotid arteries and to identify possible predisposing factors. Statistical significance o f possible risk factors was tested with Fisher's exact test for two degrees of freedom. SURGICAL TECHNIQUE All operations were performed under general anesthesia, with continuous electroencephalographic monitoring by means of a compressed spectral array 511
Journal of VASCULAR SURGERY
512 Atnip et al.
device in most instances. Carotid shunts were used selectively. Carotid bifurcation endarterectomy was carried out with loupe magnification in the plane between the media and adventitia, removing all visible media that were not tightly adherent. The arteriotomy was extended onto the internal carotid artery beyond the distal termination of the plaque as far as necessary to achieve an easily visualized smooth endpoint. Tacking sutures were rarely required. Vein patches were used selectively for arteriotomy closure, usually in patients with carotid vessels judged to be abnormally small or in those requiring intraoperative revision of the endarterectomy for any reason. Vein was variably obtained from either groin or ankle. All closures were done with loupe magnification and 60 and 7-0 polypropylene sutures. Intraoperative assessment of the endarterectomy was performed by use of either contrast arteriography or high-frequency (20 MHz) Doppler velocity spectral analysis. Arteriography was always performed if the velocity spectra were abnormal. All patients were systemically heparinized during operation and given low-molecular weight dextran at 25 ml/hr for 48 hours after operation. RESULTS Among all patients three deaths (1.6%) and two permanent neurologic deficits (1.1%) occurred within the perioperative period. No immediate carotid occlusions and no stroke deaths occurred. All surviving patients were followed at least 1 year after surgery (range 12 to 56 months, mean 35 months). Recurrent stenosis of the operated carotid resulting in at least 50% diameter reduction by duplex criteria occurred in 11 patients (6.0%) at a mean of 10.2 _+ 7.8 months after surgery. Three of the 11 patients had ipsilateral transient ischemic attacks (two amaurosis, one arm/face paresis), yielding a symptomatic recurrent stenosis rate of only 1.6%. No patient with recurrent stenosis suffered a stroke, and the remaining eight have been free of symptoms. Reoperation was performed in the three patients with symptoms and in two other patients with preocelusive lesions. Duplex findings were confirmed angiographically in all of these patients before reoperation. Disease was myointimal hyperplasia in four patients and atheroma in one. In all cases an endarterectomy plane could be developed to excise the bulk of the recurrent lesion. Saphenous vein patches were used for closure in all instances. There were no complications of reoperation, and all five reoperated vessels remain patent from 3 to 23 months of follow-up.
Among the six patients not undergoing reoperation one stenosis has remained stable, and three have regressed at a mean of 34 months of follow-up. One patient progressed to asymptomatic internal carotid occlusion after refusing reoperation for a severe stenosis. A final patient with a preocclusive restenosis has remained free of symptoms but has declined any further intervention. Restenosis was more frequent in men (10 of 128; 7.8%) than in women (1 of 56; 1.8%), but this difference did not achieve statistical significance. Ten of the 11 patients with restenosis had originally undergone operation for carotid territory ischemic symptoms, whereas only one had operation for asymptomatic stenosis. The rate of restenosis was thus significantly higher in patients with symptomatic primary carotid stenosis than in those with asymptomatic primary stenosis (11.0% vs 1.5%; p < 0.02). Vein patch closure had been used in only six,~f the 184 primary procedures, and none of these arteries restenosed. The small number of patched arteries rendered statistical comparison invalid. Intraoperative high-frequency Doppler or arteriography had been performed at the primary operation in 10 of 11 patients with restenosis, but residual stenosis or technical error had not been shown in any. With regard to other possible factors predisposing to recurrent carotid stenosis, comparative data were available in 154 patients (Table I). No statistically significant association could be demonstrated between the risk ofrestenosis and either hypertension or cigarette smoking. Blood lipid values were not available for enough patients to make a comparison. It is surprising to note that there was a strong st• tistical association between diabetes mellitus and carotid restenosis. Six of 11 patients (54%) with restenosis were diabetic, compared to only 27% of patients without restenosis. Expressed differently, the risk of restenosis was 6/45 (13.3%) among diabetic patients, compared to 5/109 (4.5%) among nondiabetic patients (p < 0.05). DISCUSSION
Recurrent carotid stenosis is a proven entity whose existence has been recognized for at least a decade. Although first reported in the carotid artery by Stoney and String in 1976,1 the phenomenon of arterial restenosis had been recognized years earlier as contributing to vascular reconstructive failures, z,3 Unfortunately, surgeons have not always taken full advantage of the methods available for objective
Volume 11 Number 4 April 1990
Recurrent carotid stenosis 513
Table I. Relationship of clinical variables to carotid restenosis Number ofpatients with restenosis
Incidence of restenosis (%)
Factor
Number ofpatients
Significance~
Hypertension Normotension Diabetic Nondiabetic Smoker Nonsmoker Original stenosis Symptomatic Asymptomatic Gender Male Female
101 53 45 109 53 101
6 5 6 5 5 6
5.9 9.4 13.3 4.5 9.4 5.9
91 66
10 1
11.0 1.5
