Objective: This analysis of the outcome of carotid endarterectomy (CEA) was performed during a period of transition from the frequent: use of autologous greater saphenous vein grafting to the frequent use of Dacron graft patch reconstruction and from the infrequent use to the moderate use of eversion plication shortening of the endarterectomized internal carotid artery segment. Methods: From 1990 to 1997, 697 consecutive primary CEAs were performed on 326 men (61 bilateral CEAs) and 272 women (38 bilateral CEAs) with a mean age (+/- SD) of 68 +/- 9 years. The indications were transient ischemic attack in 31% of the patients, stroke or reversible ischemic neurologic deficit in 18%, global ischemia in 12%, and asymptomatic stenosis greater than or equal to 70% in 39%. Patch reconstruction was performed in the 678 CEAs in which the arteriotomy extended distal to the internal carotid artery bulb (97%; 370 saphenous vein grafts, 308 Dacron grafts). Primary closure was used in the other 19 CEAs. Early in this series, saphenous vein patching frequency was performed, with. a gradual transition to the frequent use of knitted Dacron grafts. Concurrent with this, the frequency of the shortening of the internal carotid artery increased from 7% to 40%. Postoperative duplex scans were obtained on 619 CEAs (91%). Results: There were four deaths (0.6%) in 30 days-three from myocardial infarction and one from hyperperfusion stroke. Thirteen strokes (1.9%), nine ipsilateral and four contralateral, occurred in 30 days. Four nonfatal strokes and one death occurred in the saphenous vein group (3.2%), and eight strokes and two deaths occurred in the Dacron graft group (1.4%; P = .16). The combined 30-day stroke or death rate was 2.3% (16/697), and the hospital rate was 1.7% (12/697). Of the three internal carotid artery occlusions, two were identified at 2 months tone Dacron graft, one saphenous vein) and one was identified at 1 year (Dacron graft). Nonocclusive (greater than or equal to 50%) restenosis was identified in 16 CEAs. Fifteen of these were in the internal carotid artery The cumulative Kaplan-Meier method of life-table analysis for the greater than or equal to 50% restenosis rate at 2 months, 6 months, 1 gear, and 3 years for Dacron graft patched CEA was 1.7%, 2.3%, 8.8%, and 12.3% and for saphenous vein patched CEA was 0.3%, 0.3%, 0.3%, and 1.1% (P < .0001). At the same time intervals, the greater than or equal to 50% restenosis rate-for internal carotid artery shortening was 1.0%, 2.5%, 13.7%, and 19.5%, and, when shortening was not done, the rate was 0.8%, 0.8%, 1.1%, and 3.1% (P < .0001). The greater than or equal to 50% restenosis rate at the same intervals for women was 0.8%, 1.3%;,, 5.2%, and 8.9%, and, for men, the rate was 0.9%, 0.9%, 1.8%, and 2.5% (P = .11). Univariate analysis of the rate of greater than or equal to 50% restenosis in 3 years for the 346 vein patched (2; 0.6%) and 186 Dacron graft patched (7; 3.8%) CEAs that did not have internal carotid artery shortening gave a P value of .015. Similarly, Kaplan-Meier method analysis of this subset of nonshortened CEAs gave a higher restenosis rate with Dacron graft patching (P = .012). With multiple logistic regression, the greater than or equal to 50%;, restenosis rate was significantly associated with Dacron graft patching (P = .023; odds ratio, 4.5) and internal carotid artery shortening (P = .025; odds ratio, 3.1) and was weakly associated with female gender (P = .15; odds ratio, 2.0). Cox proportional hazards model analyis for greater than or equal to 50% restenosis gave relative risk ratios of 3.0 (1.6 to 6.8; 95% cofidence interval [CI]) for Dacron graft versus vein patching, 2.0 (1.2 to 3.3; 95% CI) for shortening versus not shortening, and 1.5 (0.9 to 2.4; 95% CI) for female versus male gender. Conclusion: CEA patching with Dacron grafts and saphenous vein grafts have similar low perioperative thrombosis, stroke, and death rates, although the stroke and death rates were slightly higher but not statistically different when Dacron grafts were used. Dacron graft patched CEAs are more likely to develop greater than or equal to 50% restenosis than are those that are patched with saphenous vein grafts. Eversion plication shortening of the internal carotid artery predisposes the graft to restenosis.
