There are different methods of performing carotid endarterectomy (CEA). Current European guidelines recommend eversion CEA or conventional CEA with patch closure, as previous studies have shown advantages and disadvantages with both methods. There are different types of patches available: Vein, Dacron and PTFE has been used for a long time. More recently, Bovine pericardium has been used. There are few studies comparing long-term results between bovine pericardium and other patch types. The aim of this study was to investigate the short- and long-term results after CEA depending on surgical technique and patch material.

Data on all carotid procedures performed in a national vascular registry with high external validity between July 2008 and December 2019, were obtained. Data were cross-linked to the population registry and the national stroke registry which covers approximately 96% of all strokes. Perioperative results were analysed by mixed logistic regression using Generalized Estimating Equation adjusted for demographics and risk factor profile with random operating centre effect. Time to ipsilateral stroke, all stroke and death were analysed by Cox regression adjusted for demographics and risk factor profile. Charts were obtained for all patients with a stroke to determine the brain territory affected and type of stroke. Primary endpoint was any stroke or death within 30 days of surgery. Secondary endpoints were reoperations due to neck haematoma, ipsilateral ischemic stroke>30 days, all stroke>30 days and all-cause mortality>30 days.

In total, 9205 patients were included, all treated for symptomatic carotid stenosis. Mean age was 72.5 (SD 8.2) years, 32.9% were female and 20.3% had diabetes. The majority had TIA or minor stroke as qualifying neurologic event, 19.5% presented with amaurosis fugax. 11.2% of the patients had undergone preoperative thrombolysis. There were 4.3% with contralateral carotid occlusion, 65.8% of the patients had an ipsilateral degree of stenosis of 70-99% (NASCET). Average time from qualifying neurologic event to operation was 8.0 days (IQR 9), and 76.6% of the patients were operated within 14 days from the index neurologic event. The 9205 patients operated with CEA during the study period were divided according to type of endarterectomy: eversion (n=2495), or conventional (n=6710), and the latter group were subdivided depending on type of patch or primary suture. Dacron was most frequently used as patch material (n=3921), followed by PTFE (n=588). Bovine pericardium was introduced in year 2013 and has become more frequently used over time. Currently, 24.0% of the patches used are from bovine pericardium. Primary closure was used in 19.4% of the patients (n=1788), equivalent to 26.6% of the conventional CEAs. A larger proportion of men were treated with primary closure, 74.6% compared to 65.7% in the eversion group, 65.5% in the Dacron group, 61.9% in the PTFE group and 64.9% in the Bovine pericardium group, indicating primary closure selectively used in larger carotids. 30-day outcome Major complications; stroke <30 days, and, stroke or death <30 days for the whole cohort was 3.0% (95%CI 2.6-3.3), and 3.4% (95%CI 3.0-3.8) respectively. In a risk-factor adjusted analysis, primary closure was associated with 50% increased risk of stroke or death <30 days, 1.5 (95%CI 1.2-2.0), as compared to Dacron patch closure, see table 1 for details. Long-term outcome Patients operated between June 1st 2008 and September 30th 2017 (n=7694) and surviving 30 days, were followed until censoring for end-point or until study-end. Median follow-up time was 4.2 years, equivalent to a total follow-up of 32,084 person years. During follow up, 592 patients (7.7%) had a stroke, and 1492 died (19.4%). In the long-term follow up, there was no statistically significant difference with respect to ipsilateral stroke after 30 days in crude or adjusted analysis for; all stroke or all-cause mortality, between patients operated with eversion, primary closure or patch closure after multivariate adjustment. Additionally, there was no difference in various patch types with respect to ipsilateral stroke, all stroke or all-cause mortality after the perioperative period. see table 2 and figure 2

In this nationwide cohort study, including all CEA performed for symptomatic carotid stenosis, patch closure is less likely than primary closure to cause perioperative stroke. Patching with bovine pericardium is durable and, in this series, there was no difference in outcome between different patch types in long-term follow-up.