A Ten‐Year Single‐Center Evaluation of In‐Hospital Versus Seven‐ Versus Thirty‐Day Outcomes as Quality Standards After Carotid Endarterectomy

Objective Carotid endarterectomy (CEA) remains a cornerstone for stroke prevention. The 2023 European Society for Vascular Surgery (ESVS) guidelines recommend performing CEA only if the 30‐day risk of stroke or death is < 3% for asymptomatic and < 6% for symptomatic patients. While clinical trials establish efficacy based on 30‐day outcomes, reporting in‐hospital complication rates offers greater feasibility in clinical routine. This study evaluates institutional compliance with ESVS thresholds and compares the reliability of in‐hospital and 7‐day outcomes versus the 30‐day follow‐up interval. Design Retrospective single‐center cohort study. Methods A retrospective analysis was conducted at the Medical University of Graz, encompassing 2496 patients undergoing primary CEA between 2014 and 2023. Data were retrieved from the institutional information system and cross‐referenced with national mortality registries (Statistik Austria) to ensure comprehensive follow‐up. Primary endpoints included stroke and death within 7 and 30 days, stratified by symptomatic status and surgical technique—Eversion Endarterectomy (EEA) versus Conventional Endarterectomy (TEA) with patch angioplasty. Results The final cohort comprised 2496 eligible procedures. Notably, a substantial proportion of events that occurred after hospital discharge, accounting for 17.9% (10/56) of all 30‐day strokes and 60.0% (9/15) of all 30‐day deaths were observed between hospital discharge and postoperative day 30. Overall, the in‐hospital combined stroke and death rate was 1.5% for asymptomatic patients and 3.8% for symptomatic patients. Following discharge, additional events were recorded in 10 asymptomatic and 9 symptomatic patients, resulting in a cumulative 30‐day combined stroke and death rate of 2.0% [95% CI: 1.4%–2.7%] for asymptomatic patients and 5.3% [95% CI: 3.5%–7.2%] for symptomatic patients, thereby meeting ESVS safety standards. While symptomatic status was a significant predictor of complications ( p < 0.001), neither sex nor surgical technique (EEA vs. TEA‐patch) exerted a significant influence on outcomes. Conclusion CEA is performed at ESVS‐recommended safety levels in Graz. Contrasting a clinical tendency to favor more convenient in‐hospital metrics, this study demonstrates that “in‐hospital” data cannot be standardized due to highly variable discharge windows. Furthermore, while fixed 7‐day complication rates provide a standardized timeline, they still significantly underestimate perioperative risks. The substantial rate of post‐discharge perioperative strokes and deaths observed in our cohort underscores the clinical necessity of extending quality control to at least 30 days, which remains the superior metric for clinical audit.