Limb Related Outcomes of Endovascular vs. Open Surgical Revascularisation in Patients with Peripheral Arterial Occlusive Disease: A Report from the Prospective GermanVasc Cohort Study

Abstract

The goal of this study was to compare clinical outcomes in patients with peripheral arterial occlusive disease undergoing revascularisation by peripheral endovascular intervention (EVI), bypass surgery, endarterectomy (EA), and hybrid surgery in an unselected real world setting. This was a German, prospective, multicentre, comparative cohort study, enrolling patients at hospital admission for revascularisation at 35 vascular centres with 12 months of follow up. Primary composite endpoints were major amputation or death, major adverse limb events, and any minor or major amputation. Twelve month incidences and hazard ratios (HRs) for the four subgroups and 95% confidence intervals (CIs) were estimated using Kaplan-Meier functions and Cox proportional hazard models. Sociodemographic and clinical characteristics, pharmacological treatment, and comorbidities were used to adjust for patient differentials (unique identifier ClinicalTrials.gov: NCT03098290. In total, 4 475 patients were analysed (mean age 69 years, 69.1% males, and 31.5% suffering from chronic limb threatening ischaemia). After 12 months of follow up, 5.3% (95% CI 3.6 - 6.9%) of the patients experienced either death or major amputation, 7.2% (95% CI 4.8 - 9.6%) major adverse limb event, and 6.6% (95% CI 5.0 - 8.2%) any minor or major amputation. Compared with EVI, bypass surgery was associated with an increased risk of amputation or death (HR 2.59, 95% CI 1.75 - 3.85), major adverse limb event (HR 1.93, 95% CI 1.11 - 3.36), and any minor or major amputation (HR 2.12, 95% CI 1.42 - 3.16), and hybrid surgery with an increased risk of amputation or death (HR 2.29, 95% CI 1.27 - 4.13) and major adverse limb event (HR 1.62, 95% CI 1.03 - 2.54). After adjusting for patient differentials, no significant differences among study groups remained. More favourable outcomes after EVI were completely attributed to differentials in patient characteristics and not procedure type. The current study emphasised that all competing approaches performed similarly in a real world setting.