Summary

Patients with peripheral arterial disease (PAD) are at high risk of major ischaemic events. Long-term data of all major ischaemic events in PAD patients are scarce and outdated, especially for patients with severe PAD requiring bypass surgery. Our objective was to define their long-term prognosis and develop a prediction model which quantifies this risk up to a decade after surgery. We conducted a retrospective cohort study in patients from the Dutch Bypass Oral anticoagulants or Aspirin (BOA) Study; a multicentre randomised trial comparing oral anticoagulants with aspirin after infrainguinal bypass surgery. The primary outcome was the composite event of non-fatal myocardial infarction, non-fatal ischaemic stroke, major amputation, and vascular death. Cumulative risks were assessed by Kaplan-Meier analysis and independent determinants by multivariable Cox regression models. From 1995 until 2009, 482 patients were followed for a median period of 7.8 years. Follow-up was complete in 94%. Overall 60% of patients experienced a primary outcome event, of which the majority was a vascular death (30%), followed by major amputations (12%). The primary cause of vascular death was a cardiovascular event (29%), whereas the minority was due to complications directly related to PAD (6%). Within five years after bypass surgery vascular death occurred in about a quarter of patients and within 10 years in nearly half of patients. This was double the rate as for non-vascular death. The primary outcome event occurred in over a third and over half of patients in 5 and 10 years after bypass surgery, respectively. From four independent determinants for the primary outcome event: age, diabetes, critical limb ischaemia, and prior vascular interventions, we developed a risk chart, which systematically classifies the 10-year risks of the primary outcome event, ranging from 25% to 85%. This study provided a detailed insight in the course of PAD long after peripheral bypass surgery and enables individual risk assessment of major fatal and non-fatal ischaemic events by means of cumulative incidences and a risk chart.