Long-term outcomes of endovascular aneurysm repair according to instructions for use adherence status

Abstract

ObjectiveEndovascular aneurysm repair (EVAR) has become the standard treatment for abdominal aortic aneurysms (AAAs). Endovascular device manufacturers have defined specific anatomic criteria for the aneurysm characteristics that should be observed as instructions for use (IFU) for specific grafts. In clinical practice, the prevalence of performing EVAR outside the IFU has been high. In the present study, we aimed to determine the effects of nonadherence to the IFU on the outcomes. MethodsPatients who had undergone EVAR for an infrarenal AAA between 2005 and 2013 were included. IFU nonadherence was defined as any violation of device-specific IFU criteria and was compared with IFU adherence. The primary outcomes were all-cause mortality, aneurysm-related mortality, AAA rupture, graft-related adverse events (GRAEs), including limb-related adverse events, and type Ia endoleaks. A second aim was to study whether the prevalence of EVAR performed outside the IFU has increased over time. ResultsA total of 258 patients were included, 144 (55.8%) of whom had been treated according to the IFU and 114 (44.2%) outside the IFU. In the IFU nonadherence group, all-cause mortality (hazard ratio [HR], 1.39; 95% confidence interval [CI], 1.02-1.89; P = .037) and aneurysm-related mortality (HR, 5.1; 95% CI, 1.4-18.6; P = .015), and the incidence of AAA rupture (HR, 5.4; 95% CI, 1.1-26.6; P = .036) and GRAEs (HR, 1.7; 95% CI, 1.1-2.8; P = .025). No significant association was found between the incidence of type Ia endoleaks and neck-related IFU or limb-related adverse events and iliac-related IFU. However, neck length was a risk factor for type Ia endoleaks (HR, 18.2, 95% CI, 6.3-52.2; P < .001), aneurysm-related mortality (HR, 8.7; 95% CI, 1.8-41.6; P = .007), AAA rupture (HR, 21.7; 95% CI, 2.8-166; P = .003), and GRAEs (HR, 4.4; 95% CI, 2.0-9.7; P < .001). An IFU violation regarding neck angulation was also a risk factor for all-cause mortality (HR, 2.0; 95% CI, 1.1-3.7; P = .032), aneurysm-related mortality (HR, 7.6; 95% CI, 1.4-42.8; P = .021), AAA rupture (HR, 79.4; 95% CI, 6.3-999; P = .001), and GRAEs (HR, 4.3; 95% CI, 1.9-9.5; P < .001). The prevalence of EVAR performed outside the IFU did not increase over time. ConclusionsPerforming EVAR outside the IFU had a negative effect on outcomes, including all-cause mortality, aneurysm-related mortality, AAA rupture, and GRAEs. Neck angulation and neck length seemed to be the most crucial aneurysm characteristics.