Hyperhomocysteinemia, anticardiolipin antibody status, and risk for vascular access thrombosis in hemodialysis patients

Background. Vascular access failure is an important cause of morbidity in e nd-stage renal failure patients on hemodialysis. Currently, little is known about risk factors that predispose certain hemodialysis patients to recurr ent access thrombosis. Hyperhomocysteinemia (common in patients with renal failure) predisposes people with normal renal function to recurrent and ear ly-onset venous thrombosis, although the effect on vascular access thrombos is is currently unknown. Previous studies have suggested that high titers o f IgG anticardiolipin antibody (IgG-ACA) predispose hemodialysis patients t o access thrombosis. This cross sectional study was designed to assess for an association between two predictive variables, hyperhomocysteinemia and e levated titers of IgG-ACA, and vascular access thrombosis in patients under going chronic hemodialysis. Methods. Risk factors for vascular access thrombosis were documented, and t he number of episodes of access thrombosis was recorded for the previous th ree years in patients undergoing hemodialysis. Midweek predialysis total ho mocysteine and IgG-ACA levels were measured in all subjects. Results. Of the 118 patients who were enrolled, 75.4% had a native arteriov enous fistula. Episodes of vascular access thrombosis were recorded for the previous three years; 34 (28.8%, 95% CI 20.9 to 37.9%) patients had 72 epi sodes of access thrombosis over the period of risk. Mean homocysteine level s were not significantly different between these 34 patients (28.6 mu mol/l iter, 95% CI 24.5 to 32.7) and the patients who had no episodes of graft th rombosis (29.8 mu mol/liter, 95% CI 26.7 to 32.9). Sixty-seven unselected p atients had IgG-ACA levels drawn for analysis, and all assays were negative . The only variable that was associated with a higher risk for graft thromb osis was the type of vascular access placed (odds ratio 4.0, 95% CI 1.6 to 9.6 for patients with a synthetic graft compared with those with an arterio venous fistula). Conclusions. No association was found between homocysteine levels or antica rdiolipin antibody and vascular access thrombosis in our patient population .