Hyperhomocysteinemia, hyperfibrinogenemia, and lipoprotein (a) excess in maintenance dialysis patients: a matched case-control study.

Maintenance dialysis patients experience an exceedingly high incidence of arteriosclerotic cardiovascular disease (CVD) events that are poorly predicted by traditional CVD risk factor indices. We evaluated the prevalence of three non-traditional CVD risk factors, i.e. hyperhomocysteinemia, hyperfibrinogenemia, and lipoprotein (a) Lp(a)) excess, and combined hyperhomocysteinemia, hyperfibrinogenemia, and Lp(a) excess, in maintenance dialysis patients. Fasting total plasma homocysteine (Hcy), fibrinogen, Lp(a), glucose, and total and HDL cholesterol levels, and traditional CVD risk factor (i.e. glucose tolerance, smoking, hypertension, dyslipidemia) prevalences were assessed in 71 dialysis patients and 71 age, sex, and race matched Framingham Study controls free of clinical renal disease, with normal serum creatinine (< or = 1.5 mg/dl). Mean plasma Hcy 23.7 vs. 9.9 microM, P = 0.0001), fibrinogen (457 vs. 309 mg/dl, P = 0.0001), and Lp(a) (30 vs. 17 mg/dl, P = 0.0070) levels were substantially increased in the dialysis patients. Matched odds ratios (with 95% confidence intervals), dialysis patients/controls, for hyperhomocysteinemia, hyperfibrinogenemia, and Lp(a) excess, alone or combined, were markedly greater in the dialysis patients, with no evidence of confounding by the traditional CVD risk factors: hyperhomocysteinemia, 105.0 (29.9-368.9); hyperfibrinogenemia, 16.6 (6.6-42.0); Lp(a) excess, 3.5 (1.5-8.4); all three combined 35.0 (5.7-199.8). Given in vitro evidence that Hcy, Lp(a), and fibrinogen interact to promote atherothrombosis, combined hyperhomocysteinemia, hyperfibrinogenemia, and Lp(a) excess may contribute to the high incidence of vascular disease sequelae experienced by dialysis patients, which is inadequately explained by traditional CVD risk factors. Controlled, prospective studies of well-characterized maintenance dialysis cohorts are urgently required to substantiate this hypothesis.