Plasma Homocysteine and Its Determinants in Diabetic Retinopathy

Plasma Homocysteine and Its Determinants in Diabetic Retinopathy

Olga Vaccaro

Whether hyperhomocysteinemia contributes to the development of diabetic microangiopathy is still debated. Most of the older studies are difficult to interpret because of the insufficient characterization of the type of diabetes and status of complications. More recent works conducted on better-characterized patients are fairly consistent in showing a relation between hyperhomocysteinemia and diabetic nephropathy; however, it still remains unclear whether this association is causal [1-3]. Furthermore, the relationship of plasma homocysteine with retinopathy is little explored to date. We contribute data on the relation of retinopathy with fasting plasma homocysteine and also explore some of the potential mechanisms of this association.

Sixty-nine patients with type 1 diabetes of [geq] years’ duration, consecutively seen on an outpatient basis, participated in the study All of the patients were normotensive (blood pressure [less than]140/90) and free from cardiovascular diseases as evaluated by the World Health Organization questionnaire, an electrocardiogram, and ankle/brachial pressure. To reduce to the minimum the confounding effect of renal injury patients with macroalbuminuria and/or serum creatinine [greater than]1.3 mg/dl were excluded from the study According to 45[degrees] fundus photography performed and evaluated following a standard protocol, participants were assigned to 1 of 3 groups: no retinopathy (n = 34), nonproliferative diabetic retinopathy (NPDR) (n = 20), or proliferative diabetic retinopathy (PDR) (n = 10). Plasma homocysteine was measured together with vitamin [B.sub.12] and folate, the major environmentally determined factors influencing homocysteine metabolism. Furthermore, the C677T mutation in the methylenetetrah ydrofolate reductase (MTHFR) gene, the most common genetic determinant of moderate hyperhomocysteinemia in the general population, was also studied (4).

Plasma homocysteine progressively increased with a significant linear trend (P [less than] 0.03) from the stage of no retinopathy to the stage of PDR (7.3 [pm] 3.0 vs. 8.2 [pm] 2.6 vs. 9.5 [pm] 2.6 [mu]ol/1, respectively). Post hoc Duncan’s test indicated significantly (P [less than] 0.05) higher levels of fasting plasma homocysteine in patients with PDR as compared with those without any sign of retinopathy The 3 groups were not significantly different in terms of sex distribution, age, and smoking habits; blood pressure was also comparable (118 [pm] 14 vs. 118 [pm] 16 vs. 122 [pm] 11 mmHg, respectively) as were serum creatinine (0.8 [pm] 0.1 vs. 0.9 [pm] 0.1 vs. 0.9 [pm] 0.2 mg/dl and creatinine clearance, evaluated by the Cockroft formula (98 [pm] 16 vs. 105 [pm] 21 vs. 91 [pm] 12). Furthermore, plasma concentrations of vitamin [B.sub.12] and folate were also comparable in the 3 groups (405 [pm] 102 vs. 449 [pm] 141 vs. 430 [pm] 105 pmol/l and 14.5 [pm] 4.5 vs. 16.8 [pm] 5.6 vs. 13.8 [pm] 3.6 nmol/l, resp ectively).

The allelic frequency of the C667T mutation in MTHFR gene was similar in the group of patients with no retinopathy and NPDR (38 vs. 33%) but was significantly higher in the patients with PDR as compared with those with no retinopathy (75 vs. 38%, P [less than] 0.01). Accordingly the genotype distribution of the mutated gene was significantly different in the 2 groups with PDR or no retinopathy with a significantly higher frequency of homozygosity in patients with PDR (70 vs. 18%, odds ratio 10.3, 95% CI 1.7-69.8)

The data reported indicate a relationship between PDR and plasma homocysteine levels independent of some obvious confounders and coexisting conditions associated with the elevation of plasma homocysteine or retinopathy (i.e., cardiovascular disease, impaired renal function, vitamin status, and blood pressure). An association of diabetic retinopathy with the C677T mutation in the gene coding for the MTHFR, a key enzyme in the homocysteine catabolism, has been reported previously by Neugebauer et al. [5] in type 2 diabetic patients; in this study however, plasma homocysteine was not measured. We subsequently reported moderate hyper-homocysteinemia in a small group of type 1 diabetic patients with retinopathy and normal serum creatinine, but the major determinants of homocysteine metabolism were not measured [6]. To our knowledge, this is the first report that explores the relationship of plasma homocysteine and some of its major determinants with retinopathy.

Based on these data, a role for homocysteine in the development of PDR can be hypothesized, at least in selected groups of patients. This hypothesis is also supported by the results of in vitro experiments showing a synergistic effect of plasma homocysteine and hyperglycemia in inducing cell damage in the vascular endothelium [2]. This study is also relevant inasmuch as it singles out a possible genetic marker for PDR. A genetic predisposition to retinopathy is not well documented; however, it is known that although almost the all of type 1 diabetic patients with longstanding diabetes develop some degree of retinopathy, relatively few progress toward PDR [7]. To explain this finding, among others, genetic differences in response to hyperglycemia have been hypothesized, and according to emerging knowledge, homocysteine may also play a role. Of course, the cross-sectional nature of the study makes results compatible with the alternative hypothesis that higher plasma homocysteine levels may be a marker rather t han a determinant of tissue damage in diabetic retinopathy, similar to what has been hypothesized in ischemia because of large vessels occlusions [8]. However, it would be more difficult to explain on this basis the association of PDR with the C677T mutation in the MTHFR gene.

From the Departments of Clinical and Experimental Medicine (O.V., V.C., M.S., A.A.R., G.R.) and Biochemistry and Medical Biotechnologies (F.P.M., A.T.), School of Medicine, Federico II University; and the Institute of Biochemistry of Macromolecules (A.F.P., D.I.), School of Medicine, Second University of Naples, Naples, Italy.

Address correspondence to Olga Vaccaro, Department of Clinical and Experimental Medicine, Policlinico dell’Universit[grave{a}] degli Studi “Federico II,” via S. Pansini 5, 80131 Napoli, Italy. E-mail:

Acknowledgments — This work was supported by a grant from the Italian Association for the Study of Diabetes (SID).


(1.) Chico A, Perez A, Cordoba A, Arcelus R, Carreras G, de Leiva A, Gonzalez-Sastre F, Blanco-Vaca F: Plasma homocysteine is related to albumin excretion rate in patients with diabetes mellitus: a new link between diabetic nephropathy and cardiovascular disease? Diabetologia 41:684-693, 1998

(2.) Hofmann MA, Kohl B, Zubach MS, Borcea V, Bierhaus A, Henkels M, Amiral J, Fiehn W, Ziegler R, Wahl P, Nawroth PP: Hyperhomocysteinemia and endothelial dysfunction in IDDM. Diabetes Care 21:841-848, 1998

(3.) Cronin CC, McPartlin JM, Barry DG, Ferriss JB, Scott JM, Weir DG: Plasma homocysteine concentrations in patients with type 1 diabetes. Diabetes Care 21:1843-1847, 1998

(4.) Lentz SR: Mechanism of thrombosis in hyperhomocysteinemia. Curr Opin Hematol 5:343-349, 1998

(5.) Neugebauer S, Tsuneharu B, Kurokawa K, Watanabe T: Defective homocysteine metabolism as a risk factor for diabetic retinopathy. Lancet 349:473-474, 1997

(6.) Vaccaro O, Ingrosso D, Rivellese A, Greco G, Riccardi G: Moderate hyperhomocysteinemia and retinopathy in insulin-dependent diabetes. Lancet 349:1102-1103, 1997

(7.) Klein R: Hyperglycemia and microvascular and macrovascular disease in diabetes. Diabetes Care 18:258-268, 1995

(8.) Dudman NPB: An alternative view of homocysteine. Lancet 354:2072-2074, 1999

COPYRIGHT 2000 American Diabetes Association

COPYRIGHT 2000 Gale Group