- Email: [email protected]
Total homocysteine levels relation with chronic complications of diabetes, body composition, and other cardiovascular risk factors in a population of patients with diabetes mellitus type 2
Journal of Diabetes and Its Complications 19 (2005) 42 – 46
Total homocysteine levels relation with chronic complications of diabetes, body composition, and other cardiovascular risk factors in a population of patients with diabetes mellitus type 2 D.A. de Luis*, N. Fernandez, M.L. Arranz, R. Aller, O. Izaola, E. Romero Institute of Endocrinology and Nutrition, Medicine School and Hospital Rio Hortega, University of Valladolid, C/Los perales 16 (URB Las Acen˜as), Simancas 47130, Valladolid, Spain Received 5 June 2003; received in revised form 23 September 2003; accepted 15 December 2003
Abstract Objective: The significance of hyperhomocysteinemia in type 2 diabetes is further complicated by the multiple ways of considering impaired renal function and vitamin status. The aim of our study was to analyze the relationship between total homocysteine (tHcy) in a population of type 2 diabetic patients and chronic complications. We also analyzed the relationship between tHcy and the body composition of these patients and other cardiovascular risk factors. Design: In a cross-sectional study, a total of 155 patients with diabetes mellitus attending in our diabetes service (90 females/65 males) were enrolled in a consecutive way. Material and methods: All enrolled patients underwent the following examinations; (i) biochemical cardiovascular risk factors including total cholesterol, triglyceride, lipoprotein (a), low-density lipoprotein (LDL-cholesterol), high-density lipoprotein (HDL-cholesterol), glucose, HbA1c, fibrinogen, homocysteine, vitamin B12, folate, and microalbuminuria; and (ii) fat mass assessed by body mass index, weight, percentage of fat mass, and tricipital skinfold. Results: Patients were divided in two groups (Group I: tHcyz15 Amol/l; Group II: tHcy<15 Amol/l). Smoking habit was similar in both groups. A prevalence of cerebrovascular accident was present in 3.3% in the total group. This prevalence was not different in both groups (7.4% vs. 2.3%; ns) (OR 3.3; 95% CI 0.49 – 19.68). The prevalence of coronary heart disease in the total group was 5.8% without statistical differences between groups (3.5% vs. 6.3%; ns) (OR 0.57; 95% CI 0.065 – 4.53). Concerning macrovascular complications, only peripheral vascular disease prevalence was higher in Group I (16% vs. 3.1%; P < 0.05; OR 5.33; 95% CI 1.18 – 21.5). A prevalence of nephropathy was higher in Group I (93.3% vs. 12.8%; P < 0.05; OR 7.15; 95% CI 2.9 – 17.9). No statistical differences were detected in prevalence of retinopathy (global group 41.9%) (42.5% vs. 40.9%; ns) (OR 1.75; 95% CI 0.78 – 3.9). Also, peripheral neuropathy was similar in both groups (7.1% vs. 6.5%; ns) (OR 1.1; 95% CI 0.15 – 8.2). No correlation was detected among homocysteine and anthropometric parameters (body mass index, weight, percentage of fat mass, fat mass, and tricipital skinfold). Elevated levels of fibrinogen, lipoprotein (a), microalbuminuria, and blood pressure were detected in Group I. Conclusion: The present study shows that elevation of plasma tHcy levels in type 2 diabetic patients is associated with a higher prevalence of peripheral arteriopathy and nephropathy. Our data suggest that hyperhomocysteinemia is not associated with fat mass but it is associated with high levels of fibrinogen, lipoprotein (a), microalbuminuria, and blood pressure levels. D 2005 Elsevier Inc. All rights reserved. Keywords: Diabetes mellitus; Homocysteine; Macroangiopathy; Microangiopathy
1. Introduction It is important to search for modificable risk factors in patients with diabetes mellitus type 2. The aminoacid homocysteine may be such a risk factor (Rosenberg & Miller, 1992). A lot of longitudinal and cross-sectional
* Corresponding author. Fax: +34-98-333-1566. E-mail address: [email protected] (D.A. de Luis). 1056-8727/05/$ – see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jdiacomp.2003.12.003
studies have shown that homocysteine is associated with an increased risk of stroke (Coull, Malinow, & Beamer, 1990; Perry, Refsum, & Morris, 1995; Verhoef, Hennekens, & Malinow, 1994), carotid artery atherosclerosis (Malinow, Nieto, & Szklo, 1993; Selhub, Jacques, & Bostom, 1995), and coronary heart disease (Arnesen, Refsum, & Bonaa, 1995; Stampfer, Malinow, & Willett, 1992). Although type 2 diabetes is definitely associated with premature atherosclerosis and microvascular complications, only a handful of studies have dealt with the between hyperhomocysteinemia and macro- or microangiopathic complications. The
D.A. de Luis et al. / Journal of Diabetes and Its Complications 19 (2005) 42 – 46
significance of hyperhomocysteinemia in type 2 diabetes is further complicated by the multiple ways of considering impaired renal function and vitamin status. Therefore, we measured total homocysteine (tHcy) in a population of type 2 diabetic patients and assessed whether high levels were related with chronic complications. We also analyzed the relationship between tHcy, body composition, and other cardiovascular risk factors.
2. Material and methods 2.1. Subjects In a cross-sectional study, a total of 155 patients attending in our diabetes service (90 females/65 males) with diabetes mellitus were enrolled in a consecutive way. Exclusion criteria were: treatments such as medical conditions known to increase tHcy (e.g., dysthyroidism, and drugs with interaction with folate and vitamin B12) and diabetes type 1. The group was divided in two groups according to their fasting plasma tHcy: Group I had increased tHcy (z15 Amol/l, n = 22), whereas Group II had normal laboratory values (<15 Amol/l, n =133). This value was selected from published data that defined high and normal tHcy levels (Welch & Loscalzo, 1998). The clinical characteristics of these patients included age of 64.6F8.6 years (range: 49 – 78), body mass index (BMI) of 29.6F6.3 kg/m2, and duration of diabetes 8.9F6.7 years (Table 1). Drugs in both groups were similar (antihypertensive drugs, hypoglycemic agents, fibrates, and statins). Smoking habit was similar in both groups (Group I 12.5% vs. Group II 16.4%; ns). The study was approved by the local ethical committee and each patient gave informed consent to participate in the study.
3. Design All enrolled patients underwent the following examinations: (i) biochemical cardiovascular risk factors including total cholesterol, triglyceride, lipoprotein (a), low density lipoprotein (LDL-cholesterol), high density lipoprotein (HDL-cholesterol), glucose, HbA1c fibrinogen, homocysteine, vitamin B12, folate, and microalbuminuria; and (ii) fat mass assessed by body mass index, weight, percentage of fat mass, and tricipital skinfold. 3.1. Chronic diabetic complications assessment All diabetic patients were checked in the clinic for chronic complications. Ischemic heart disease was clinically assessed (history of a cardiovascular event and/or presence of angina) and in addition, a 12-lead resting electrocardiogram was recorded supine (Mac PC Electrocardiograph, Marquette Electronics, Los Angeles, CA) and evaluated by a cardiologist. The presence of any of the following findings was
43
considered suggestive of coronary heart disease: T-wave inversion, ST-segment depression, and Q waves. Patients with indicative signs or symptoms of cerebrovascular disease were evaluated with a CNS computerized tomography. The final diagnosis was reviewed by a neurologist. Peripheral vascular disease was clinically defined by the presence of intermittent claudication, absent or weakened peripheral pulses, or both (no amputations were detected in our population). Retinopathy was documented by standard fundus eye examination and diagnosed on the presence of microaneurysms, venous dilatation, cotton-wool spots, neovascularization, or hemorrhages. Clinical neuropathy was defined by an abnormal neurologic examination, consistent with the presence of peripheral sensorimotor neuropathy. Nephropathy was defined by the presence of urinary albumin excretion of > 30 mg 24-h and/or high plasma creatinine levels. 3.2. Laboratory determinations Serum total cholesterol, HDL-cholesterol, and triglyceride concentrations were determined by enzymatic colorimetric assay (Hitachi 917, Roche Diagnostics, Mannheim, Germany), while HDL-cholesterol was determined enzymatically in the supernatant after precipitation of other lipoproteins with dextran sulfate-magnesium. LDL cholesterol was calculated using Friedewald formula. Lipoprotein (a) was determined by nephelometry (Beckman Coulter, Fullerton, CA, USA). Glycated haemoglobin was measured as HbA1c by HPLC (Menarini, Florence, Italy). Plasma glucose levels were determined by using an automated glucose oxidase method (Hitachi 917, Roche Diagnostics, Mannheim, Germany). Homocysteine levels were determined by immunoassay (Abbott, Wiesbaden, Germany) (normal value: 5 – 13 Amol/l). Fasting serum samples were obtained at baseline, put on ice immediately, and were processed within 60 min, which has been shown to be sufficient to prevent increases in tHcy concentration due to ex vivo generation. Serum was kept frozen at 20jC until determination of tHcy. The
Table 1 Clinical characteristics of diabetic patients studied (meanFS.D.) n Age (years) Sex (male/female) BMI (kg/m2) Diabetes course (years) Glucose (mmol/l) Retinopathy (%) Nephropathy (%) Neuropathy (%) Stroke (%) Peripheral vascular disease (%) Coronary heart disease (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
155 64.6F8.6 65/90 29.6F6.3 8.4F6.7 9.72F3.1 41.9 18.7 6.8 3.3 5.2 5.8 142.9F18.9 81.7F9
44
D.A. de Luis et al. / Journal of Diabetes and Its Complications 19 (2005) 42 – 46
coefficients of variation (inter-assay and intra-assay) were 5% and 6%, respectively. Fibrinogen levels were determined by a quantitative assay (Boehringer-Mannheim, Mannheim, Germany) (normal value: < 300 mg/dl). Vitamin B12 (normal value: 200– 970 pg/ml) and folate (normal value: 2.2– 17.5 ng/ml) were measured using an immunoassay (Abbott). 3.3. Body composition Body weight was measured to an accuracy of 0.5 kg and body mass index was computed as body weight/(height2). Bipolar body electrical bioimpedance was used to determine body composition (Pichard, Slosman, Hirschel, & Kyle, 1997). An electric current of 0.8 mA and 50 kHz was produced by a calibrated signal generator (Biodynamics Model 310e, Seattle, WA, USA) and applied to the skin using adhesive electrodes placed on right-side limbs. Resistance and reactance were used to calculate total body water, fat, and fat-free mass. Regional changes in body mass were estimated by measuring tricep skinfold of the midarm.
4. Statistical analysis The results were expressed as meanFstandard deviation. The distribution of variables was analyzed with the Kolmogorov – Smirnov test. Quantitative variables with normal distribution were analyzed with a two-tailed, paired Student’s t test. Nonparametric variables were analyzed with the Mann – Whitney U test. Qualitative variables were analyzed with the chi-square test, with Yates correction as necessary, and Fisher’s test. Pearson and Spearman tests were used in correlational analysis. Odds ratio and 95% CI were used to studied association among complications of diabetes mellitus and high levels of homocysteine, adjusted by age, sex, and smoking habit. A P value under .05 was considered statistically significant.
5. Results Clinical characteristics of the patients studied are shown in Table 1. Mean homocysteine levels in patients was Table 2 Anthropometric parameters (t test) r ( P) .07 .12 .04 .02 .05
Parameter (.9) (.7) (.8) (.6) (.9)
BMI Weight (kg) Fat mass (kg) % fat mass Tricipital skinfold (mm)
Group I (tHcy >15)
Group II (tHcy <1)
30.5F7.4 75.9F14.2 22.8F9.4 29.3F9.6 25.1F7
29.1F5.6 73.5F13.2 20.1F9.5 26.8F10.1 25.2F6.1
Table 3 Cardiovascular risk factors Parameter Glucose (mmol/l) HbA1c (%) Total cholesterol (mmol/l) LDL cholesterol (mmol/l) HDL cholesterol (mmol/l) Lipoprotein (a) (mg/dl) Fibrinogen (mg/dl) Triglycerides (g/l) Microalbuminuria (mg/day) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
Group I (tHcy z15) 8.4F3.5 8.4F3.5 5.9F1.3
Group II (tHcy <15) 10.2F2.9 (.1) 9.2F2.9 (.8) 5.94F1.13 (.9)
3.33F0.28 (.9)
3.34F1
1.67F0.66 (.4)
1.67F0.4
62F89 (<.05) 408.3F106 (<.05) 1.33F0.73 (.3) 563.16F674.7 152.7F20.1 (<.05) 85F9 (<.05)
31.7F36.7 358F69 1.23F0.59 67.8F166.7 (<.05) 140.6 +19 81.4F11
P values are in parentheses.
10.5F4.3 Amol/l. No statistical differences were detected between homocysteine levels in patients without and with macroangiopathy (cardiovascular disease and/or peripheral vascular disease and/or cerebrovascular accident) (10.13F 3.67 Amol/l vs. 11.68F4.1 Amol/l). Patients without and with microangiopathy (retinopathy and/or nephropathy and/or neuropathy) showed statistical differences in homocysteine levels (9.7F4.2 Amol/l vs. 11.9F4.5 Amol/l). Prevalence of cerebrovascular accident, as shown by clinical examination, imaging techniques, or both, was present in 3.3% in the total group. This prevalence was not different in both groups (7.4% vs. 2.3%; ns) (OR 3.3; 95% CI 0.49 –19.68). The prevalence of coronary heart disease in the total group was 5.8% without statistical differences between groups (3.3% vs. 6.3%; ns) (OR 0.57; 95% CI 0.065 – 4.53). Only peripheral vascular disease prevalence was higher in Group I (16% vs. 3.1%; P <.05), (OR 5.33; 95% CI 1.18 – 21.5). Prevalence of nephropathy was higher in Group I (93.3% vs. 12.8%; P < .05; OR 7.15; 95% CI 2.9– 17.9). No statistical differences were detected in prevalence of retinopathy (global group 41.9%) (42.5% vs. 40.9%; ns) (OR 1.75; 95% CI 0.78 –3.9). Also, peripheral neuropathy was similar in both groups (7.1% vs. 6.5%; ns) (OR 1.1; 95% CI 0.15 –8.2). No correlations were detected among homocysteine and anthropometric parameters (body mass index, weight, percentage of fat mass, fat mass, and tricipital skinfold). Table 2 shows the mean values of these parameters in Groups I and II, without statistical differences. Concerning biochemical results, only microalbuminuria (r =.47; P < .001) and fibrinogen (r =.3; P < .05) showed a direct correlation with homocysteine. No correlation was
D.A. de Luis et al. / Journal of Diabetes and Its Complications 19 (2005) 42 – 46
detected among homocysteine and other cardiovascular risk factors, lipid parameters, and glucose control. Table 3 shows the statistical differences in levels of fibrinogen, lipoprotein (a), and microalbuminuria in Groups I and II. Other cardiovascular parameters were similar in both groups. High levels of blood pressure were detected in Group I. Levels of vitamin B12 and folate were higher in Group II (Hcy <15 Amol/l) (529.8F368 pg/ml vs. 609.9F539.9 pg/ml; P < .05) and (9.8F7.8 ng/ml vs. 10.8F5.2 ng/ml; P <.05), respectively.
6. Conclusions A newly recognised cardiovascular risk factor is the sulphur amino acid homocysteine. High homocysteine levels (tHcy) have been associated with an increased risk of stroke and other cardiovascular events (Boushey, Beresford, & Omenn, 1995; Smulders et al., 1999). Also high homocysteine levels have been associated with microangiopathy (Stabler et al., 1999). This association between homocysteine and chronic complications of diabetes mellitus could be explained by different mechanisms; direct toxic effect on vascular endothelial and indirect effect on the normal methylation in endothelial cells (Weir & Molloy, 2000). Direct toxic effect of homocysteine could be mediated by damage to vascular endothelial cells, resulting in vascular events, such as microvascular disease and cerebrovascular accidents. Our data showed high prevalence of peripheral arteriopathy in patients with homocysteine z15 Amol/l. No differences were detected in stroke and coronary heart diseases in our population. Stabler et al. (1999) showed that tHcy was not increased in subjects with cardiovascular disease. However, in other study, a correlation between homocysteine levels with extension of coronary atherosclerosis was detected (Tsai et al., 2000). Almost a twofold increased likelihood of myocardial infarction among persons with a tHcy concentration z15 Amol/l was noted in the third National Health and Nutrition Examination Survey (Giles, Croft, Greenlund, Ford, & Kittner, 2000). These different results could be due to different characteristics of studied population such as ethnicity (blacks and Hispanics) (Giles et al., 2000) or other associated cardiovascular risk factors in these populations. For example, in the study of Fiorina et al. (1998), Caucasian patients with elevated tHcy levels had significantly higher diastolic pressure and mean arterial pressure. These results are similar to our data. Other populations (Indians) have shown correlation between homocysteine concentrations and body weight (Das et al., 1999). Our results did not show correlations among tHcy and different anthropometric parameters (body mass index, weight, percentage of fat mass, and tricipital skinfold). Other possible effect is the association of different cardiovascular risk factors with elevated tHcy such as lipoprotein (a). In our
45
study, this parameter was elevated in patients of Group I (Foody et al., 2000). However, the relationship between homocysteine and cardiovascular disease in diabetic patients remained unclear with different factors that could influence this association such as ethnicity, levels of lipoprotein (a) and fibrinogen. Our data showed a good correlation between microalbuminuria and homocysteine, perhaps renal function plays a central role for clearance of homocysteine (Guttormsen, Ueland, Svarstad, & Refsum, 1995). Recent studies showed that microalbuminuria is strongly related to homocysteine in diabetic patients (Hoogeveen, Kostense, & Beks, 1998), although the relation may be independent of the diabetes per se. The Hoorn Study showed that a 5-Amol/l increase of the homocysteine levels was associated with an increased risk of developing microalbuminuria (Jeger et al., 2001). Although some authors did not find an association between homocysteine and microalbuminuria (Buysschaert, Dramais, Wallemacq, & Hermans, 2000), this relationship remains confusing in the literature. As far as other complications of diabetes are concerned, we found no correlation between tHcy levels and retinopathy, which is in agreement with other data reported in type 2 diabetes (Chico et al., 1999). In a Caucasian population, Stabler et al. (1999) showed a correlation between homocysteine and neuropathy in patients with diabetes type 2, in our population these data have not been confirmed. Perhaps hyperhomocysteinemia could cause macroangiopathy directly or indirectly (via hypertension, lipoprotein (a), fibrinogen, and/or renal dysfunction) through the hypothesized mechanism of endothelium damage (Coldwell, 1997). Our data showed high levels of blood pressure in Group I. Fiorina et al. (1998) postulated a particular propensity of patients with diabetes type 2 towards endothelial dysfunction and this fact could explain the presence of correlations between these metabolic parameters and arterial blood pressure. Although statistic differences have been detected in our study in Hcy levels and chronic complications in diabetic patients, the differences between results in different studies may be explained too by the preanalytical variations that could have an effect on Hcy determinations. These variations have been minimized in our work using an adequate protocol for sample extraction, but Hcy ex vivo generation cannot be completely discarded. In conclusion, the present study shows that elevation of plasma tHcy levels in type 2 diabetic patients is associated with a higher prevalence of peripheral arteriopathy and nephropathy. The precise mechanisms by which hyperhomocysteinemia is associated with chronic complications of diabetes remained unclear. Our data suggest that hyperhomocysteinemia is not associated with fat mass but it is associated with high levels of fibrinogen and lipoprotein (a). Further prospective studies are needed to define the causality direction for this association.
46
D.A. de Luis et al. / Journal of Diabetes and Its Complications 19 (2005) 42 – 46
References Arnesen, E., Refsum, H., & Bonaa, K. H. (1995). Serum total homocysteine and coronary heart disease. International Journal of Epidemiology, 24, 704 – 709. Boushey, C. J., Beresford, S. A. A., & Omenn, G. S. (1995). A quantitative assessment of plasma homocysteine as a risk factor for vascular disease. Probable benefits of increasing folic acid intakes. JAMA, 274, 1049 – 1057. Buysschaert, M., Dramais, A. S., Wallemacq, P., & Hermans, M. (2000). Hyperhomocysteinemia in type 2 diabetes. Diabetes Care, 23, 1816 – 1822. Chico, A., Perez, A., Cordoba, A., Arcelus, R., Carreras, G., de Leiva, A., et al (1999). 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. Coldwell, J. A. (1997). Elevated plasma homocysteine and diabetic vascular disease (Editorial). Diabetes Care, 20, 1805 – 1806. Coull, B. M., Malinow, M. R., & Beamer, N. (1990). Elevated plasma homocystein concentration as a possible independent risk factor for stroke. Stroke, 21, 572 – 576. Das, S., Reynolds, T., Patnaik, A., Rais, N., Fink, L. M., & Fonseca, V. A. (1999). Plasma homocysteine concentrations in type II diabetic patients in India: Relationship to body weight. Journal of Diabetes Complications, 13, 200 – 203. Fiorina, P., Lanfredini, M., Montanari, A., Peca, M. G., Veronelli, A., Mello, A., et al (1998). Plasma homocysteine and folate are related to arterial boold pressure in type 2 diabetes mellitus. American Journal of Hypertension, 11, 1100 – 1107. Foody, J. M., Milberg, J. A., Robinson, K., Pearce, G. L., Jacobsen, D. W., & Sprecher, D. L. (2000). Homcysteine and lipoprotein (a) interact to increase CAD risk in young men and women. Arteriosclerosis, Thrombosis and Vascular Biology, 20, 493 – 499. Giles, W. H., Croft, J. B., Greenlund, K. J., Ford, E. S., & Kittner, S. J. (2000). Association between total homocysteine and the likelihood for a history of acute myocardial infarctation by race and ethnicity: Results from the Third National Health and Nutrition Examination Survey. American Heart Journal, 139, 446 – 453. Guttormsen, A. B., Ueland, P. M., Svarstad, E., & Refsum, H. (1995). Kinetic basis of hyperhomocisteinemia in patients with endstage renal failure. Irish Journal of Medical Science, 164, 8. Hoogeveen, E. K., Kostense, P. J., & Beks, P. J. (1998). Hyperhomocysteinemia is associated with an increased risk of cardiovascular disease, especially in non-insulin-dependent diabetes mellitus: A population-
based study. Arteriosclerosis, Thrombosis and Vascular Biology, 18, 133 – 138. Jeger, A., Kostense, P. J., Nijpels, G., Dekker, J. M., Heine, R. J., Bouter, L. M., et al (2001). Serum homocysteine levels are associated with the development of microalbuminuria: The Hoorn stuy. Arteriosclerosis, Thrombosis and Vascular Biology, 21, 74 – 81. Malinow, M. R., Nieto, F. J., & Szklo, M. (1993). Carotid artery intimalmedial wall thickening and plasma homocystein in asymptomatic adults. The atherosclerosis Risk in Communities Study. Circulation, 87, 1107 – 1113. Perry, I. J., Refsum, H., & Morris, R. W. (1995). Prospective study of serum homocysteine concentration and risk of stroke in middle-aged British men. Lancet, 346, 1395 – 1398. Pichard, C., Slosman, D., Hirschel, B., & Kyle, U. (1997). Bioimpedance analysis in AIDS patients: An improved method for nutritional follow up. Clinical Research, 41, 53a. Rosenberg, I. H., & Miller, J. W. (1992). Nutrition factors in physical and cognitive functions of elderly people. American Journal of Clinical Nutrition, 55, 1237s – 1243. Selhub, J., Jacques, P. F., & Bostom, A. G. (1995). Association between plasma homocystein concentrationsand extracranial carotid-artery stenosis. New England Journal of Medicine, 332, 286 – 291. Smulders, Y. M., Rakic, M., Slaats, E. H., Treskes, M., Sijbrands, E. J., Odekerken, D. A., et al (1999). Fasting and post-methionine homocysteine levels in NIDDM. Determinants and correlations with retinopathy, albuminuria and cardiovascular disease. Diabetes Care, 22, 125 – 132. Stabler, S. P., Estacio, R., Jeffers, B. W., Cohen, J. A., Allen, R. H., & Schrier, R. W. (1999). Total homocysteine is associated with nephropathy in non-insulin-dependent diabetes mellitus. Metabolism, 48, 1096 – 1101. Stampfer, M. J., Malinow, R., & Willett, W. C. (1992). A prospective study of plasma homocysteine and risk of myocardial infarctation US physicians. JAMA, 268, 877 – 881. Tsai, W. C., Li, Y. H., Tsai, L. M., Chao, T. H., Lin, L. J., Chen, T. Y., et al (2000). Correlation of homocysteine levels with the extent of coronary atherosclerosis in patients with low cardiovascular risk profiles. American Journal of Cardiology, 85, 49 – 52. Verhoef, P., Hennekens, C. H., & Malinow, M. R. (1994). A prospective study of plasma homocysteine and risk of ischemic stroke. Stroke, 25, 1924 – 1930. Weir, D., & Molloy, A. (2000). Microvascular disease and dementia in the elderly: Are they related to hyperhomocysteinemia? American Journal of Clinical Nutrition, 71, 859 – 860. Welch, G., & Loscalzo, J. (1998). Homocysteine and atherothrombosis (Review). New England Journal of Medicine, 338, 1042 – 1050.
Total homocysteine levels relation with chronic complications of diabetes, body composition, and other cardiovascular risk factors in a population of patients with diabetes mellitus type 2 D.A. de Luis*, N. Fernandez, M.L. Arranz, R. Aller, O. Izaola, E. Romero Institute of Endocrinology and Nutrition, Medicine School and Hospital Rio Hortega, University of Valladolid, C/Los perales 16 (URB Las Acen˜as), Simancas 47130, Valladolid, Spain Received 5 June 2003; received in revised form 23 September 2003; accepted 15 December 2003
Abstract Objective: The significance of hyperhomocysteinemia in type 2 diabetes is further complicated by the multiple ways of considering impaired renal function and vitamin status. The aim of our study was to analyze the relationship between total homocysteine (tHcy) in a population of type 2 diabetic patients and chronic complications. We also analyzed the relationship between tHcy and the body composition of these patients and other cardiovascular risk factors. Design: In a cross-sectional study, a total of 155 patients with diabetes mellitus attending in our diabetes service (90 females/65 males) were enrolled in a consecutive way. Material and methods: All enrolled patients underwent the following examinations; (i) biochemical cardiovascular risk factors including total cholesterol, triglyceride, lipoprotein (a), low-density lipoprotein (LDL-cholesterol), high-density lipoprotein (HDL-cholesterol), glucose, HbA1c, fibrinogen, homocysteine, vitamin B12, folate, and microalbuminuria; and (ii) fat mass assessed by body mass index, weight, percentage of fat mass, and tricipital skinfold. Results: Patients were divided in two groups (Group I: tHcyz15 Amol/l; Group II: tHcy<15 Amol/l). Smoking habit was similar in both groups. A prevalence of cerebrovascular accident was present in 3.3% in the total group. This prevalence was not different in both groups (7.4% vs. 2.3%; ns) (OR 3.3; 95% CI 0.49 – 19.68). The prevalence of coronary heart disease in the total group was 5.8% without statistical differences between groups (3.5% vs. 6.3%; ns) (OR 0.57; 95% CI 0.065 – 4.53). Concerning macrovascular complications, only peripheral vascular disease prevalence was higher in Group I (16% vs. 3.1%; P < 0.05; OR 5.33; 95% CI 1.18 – 21.5). A prevalence of nephropathy was higher in Group I (93.3% vs. 12.8%; P < 0.05; OR 7.15; 95% CI 2.9 – 17.9). No statistical differences were detected in prevalence of retinopathy (global group 41.9%) (42.5% vs. 40.9%; ns) (OR 1.75; 95% CI 0.78 – 3.9). Also, peripheral neuropathy was similar in both groups (7.1% vs. 6.5%; ns) (OR 1.1; 95% CI 0.15 – 8.2). No correlation was detected among homocysteine and anthropometric parameters (body mass index, weight, percentage of fat mass, fat mass, and tricipital skinfold). Elevated levels of fibrinogen, lipoprotein (a), microalbuminuria, and blood pressure were detected in Group I. Conclusion: The present study shows that elevation of plasma tHcy levels in type 2 diabetic patients is associated with a higher prevalence of peripheral arteriopathy and nephropathy. Our data suggest that hyperhomocysteinemia is not associated with fat mass but it is associated with high levels of fibrinogen, lipoprotein (a), microalbuminuria, and blood pressure levels. D 2005 Elsevier Inc. All rights reserved. Keywords: Diabetes mellitus; Homocysteine; Macroangiopathy; Microangiopathy
1. Introduction It is important to search for modificable risk factors in patients with diabetes mellitus type 2. The aminoacid homocysteine may be such a risk factor (Rosenberg & Miller, 1992). A lot of longitudinal and cross-sectional
* Corresponding author. Fax: +34-98-333-1566. E-mail address: [email protected] (D.A. de Luis). 1056-8727/05/$ – see front matter D 2005 Elsevier Inc. All rights reserved. doi:10.1016/j.jdiacomp.2003.12.003
studies have shown that homocysteine is associated with an increased risk of stroke (Coull, Malinow, & Beamer, 1990; Perry, Refsum, & Morris, 1995; Verhoef, Hennekens, & Malinow, 1994), carotid artery atherosclerosis (Malinow, Nieto, & Szklo, 1993; Selhub, Jacques, & Bostom, 1995), and coronary heart disease (Arnesen, Refsum, & Bonaa, 1995; Stampfer, Malinow, & Willett, 1992). Although type 2 diabetes is definitely associated with premature atherosclerosis and microvascular complications, only a handful of studies have dealt with the between hyperhomocysteinemia and macro- or microangiopathic complications. The
D.A. de Luis et al. / Journal of Diabetes and Its Complications 19 (2005) 42 – 46
significance of hyperhomocysteinemia in type 2 diabetes is further complicated by the multiple ways of considering impaired renal function and vitamin status. Therefore, we measured total homocysteine (tHcy) in a population of type 2 diabetic patients and assessed whether high levels were related with chronic complications. We also analyzed the relationship between tHcy, body composition, and other cardiovascular risk factors.
2. Material and methods 2.1. Subjects In a cross-sectional study, a total of 155 patients attending in our diabetes service (90 females/65 males) with diabetes mellitus were enrolled in a consecutive way. Exclusion criteria were: treatments such as medical conditions known to increase tHcy (e.g., dysthyroidism, and drugs with interaction with folate and vitamin B12) and diabetes type 1. The group was divided in two groups according to their fasting plasma tHcy: Group I had increased tHcy (z15 Amol/l, n = 22), whereas Group II had normal laboratory values (<15 Amol/l, n =133). This value was selected from published data that defined high and normal tHcy levels (Welch & Loscalzo, 1998). The clinical characteristics of these patients included age of 64.6F8.6 years (range: 49 – 78), body mass index (BMI) of 29.6F6.3 kg/m2, and duration of diabetes 8.9F6.7 years (Table 1). Drugs in both groups were similar (antihypertensive drugs, hypoglycemic agents, fibrates, and statins). Smoking habit was similar in both groups (Group I 12.5% vs. Group II 16.4%; ns). The study was approved by the local ethical committee and each patient gave informed consent to participate in the study.
3. Design All enrolled patients underwent the following examinations: (i) biochemical cardiovascular risk factors including total cholesterol, triglyceride, lipoprotein (a), low density lipoprotein (LDL-cholesterol), high density lipoprotein (HDL-cholesterol), glucose, HbA1c fibrinogen, homocysteine, vitamin B12, folate, and microalbuminuria; and (ii) fat mass assessed by body mass index, weight, percentage of fat mass, and tricipital skinfold. 3.1. Chronic diabetic complications assessment All diabetic patients were checked in the clinic for chronic complications. Ischemic heart disease was clinically assessed (history of a cardiovascular event and/or presence of angina) and in addition, a 12-lead resting electrocardiogram was recorded supine (Mac PC Electrocardiograph, Marquette Electronics, Los Angeles, CA) and evaluated by a cardiologist. The presence of any of the following findings was
43
considered suggestive of coronary heart disease: T-wave inversion, ST-segment depression, and Q waves. Patients with indicative signs or symptoms of cerebrovascular disease were evaluated with a CNS computerized tomography. The final diagnosis was reviewed by a neurologist. Peripheral vascular disease was clinically defined by the presence of intermittent claudication, absent or weakened peripheral pulses, or both (no amputations were detected in our population). Retinopathy was documented by standard fundus eye examination and diagnosed on the presence of microaneurysms, venous dilatation, cotton-wool spots, neovascularization, or hemorrhages. Clinical neuropathy was defined by an abnormal neurologic examination, consistent with the presence of peripheral sensorimotor neuropathy. Nephropathy was defined by the presence of urinary albumin excretion of > 30 mg 24-h and/or high plasma creatinine levels. 3.2. Laboratory determinations Serum total cholesterol, HDL-cholesterol, and triglyceride concentrations were determined by enzymatic colorimetric assay (Hitachi 917, Roche Diagnostics, Mannheim, Germany), while HDL-cholesterol was determined enzymatically in the supernatant after precipitation of other lipoproteins with dextran sulfate-magnesium. LDL cholesterol was calculated using Friedewald formula. Lipoprotein (a) was determined by nephelometry (Beckman Coulter, Fullerton, CA, USA). Glycated haemoglobin was measured as HbA1c by HPLC (Menarini, Florence, Italy). Plasma glucose levels were determined by using an automated glucose oxidase method (Hitachi 917, Roche Diagnostics, Mannheim, Germany). Homocysteine levels were determined by immunoassay (Abbott, Wiesbaden, Germany) (normal value: 5 – 13 Amol/l). Fasting serum samples were obtained at baseline, put on ice immediately, and were processed within 60 min, which has been shown to be sufficient to prevent increases in tHcy concentration due to ex vivo generation. Serum was kept frozen at 20jC until determination of tHcy. The
Table 1 Clinical characteristics of diabetic patients studied (meanFS.D.) n Age (years) Sex (male/female) BMI (kg/m2) Diabetes course (years) Glucose (mmol/l) Retinopathy (%) Nephropathy (%) Neuropathy (%) Stroke (%) Peripheral vascular disease (%) Coronary heart disease (%) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
155 64.6F8.6 65/90 29.6F6.3 8.4F6.7 9.72F3.1 41.9 18.7 6.8 3.3 5.2 5.8 142.9F18.9 81.7F9
44
D.A. de Luis et al. / Journal of Diabetes and Its Complications 19 (2005) 42 – 46
coefficients of variation (inter-assay and intra-assay) were 5% and 6%, respectively. Fibrinogen levels were determined by a quantitative assay (Boehringer-Mannheim, Mannheim, Germany) (normal value: < 300 mg/dl). Vitamin B12 (normal value: 200– 970 pg/ml) and folate (normal value: 2.2– 17.5 ng/ml) were measured using an immunoassay (Abbott). 3.3. Body composition Body weight was measured to an accuracy of 0.5 kg and body mass index was computed as body weight/(height2). Bipolar body electrical bioimpedance was used to determine body composition (Pichard, Slosman, Hirschel, & Kyle, 1997). An electric current of 0.8 mA and 50 kHz was produced by a calibrated signal generator (Biodynamics Model 310e, Seattle, WA, USA) and applied to the skin using adhesive electrodes placed on right-side limbs. Resistance and reactance were used to calculate total body water, fat, and fat-free mass. Regional changes in body mass were estimated by measuring tricep skinfold of the midarm.
4. Statistical analysis The results were expressed as meanFstandard deviation. The distribution of variables was analyzed with the Kolmogorov – Smirnov test. Quantitative variables with normal distribution were analyzed with a two-tailed, paired Student’s t test. Nonparametric variables were analyzed with the Mann – Whitney U test. Qualitative variables were analyzed with the chi-square test, with Yates correction as necessary, and Fisher’s test. Pearson and Spearman tests were used in correlational analysis. Odds ratio and 95% CI were used to studied association among complications of diabetes mellitus and high levels of homocysteine, adjusted by age, sex, and smoking habit. A P value under .05 was considered statistically significant.
5. Results Clinical characteristics of the patients studied are shown in Table 1. Mean homocysteine levels in patients was Table 2 Anthropometric parameters (t test) r ( P) .07 .12 .04 .02 .05
Parameter (.9) (.7) (.8) (.6) (.9)
BMI Weight (kg) Fat mass (kg) % fat mass Tricipital skinfold (mm)
Group I (tHcy >15)
Group II (tHcy <1)
30.5F7.4 75.9F14.2 22.8F9.4 29.3F9.6 25.1F7
29.1F5.6 73.5F13.2 20.1F9.5 26.8F10.1 25.2F6.1
Table 3 Cardiovascular risk factors Parameter Glucose (mmol/l) HbA1c (%) Total cholesterol (mmol/l) LDL cholesterol (mmol/l) HDL cholesterol (mmol/l) Lipoprotein (a) (mg/dl) Fibrinogen (mg/dl) Triglycerides (g/l) Microalbuminuria (mg/day) Systolic blood pressure (mmHg) Diastolic blood pressure (mmHg)
Group I (tHcy z15) 8.4F3.5 8.4F3.5 5.9F1.3
Group II (tHcy <15) 10.2F2.9 (.1) 9.2F2.9 (.8) 5.94F1.13 (.9)
3.33F0.28 (.9)
3.34F1
1.67F0.66 (.4)
1.67F0.4
62F89 (<.05) 408.3F106 (<.05) 1.33F0.73 (.3) 563.16F674.7 152.7F20.1 (<.05) 85F9 (<.05)
31.7F36.7 358F69 1.23F0.59 67.8F166.7 (<.05) 140.6 +19 81.4F11
P values are in parentheses.
10.5F4.3 Amol/l. No statistical differences were detected between homocysteine levels in patients without and with macroangiopathy (cardiovascular disease and/or peripheral vascular disease and/or cerebrovascular accident) (10.13F 3.67 Amol/l vs. 11.68F4.1 Amol/l). Patients without and with microangiopathy (retinopathy and/or nephropathy and/or neuropathy) showed statistical differences in homocysteine levels (9.7F4.2 Amol/l vs. 11.9F4.5 Amol/l). Prevalence of cerebrovascular accident, as shown by clinical examination, imaging techniques, or both, was present in 3.3% in the total group. This prevalence was not different in both groups (7.4% vs. 2.3%; ns) (OR 3.3; 95% CI 0.49 –19.68). The prevalence of coronary heart disease in the total group was 5.8% without statistical differences between groups (3.3% vs. 6.3%; ns) (OR 0.57; 95% CI 0.065 – 4.53). Only peripheral vascular disease prevalence was higher in Group I (16% vs. 3.1%; P <.05), (OR 5.33; 95% CI 1.18 – 21.5). Prevalence of nephropathy was higher in Group I (93.3% vs. 12.8%; P < .05; OR 7.15; 95% CI 2.9– 17.9). No statistical differences were detected in prevalence of retinopathy (global group 41.9%) (42.5% vs. 40.9%; ns) (OR 1.75; 95% CI 0.78 –3.9). Also, peripheral neuropathy was similar in both groups (7.1% vs. 6.5%; ns) (OR 1.1; 95% CI 0.15 –8.2). No correlations were detected among homocysteine and anthropometric parameters (body mass index, weight, percentage of fat mass, fat mass, and tricipital skinfold). Table 2 shows the mean values of these parameters in Groups I and II, without statistical differences. Concerning biochemical results, only microalbuminuria (r =.47; P < .001) and fibrinogen (r =.3; P < .05) showed a direct correlation with homocysteine. No correlation was
D.A. de Luis et al. / Journal of Diabetes and Its Complications 19 (2005) 42 – 46
detected among homocysteine and other cardiovascular risk factors, lipid parameters, and glucose control. Table 3 shows the statistical differences in levels of fibrinogen, lipoprotein (a), and microalbuminuria in Groups I and II. Other cardiovascular parameters were similar in both groups. High levels of blood pressure were detected in Group I. Levels of vitamin B12 and folate were higher in Group II (Hcy <15 Amol/l) (529.8F368 pg/ml vs. 609.9F539.9 pg/ml; P < .05) and (9.8F7.8 ng/ml vs. 10.8F5.2 ng/ml; P <.05), respectively.
6. Conclusions A newly recognised cardiovascular risk factor is the sulphur amino acid homocysteine. High homocysteine levels (tHcy) have been associated with an increased risk of stroke and other cardiovascular events (Boushey, Beresford, & Omenn, 1995; Smulders et al., 1999). Also high homocysteine levels have been associated with microangiopathy (Stabler et al., 1999). This association between homocysteine and chronic complications of diabetes mellitus could be explained by different mechanisms; direct toxic effect on vascular endothelial and indirect effect on the normal methylation in endothelial cells (Weir & Molloy, 2000). Direct toxic effect of homocysteine could be mediated by damage to vascular endothelial cells, resulting in vascular events, such as microvascular disease and cerebrovascular accidents. Our data showed high prevalence of peripheral arteriopathy in patients with homocysteine z15 Amol/l. No differences were detected in stroke and coronary heart diseases in our population. Stabler et al. (1999) showed that tHcy was not increased in subjects with cardiovascular disease. However, in other study, a correlation between homocysteine levels with extension of coronary atherosclerosis was detected (Tsai et al., 2000). Almost a twofold increased likelihood of myocardial infarction among persons with a tHcy concentration z15 Amol/l was noted in the third National Health and Nutrition Examination Survey (Giles, Croft, Greenlund, Ford, & Kittner, 2000). These different results could be due to different characteristics of studied population such as ethnicity (blacks and Hispanics) (Giles et al., 2000) or other associated cardiovascular risk factors in these populations. For example, in the study of Fiorina et al. (1998), Caucasian patients with elevated tHcy levels had significantly higher diastolic pressure and mean arterial pressure. These results are similar to our data. Other populations (Indians) have shown correlation between homocysteine concentrations and body weight (Das et al., 1999). Our results did not show correlations among tHcy and different anthropometric parameters (body mass index, weight, percentage of fat mass, and tricipital skinfold). Other possible effect is the association of different cardiovascular risk factors with elevated tHcy such as lipoprotein (a). In our
45
study, this parameter was elevated in patients of Group I (Foody et al., 2000). However, the relationship between homocysteine and cardiovascular disease in diabetic patients remained unclear with different factors that could influence this association such as ethnicity, levels of lipoprotein (a) and fibrinogen. Our data showed a good correlation between microalbuminuria and homocysteine, perhaps renal function plays a central role for clearance of homocysteine (Guttormsen, Ueland, Svarstad, & Refsum, 1995). Recent studies showed that microalbuminuria is strongly related to homocysteine in diabetic patients (Hoogeveen, Kostense, & Beks, 1998), although the relation may be independent of the diabetes per se. The Hoorn Study showed that a 5-Amol/l increase of the homocysteine levels was associated with an increased risk of developing microalbuminuria (Jeger et al., 2001). Although some authors did not find an association between homocysteine and microalbuminuria (Buysschaert, Dramais, Wallemacq, & Hermans, 2000), this relationship remains confusing in the literature. As far as other complications of diabetes are concerned, we found no correlation between tHcy levels and retinopathy, which is in agreement with other data reported in type 2 diabetes (Chico et al., 1999). In a Caucasian population, Stabler et al. (1999) showed a correlation between homocysteine and neuropathy in patients with diabetes type 2, in our population these data have not been confirmed. Perhaps hyperhomocysteinemia could cause macroangiopathy directly or indirectly (via hypertension, lipoprotein (a), fibrinogen, and/or renal dysfunction) through the hypothesized mechanism of endothelium damage (Coldwell, 1997). Our data showed high levels of blood pressure in Group I. Fiorina et al. (1998) postulated a particular propensity of patients with diabetes type 2 towards endothelial dysfunction and this fact could explain the presence of correlations between these metabolic parameters and arterial blood pressure. Although statistic differences have been detected in our study in Hcy levels and chronic complications in diabetic patients, the differences between results in different studies may be explained too by the preanalytical variations that could have an effect on Hcy determinations. These variations have been minimized in our work using an adequate protocol for sample extraction, but Hcy ex vivo generation cannot be completely discarded. In conclusion, the present study shows that elevation of plasma tHcy levels in type 2 diabetic patients is associated with a higher prevalence of peripheral arteriopathy and nephropathy. The precise mechanisms by which hyperhomocysteinemia is associated with chronic complications of diabetes remained unclear. Our data suggest that hyperhomocysteinemia is not associated with fat mass but it is associated with high levels of fibrinogen and lipoprotein (a). Further prospective studies are needed to define the causality direction for this association.
46
D.A. de Luis et al. / Journal of Diabetes and Its Complications 19 (2005) 42 – 46
References Arnesen, E., Refsum, H., & Bonaa, K. H. (1995). Serum total homocysteine and coronary heart disease. International Journal of Epidemiology, 24, 704 – 709. Boushey, C. J., Beresford, S. A. A., & Omenn, G. S. (1995). A quantitative assessment of plasma homocysteine as a risk factor for vascular disease. Probable benefits of increasing folic acid intakes. JAMA, 274, 1049 – 1057. Buysschaert, M., Dramais, A. S., Wallemacq, P., & Hermans, M. (2000). Hyperhomocysteinemia in type 2 diabetes. Diabetes Care, 23, 1816 – 1822. Chico, A., Perez, A., Cordoba, A., Arcelus, R., Carreras, G., de Leiva, A., et al (1999). 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. Coldwell, J. A. (1997). Elevated plasma homocysteine and diabetic vascular disease (Editorial). Diabetes Care, 20, 1805 – 1806. Coull, B. M., Malinow, M. R., & Beamer, N. (1990). Elevated plasma homocystein concentration as a possible independent risk factor for stroke. Stroke, 21, 572 – 576. Das, S., Reynolds, T., Patnaik, A., Rais, N., Fink, L. M., & Fonseca, V. A. (1999). Plasma homocysteine concentrations in type II diabetic patients in India: Relationship to body weight. Journal of Diabetes Complications, 13, 200 – 203. Fiorina, P., Lanfredini, M., Montanari, A., Peca, M. G., Veronelli, A., Mello, A., et al (1998). Plasma homocysteine and folate are related to arterial boold pressure in type 2 diabetes mellitus. American Journal of Hypertension, 11, 1100 – 1107. Foody, J. M., Milberg, J. A., Robinson, K., Pearce, G. L., Jacobsen, D. W., & Sprecher, D. L. (2000). Homcysteine and lipoprotein (a) interact to increase CAD risk in young men and women. Arteriosclerosis, Thrombosis and Vascular Biology, 20, 493 – 499. Giles, W. H., Croft, J. B., Greenlund, K. J., Ford, E. S., & Kittner, S. J. (2000). Association between total homocysteine and the likelihood for a history of acute myocardial infarctation by race and ethnicity: Results from the Third National Health and Nutrition Examination Survey. American Heart Journal, 139, 446 – 453. Guttormsen, A. B., Ueland, P. M., Svarstad, E., & Refsum, H. (1995). Kinetic basis of hyperhomocisteinemia in patients with endstage renal failure. Irish Journal of Medical Science, 164, 8. Hoogeveen, E. K., Kostense, P. J., & Beks, P. J. (1998). Hyperhomocysteinemia is associated with an increased risk of cardiovascular disease, especially in non-insulin-dependent diabetes mellitus: A population-
based study. Arteriosclerosis, Thrombosis and Vascular Biology, 18, 133 – 138. Jeger, A., Kostense, P. J., Nijpels, G., Dekker, J. M., Heine, R. J., Bouter, L. M., et al (2001). Serum homocysteine levels are associated with the development of microalbuminuria: The Hoorn stuy. Arteriosclerosis, Thrombosis and Vascular Biology, 21, 74 – 81. Malinow, M. R., Nieto, F. J., & Szklo, M. (1993). Carotid artery intimalmedial wall thickening and plasma homocystein in asymptomatic adults. The atherosclerosis Risk in Communities Study. Circulation, 87, 1107 – 1113. Perry, I. J., Refsum, H., & Morris, R. W. (1995). Prospective study of serum homocysteine concentration and risk of stroke in middle-aged British men. Lancet, 346, 1395 – 1398. Pichard, C., Slosman, D., Hirschel, B., & Kyle, U. (1997). Bioimpedance analysis in AIDS patients: An improved method for nutritional follow up. Clinical Research, 41, 53a. Rosenberg, I. H., & Miller, J. W. (1992). Nutrition factors in physical and cognitive functions of elderly people. American Journal of Clinical Nutrition, 55, 1237s – 1243. Selhub, J., Jacques, P. F., & Bostom, A. G. (1995). Association between plasma homocystein concentrationsand extracranial carotid-artery stenosis. New England Journal of Medicine, 332, 286 – 291. Smulders, Y. M., Rakic, M., Slaats, E. H., Treskes, M., Sijbrands, E. J., Odekerken, D. A., et al (1999). Fasting and post-methionine homocysteine levels in NIDDM. Determinants and correlations with retinopathy, albuminuria and cardiovascular disease. Diabetes Care, 22, 125 – 132. Stabler, S. P., Estacio, R., Jeffers, B. W., Cohen, J. A., Allen, R. H., & Schrier, R. W. (1999). Total homocysteine is associated with nephropathy in non-insulin-dependent diabetes mellitus. Metabolism, 48, 1096 – 1101. Stampfer, M. J., Malinow, R., & Willett, W. C. (1992). A prospective study of plasma homocysteine and risk of myocardial infarctation US physicians. JAMA, 268, 877 – 881. Tsai, W. C., Li, Y. H., Tsai, L. M., Chao, T. H., Lin, L. J., Chen, T. Y., et al (2000). Correlation of homocysteine levels with the extent of coronary atherosclerosis in patients with low cardiovascular risk profiles. American Journal of Cardiology, 85, 49 – 52. Verhoef, P., Hennekens, C. H., & Malinow, M. R. (1994). A prospective study of plasma homocysteine and risk of ischemic stroke. Stroke, 25, 1924 – 1930. Weir, D., & Molloy, A. (2000). Microvascular disease and dementia in the elderly: Are they related to hyperhomocysteinemia? American Journal of Clinical Nutrition, 71, 859 – 860. Welch, G., & Loscalzo, J. (1998). Homocysteine and atherothrombosis (Review). New England Journal of Medicine, 338, 1042 – 1050.