Fasting Plasma Homocysteine Levels in the Insulin Resistance Syndrome The Framingham Offspring Study JAMES B. MEIGS, MD, MPH 1 PAUL F. JACQUES, PHD 2 JACOB SELHUB, PHD 2 DANIEL E. SINGER, MD 1 DAVID M. NATHAN, MD 3 NADER RIFAI, PHD 4 RALPH B. D’AGOSTINO,SR., PHD 5 PETER W.F. WILSON, MD 6 OBJECTIVE — Insulin resistance, associated metabolic abnormalities, and elevated homo- cysteine levels are risk factors for cardiovascular disease (CVD). We examined relationships between homocysteine levels and features of insulin resistance syndrome (IRS). RESEARCH DESIGN AND METHODS — We measured clinical characteristics, plasma levels of fasting homocysteine, folate, B vitamins, creatinine, and fasting and 2-h insulin and glucose levels after a 75-g oral glucose tolerance test in 2,214 subjects without CVD at the fifth examination (1991–1995) of the Framingham Offspring Study. After excluding 203 sub- jects with diabetes, the remaining 2,011 subjects were categorized as having none, one, two, or all three of the phenotypes of IRS: impaired glucose tolerance, hypertension, and/or a central metabolic syndrome (two or more traits: obesity, dyslipidemia, or hyperinsulinemia). In addi- tion, in 1,592 subjects attending the sixth examination (1995–1998), we measured the urine albumin/creatinine ratio (UACR). Age-, sex-, creatinine-, vitamin-, and UACR-adjusted mean homocysteine levels or proportions with homocysteine 14 mol/l in each phenotypic category and differences between categories were assessed with regression models. RESULTS — The mean age of the subjects was 54 years (range 28 – 82); 55% were women, 12.3% had hyperinsulinemia, and 15.9% had two or more of the IRS phenotypes. Adjusted mean homocysteine levels were higher comparing those with hyperinsulinemia (9.8 mol/l) and those without (9.4 mol/l, P = 0.04) and were higher among subjects with two or more IRS pheno- types (9.9 mol/l) compared with those with 1 or no phenotype (9.3 mol/l, P = 0.003). Mean UACR levels were also higher among subjects with two or more IRS phenotypes (7.2 mg/g) compared with those with 1 or no phenotype (5.5 mg/g, P = 0.007). CONCLUSIONS — Hyperhomocysteinemia and abnormal urinary albumin excretion are both associated with hyperinsulinemia and may partially account for increased risk of CVD associated with insulin resistance. Because hyperhomocysteinemia and microalbuminuria also reflect endothelial injury, these observations also support the hypothesis that endothelial dys- function is associated with expression of the IRS. Diabetes Care 24:1403–1410, 2001 I nsulin resistance is a fundamental ab- normality in the pathogenesis of type 2 diabetes and may also play a role in the development of atherosclerotic cardio- vascular disease (CVD) (1,2). Insulin re- sistance may be an expression of diffuse arterial endothelial dysfunction contrib- uting to atherosclerosis (3), may lead di- rectly to arterial damage through toxic effects of hyperinsulinemia (4), or may act indirectly through atherogenic effects of the constellation of risk factors associated with the insulin resistance syndrome (IRS) (5,6). Insulin resistance or hyperin- sulinemia has been associated with CVD in some but not all studies, suggesting that previously unmeasured factors may be involved in the link between hyperin- sulinemia and atherosclerosis (7–12). One potential factor may be variation in plasma levels of homocysteine. Homocysteine is a sulfur-containing amino acid formed during the metabo- lism of methionine. Elevated levels of homocysteine are toxic to vascular en- dothelium (13), inducing endothelial dysfunction and contributing to develop- ment of atherosclerosis independent of standard CVD risk factors in diabetic (14 –17) and nondiabetic subjects (18 – 20). Plasma levels of insulin seem to in- fluence homocysteine metabolism, possibly through effects on glomerular fil- tration or by influencing activity of key enzymes in homocysteine metabolism, including 5,10-methylenetetrahydrofo- late reductase (MTHFR) or cystathione -synthase (CBS). (21–25) However, most data suggest that fasting homocysteine levels are similar in subjects with type 2 diabetes and nondiabetic control subjects (26 –30). Although insulin resistance (or hyper- insulinemia) and hyperhomocysteinemia may both be associated with endothelial dysfunction and CVD, associations be- tween fasting levels of insulin and homo- cysteine in nondiabetic or prediabetic subjects are not well characterized. In this study, we assessed the relationship be- tween hyperinsulinemia, phenotypes of ●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●●● From the 1 General Medicine Division and Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; the 2 U.S. Department of Agriculture Human Nutrition Center on Aging, Tufts University, Boston, Massachusetts; 3 Diabetes Unit and Department of Medicine, Massachusetts General Hospital and Harvard Medical School, Boston, Massachusetts; 4 the Children’s Hos- pital Medical Center and Harvard Medical School, Boston, Massachusetts; the 5 Department of Mathematics, Statistics, and Consulting Unit, Boston University, Boston, Massachusetts; and the 6 Framingham Heart Study, Boston University School of Medicine, Framingham, Massachusetts. Address correspondence and reprint requests to James B. Meigs, MD, MPH, General Medicine Division, Massachusetts General Hospital, 50 Staniford St., 9th Floor, Boston, MA 02114. E-mail: jmeigs@partners. org. Received for publication 21 November 2000 and accepted in revised form 12 April 2001. Abbreviations: CBS, cystathione -synthase; CVD, cardiovascular disease; IGT, impaired glucose toler- ance; IRS, insulin resistance syndrome; MTHFR, 5,10-methylenetetrahydrofolate reductase; OGTT, oral glucose tolerance test; PLP, plasma pyridoxal 5'-phosphate; UACR, urine albumin/creatinine ratio. Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the authors and do not necessarily reflect the view of the U.S. Department of Agriculture. A table elsewhere in this issue shows conventional and Syste `me International (SI) units and conversion factors for many substances. Epidemiology/Health Services/Psychosocial Research O R I G I N A L A R T I C L E DIABETES CARE, VOLUME 24, NUMBER 8, AUGUST 2001 1403