Abstract. Aim of the study was to clarify the relationship between metformin-induced vitamin B12 (B12) deficiency, hyperhomocysteinemia and vascular complications in patients with type 2 diabetes. Serum B12 concentrations, homocysteine plasma levels, the presence of retinopathy and history of macroangiopathy (stroke or coronary heart disease) were analyzed in patients without renal dysfunction (serum creatinine<115 µmol/L). Firstly, B12 status was analyzed in 62 consecutive metformin-treated patients. Secondly, the relationship between B12, homocysteine and vascular complications was analyzed in 46 metformin-treated and 38 age-and sex-matched non-metformin-treated patients. Among the 62 consecutive metformin-treated patients, B12 was deficient (<150 pmol/L) in 8 (13%) and borderline-deficient (150-220 pmol/L) in 18 (29%): the larger the metformin dosage, the lower the B12 (P=0.02, Spearman's ρ=-0.30). There were independent correlations between metformin use and B12 lowering (P = 0.02, r = -0.25), and B12 lowering and elevation of homocysteine (P<0.01, r =-0.34). Elevation of homocysteine was a risk for retinopathy (P =0.02, OR 1.26, 95%CI 1.04-1.52). There was no significant relation between homocysteine and macroangiopathy. Correlation between B12 and homocysteine was stronger in metformin-treated (P<0.01, r =-0.48) than non-metformin-treated (P=0.04, r = -0.38) patients. In ten B12 deficient patients, B12 supplementation (1,500 µg/day) for 2.2±1.0 months with continued use of metformin raised B12 levels: 152±42 and 299±97 pmol/L before and after treatment, respectively (P<0.01). Metformininduced B12 lowering in diabetes was associated with elevation of homocysteine, and hyperhomocysteinemia was independently related to retinopathy. Metformin-induced B12 deficiency was correctable with B12 supplementation.Key words: Metformin, Homocysteine, Diabetes, Retinopathy VITAMIN B12 (B12) DEFICIENCY in metformintreated patients with type 2 diabetes mellitus (T2DM) is well known [1][2][3][4][5]. The drug may interfere with the calcium-dependent absorption of B12 [6,7]. In general, B12 deficiency is a cause of hyperhomocysteinemia [8], which is a risk factor for micro-and macrovascular complications [9][10][11]. However, as far as we are aware, no systematic analysis of the status of B12, homocysteine (HC) and vascular complications in relation to metformin use has been performed in patients with T2DM. We hypothesized that metformin-induced B12 deficiency may be causing an elevation of HC in patients with T2DM, which may in turn have a deleterious effect on diabetic vascular complications. Because metformin is the first choice oral hypoglycemic agent (OHA) [12] and is prescribed in a large number of patients worldwide, the issue is a very important one. Materials and Methods ParticipantsThe study was approved by the Aizawa Hospital Review Board and written informed consent was obtained from the participants. Patients with renal dysfunction (serum creatinine ≥115 µmol/L), or a history of gastrectomy, and those receiving v...

No abstract

Aim of this study was to formulate an index for glucose effectiveness (Sg), SgIo, based on 3-point (0, 30 and 120 min) 75 g oral glucose tolerance test (OGTT). The equation for SgI(O) was developed in the Chikuma cohort (n = 502). Firstly, post-loading plasma glucose without insulin action and Sg (PPG-without insulin and Sg) was calculated as follows: fasting plasma glucose (mg/dl) + [0.75 × 75,000]/[0.19 × BW(kg) × 10]. Secondly, 'PPG-without insulin/with Sg' was obtained from inverse correlation between log(10)DI(O) and 2-h post-glucose plasma glucose at OGTT (2hPG) in each glucose tolerance category: DI(O) denotes oral disposition index, a product of the Matsuda Index and δIRI(0-30)/δPG(0-30). Thirdly, expected 2hPG (2hPG(E)) of a given subject was obtained from the regression, and the ratio of 2hPG to 2hPG(E) (2hPG/2hPG(E)) was determined as an adjustment factor. Lastly, SgI(O) ([mg/dl]/min) was calculated as [PPG-without insulin and Sg]-[PPG-without insulin / with Sg] x [(2hPG) / 2hPG(E)]. SgI(O) was validated against Sg obtained by frequently sampled intravenous glucose tolerance test in the Jichi cohort (n = 205). Also, the accuracy of prediction of Sg by SgIo was tested by the Bland-Altman plot. SgI(O) was 3.61 ± 0.73, 3.17 ± 0.74 and 2.15 ± 0.60 in subjects with normal glucose tolerance (NGT), non-diabetic hyperglycemia and diabetes, respectively, in the Chikuma cohort. In the Jichi cohort, SgI(O) was significantly correlated with Sg in the entire group (r = 0.322, P < 0.001) and in subjects with NGT (r = 0.286, P < 0.001), and SgIo accurately predicted Sg. In conclusion, SgI(O) could be a simple, quantitative index for Sg.

No abstract

No abstract