Diabetologia (2007) 50:2164–2170 DOI 10.1007/s00125-007-0771-4 ARTICLE High prevalence of low plasma thiamine concentration in diabetes linked to a marker of vascular disease P. J. Thornalley & R. Babaei-Jadidi & H. Al Ali & N. Rabbani & A. Antonysunil & J. Larkin & A. Ahmed & G. Rayman & C. W. Bodmer Received: 23 February 2007 /Accepted: 22 June 2007 /Published online: 4 August 2007 # Springer-Verlag 2007 Abstract 0.001, and type 2 diabetes 16.3 (95% CI 13.0–9.6) nmol/l, Aims/hypothesis To assess thiamine status by analysis of p<0.001. Renal clearance of thiamine was increased 24- plasma, erythrocytes and urine in type 1 and type 2 diabetic fold in type 1 diabetic patients and 16-fold in type 2 patients and links to markers of vascular dysfunction. diabetic patients. Plasma thiamine concentration correlated Methods Diabetic patients (26 type 1 and 48 type 2) with negatively with renal clearance of thiamine (r=−0.531, p< and without microalbuminuria and 20 normal healthy 0.001) and fractional excretion of thiamine (r=−0.616, p< control volunteers were recruited. Erythrocyte activity of 0.001). Erythrocyte transketolase activity correlated nega- transketolase, the concentrations of thiamine and related tively with urinary albumin excretion (r=−0.232, p<0.05). phosphorylated metabolites in plasma, erythrocytes and Thiamine transporter protein contents of erythrocyte mem- urine, and markers of metabolic control and vascular branes of type 1 and type 2 diabetic patients were increased. dysfunction were determined. Plasma thiamine concentration and urinary excretion of Results Plasma thiamine concentration was decreased 76% thiamine correlated negatively with soluble vascular adhe- in type 1 diabetic patients and 75% in type 2 diabetic sion molecule-1 (r=−0.246, p<0.05, and −0.311, p<0.01, patients: normal volunteers 64.1 (95% CI 58.5–69.7) nmol/l, respectively). type 1 diabetes 15.3 (95% CI 11.5–19.1) nmol/l, p< Conclusions/interpretation Low plasma thiamine concen- tration is prevalent in patients with type 1 and type 2 diabetes, associated with increased thiamine clearance. The conventional assessment of thiamine status was masked by P. J. Thornalley : R. Babaei-Jadidi : H. Al Ali : N. Rabbani : increased thiamine transporter content of erythrocytes. A. Antonysunil : J. Larkin Department of Biological Sciences, University of Essex, . Colchester, Essex, UK Keywords Adhesion molecules Microalbuminuria Thiamine . Type 1 diabetes . Type 2 diabetes P. J. Thornalley (*) : N. Rabbani : A. Antonysunil : J. Larkin Clinical Sciences Research Institute, Warwick Medical School, Abbreviations University of Warwick, University Hospital, Clifford Bridge Road, ARB angiotensin receptor blocker Coventry CV2 2DX, UK ClThiamine renal clearance of thiamine e-mail: [email protected] FEThiamine fractional excretion of thiamine MNL mononuclear leucocyte A. Ahmed : C. W. Bodmer Department of Diabetes and Endocrinology, RFC-1 reduced folate carrier-1 Colchester General Hospital, sVCAM-1 soluble vascular adhesion molecule-1 Colchester, Essex, UK THTR-1 thiamine transporter-1 THTR-2 thiamine transporter-2 G. Rayman Ipswich Diabetic Foot Unit and Diabetes Centre, TK transketolase Ipswich Hospital NHS Trust, TMP thiamine monophosphate Ipswich, UK TPP thiamine pyrophosphate Diabetologia (2007) 50:2164–2170 2165 Introduction Methods Diabetes is increasing in incidence in the UK and elsewhere Patients and normal healthy volunteers Diabetic patients to epidemic proportions. The major health concerns of were recruited from patients attending the Diabetes Clinic at diabetes are achieving good control of blood glucose, BP Colchester General Hospital, and normal healthy control and lipids to suppress the development of associated volunteers from partners and friends of the patients and vascular complications—microvascular complications (ne- investigators. Inclusion criteria were: diabetic patients with phropathy, retinopathy and peripheral neuropathy) and normoalbuminuria (AER <30 mg/24 h) and microalbuminuria macrovascular complications (cardiovascular disease and (AER 30–300 mg/24 h); matched for age and sex (18– stroke) [1, 2]. This is not always achievable because of 65 years); diabetes duration of ≥5 years; HbA1c <10%; and limitations of current drug therapy, patient compliance and BMI 19–40 kg/m2. Exclusion criteria were: patients with other factors linked to the development of vascular end-stage renal disease; severe excess alcohol consumption complications [3]. Secondarily to the global epidemic of (>50 U per week; 1 U is 8 g alcohol); significant comor- diabetes, therefore, is a global burden of vascular compli- bidities; known allergy or intolerance to thiamine; use of cations [4]. The significant residual risk of development of thiamine supplements or goldenseal (Hydrastis canadensis,a microvascular and macrovascular complications in prospec- multipurpose herbal remedy); participating in an intervention tive clinical trials where best efforts have been made to study within 30 days; recipients of renal and/or pancreatic optimise glycaemic control [5, 6], and the failure of major transplants; and women who were pregnant or breastfeeding genetic susceptibility factors for diabetic complications to or of child-bearing potential not using adequate contracep- emerge from intensive studies [7] (although some may be tive precautions. Twenty-four hour urine collections and found with further study [8, 9]), encourage the search for blood samples (fasting) were taken with informed consent of other metabolic and nutritional factors that may predispose the patients and volunteers. Plasma, erythrocyte and mono- to the development of vascular complications in diabetes. nuclear leucocyte (MNL) fractions were prepared immedi- Vascular complications develop progressively from 5– ately and stored at −80°C until analysis. Ethical approval for 40 years after the onset of diabetes, although they may appear the study was given by the local ethics committee (North and earlier in type 2 diabetes when associated with a period of Mid-Essex Local Research Ethics Committee). undiagnosed diabetes and impaired glucose tolerance. A strategy to counter multiple pathways of biochemical Assay of thiamine and phosphorylated metabolites Thi- dysfunction linked to the development of vascular complica- amine status was assessed by measurement of thiamine, tions emerged from recent advances in understanding of the thiamine monophosphate (TMP) and TPP, determined by cell biology of diabetic complications: high-dose therapy HPLC with fluorimetric detection (pre-column derivatisa- with thiamine and related derivatives such as benfotiamine. tion to thiachromes) [13]. The retention times, limits of High-dose thiamine therapy prevented the development of detection, interbatch CV values and recoveries for these microvascular complications in experimental diabetes with- metabolites were: thiamine 13.1 min, 36 fmol, 1.1 and out improvement of metabolic control, as reviewed in [10]. 97%; TMP 6.0 min, 52 fmol, 2.1 and 92%; and TPP Experimental diabetes was also associated with tissue- 4.5 min, 51 fmol, 2.9 and 94%. There was no interference specific thiamine deficiency characterised by a marked from glucose in the assay and validations were not decrease of plasma thiamine concentration and decreased significantly different in analysis of samples from normal activity of the thiamine-dependent enzyme of transketolase healthy volunteers and diabetic patients. Sample storage (TK) and decreased levels of TK protein in renal glomeruli studies indicated that analyte content was stable for plasma, linked to a profound increase in renal clearance of thiamine erythrocytes and urine stored at −80°C for at least 6 weeks (ClThiamine)[11]. Thiamine deficiency was not detected by and storage of plasma, erythrocyte haemolysate and urine the conventional indicator of thiamine status, the erythrocyte and related de-proteinised extracts for 8 h. Stock solutions ‘thiamine effect’ (the percentage unsaturation of TK with of thiamine, TMP and TPP were calibrated by spectropho- thiamine pyrophosphate [TPP] cofactor) [12]. The preva- tometry assuming molar extinction coefficients of ɛ233= −1 −1 lence of similar low plasma thiamine concentrations and 14.2, ɛ247=15.3 and ɛ247=13.0 (mmol/l) cm , respec- impaired renal handling of thiamine in clinical diabetes has tively [14]. Shewhart analysis assessing the stability of not been addressed previously. analyte estimates in samples every day over a period of ten In this study, we sought to establish whether there is a consecutive days indicated the analysis had acceptable disturbance of thiamine homeostasis prevalent in type 1 and quality control (all estimates within mean±2 SD) [15]. The type 2 diabetic patients in the UK and if this is linked to concentration of thiamine metabolites was determined in incipient nephropathy and markers of vascular dysfunction plasma, erythrocytes and urine; ClThiamine and fractional and early stage renal dysfunction. excretion of thiamine (FEThiamine) were deduced. 2166 Diabetologia (2007) 50:2164–2170 Other biochemical measurements Erythrocyte TK activity clearance. Fifty diabetic patients (17 type 1 and 33 type 2) and percentage unsaturation with TPP (thiamine effect) had urinary albumin excretion within the normal range were determined. Erythrocyte membrane content of thia- (median 11.5, 2.3–29.1 mg/24 h) and 24 had urinary mine transporter-1 (THTR-1) [16]andreducedfolate albumin excretion characteristic of microalbuminuria and carrier-1