Fibrinogen and Von Willebrand Factor in IDDM: Relationships to Lipid
Total Page:16
File Type:pdf, Size:1020Kb
Diabetologia (1997) 40: 698–705 Springer-Verlag 1997 Fibrinogen and von Willebrand factor in IDDM: relationships to lipid vascular risk factors, blood pressure, glycaemic control and urinary albumin excretion rate: the EURODIAB IDDM complications study M. Greaves1, R. G. Malia2, K. Goodfellow2, M. Mattock3, L. K. Stevens4, J.M. Stephenson4, J.H. Fuller4, and the EURODIAB IDDM Complications Study Group* 1 Department of Medicine and Therapeutics, University of Aberdeen, Scotland, UK 2 Department of Haematology, Royal Hallamshire Hospital, Sheffield, UK 3 Department of Chemical Pathology, UMDS, St. Thomas’s Hospital, London, UK 4 Department of Epidemiology and Public Health, University College, London, UK Summary The interrelationships between fibrinogen, fibrinogen and high density lipoprotein cholesterol von Willebrand factor, a marker of vascular endothe- was also apparent in males. A prominent feature was lial cell damage, and serum lipids were explored in a positive relationship between both fibrinogen and well-characterised subjects with insulin-dependent von Willebrand factor and albumin excretion rate diabetes mellitus. The 2091 subjects were enrolled (p < 0.001 and p < 0.003 respectively) in those with into a cross-sectional, clinic-based study of complica- retinopathy but not in those without this complica- tions, from 16 European countries: the EURODIAB tion. In view of previous observations on blood pres- IDDM Complications study. The anticipated signifi- sure and albuminuria in these subjects the findings cant relationships between both plasma fibrinogen are consistent with the hypothesis that microalbumin- and plasma von Willebrand factor concentrations uria and increased plasma von Willebrand factor are and age and glycaemic control, and between fibrino- due to endothelial cell perturbation in response to gen and body mass index, were noted. Fibrinogen, ad- mildly raised blood pressure in subjects with retinop- justed for age and glycated haemoglobin concentra- athy. Fibrinogen may also contribute to microvascu- tion, was also related to smoking habits and was high- lar disease and its relationships to lipid vascular risk er in the quartiles with highest systolic and diastolic factors suggest a possible pathogenic role in arterial blood pressures. There was a clustering of vascular disease in diabetes. [Diabetologia (1997) 40: 698–705] risk factors, with a positive relationship between plas- ma fibrinogen and serum triglyceride concentrations Keywords Fibrinogen, von Willebrand factor, albu- in both genders and between fibrinogen and total minuria, insulin-dependent diabetes mellitus, vascu- cholesterol in males. An inverse relationship between lar disease. The pathogenic mechanisms underlying the develop- ment of microvascular complications in insulin-de- Received: 10 December 1996 and in revised form: 14 March pendent diabetes mellitus (IDDM) remain obscure 1997 despite the fact that retinopathy develops in the ma- Corresponding author: Dr. M. Greaves, Department of Medi- jority [1–3], and nephropathy in around 30% [3, 4]. cine and Therapeutics, University of Aberdeen, Polwarth Although duration of diabetes and poor glycaemic Building, Foresterhill, Aberdeen AB25 2ZD, UK control are established risk factors, they appear to ac- Abbreviations: AER, Albumin excretion rate; ApoA1, apolipo- count for only a proportion of an individual’s risk [5]. protein-A1; BMI, body mass index; CVD, cardiovascular dis- We have recently demonstrated that, in subjects with ease; ELISA, enzyme-linked immunosorbent assay; HDL, retinopathy but not those without, mean urinary al- high density lipoprotein; IDDM, insulin-dependent diabetes mellitus; LDL, low density lipoprotein; vWF, von Willebrand bumin excretion rate increases steeply when the factor. mean diastolic blood pressure rises above 75 mmHg. It thus appears that members of a subgroup of dia- * see acknowledgements betic subjects may have abnormal renal vulnerability M. Greaves et al.: Fibrinogen and vWF in IDDM 699 to mildly raised blood pressure and that retinopathy Table 1. Demographic and clinical characteristics of the pa- is a close correlate of this risk [6]. tients von Willebrand factor (vWF) is synthesised by vas- Men Women cular endothelial cells and megakaryocytes and an in- n = 1064 n = 1027 creased concentration in plasma can serve as a mar- Age (years) 33.0 ± 10.2 32.8 ± 10.1 ker of endothelial cell damage [7–9]. Such an increase Duration (years) 14.5 ± 9.7 15.0 ± 9.3 ± ± is a feature of diabetic nephropathy [10]. Recent HbA1c (%) 6.6 1.8 6.7 1.9 Body mass index (kg/m2) 23.6 ± 2.7 23.5 ± 3.2 studies in small numbers of subjects with IDDM sug- Systolic blood pressure (mmHg) 124.3 ± 16.4 119.0 ± 18.4 gest that vWF is a sensitive marker of incipient neph- Diastolic blood pressure (mmHg) 77.0 ± 11.3 74.0 ± 11.4 ropathy and that a rise in the plasma concentration Total cholesterol (mmol/l) 5.20 ± 1.14 5.5 ± 1.12 may even precede the development of microalbumin- HDL-cholesterol (mmol/l) 1.37 ± 0.38 1.62 ± 0.44 LDL-cholesterol (mmol/l) 3.34 ± 0.98 3.41 ± 1.02 uria [11, 12]. Information regarding any relationship Fasting triglyceridea (mmol/l) 1.03 ± 3.59 0.94 ± 3.11 between vWF and retinopathy is, however, conflict- Fibrinogen (g/l) 3.07 ± 0.93 3.37 ± 0.96 ing [13–15]. vWF is an essential component in platelet von Willebrand factora (U/l) 1.09 ± 3.79 1.09 ± 3.46 adhesion to subendothelium and a potential role in n (%) n (%) the pathogenesis of diabetic microvascular disease Ex-smoker 238 (22) 149 (15) cannot be discounted [7]. Current smoker 368 (35) 291 (28) Neuropathy present 291 (29) 241 (25) Several epidemiological studies have confirmed Microalbuminuria present 244 (24) 190 (19) the role of plasma fibrinogen as an independent risk Macroalbuminuria present 109 (11) 84 (8) factor for myocardial infarction and stroke in the gen- Retinopathy present 393 (47) 366 (45) eral population [16–18]. Although microalbuminuria CVD present 109 (10) 103 (10) in diabetes predicts for an increased risk of cardiovas- a Geometric mean (range) cular death [19–21], the relationships between fibrin- Data are mean ± SD or n (%) ogen and microalbuminuria have only been studied in a limited number of subjects [22, 23]. In addition, method using goat anti-human albumin antiserum and human while complex interrelationships between plasma fi- serum albumin standards. Microalbuminuria was defined as brinogen concentration and lifestyle and lipid risk- AER between 20 and 200 mg/min. HbA1c was measured in a factors have been identified in population studies central laboratory by immunoassay using monoclonal anti- HbA1c. Plasma fibrinogen and vWF were also measured cen- [24–27], these surveys included few IDDM patients. trally on citrated plasma prepared from fresh blood obtained The EURODIAB IDDM Complications Study is a by venepuncture. Plasma samples were stored at less than –20 cross-sectional clinic-based study of complications in degrees C° at all times until assay. Fibrinogen was assayed by randomly selected IDDM patients attending 31 clinics a clotting assay based on the prothrombin time, on an Instru- in 16 European countries which was established to mentation Laboratories Series 300(R) Autoanalyser and vWF measure the prevalence and severity of diabetic com- antigen by ELISA, as previously described [28]; a pooled plas- plications and to examine specific relations between ma standard, calibrated against the UK (NIBSAC) standard, was employed. Hidden duplicate samples were included as a putative risk factors [3]. In order to explore the poten- check on reproducibility. Satisfactory performance of these as- tially important interrelationships between a vascular says in UK National Quality Assurance schemes was recorded. risk factor for macrovascular disease, and a marker of Triglyceride, cholesterol and the cholesterol content of HDL endothelial cell damage, and complications in subjects were determined on fasting serum, after manganese and hepa- with IDDM, we have included assays for plasma fi- rin precipitation, by standard enzymatic techniques; LDL brinogen and vWF in the study programme. cholesterol was calculated. Presence of cardiovascular disease (CVD) was assessed by the clinical history of myocardial infarction, angina pectoris, stroke or coronary artery bypass graft surgery, and by Minne- Subjects and methods sota-coding of 12-lead resting ECGs [29]. Sitting blood pres- sure was measured to the nearest 2 mmHg. with a random Details of patient selection have been previously described zero sphygmomanometer after 5 min of rest; the mean of two (Eurodiab IDDM Complications Study Group, 1994). In sum- readings was used for analysis. Retinopathy was assessed from mary, each centre selected a stratified random sample of pa- photographs of two retinal fields per eye graded by a single ob- tients attending the hospital in 1 year. IDDM was defined as server, as previously described [30]. onset before age 36 years with continuous insulin treatment initiated less than 1 year from diagnosis. From the entire study cohort of 3250 patients blood samples for analysis of haemo- Statistical analysis static and lipid risk factors were available on 2091 (64%). The demographic and clinical characteristics are listed in Table 1. To examine the association of fibrinogen and vWF with other The composition of the study group was representative of the risk factors for diabetes complications and also with the diabe- entire cohort, there being no significant difference in any of tes complications themselves,