Healthy first-degree relatives of patients with Type 2 diabetes are at increased risk of Type 2 diabetes and cardiovascular disease (CVD). Alterations in plasma levels of inflammatory and thrombotic markers are associated with both Type 2 diabetes and CVD. Increased clot density with prolonged lysis times are associated with increased cardiovascular risk.[1–6] The aim of the present study was to investigate whether complement C3, C-reactive protein (CRP) and fibrin clot characteristics are altered in relatives of patients with Type 2 diabetes and are associated with a family history of diabetes.
Patients with Type 2 diabetes, their first-degree relatives and healthy control subjects, age- and sex-matched to relatives, were recruited as described previously [7,8]. All subjects were White European and gave informed consent according to protocols approved by the Leeds Teaching Hospitals Trust Research Ethics Committee. Metabolic and haemostatic factors were determined as described previously [7,8]. The relatives underwent an oral glucose tolerance test to exclude the possibility of undiagnosed diabetes.
Plasma levels of complement C3 and CRP were both measured by in-house ELISAs  (inter-assay coefficients of variance were 7.7% and 2.8% for the C3 and CRP ELISAs, respectively). Clot formation and lysis were measured by turbidimetric assays as described previously  (inter-assay coefficients of variance were 3.7% for lag time, 1.9% for maximum absorbance, and 3.3% for lysis time).
Continuous variables were tested for normal distribution and analysed accordingly (normally distributed variables: one-way anova with post-hoc Scheffé analysis for pairwise comparison; not normally distributed variables: Kruskal–Wallis with Mann–Whitney U-test for pairwise comparison, with Bonferroni adjustment).
The contribution of demographic, metabolic and hemostatic variables to group differences in inflammatory and turbidimetric variables was estimated by univariate general linear regression analyses. Independent associations with a family history of Type 2 diabetes were investigated by stepwise logistic regression analysis including the variables body mass index (BMI), glucose, HbA1c, triglycerides, total cholesterol, high-density lipoprotein (HDL), fibrinogen, plasminogen activator inhibitor-1 (PAI-1), C3, CRP and lysis time. All analyses were carried out using spss v12.0 (SPSS Inc., Chicago, IL, USA).
The characteristics of first-degree relatives, healthy controls and Type 2 diabetes patients are shown in Table 1. The patients were older, but relatives and controls were age matched. The relatives showed intermediate levels between controls and patients for most parameters, with differences maintained after adjustment for age and gender (data not shown).
|Controls n = 60||Relatives n = 60||Patients n = 60||P|
|Sex (m:f) (%m)||27:33 (45%)||27:33 (45%)||29:31(48%)||0.946a)|
|Age (years)||43.6 (40.9–46.2)||43.4* (40.7–46.1)||62.9† (60.2–65.6)||< 0.001b)|
|BMI (kg m−2)||25.2 (24.1–26.3)||27.0* (25.9–28.2)||30.5† (29.1–31.9)||< 0.001b)|
|Hypertensive‡ (%)||2 (3%)||5* (8%)||32† (53%)||< 0.001c)|
|Glucose (mmol L−1)||5.1 (4.9–5.4)||5.2* (4.8–5.5)||9.9† (9.1–10.8)||< 0.001d)|
|HbA1c (%)||4.8 (4.6–5.0)||5.0*† (4.8–5.2)||7.3† (6.9–7.6)||< 0.001)d|
|Insulin (μU mL−1)||6.6 (5.5–7.8)||9.6*† (8.5–10.8)||21.0† (17.9–24.6)||< 0.001b)|
|HOMA||1.0 (0.8–1.1)||1.4*† (1.2–1.6)||3.5† (3.0–4.0)||< 0.001d)|
|Cholesterol (mmol L−1)||5.3 (5.1–5.6)||5.9† (5.6–6.1)||6.1† (5.8–6.3)||< 0.001b)|
|Triglycerides (mmol L−1)||1.2 (1.0–1.3)||1.4* (1.3–1.6)||2.2† (2.0–2.5)||< 0.001b)|
|HDL (mmol L−1)||1.4 (1.3–1.5)||1.3*† (1.1–1.4)||1.0† (1.0–1.1)||< 0.001b)|
|Fibrinogen (g L−1)||2.7 (2.5–2.8)||3.0† (2.9–3.2)||3.2† (3.0–3.4)||< 0.001b)|
|FXIII (%)||110.8 (105.2–116.7)||105.3 (95.9–115.6)||115.6 (108.5–123.1)||0.207b)|
|PAI-1 (ng mL−1)||6.7 (5.2–8.5)||13.0*† (10.4–16.2)||24.6† (20.0–30.3)||< 0.001b)|
|C3 (mg mL−1)||1.06 (1.00–1.13)||1.23† (1.17–1.30)||1.34† (1.28–1.41)||< 0.001b)|
|CRP (μg mL−1)||0.60 (0.44–0.84)||1.07† (0.75–1.52)||1.81† (1.41–2.32)||< 0.001b)|
|LagC (s)||479.1 (460.5–497.6)||478.3 (453.2–503.3)||504.0 (484.5–523.5)||0.165b)|
|MaxAbsC (au)||0.28 (0.26–0.30)||0.31 (0.29–0.33)||0.32 (0.29–0.34)||0.049b)|
|Lys50t0 (s)||1606 (1559–1654)||1750† (1687–1816)||1840† (1772–1910)||< 0.001b)|
|AUC||171.7 (151.3–194.9)||202.9* (178.6–230.5)||261.4† (234.2–291.7)||< 0.001b)|
C3 and CRP (Table 1) were significantly higher in relatives and patients compared to controls, however, C3 and CRP were not significantly different between relatives and patients. There were no significant differences in clot lag time (LagC) between the three groups, whereas significant trends across the groups were found for clot density (MaxAbsC), lysis time and area under the clot formation/lysis curve (AUC) (Table 1), with the differences remaining significant after adjustment for age and gender (data not shown).
C3, CRP and lysis time significantly correlated with metabolic and hemostatic parameters (correlation coefficients ≥ 0.4) as follows: C3 correlated with BMI (controls, relatives and patients), insulin (relatives and patients), insulin resistance estimated by homeostasis model assessment (HOMA) (relatives), triglycerides (controls and patients), fibrinogen and lysis time (controls), and with CRP (in all three groups). CRP correlated with BMI (controls), insulin (patients) and fibrinogen (controls and relatives). Lysis time correlated with fibrinogen, C3 (controls) and PAI-1 (relatives). In linear regression analyses for C3 and CRP, including the variables age, gender, BMI, HOMA, cholesterol, triglycerides and HDL, independent predictors of C3 were BMI, insulin resistance (HOMA) and triglycerides accounting for 21%, 9.7%, and 3.2%, respectively, of variance. After accounting for these variables the group differences in C3 remained significant (P = 0.011). Independent predictors of CRP were BMI, HOMA and triglycerides accounting for 16%, 5.8% and 4.5%, respectively, of variance. These variables accounted for the observed between-group differences in CRP. In linear regression analysis for lysis time, including the variables age, gender, fibrinogen, factor (F)XIII, PAI-1, C3, CRP, BMI, HOMA, cholesterol, triglycerides, and HDL, independent predictors of lysis time were fibrinogen, FXIII, PAI-1, C3 and total cholesterol accounting for 6.7%, 5.8%, 5.1%, 1.5% and 2.0%, respectively, of variance. These variables accounted for the between-group differences in lysis time. In a stepwise logistic regression model comparing relatives and controls, including BMI, glucose, HbA1c, triglycerides, total cholesterol, HDL, fibrinogen, PAI-1, C3, CRP and lysis time, only PAI-1 (odds ratio for a 1SD increase: 1.81 (95% CI 1.17–2.80), P = 0.008) and C3 [odds ratio for a 1SD increase: 1.75 (1.13–2.71), P = 0.012] were independently associated with a family history of Type 2 diabetes.
This is the first study to investigate fibrin clot formation and lysis in first-degree relatives of Type 2 diabetes patients. Similar to diabetes patients, hemostatic and metabolic factors clustered in relatives and contributed to an adverse fibrin phenotype, characterized by prolonged fibrinolysis times. This may in part explain the increased cardiovascular risk in relatives. Consistent with our previous results , we found that fibrinogen, FXIII and PAI-1 were significant determinants of lysis time, accounting for ∼ 17% of its variance. Interestingly, C3 was also an independent predictor of lysis time. We have recently identified C3 as a component of plasma clots and shown a direct influence of C3 on fibrin clot structure/function providing evidence for a functional link between elevated C3 levels and prolonged fibrinolysis . These results therefore suggest that a proinflammatory phenotype in first-degree relatives may contribute to thrombotic risk.
C3 and CRP were elevated in relatives and Type 2 diabetes patients and clustered with metabolic and hemostatic cardiovascular risk factors, which accounted for the increased CRP levels in the relatives. Metabolic and hemostatic cardiovascular risk factors did not fully account for elevated C3 in relatives and patients. In addition, elevated C3 was independently associated with a family history of diabetes. These data suggest that C3 may be more specifically related to the inflammatory processes contributing to the development of Type 2 diabetes and CVD than CRP, which appears to reflect non-specific inflammatory processes in the present study. The results are consistent with our previous studies in which C3, but not CRP, was independently associated with CVD after accounting for conventional cardiovascular risk factors [5,9]. Our results contradict those of Kriketos et al. who reported no difference in C3 levels between relatives of Type 2 diabetes patients and controls . This discrepancy most likely reflects the smaller group sizes (19 relatives and 22 controls) resulting in a lack of power to detect significant difference, despite a similar trend in that study.
A potential limitation of the present study is the disparity in the ages of the patients compared with the relatives and matched controls. However, the similarity in inflammatory and hemostatic risk factor profiles in relatives and patients despite the disparity in age emphasises the importance of these observations which suggest abnormalities in inflammation and hemostasis predate the development of diabetes, perhaps at the time when insulin resistance achieves maximal levels, but compensatory hyperinsulinemia maintains euglycemia prior to the development of overt diabetes.
In conclusion, healthy first-degree relatives of Type 2 diabetes patients showed a proinflammatory and hypofibrinolytic phenotype similar to that seen in their older relatives with diabetes. Moreover, complement C3 was both related to fibrinolytic activity and independently associated with a family history of Type 2 diabetes. This proinflammatory hypofibrinolytic phenotype may contribute to the increased risk of Type 2 diabetes and CVD in relatives and these data lend further support for exploring C3 as a specific marker for disease development.