Glycemic markers and relation with arterial stiffness in Caucasian subjects of the MARK study

Abstract

Background: Effect of prediabetes and normal glucose on arterial stiffness remains controversial. The primary aim of this study was to investigate the relationship of fasting plasma glucose (FPG), postprandial glucose (PG) and glycosylated haemoglobin (HbA1c) with brachial-ankle pulse wave velocity (baPWV) and cardio-ankle vascular index (CAVI) in Caucasian adults. The secondary aim was to analyse this relationship by glycaemic status.

Methods: Cross-sectional study. Setting: Primary care. Participants: 2,233 subjects, 35-74 years. Measures: FPG (mg/dL) and HbA1c (%) of all subjects were measured using standard automated enzymatic methods. PG (mg/dL) was self-measured at home two hours after meals (breakfast, lunch and dinner) for one day using an Accu-chek ® glucometer. CAVI was measured using a VaSera VS-1500® device (Fukuda Denshi), and baPWV was calculated using a validated equation.

Results: CAVI and baPWV values were significantly higher in subjects with diabetes mellitus than in glucose normal and prediabetes groups (p<0.001). FPG, PG and HbA1c were positively associated with CAVI and baPWV. The β regression coefficient for: HbA1c was 0.112 (CI 95% 0.068 to 0.155) with CAVI, 0.266 (CI 95% 0.172 to 0.359) with baPWV; for PG was 0.006 (CI 95% 0.004 to 0.009 and for FPG was 0.005 (CI 95% 0.002 to 0.008) with baPWV; and for PG was 0.002 (CI 95% 0.001 to 0.003) and 0.003 (CI 95% 0.002 to 0.004) with CAVI (p<0.01 in all cases). When analysing by hyperglycaemic status, FPG, PG and HbA1c were positively associated with CAVI and baPWV in subjects with type 2 diabetes mellitus.

Conclusion: FPG, PG and HbA1c show a positive association with CAVI and baPWV, in Caucasian adults with intermediate cardiovascular risk factors. When analysing by hyperglycaemic status, the association is only maintained in subjects with type 2 diabetes mellitus.

Trial registration: Clinical Trials.gov Identifier: NCT01428934. Registered 2 September 2011. Retrospectively registered. Last updated September 8, 2016.

Fig 1. Flow chart of this MARK substudy.

N, number; CAVI, cardio-ankle vascular index; baPWV, brachial-ankle pulse wave velocity; ABI, ankle-brachial index; HbA1c, glycosylated hemoglobin.

Fig 2. Estimated averages and standard deviations of CAVI (a) and baPWV (b) in diabetic subjects, prediabetic subjects and subjects with normal blood glucose.

Adjusted for Age (years). Gender (0 = male and 1 = female). Smoking (0 = Not and 1 = Yes). Body mass index. Mean arterial pressure. Atherogenic index. Alcohol consumption. Antihypertensive drugs (0 = Not and 1 = Yes). Antidiabetic drugs and lipid lowering drugs (0 = Not and 1 = Yes). Differences among groups: analysis of variance and post hoc using the Bonferroni tests. CAVI cardio-ankle vascular index. baPWV brachial-ankle pulse wave velocity.

Fig 3. Logistic regression analysis, OR of HbA1c with CAVI (a), of HbA1c with baPWV (b), of PG with CAVI (c), of PG with baPWV (d), of FPG with CAVI (e) and of FPG with baPWV (f).

Adjusted for Age (years). Gender (0 = male and 1 = female). Smoking (0 = Not and 1 = Yes). Body mass index. Mean arterial pressure. Atherogenic index. Alcohol consumption. Antihypertensive drugs (0 = Not and 1 = Yes). Antidiabetic drugs and lipid lowering drugs (0 = Not and 1 = Yes). CAVI cardio-ankle vascular index. baPWV brachial-ankle pulse wave velocity. OR odds ratio. CI confidence interval. HbA1c glycosylated hemoglobin. PG postprandial glucose. FPG fasting plasma glucose.