OBJECTIVE: To our knowledge, few studies compared the association of brachial-ankle pulse wave velocity (baPWV) with cardiovascular risk factors among Chinese with type-2 diabetes mellitus and non-diabetic controls. This study addresses this issue. METHODS: We measured baPWV (OMRON VP1000) in 413 diabetic patients from Shanghai city (mean age: 58.7 years; 57% women) and 354 controls randomly recruited from the population of JingNing county, Zhejiang Province. We used stepwise multiple regression to identify covariables of baPWV and introduced interaction terms in the models to compare slopes. We expressed association sizes for continuous variables for a 1-SD increase in the dependent variable. RESULTS: Adjusted baPWV was higher in diabetic patients than controls (1678 vs 1583 cm/sec; P= 0.018). In diabetic patients, baPWV was independently correlated with female sex (-61 cm/sec; P = 0.061), age (+107 cm/sec; P < 0.0001), height (-51 cm/sec; P = 0.012), systolic pressure (+99 cm/sec; P < 0.0001), and the HDL-to-total cholesterol ratio (-38 cm/sec; P = 0.0004). In controls, the explanatory variables were female sex (-74 cm/sec; P = 0.045), age (+138 cm/sec; P < 0.0001), height (-262 cm/sec; P < 0.0001), systolic pressure (+202 cm/sec; P < 0.0001), but not the HDL-to-total cholesterol ratio (P = 0.48). Explained variance of baPWV was 34% and 61% in diabetic patients and controls, respectively. The associations of baPWV with age, height and systolic pressure were steeper (P < 0.04) in controls than diabetic patients. CONCLUSION: Sex, age, height and systolic pressure were the main determinants of baPWV in Chinese, irrespective of whether they had diabetes or not, but these associations were tighter in population-based controls without diabetes.
OBJECTIVE: To our knowledge, few studies compared the association of brachial-ankle pulse wave velocity (baPWV) with cardiovascular risk factors among Chinese with type-2 diabetes mellitus and non-diabetic controls. This study addresses this issue. METHODS: We measured baPWV (OMRON VP1000) in 413 diabeticpatients from Shanghai city (mean age: 58.7 years; 57% women) and 354 controls randomly recruited from the population of JingNing county, Zhejiang Province. We used stepwise multiple regression to identify covariables of baPWV and introduced interaction terms in the models to compare slopes. We expressed association sizes for continuous variables for a 1-SD increase in the dependent variable. RESULTS: Adjusted baPWV was higher in diabeticpatients than controls (1678 vs 1583 cm/sec; P= 0.018). In diabeticpatients, baPWV was independently correlated with female sex (-61 cm/sec; P = 0.061), age (+107 cm/sec; P < 0.0001), height (-51 cm/sec; P = 0.012), systolic pressure (+99 cm/sec; P < 0.0001), and the HDL-to-total cholesterol ratio (-38 cm/sec; P = 0.0004). In controls, the explanatory variables were female sex (-74 cm/sec; P = 0.045), age (+138 cm/sec; P < 0.0001), height (-262 cm/sec; P < 0.0001), systolic pressure (+202 cm/sec; P < 0.0001), but not the HDL-to-total cholesterol ratio (P = 0.48). Explained variance of baPWV was 34% and 61% in diabeticpatients and controls, respectively. The associations of baPWV with age, height and systolic pressure were steeper (P < 0.04) in controls than diabeticpatients. CONCLUSION: Sex, age, height and systolic pressure were the main determinants of baPWV in Chinese, irrespective of whether they had diabetes or not, but these associations were tighter in population-based controls without diabetes.