Factors influencing brachial-ankle pulse wave velocity in people undergoing health examination and model evaluation: Report of 4 159 case / 第二军医大学学报

ABSTRACT

Objective To analyze the factors influencing brachial-ankle pulse wave velocity (baPWV) and to establish a logit model for predicting baPWV. Methods The data of 4 159 cases who underwent health examination from 2010 to 2014 in our hospital were retrospectively analyzed. The parameters included gender, age, systolic blood pressure, diastolic blood pressure, pulse, fasting glucose, triglyceride, cholesterol, alanine aminotransferase, glutamyl transpeptidase and cholesterol, low density lipoprotein (LDL), high density lipoprotein (HDL), and uric acid. Logistic regression was used to explore the influencing factors of baPWV, and a regression model was established to predict baPWV and it was evaluated. Results Univariate analysis showed that, except for HDL, all the other parameters above were significantly different between normal baPWV group and abnormal baPWV group (P<0.05). Multivariable analysis yielded the following logit model: logit (p)=-17.888+0.001×uric acid 0.004×alanine aminotransferase 0.105×fasting glucose 0.023×pulse 0.032×diastolic blood pressure 0.061×systolic blood pressure 0.092×age 0.411×sex, which showed a correct predicting rate of 79.6% for baPWV in health examination population, with the ROC area being 0.869 (95%CI:0.859-0.879). Conclusion The baPWV values can be influenced by gender, age, systolic blood pressure, diastolic blood pressure, pulse, fasting blood glucose, alanine aminotransferase and uric acid levels, and logit model may serve as a satisfactory model for these types of study.