Development and validation of a nomogram for arterial stiffness

Abstract

Even though as a gold standard for noninvasive measurement of arterial stiffness, carotid‐femoral pulse wave velocity (cfPWV) is not widely used in primary healthcare institutions due to time‐consuming and unavailable equipment. The aim of this study was to develop a convenient and low‐cost nomogram model for arterial stiffness screening. A cross‐sectional study was undertaken in the department of general practice, the First Affiliated Hospital of Fujian Medical University. Arterial stiffness was defined as cfPWV ≥ 10 m/s. A total of 2717 participants were recruited to construct the nomogram using the least absolute shrinkage and selection operator and logistic regressions. Receiver operating characteristic (ROC) curve, calibration curve, decision curve analysis, clinical impact curve were used to evaluate the performance of the model. The model was validated internally and externally (399 participants) by bootstrap method. Arterial stiffness was identified in 913 participants (33.60%). Age, sex, waist to hip ratio, systolic blood pressure, duration of diabetes, heart rate were selected to construct the nomogram model. Good discrimination and accuracy were exhibited with area under curve of 0.820 (95% CI 0.803–0.837) in ROC curve and mean absolute error = 0.005 in calibration curve. A positive net benefit was shown in decision curve analysis and clinical impact curve. A satisfactory agreement was displayed in internal validation and external validation. The low cost and user‐friendly nomogram is suitable for arterial stiffness screening in primary healthcare institutions.