Abstract: 【Abstract】 Objective   To develop a nomogram for predicting arteriovenous fistula (AVF) maturity in hemodialysis patients. Methods   A total of 238 hemodialysis patients treated in the period from January 1, 2016 to January 1, 2018 were enrolled in this study as the model group, and a total of 80 uremic patients treated in the period from January 1, 2018 to July 8, 2018 were recruited as the validation group. Baseline characteristics and ultrasound parameters were collected for analyses. We performed logistic regression analysis to identify the independent risk factors for AVF maturity, and then constructed a nomogram according to a regression equation to predict the results. Concordance index (C-index) and calibration curve were used to verify the predictive and discriminative abilities of the nomogram. Results A total of 176 patients (74.8%) achieved AVF maturation during the 8- week study period. Multivariate logistic regression showed that the risk factors for AVF maturity included systolic pressure (OR=0.987, 95% CI: 0.975～0.999, P=0.032), serum phosphate (OR = 0.012, 95% CI: 0.312～0.866, P=0.012), serum cholesterol (OR=0.694, 95% CI: 0.577～0.835,P＜0.001), diabetes (OR=0.226, 95% CI: 0.107～0.477, P＜0.001) and hypertension (OR=0.121, 95% CI: 0.059～0.249, P＜0.001); and that the protective factors significantly associated with the maturity included cephalic vein diameter (OR=7.301, 95% CI: 3.369～15.821, P＜0.001), radial artery diameter (OR=3.225, 95% CI=1.441～7.218, P=0.004), radial artery blood flow (OR=1.086, 95% CI: 1.058～1.115, P＜0.001) and cephalic vein diameter after pressure (OR=2.955, 95% CI: 1.747～4.998, P＜0.001). A nomogram, which included the variances of cephalic vein diameter (β =1.700, P＜0.001), radial artery blood flow (β =0.054, P＜ 0.001), serum cholesterol (β=-0.326, P=0.034), diabetes (β=-1.183, P=0.017) and hypertension (β=-1.624, P= 0.002), was constructed. This nomogram has better discrimination and calibration abilities to predict AVF maturation. Receiver operator curves (ROC) indicated that the area under the curve (AUC) of the nomogram model was higher than that of physical and routine color Doppler ultrasound examinations for predicting AVF maturity among 80 patients. Conclusion The nomogram established in this study is better than preoperative routine vascular ultrasound and physical examinations for predicting AVF maturity. This nomogram provides a simple and direct method for personalized prediction of AVF maturity.

Key words: uremia, arteriovenous fistulas, maturity, nomogram