Development of a Simple Risk Score Model to Predict Renal Artery Stenosis

Background/Aims: Renal artery stenosis (RAS) is a known cause of secondary hypertension and renal failure. Although renal artery angiography is the gold standard for diagnosing RAS, a simple method to estimate if patients will develop RAS is required. The aim of this retrospective study was to develop a simple risk score to predict significant RAS. Methods: Four thousand one hundred seventy-seven patients who underwent renal angiography between 2002 and 2016 at Tehran Heart Center were included. Significant RAS was defined as narrowing of the renal artery by at least 70%. Multiple predictors of the RAS were determined using multivariable logistic regression with a backward elimination method. The scoring system obtained from the final model was presented as nomogram. The possible nonlinear effect of continuous variables was evaluated using restricted cubic splines. Overfitting of the final model was assessed applying the tenfold cross-validation method. Model performance was checked using calibration plot as well as Hosmer-Lemeshow goodness of fit test, and area under the receiver operating characteristics (ROC) curve. Results: The prevalence of RAS was 14.1%. Female sex (OR [95% CI]: 1.53 [1.26–1.85]), hypertension (OR [95% CI]: 1.38 [1.08–1.77]), estimated glomerular filtration rate (OR [95% CI]: 0.98 [0.97–0.98]), body mass index (OR [95% CI]: 0.97 [0.95–0.99]), and age (OR [95% CI]: 1.01 [1.00–1.02]) were determined as the multiple predictors of RAS. The area under the ROC curve of the final predictive model was 0.702 (95% CI: 0.679–0.725). Conclusion: This model assesses the risk of RAS using available information. This model can be used for both clinical and research purposes.