A nomogram for predicting aldosterone-renin ratio in patients with hypertension

Background : It’s necessary to assess activity of renin-angiotensin-aldosterone system(RAAS) in patients with hypertension by means of orthostatic aldosterone-renin ratio(ARR) which is demanding and not available to those primary hospitals. A novel and portable prediction tool is highly desirable to distinguish abnormal ARR in those patients and guide hypertension therapy to some degree. Methods : Clinical characteristics and laboratory data of 1,212 patients with hypertension were collected for modeling and randomly divided into a training cohort (865 of 1,212, 70%) and an internal validation cohort (347 of 1,212, 30%). Then, predictors for ARR were extracted to construct a nomogram model based on regression analysis of the training set. Receiver operating characteristics (ROC), calibration plots and decision curve analyses (DCA) were applied to evaluate the model. Conclusions : Five predictors were adopted to the nomogram including Na-K ratio, gender, serum chloridion(Cl ^- ), estimated glomerular filtration rate(eGFR) and urinary pH. Based on this nomogram, the area under the curve(AUC) was 0.756 (95% CI: 0.71-0.80, p < 0.05) in the training set and 0.725 (95% CI: 0.64-0.81, p < 0.05) in the validation set. The calibration curves exhibited great agreement between the predictive risk of the model and the actual risk and the DCA also showed good clinical benefit. Conclusion : We have firstly developed a novel nomogram to predict abnormal ARR in hypertensive individuals based on routine biochemical variables.