A risk prediction model for renal damage in a hypertensive Chinese Han population

Backgroud: While numerous risk factors for renal damage in the hypertensive population have been reported, there is no single prediction model. The purpose of this study was to develop a model to comprehensively evaluate renal damage risk among hypertensive patients. Methods: We analyzed the data of 582 Chinese hypertensive patients from 1 January 2013 to 30 June 2016. Basic patient information was collected along with laboratory test results. According to the albumin-to-creatinine ratio, the subjects were divided into a hypertension with renal damage group and a hypertension without renal damage group. The prediction model was established by logistic regression based on principal component analysis, and the area under the receiver operating characteristic curve was used to evaluate the predictive performance of the model.Results: There are 11 indicators have statistically significant difference between the two groups (P < 0.05); The equation expressed including all 11 risk factors was as follows: Y = (-0.236) - 0.1705 (sex) - 0.0098 (age) - 0.1067 (smoking history) + 0.0303 (drinking history) - 0.3031 (CHD) + 0.1276 (diabetes history) - 0.0596 (CRP level) - 0.0732 (CysC level) + 0.0949 (β2-MG level) + 0.5407 (blood pressure type) + 0.6470 (RRI). The calculated AUC was 74.4%; The risk in males was much higher than that in females of the same age. However, with increasing age, the male:female risk ratio gradually decreased. Conclusion: Eleven indicators (including sex, age, smoking history, drinking history, coronary heart disease, diabetes history, C-reactive protein, CystatinC, β2-microglobulin protein, blood pressure type, renal artery resistance index) may be the risk factors of renal damage in hypertension. Our regression equation provides a feasible means of predicting renal damage in Chinese hypertensive populations, and the model showed good predictive power. In addition, estrogen may confer a protective effect on the kidney. Abbreviations: PCA: principal component analysis; SLPs: synthetic latent predictors; CKD: chronic kidney disease; RRI: renal artery resistance index; MLR: multivariate logistic regression; CHD: coronary heart disease; UACR: urine trace albumin/uric creatinine ratio; CysC: CystatinC; TG: Triglyceride; CHO: cholesterol; HDL: high-density lipoprotein cholesterol; LDL: low-density lipoprotein cholesterol; CRP: C-reactive protein; HCY: homocysteine; UA: uric acid; AUC: area under the ROC curve; CVE: cardiovascular events; RFF: renal function related factor; PHF: personal history related factor; CVF: cardiovascular factor; GMF: glucose metabolism factor; IF: inflammatory factor; BPF: blood pressure factor.