End-stage renal disease risk equations for Hong Kong Chinese patients with type 2 diabetes: Hong Kong Diabetes Registry.
Yang XL., So WY., Kong APS., Clarke P., Ho CS., Lam CWK., Ng MHL., Lyu RR., Yin DD., Chow CC., Cockram CS., Tong PCY., Chan JCN.
AIMS/HYPOTHESIS: The objective of the study was to investigate risk factors and develop risk equations for end-stage renal disease (ESRD) in Chinese patients with type 2 diabetes. SUBJECTS AND METHODS: A prospective cohort of 4,438 patients with type 2 diabetes mellitus and without ESRD (median observation period 2.9 years, interquartile range 1.6-4.1 years) was included in the analysis. The end-point (ESRD) was defined by: (1) death due to diabetes with renal manifestations or renal failure; (2) hospitalisation due to renal failure; (3) estimated GFR (eGFR) <15 ml min(-1) 1.73 m(-2). Cox proportional hazards regression was used to develop risk equations. The data were randomly and evenly divided into the training data for development of the risk equations and the test data for validation. The validation was performed using the area under the receiver operating characteristic curve (aROC), which takes into account follow-up time and censoring. RESULTS: During the observation period, 159 patients or 12.45 per 1,000 person-years (95% CI 10.52-14.37 per 1,000 person-years) developed ESRD. Known duration of diabetes, systolic blood pressure, log(10) total cholesterol:HDL cholesterol ratio and retinopathy were significant predictors of ESRD. After further adjusting for eGFR, log(10) spot albumin:creatinine ratio (ACR) and haematocrit, only eGFR, haematocrit and log(10) ACR remained as independent predictors of ESRD. The risk equation derived from these three independent predictors had good discrimination, with an aROC of 0.97. CONCLUSIONS/INTERPRETATION: Estimated GFR, haematocrit and ACR were independent predictors of ESRD and the derived risk equation performed well in Chinese patients with type 2 diabetes.