Scores may help predict renal risk in type 2 diabetes

Five-year models were useful in predicting renal risk among patients with type 2 diabetes, according to a recent study.

Researchers developed risk models for predicting end-stage renal disease (ESRD) events in type 2 diabetes and used patients in the nationwide New Zealand Diabetes Cohort Study as the derivation cohort. The New Zealand Diabetes Cohort Study included patients with type 2 diabetes, excluding those with ESRD, who were first assessed in 2000 to 2006 and were then followed until December 2010. The main outcome of the New Zealand Diabetes Cohort Study was a fatal or nonfatal ESRD event, that is, peritoneal dialysis or hemodialysis for ESRD, renal transplant, or ESRD death. The researchers developed risk models with Cox proportional hazards models and assessed them in a separate validation cohort, which included 5,877 patients with type 2 diabetes in the Diabetes Care Support Services audit database who had been followed for at least five years. Results were published online June 25 by Diabetes Care.

A total of 25,736 patients were included in the derivation cohort and were followed for up to 11 years; 86% were followed for more than five years. Mean patient age at baseline was 62 years, and the median diabetes duration was five years. Median hemoglobin A1c level and median estimated glomerular filtration rate were 7.2% and 77 mL/min/1.73 m², respectively, and 37% of patients had albuminuria. Six hundred thirty-seven ESRD events occurred during follow-up in the derivation cohort, and 121 renal events occurred during five-year follow-up in the validation cohort. The authors found that models including sex, ethnicity, age, diabetes duration, albuminuria, serum creatinine level, systolic blood pressure, hemoglobin A1c, smoking status, and history of cardiovascular disease did well at predicting renal risk in both the derivation and validation cohorts and that their predictive performance was better than in previous models.

Complete data were not available for analysis in all patients, and data on diet and physical activity were not accessible for any patients, the authors noted. In addition, they said, risk was predicted using only baseline data, and many of the patients at high risk for renal problems died of other causes before having an ESRD event. However, the authors concluded that the models assessed in their study performed well in predicting renal risk among patients with type 2 diabetes treated in primary care. They called for further studies to continue to refine risk assessment in this population. “Being able to enumerate the risk will also be important in treatment pathways, and the use of (externally validated) risk models will be able to be incorporated to ensure [that] escalation of care…can be based on risk rather than clinical factors, (in)equities, or less precise clinical descriptions,” they wrote.