For their research, the team led by Rainer Oberbauer, Head of the Division of Nephrology and Dialysis at MedUni Vienna's Department of Medicine III, and Mariella Gregorich from MedUni Vienna's Center for Medical Data Science drew on data from major international studies. This made it possible to include 13 routinely collected baseline values from 4,637 18- to 75-year-olds with type 2 diabetes with slightly to moderately impaired kidney function. In addition to the most important value for assessing kidney function (estimated glomerular filtration rate, eGFR), age, gender, body mass index, smoking behaviour, haemoglobin and cholesterol levels as well as medication intake were selected as predictors. On this basis, the research team developed a predictive model based on tested statistical methods, which is already being prepared for clinical use.
"The strength of our study compared to previous research on the topic lies not only in the refined methodology, but also in the large amount of data. This allows us to attain a high level of confidence in our findings," says first author Mariella Gregorich. "Accordingly, the prediction tool proves to be reliable and able to predict a decline in kidney function based on eGFR for up to five years after baseline." However, the study has also revealed that the individual course depends on other, still unknown factors.
Early recognition and therapy management
Chronic kidney disease (CKD) is one of the most common complications of diabetes and the most common cause of kidney failure that requires dialysis. Since CKD does not have any symptoms in the early stages, it is often recognised only when the decline in kidney function is already very advanced. Through early detection and consistent therapy management, especially in diabetic metabolic and blood pressure control, kidney damage can be delayed or prevented. Currently, kidney function in persons with diabetes is mainly monitored by the regular measurement of eGFR. "Our prediction tool can assist in the continuous monitoring of disease progression and allow the identification of patients with an increased risk of worsening kidney function in the coming years," says study leader Rainer Oberbauer, highlighting the significant clinical relevance of the prediction tool. A web-adapted version of the model is already under construction and will soon be available for further, independent validation: https://beatdkd.shinyapps.io/shiny/
