Predicting Future eGFR in People With Type 2 Diabetes and CKD

People with type 2 diabetes have higher amounts of sugar in their blood. If type 2 diabetes is left untreated, it can cause other problems over time. For example, type 2 diabetes can affect the kidneys, causing chronic kidney disease (CKD). About 40 percent of people with type 2 diabetes also have CKD.¹

The kidneys are responsible for removing extra water and waste from the blood. In people with CKD, the kidneys do not work as well. Over time, CKD can get worse and lead to kidney failure. This is a serious condition where the kidneys have mostly stopped working. People with kidney failure need dialysis or a kidney transplant to stay healthy.²

CKD is common among people with type 2 diabetes. So, researchers are interested in finding ways to predict how likely it is that a person with type 2 diabetes will develop CKD. If we know who is at a higher risk, we may be able to use different ways to prevent CKD.¹

How is kidney function measured?

Estimated glomerular filtration rate (eGFR) is a measure of how well your kidneys are working. The kidneys use tiny filters called glomeruli. Glomeruli remove waste from the blood. In people with CKD, the glomeruli filter a smaller amount of blood. So, some of the waste remains in the blood.³

Doctors can figure out the eGFR by measuring the amount of waste in the blood. Using a blood test, your doctor will look for a certain waste product, called creatinine. When the kidneys are not working very well, there will be more creatinine in the blood.³

Along with creatinine levels, your age, sex, and race are used to find out the eGFR value. Higher eGFR values are good. Lower eGFR values mean there is kidney damage.³

Can eGFR be predicted?

Being able to predict eGFR values can be very useful in diagnosing CKD. If your doctor can predict your eGFR values for the next few years, that can play an important role in your treatment plan. If your predicted eGFR value is low, your doctor can create a treatment plan that especially focuses on keeping your kidneys healthy.⁴

In 2023, a few researchers came up with a prediction model that can estimate eGFR in people with type 2 diabetes. They created this model by using data from people with both type 2 diabetes and CKD. Using this data, they trained their model to predict eGFR. Then, they checked how well their model worked. Their results show that the model can give a good estimate of eGFR for up to 5 years into the future.¹

What goes into the eGFR prediction model?

The prediction model uses a few inputs to create an output. Inputs are values or information that you give to the model. They are also called predictors. Outputs are what the model returns as the answer, in this case eGFR value.¹

For this model, the researchers used 13 predictors. The predictors were health-related information about the person, such as:^1,5

• Age
• Sex
• Height
• Weight
• Smoking status
• What medicines they take
• Results from certain blood and urine tests

The researchers chose these predictors because they are commonly reported during doctor visits.^1,5

Some researchers are looking into other predictors that might be important. For example, having certain proteins in the blood can be a sign of certain diseases. Finding proteins related to CKD and including them in the model may improve its accuracy.⁶

There are a few other models for predicting the progress of kidney disease. Different models may use different inputs or predictors. Depending on what information is available, 1 model may be favored because it requires less information. These prediction models are just 1 of the tools that may help manage CKD.⁴