Type 2 Diabetes: Validating a Renal Risk Score

MedicalResearch.com Interview with: C Raina Elley

C Raina Elley  Associate Professor and General Practitioner, Acting Head, Dept General Practice & Primary Health Care, Faculty Medical and Health Sciences, University of Auckland, Private Bag 92019, Auckland, New ZealandAssociate Professor and General Practitioner,
Acting Head, Dept General Practice & Primary Health Care,
Faculty Medical and Health Sciences,
University of Auckland, Private Bag 92019,
Auckland, New Zealand

MedicalResearch.com: What are the main findings of the study?

Answer: Type 2 Diabetes is the leading cause of end-stage renal failure and dialysis in many countries. Early identification of those who are at risk within primary care could prompt more intensive intervention to control glycaemia and blood pressure and use of ACE inhibitors or angiotensin receptor blockers to slow progression. Traditionally estimated glomerular filtration rate and/or urine albumin creatinine ratio have been used to alert clinicians of deteriorating renal function in people with diabetes. However, a far more accurate renal risk score has been developed that combines serum creatinine, demographic characteristics, albuminuria, glycaemia, blood pressure, cardiovascular co-morbidity and duration of diabetes.

The 5-year renal risk score was developed by following more than 25,000 people with type 2 diabetes in New Zealand for a median of 7.3 years (equivalent to 180,497 person-years). The study identified those who commenced dialysis for end-stage renal disease, received a renal transplant or died from renal failure to derive the risk score.

MedicalResearch.com: Were any of the findings unexpected?

Answer: The accuracy with which the risk score predicted renal events in a separate validation cohort was unexpected, with areas under the ROC of 0.91-0.92, compared with eGFR (0.77). People with the same eGFR (e.g. a 55 year old man with an eGFR of 58ml/min/1.73m2) could have a 5 year risk of an end stage renal event of anywhere between 0.4% and 28%.

The significant contribution of ethnicity to risk (after controlling for other known risk factors) was also surprising. There are possibly genetic components contributing to risk, but ethnicity may also be a marker of other risk factors, such as long-term exposure to poverty, smoke or other differences in lifestyle risk factors.

MedicalResearch.com: What should clinicians and patients take away from your report?

Answer: With electronic medical records, routinely entered measures can be combined to produce risk profiles and prompting early preventive therapy of those at greater risk. Our new renal risk score is being added to the online cardiovascular risk calculator for people with type 2 diabetes http://www.nzssd.org.nz/cvd/ so cardiovascular and renal risk can be assessed together. Risk scores can also be used as an educational tool for patients. Their assessment of 5-year risk or pictorial depiction of risk can highlight an issue and help motivate change or adherence with interventions.

MedicalResearch.com: What recommendations do you have for future research as a result of this study?

Answer: Future research could assess the effectiveness of use of risk prediction and early intervention in randomized controlled trials. Furthermore, with the paucity of interventions to slow progression to renal failure, partly due to the time it takes to reach hard endpoints like dialysis or doubling of serum creatinine in efficacy trials, using significant change in a risk profile that at least combines eGFR and albuminuria could allow more timely assessment of effectiveness of new interventions. Lastly, as the racial or ethnicity profiles vary from country to country, similar studies carried out within national or regional contexts may produce locally valid risk calculators.


Derivation and Validation of a Renal Risk Score for People with Type 2 Diabetes” in Diabetes Care, 2013, doi:10.2337/dc13-0190 by C. Raina Elley, Tom Robinson, Simon A Moyes, Tim Kenealy, John Collins, Elizabeth Robinson, Brandon Orr-Walker, Paul L Drury