Abstract: PO0454
Indexing Proteinuria to Renal Function Improves Prediction for Renal Events in Individuals with CKD
Session Information
- CKD Epidemiology, Biomarkers, Predictors
October 22, 2020 | Location: On-Demand
Abstract Time: 10:00 AM - 12:00 PM
Category: CKD (Non-Dialysis)
- 2101 CKD (Non-Dialysis): Epidemiology, Risk Factors, and Prevention
Authors
- Parsa, Afshin, Division of Kidney, Urologic, and Hematologic Diseases, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Behtesda, Maryland, United States
- Schulman, Ivonne Hernandez, Division of Kidney, Urologic, and Hematologic Diseases, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Behtesda, Maryland, United States
- Wilkins, Kenneth J., Division of Kidney, Urologic, and Hematologic Diseases, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Behtesda, Maryland, United States
- Mendley, Susan R., Division of Kidney, Urologic, and Hematologic Diseases, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Behtesda, Maryland, United States
- Kimmel, Paul L., Division of Kidney, Urologic, and Hematologic Diseases, National Institute of Diabetes and Digestive and Kidney Diseases, NIH, Behtesda, Maryland, United States
- Becerra, Adan Z., Social & Scientific Systems, Siver Spring, Maryland, United States
Background
Identifying the optimal measurement of proteinuria in clinical settings has been challenging. To determine the potential consequence of varied measures of proteinuria, we contrasted their clinical significance primarily in relation to renal events and secondarily to cardiovascular (CVD) and mortality events.
Methods
We compared the predictive ability of four measures of proteinuria and albuminuria, among 3592 CKD participants from the Chronic Renal Insufficiency Cohort (CRIC) Study, for incident renal events (halving of glomerular filtration rate [GFR] or end-stage renal disease), CVD events (myocardial infarction, congestive heart failure, stroke and peripheral arterial disease) and mortality over 3 years. The four measures included timed urinary albumin and protein excretion rates (AER, PER), albumin/protein: creatinine ratios (ACR, PCR), albumin/protein: adjusted creatinine (accounting for creatinine production) ratios (eAER, ePER), and lastly albuminuria/proteinuria indexed to GFR (ACR-G, PCR-G), as an estimation of glomerular permeability. We used Harrell’s C-Statistics to measure model discrimination.
Results
Predictive performance for renal events was lowest for AER and PER. Results were generally similar for ACR vs eAER and PCR vs ePER. Notably, PCR-G showed significant improvement in predicting renal outcomes and performed better than albuminuria-based measures. C-statistics for renal events were 0.831, 0.840, 0.841, 0.846 and 0.862 for AER, eAER, ACR, PCR and PCR-G, respectively. Trends were similar for CVD and mortality events, except that ACR performed better than PCR for CVD events, but not as well as PCR-G or ACR-G. Results were overall consistent across diabetes, gender and race strata, and were validated in an additional 1443 participants from the third phase of CRIC.
Conclusion
Indexing proteinuria to GFR is a simple and economical measure, compared to albuminuria, that significantly enhances the prediction of CKD progression and associated outcomes.
Funding
- NIDDK Support