BACKGROUND: It is important to know the reliability of early changes in proteinuria in predicting late renal outcomes. The IDNT was a trial in which treatment assignment, baseline and follow-up blood pressure determinations, and albumin/creatinine ratios (ACR), and renal outcomes were recorded. METHODS:Risk of renal outcomes in the IDNT was assessed by proportional hazards modeling as a function of treatment assignment, and achieved systolic blood pressure (SBP) both without, and then with, inclusion of values for baseline proteinuria and early changes in proteinuria. RESULTS: In models without ACR variables, both treatment with irbesartan and achieved SBP during follow-up were significantly predictive of the risk of renal outcomes. Addition of ACR variables to the models reduced the apparent impact of assignment to irbesartan by 52% to 81%, and irbesartan was no longer a significant predictor of renal outcomes. Conversely, addition of ACR variables to the models attenuated the effect of achieved follow-up SBP by only 32% to 46%, and follow-up BP remained a highly significant predictor of renal outcomes. CONCLUSION: The ability of early changes in proteinuria to predict the impact of treatment on renal outcomes is a function of the specific treatment. One must use caution in using early changes in proteinuria as a surrogate for longer-term renal outcomes.
RCT Entities:
BACKGROUND: It is important to know the reliability of early changes in proteinuria in predicting late renal outcomes. The IDNT was a trial in which treatment assignment, baseline and follow-up blood pressure determinations, and albumin/creatinine ratios (ACR), and renal outcomes were recorded. METHODS: Risk of renal outcomes in the IDNT was assessed by proportional hazards modeling as a function of treatment assignment, and achieved systolic blood pressure (SBP) both without, and then with, inclusion of values for baseline proteinuria and early changes in proteinuria. RESULTS: In models without ACR variables, both treatment with irbesartan and achieved SBP during follow-up were significantly predictive of the risk of renal outcomes. Addition of ACR variables to the models reduced the apparent impact of assignment to irbesartan by 52% to 81%, and irbesartan was no longer a significant predictor of renal outcomes. Conversely, addition of ACR variables to the models attenuated the effect of achieved follow-up SBP by only 32% to 46%, and follow-up BP remained a highly significant predictor of renal outcomes. CONCLUSION: The ability of early changes in proteinuria to predict the impact of treatment on renal outcomes is a function of the specific treatment. One must use caution in using early changes in proteinuria as a surrogate for longer-term renal outcomes.
Authors: Lesley A Inker; Andrew S Levey; Kruti Pandya; Nicholas Stoycheff; Aghogho Okparavero; Tom Greene Journal: Am J Kidney Dis Date: 2014-04-29 Impact factor: 8.860