Evaluation of short-term changes in serum creatinine level as a meaningful end point in randomized clinical trials

Abstract

Observational studies have shown that acute change in kidney function (specifically, AKI) is a strong risk factor for poor outcomes. Thus, the outcome of acute change in serum creatinine level, regardless of underlying biology or etiology, is frequently used in clinical trials as both efficacy and safety end points.We performed ameta-analysis of clinical trials to quantify the relationship between positive or negative short-term effects of interventions on change in serum creatinine level and more meaningful clinical outcomes. After a thorough literature search,we included 14 randomized trials of interventions that altered risk for an acute increase in serum creatinine level and had reported between-group differences in CKD and/or mortality rate 3 months after randomization. Seven trials assessed interventions that, compared with placebo, increased risk of acute elevation in serum creatinine level (pooled relative risk, 1.52; 95% confidence interval, 1.22 to 1.89), and seven trials assessed interventions that, compared with placebo, reduced risk of acute elevation in serumcreatinine level (pooled relative risk, 0.57; 95%confidence interval, 0.44 to 0.74). However, pooled risks for CKD andmortality associated with interventions did not differ fromthose with placebo in either group. In conclusion, several interventions that affect risk of acute, mild to moderate, often temporary elevation in serum creatinine level in placebo-controlled randomized trials showed no appreciable effect on CKD or mortalitymonths later, raising questions about the value of using small to moderate changes in serum creatinine level as end points in clinical trials.