Impact of Outcome Adjudication in Kidney Disease Trials: Observations From the Study of Heart and Renal Protection

INTRODUCTION: We aimed to assess opportunities for trial streamlining and the scientific impact of adjudication on kidney and cardiovascular (CV) outcomes in chronic kidney disease (CKD). METHODS: We analyzed the effects of adjudication of approximately 2100 maintenance kidney replacement therapy (KRT) and approximately 1300 major atherosclerotic events (MAEs) recorded in the Study of Heart and Renal Protection (SHARP). We first compared outcome classification before adjudication versus after adjudication, and then reran randomized comparisons using preadjudicated follow-up data. RESULTS: For maintenance KRT, adjudication had little impact with only 1% of events being refuted (28/2115). Consequently, randomized comparisons using preadjudication reports found almost identical results (preadjudication: simvastatin/ezetimibe 1038 vs. placebo 1077; rate ratio [RR] 0.95, 95% confidence interval [CI] 0.88–1.04; postadjudicated: 1057 vs. 1084; RR = 0.97, 95% CI 0.89–1.05). For MAEs, about one-quarter of patient reports were refuted (324/1275 [25%]); and reviewing 3538 other potential vascular events and death reports identified only 194 additional MAEs. Nevertheless, randomized analyses using SHARP’s preadjudicated data alone found similar results to analyses based on adjudicated outcomes (preadjudication: 573 vs. 702; RR = 0.80, 95% CI 0.72–0.89; adjudicated: 526 vs. 619; RR = 0.83, 95% CI 0.74–0.94); and also suggested that refuted MAEs were likely to represent atherosclerotic disease (RR for refuted MAEs = 0.80, 95% CI 0.65–1.00). CONCLUSIONS: These analyses provide 3 key insights. First, they provide a rationale for nephrology trials not to adjudicate maintenance KRT. Second, when an event that mimics an atherosclerotic outcome is not expected to be influenced by the treatment under study (e.g., heart failure), the aim of adjudicating atherosclerotic outcomes should be to remove such events. Lastly, restrictive definitions for the remaining suspected atherosclerotic outcomes may reduce statistical power.