#2609 Comparison of baseline patient characteristics of the CONVINCE randomized controlled trial with existing trial and registry populations

Abstract

Abstract Background and Aims The generalizability of randomized controlled trials (RCTs) results is based on representativeness of the population involved. Trial populations may have different baseline characteristics compared to the general population that could benefit from the intervention. We compare the baseline characteristics of patients involved in the CONVINCE trial with those involved in all previous RCTs and representative registries of patients receiving hemodialysis as renal replacement therapy. Method Based on world class methods for systematic reviews, we identified all RCTs comparing hemodiafiltration (HDF) with hemodialysis (HD) published up to January 2024 and extracted data on baseline patient characteristics. Where data were missing, authors have been contacted and data so far obtained were included in this analysis (Fig. 1). Additional contact attempts are being performed. In addition, a comprehensive search was conducted to identify renal registries from all geographic regions. Those without information in English, without information on HDF and registries focusing on special patient groups (e.g. transplantation only) were excluded. This registry data will not be analyzed in this abstract as we are in progress of extracting the patient data from the latest registry annual reports available up to January 2024 for further analysis. Results We have selected 11 RCTs for further analysis, with the selection criteria and outcomes presented in Fig. 1. Key patient characteristics from RCTs were compared with those of patients in the CONVINCE trial using descriptive statistics and forest plots (Fig. 1). Comparison of differences was analyzed with the Kruskal-Wallis-Test. Interpretation of the results are as follows: • Age: Mean age of 76.5 years noted in Morena et al. (2017) because study inclusion criterion was ≥65 years. Other RCTs ranged from 54.1 to 68 years, which is consistent with 62.4 years in the CONVINCE trial (χ2 = 10, df = 10, p = 0.44). • Sex: 16.7% and 50% female noted in Selby et al. (2006) and Meert et al. (2009) respectively probably due to limited sample size (12 patients and 14 patients respectively). Other RCTs ranges from 30.4% to 42.5%, which is consistent with 36.0% in the CONVINCE trial (χ2 = 11, df = 11, p = 0.44). • Diabetes comorbidities: Ranges from 17.8% to 45%. CONVINCE trial is 35.0% (χ2 = 6, df = 6, p = 0.42). • Cardiovascular diseases comorbidities: Range from 26.4% to 45%. CONVINCE trial is 45% (χ2 = 3, df = 3, p = 0.39). • AVF as vascular access: value of 100% in Pedrini et al. (2011) because it was an inclusion criterion. Other RCTs ranged from 79.4% to 90%, which is consistent with 82% in the CONVINCE trial (χ2 = 7, df = 7, p = 0.43). There were no significant differences between CONVINCE and other RCTs data regarding all 5 patient baseline characteristics. Conclusion Our analysis showed the CONVINCE trial has no significant differences in patient baseline characteristics compared to previous RCTs. For a more comprehensive evaluation of representativeness of the trial's population, we are currently retrieving missing data from the trial authors and analyzing data from hemodialysis registries.