Abstract: TH-PO844
Marginal Structural Model (MSM) Analysis to Investigate the Causal Relationship of XOR Inhibitors With Outcomes in Hemodialysis
Session Information
- Health Maintenance, Nutrition, Metabolism
November 03, 2022 | Location: Exhibit Hall, Orange County Convention Center‚ West Building
Abstract Time: 10:00 AM - 12:00 PM
Category: Health Maintenance‚ Nutrition‚ and Metabolism
- 1400 Health Maintenance‚ Nutrition‚ and Metabolism
Authors
- Ishii, Takeo, Yokohama Daiichi Hospital, Yokohama, Kanagawa, Japan
- Taguri, Masataka, Department of Health Data Science, Tokyo Medical University, Shinjuku, Tokyo, Japan
- Wakui, Hiromichi, Department of Medical Science and Cardiorenal Medicine, Yokohama City University School of Medicine, Yokohama, Kanagawa, Japan
- Tamura, Kouichi, Department of Medical Science and Cardiorenal Medicine, Yokohama City University School of Medicine, Yokohama, Kanagawa, Japan
Group or Team Name
- Department of Medical Science and Cardiorenal Medicine, Yokohama City University School of Medicine
Background
Randomized trial of allopurinol and febuxostat has been performed in hyperuricemic patients to inhibit XOR and reduce uric acid levels. Nonetheless, the protective effect for CVD events and mortality is insufficient in dialysis patients. Furthermore, febuxostat(FEB) is also known to be a potent ATP binding cassette transporter subfamily G member2 (ABCG2) inhibitor that promotes the accumulation of uremic toxins by reduced excretion from the gut. We investigated the preventative effect of XOR inhibitors on outcomes under observation.
Methods
We observed 6791 maintenance dialysis patients from April 2016 to March 2019. Outcomes were all-cause mortality and CVD events, using longitudinal data. Baseline uric acid and outcome was examined using Cox Hazard analysis. Causal associations for outcomes of allopurinol (ALLO) or FEB were analyzed using 3 years of longitudinal data, including concomitant drugs, using MSM, which is weighted analysis that examined the causal effect of ALLO or FEB versus control at each visit.
Results
MSM indicated that ALLO estimated HR 0.35 for all-cause mortality versus un treatment (p<0.001), also indicated that FEB estimated HR 0.42 (p<0.001). In CVD events, ALLO demonstrated a preventative effect of HR 0.81 versus control (p<0.001). However, FEB did not show a preventative effect of HR 0.98 (95% CI: 0.91–1.04). ALLO versus FEB for CVD events was HR 0.83（95％ CI 0.75–0.92）.
Conclusion
FEB failed to demonstrate a preventative effect for CVD events compared with ALLO and the control but did not increase the risk of all-cause mortality or CVD events. Attenuated effect of FEB was considered to be the result of uremic toxin accumulation evoked by ABCG2 dysfunction.
CKD events