Kidney Week

Abstract: FR-PO782

Discrepant Associations of Hemodialysis Intensity and Survival in the Facility and at Home

Session Information

• Dialysis: Hospitalization and Mortality
  October 26, 2018 | Location: Exhibit Hall, San Diego Convention Center
  Abstract Time: 10:00 AM - 12:00 PM

Category: Dialysis

• 701 Dialysis: Hemodialysis and Frequent Dialysis

Authors

• Weinhandl, Eric D., NxStage Medical, Inc., Victoria, Minnesota, United States

Eric D. Weinhandl,

• Kubisiak, Kristine, NxStage Medical, Inc., Victoria, Minnesota, United States

Kristine Kubisiak,

• Ray, Debabrata, NxStage Medical, Inc., Victoria, Minnesota, United States

Debabrata Ray,

• Collins, Allan J., NxStage Medical, Inc., Victoria, Minnesota, United States

Allan J. Collins,

Background

Incremental hemodialysis (HD) has recently garnered interest. Mathew et al (Kidney Int, 2016) reported that in US incident patients on in-center HD, adjusted hazard ratios (AHRs) of death for 2 vs. 3 and ≥4 vs. 3 sessions/week were 0.88 (95% confidence interval, 0.72-1.08) and 1.56 (1.21-2.03), respectively. These data may reflect either the relative efficacy of twice-weekly HD as an initial prescription or residual confounding. We assessed whether these associations could be replicated among in-center hemodialysis (IHD) and home hemodialysis (HHD) patients.

Methods

We used data from the United States Renal Data System. We identified adult patients who initiated HD in 2011-2015, and we retained patients with an initial HD prescription of 2 to 7 sessions/week and 2-8 hours/session, according to form CMS-2728. We followed patients until death, but for a maximum of one year. Using Cox regression, we assessed associations of HD frequency, HD hours per week, and HD product (frequency² × hours/session) separately among IHD and HHD patients, with adjustment for demography, vascular access type, and disease severity, including estimated glomerular filtration rate.

Results

We identified 503,678 IHD patients and 2849 HHD patients. With IHD (vs. HHD), the percentage of patients prescribed 2 sessions/week was 1.1% (vs. 1.1%), the percentage of patients prescribed ≥4 sessions/week was 0.5% (vs. 59.7%), mean HD hours per week were 11.4 (vs. 14.0), and mean HD product was 34 (vs. 62) points. With IHD, the AHRs of death for 2 vs. 3, ≥4 vs. 3, and ≥4 vs. 2 sessions/week were 0.84 (0.80-0.89), 1.05 (0.96-1.14), and 1.24 (1.12-1.38), respectively; with HHD, the AHRs were 0.83 (0.39-1.76), 0.63 (0.52-0.75), and 0.76 (0.36-1.61), respectively. With IHD, the AHR of death for each additional HD hour per week was 1.012 (1.008-1.016); the corresponding AHR for HHD was 0.967 (0.943-0.993). Finally, with IHD, the AHR of death for each 10-point increment in HD product was 1.039 (1.028-1.051); the corresponding AHR for HHD was 0.946 (0.915-0.979).

Conclusion

Associations of HD frequency, HD hours per week, and HD product are almost entirely discrepant with IHD and HHD. In the facility, lower HD frequency and fewer HD hours are associated with better survival. In the home, higher frequency and greater HD hours are associated with better survival.

Funding

• Commercial Support –