Related Experiment Videos

Comparing mortality rates on CAPD/CCPD and hemodialysis. The Canadian experience: fact or fiction?

D E Schaubel1, H I Morrison, S S Fenton

  • 1Laboratory Centre for Disease Control, Health Canada, Ottawa.

Insights

Continuous ambulatory/cyclic peritoneal dialysis (CAPD/CCPD) shows lower mortality rates than hemodialysis (HD), especially within the first two years. This study highlights challenges in comparing dialysis modalities due to potential biases and differing methodologies.

Area of Science:

  • Nephrology
  • Renal Replacement Therapy
  • Dialysis Modalities

Background:

  • Hemodialysis (HD) and continuous ambulatory/cyclic peritoneal dialysis (CAPD/CCPD) are primary renal replacement therapies.
  • Comparing mortality rates between HD and CAPD/CCPD is crucial for patient care and treatment selection.

Purpose of the Study:

  • To compare mortality rates between hemodialysis (HD) and continuous ambulatory/cyclic peritoneal dialysis (CAPD/CCPD).
  • To contrast findings with previous investigations and explore reasons for discrepancies in reported outcomes.
  • To discuss the implications of different statistical approaches on observed mortality differences.

Main Methods:

  • Utilized patient-specific data from the Canadian Organ Replacement Register for 14,483 patients initiating renal replacement therapy (RRT) between 1990 and 1995.
  • Employed an "as-treated" (AT) analysis using Poisson regression to calculate mortality rate ratios (RR), adjusting for age, diagnosis, and comorbidities.
  • Conducted an "intent-to-treat" (ITT) analysis using Cox regression to estimate hazard ratios (HR) based on treatment modality at day 90.

Main Results:

  • The "as-treated" analysis revealed significantly lower adjusted mortality rates for CAPD/CCPD compared to HD (RR = 0.73).
  • The protective effect of CAPD/CCPD was most pronounced in the initial two years following RRT initiation.
  • The "intent-to-treat" analysis showed a substantially diminished effect, with a hazard ratio of 0.93 for CAPD/CCPD versus HD.

Conclusions:

  • CAPD/CCPD is not inferior to HD, particularly for short-term outcomes.
  • Direct comparison of mortality between CAPD/CCPD and HD is complex and prone to bias.
  • Variability in study findings is attributed to differences in patient populations, study designs, statistical methods, and covariate interactions.
Abstract

Related Concept Videos