Publications

Daily Variation in Death in Patients Treated by Long-term Dialysis: Comparison of In-Center Hemodialysis to Peritoneal and Home Hemodialysis.

ANZDATA
Article
  • Posted1 Jan 2013
  • PMID22901771

AuthorsKrishnasamy, Rathika, Badve, Sunil V, Hawley, Carmel M, McDonald, Stephen P, Boudville, Neil, Brown, Fiona G, Polkinghorne, Kevan R, Bannister, Kym M, Wiggins, Kathryn J, Clayton, Philip A, Johnson, David W

Periodical/sAmerican Journal of Kidney Diseases

View Publication

Overview

Background

There has been little study to date of daily variation in cardiac death in dialysis patients and whether such variation differs according to dialysis modality and session frequency.

Study Design

Observational cohort study using ANZDATA (Australia and New Zealand Dialysis and Transplant) Registry data.

Setting & Participants

All adult patients with end-stage kidney failure treated by dialysis in Australia and New Zealand who died between 1999 and 2008.

Predictors

Timing of death (day of week), dialysis modality, hemodialysis (HD) session frequency, and demographic, clinical, and facility variables.

Outcomes & Measurements

Cardiac and noncardiac mortality.

Results

14,636 adult dialysis patients died during the study period (HD, n = 10,338; peritoneal dialysis [PD], n = 4,298). Cardiac death accounted for 40% of deaths and was significantly more likely to occur on Mondays in in-center HD patients receiving 3 or fewer dialysis sessions per week (n = 9,503; adjusted OR, 1.26; 95% CI, 1.14-1.40; P < 0.001 compared with the mean odds of cardiac death for all days of the week). This daily variation in cardiac death was not seen in PD patients, in-center HD patients receiving more than 3 sessions per week (n = 251), or home HD patients (n = 573). Subgroup analyses showed that deaths related to hyperkalemia and myocardial infarction also were associated with daily variation in risk in HD patients. This pattern was not seen for vascular, infective, malignant, dialysis therapy withdrawal, or other deaths.

Limitations

Limited covariate adjustment. Residual confounding and coding bias could not be excluded. Possible type 2 statistical error due to limited sample size of home HD and enhanced-frequency HD cohorts.

Conclusions

Daily variation in the pattern of cardiac deaths was observed in HD patients receiving 3 or fewer dialysis sessions per week, but not in PD, home HD, and HD patients receiving more than 3 sessions per week.