Long-term Trends in Infection-Related Mortality in Adults Treated With Maintenance Dialysis

Rationale & Objective: Infection is 1 of the top 3 causes of death in patients receiving maintenance dialysis. We evaluated the trends over time and risk factors for infection-related deaths among people receiving dialysis. 

Study Design: Retrospective cohort study. 

Setting & Participants: We included all adults who began dialysis between 1980 and 2018 in Australia and New Zealand. 

Exposure: Age, sex, dialysis modality, and dialysis era. 

Outcome: Infection-related death. 

Analytical Approach: Incidence was described and standardized mortality ratios (SMR) calculated for infection-related death. Fine-Gray subdistribution hazards models were fitted, with non–infection-related death and kidney transplantation treated as competing events. 

Results: The study comprised 46,074 patients who received hemodialysis and 20,653 who were treated with peritoneal dialysis who were followed for 164,536 and 69,846 person-years, respectively. There were 38,463 deaths during the follow-up period, 12% of which were ascribed to infection. The overall rate of mortality from infection (per 10,000 person-years) was 185 and 232 for patients treated with hemodialysis and peritoneal dialysis, respectively. The rates were 184 and 219 for males and females, respectively; and 99, 181, 255, and 292 for patients aged 18-44, 45-64, 65-74, 75 years and over, respectively. The rates were 224 and 163 for those commencing dialysis in years 1980-2005 and 2006-2018, respectively. The overall SMR declined over time, from 37.1 (95% CI, 35.5-38.8) in years 1980-2005 to 19.3 (95% CI, 18.4-20.3) in years 2006-2018, consistent with the declining 5-year SMR trend (P < 0.001). Infection-related mortality was associated with being female, older age, and Aboriginal and/or a Torres Strait Islander or Māori. 

Limitations: Mediation analyses defining the causal relationships between infection type and infection-related death could not be undertaken as disaggregating the data was not feasible. 

Conclusions: The excess risk of infection-related death in patients on dialysis has improved substantially over time but remains more than 20 times higher than in the general population.