A new retrospective cohort study published in the American Journal of Kidney Diseases (AJKD) has found that women are more likely to die from infection and dialysis withdrawal, contributing to a higher risk of all-cause mortality in the first five years after beginning dialysis treatment.
Lead author Wai H Lim (Sir Charles Gairdner Hospital, Perth, Australia) and colleagues note in their introduction that cardiovascular disease and all-cause mortality in the general population is “significantly lower” for women than for men, but that this does not translate to such an advantage among those with chronic kidney disease (CKD).
They refer to a previous study carried out by Emmanuel Villar (Lyon Sud Hospital, Lyon, France) et al, which in an analysis of the data for 3,025 patients in France receiving maintenance dialysis found that female patients had a mortality that “was as much as 11 times greater” than male patients’. Pointing to that study’s lack of information on cause-specific mortality, Lim et al set out to rectify the “gap in data” they diagnosed on “potential interactive effects between sex and patient characteristics”, which included “age, ethnicity, and era of commencing dialysis” on cause-specific death for dialysis patients.
The retrospective study took data from patients in the Australia and New Zealand Dialysis and Transplant Registry (ANZDATA). The data included information on “age, sex, body mass index, race, [and] dialysis modality (haemodialysis or peritoneal)”. The total cohort included 53,414 patients, 20,876 of whom were female and 32,538 of whom were male.
The authors “constructed time-stratified proportional hazards and Fine and Gray models”, and also used a multivariable model in a separate analysis. The median follow-up period was 2.8 years (2.9 for women and 2.7 for men). Sex-based associations with both cause-specific and all-cause mortality were compared using incidence rate ratios. Men were generally “significantly older” than women when beginning dialysis. A lower proportion of women had prevalent coronary artery disease (25% vs 36%) or peripheral vascular disease (15% vs 20%) at baseline.
All-cause mortality was found to be statistically significantly higher for women, with an adjusted hazard ratio of 1.08 (95% confidence interval [CI], 1.05–1.11). They were, however, found to be more likely to die as a result of infection or dialysis withdrawal. Cause-specific mortality was also examined in a substudy of 13,966 patients who had survived for five years of maintenance dialysis without receiving a kidney transplant. Of these, 55% of women and 58% of men died, but women were less likely to die of cardiovascular disease (CVD), which was the primary cause of death in both groups. This was found to be “independent of diabetes, smoking, and prevalent vascular disease status,” and this effect was more pronounced as patients spent more time on dialysis. Nevertheless, the authors emphasise, “one in three female patients with kidney failure die of CVD”.
The study found no meaningful association between sex and “age, ethnicity, or era for cause-specific mortality”, leading them to suggest that “the observed associations between sex and mortality were not modified by these factors”. Lim et al make the case that better understanding is needed of the processes underlying kidney disease in order to explain the differences that, they suggest, are specific to biological sex. This may contribute to “novel and effective interventions… to narrow the equity gap” for outcomes among female and male patients.
The authors propose a few of their own theories to explain that gap, suggesting that the heightened risk of infection-related mortality among younger female dialysis patients may be in part a result of “the effects of sex hormones on the immune system” combined with “increased use of central venous catheters during haemodialysis”. They also noted that female patients are often “more likely to opt for a conservative pathway for management of kidney failure”.
Finally, Lim and colleagues alerted readers to a few limitations of their study. Among these was the fact that ANZDATA does not account for preference and acceptance rates of dialysis among men and women, as well as its failure to verify the accuracy of reported data. Looking forward, the authors call for further research in other countries with different healthcare systems, as well as into potential “targeted interventions” to improve outcomes for women.
