Estradiol and mortality in women with end-stage kidney disease.

ABSTRACT

BACKGROUND: Young women with end-stage kidney disease (ESKD) have early menopause compared with women in the general population and the highest mortality among the dialysis population. We hypothesized that low estrogen status was associated with death in women with ESKD. METHODS: We measured estradiol and sex hormone levels in female ESKD patients initiating hemodialysis from 2005 to 2012 in four Canadian centers. We divided women into quintiles based on estradiol levels and tested for associations between the estradiol level and cardiovascular (CV), non-CV and all-cause mortality. Participants were further dichotomized by age. RESULTS: A total of 482 women (60 ± 15 years of age, 53% diabetic, estradiol 116 ± 161 pmol/L) were followed for a mean of 2.9 years, with 237 deaths (31% CV). Estradiol levels were as follows (mean ± standard deviation) Quintile 1 19.3 ± 0.92 pmol/L; Quintile 2 34.6 ± 6.6 pmol/L; Quintile 3 63.8 ± 10.6 pmol/L; Quintile 4 108.9 ± 19.3; Quintile 5 355 ± 233 pmol/L. Compared with Quintile 1, women in Quintiles 4 and 5 had significantly higher adjusted all-cause mortality {hazard ratio [HR] 2.12 [95% confidence interval (CI) 1.38-3.25] and 1.92 [1.19-3.10], respectively}. Similarly, compared with Quintile 1, women in Quintile 5 had higher non-CV mortality [HR 2.16 (95% CI 1.18-3.96)]. No associations were observed between estradiol levels and CV mortality. When stratified by age, higher quintiles were associated with greater all-cause mortality (P for trend <0.001) and non-CV mortality (P for trend = 0.02), but not CV mortality in older women. CONCLUSIONS: In women with ESKD treated with hemodialysis, higher estradiol levels were associated with greater all-cause and non-CV mortality. Further studies are required to determine the mechanism for the observed increased risk.