An investigation into possible sex differences in Association of Haemoglobin with survival among Haemodialysis patients in the J-DOPPS cohort.

Abstract

Lower haemoglobin levels are common among females without kidney diseases. However, little is known about the sex-specific management of anaemia in haemodialysis patients. This prospective cohort study investigated the role of sex differences in the association between categorical baseline or time-varying haemoglobin levels and all-cause mortality via cox regression using data from 6890 patients in the Japan Dialysis Outcomes and Practice Patterns Study (J-DOPPS, 2005-2015). Likelihood ratio tests were used to evaluate the effect modification of sex on the relationship between haemoglobin and mortality. A total of 781 patients died during the median follow-up of 31 months. Mortality risk, adjusted for case mix, varied between five haemoglobin categories, with the highest category (≥12 g/dL) having a hazard ratio of 0.73 (0.41-1.29) for females and 2.02 (1.03-3.95) for males versus 10-10.9g/dL. Despite this difference, the p-value comparing the overall among males versus females was.35. Similar associations were observed in models stratified by patient age (<75 years), time on dialysis (≤1year), and models lagging the haemoglobin exposure. The results based on this sample of Japanese haemodialysis patients did not support the hypothesis that the association between haemoglobin and survival differed by sex. We also could not conclude that the association was identical, as the parameter estimates are consistent with male patients having a relatively greater mortality risk than female patients at higher haemoglobin levels. More detailed investigations into the effects of higher haemoglobin levels by sex might help better understand strategies for anaemia management.