著者
Takuhiro Moromizato Ryoto Sakaniwa Takamasa Miyauchi Ryuhei So Hiroyasu Iso Kunitoshi Iseki
出版者
Japan Epidemiological Association
雑誌
Journal of Epidemiology (ISSN:09175040)
巻号頁・発行日
vol.33, no.8, pp.390-397, 2023-08-05 (Released:2023-08-05)
参考文献数
31
被引用文献数
2

Background: Serial weight decrease can be a prognostic predictor in chronic hemodialysis (HD) patients. We investigated the impact of long-term post-HD body weight (BW) changes on all-cause mortality among HD patients.Methods: This longitudinal cohort study and post-hoc analysis evaluated participants of a previous randomized controlled trial conducted between 2006 and 2011 who were followed up until 2018. Weight change slopes were generated with repeated measurements every 6 months during the trial for patients having ≥5 BW measurements. Participants were categorized into four groups based on quartiles of weight change slopes; the median weight changes per 6 months were −1.02 kg, −0.25 kg, +0.26 kg, and +0.86 kg for first, second, third, and fourth quartile, respectively. Cox proportional hazard regression was used to evaluate differences in subsequent survival among the four groups. BW trajectories were plotted with a backward time-scale and multilevel regression analysis to visualize the difference in BW trajectories between survivors and non-survivors.Results: Among the 461 patients, 404 were evaluated, and 168 (41.6%) died within a median follow-up period of 10.2 years. The Cox proportional hazard regression adjusted for covariates and baseline BW showed that a higher rate of weight loss was associated with higher mortality. The hazard ratios were 2.02 (95% confidence interval [CI], 1.28–3.20), 1.77 (95% CI, 1.10–2.85), 1.00 (reference), and 1.11 (95% CI, 0.67–1.83) for the first, second, third (reference), and fourth quartiles, respectively. BW trajectories revealed a significant decrease in BW in non-survivors.Conclusion: Weight loss elucidated via serial BW measurements every 6 months is significantly associated with higher mortality among HD patients.