Fig 1.
Flowchart of study participants.
Table 1.
Demographic, clinical, and laboratory values in 142,412 hemodialysis patients.
Table 2.
HRs and 95% CIs for variables evaluated as potential predictors of mortality among all the patients.
Table 3.
Demographic, clinical, and laboratory values in 142,412 according to type of dialyzer membrane.
Fig 2.
HRs of all-cause mortality between seven types of dialyzer membranes in 142,412 patients undergoing maintenance hemodialysis using standard Cox proportional hazards regression.
White bars are adjusted for age, sex, dialysis duration, presence or absence of diabetes, and presence or absence of cardiovascular comorbidity. Gray bars are adjusted for basic factors and dialysis dose as indicated by single-pool Kt/V. Black bars are adjusted for basic factors, dialysis dose, and nutrition factors, including body mass index, serum levels of albumin, total cholesterol levels, normalized protein catabolic rate, and % creatinine generation rate. *P < 0.05, **P < 0.01, ***P < 0.001 vs. PS. CTA, cellulose triacetate; EVAL, ethylene vinyl alcohol; HR, hazard ratio; PAN, polyacrylonitrile; PEPA, polyester polymer alloy; PES, polyethersulfone; PMMA, polymethylmethacrylate; PS, polysulfone.
Table 4.
Baseline characteristics after propensity score matching between PS and other groups.
Fig 3.
HRs of all-cause mortality after propensity score matching for six types of dialyzer groups compared to the PS group using Cox proportional hazards regression.
*P < 0.01 vs. PS. CTA, cellulose triacetate; EVAL, ethylene vinyl alcohol; HR, hazard ratio; PAN, polyacrylonitrile; PEPA, polyester polymer alloy; PES, polyethersulfone; PMMA, polymethylmethacrylate; PS, polysulfone.