Objective To compare the efficacy of five disinfection methods, including peracetic acid, sodium hypochlorite, enzymatic detergent, alkaline wash, and steam, for biofilm removal in hemodialysis water supply pipelines using a multidimensional evaluation system involving "bacterial count - endotoxin quantification - electron microscopic morphology".  Methods  A mature biofilm model of Pseudomonas aeruginosa (ATCC 27853) was established on 316L stainless steel carriers. The samples were divided into six treatment groups: a control group, 0.2% peracetic acid, 0.2% sodium hypochlorite, enzymatic detergent, alkaline wash and hot steam, with each group repeated independently three times. After treatment, bacterial residues were detected using both Luria-Bertani (LB) agar and Reasoner's 2A (R2A) low-nutrient agar to calculate the Log Reduction Value (LRV); endotoxin LRV was determined by kinetic turbidimetric assay; and physical removal was verified by scanning electron microscopy (SEM).  Results  ①Bacterial removal: The bacterial count in the control group on LB media was (1.37±1.10)×107 CFU/ml. There was a statistically significant difference in the overall LB-LRV values between the groups (F=64.330, P＜0.001). No bacteria were detected in the hot steam group in either of the two culture media (LB-LRV >7.14); there were no statistically significant differences in the bactericidal efficacy of the peracetic acid group and the sodium hypochlorite group compared with the hot steam group (t=1.017, 2.580; P=0.842, 0.148).  ②Endotoxin removal efficacy: The endotoxin concentration in the control group was (1.14±1.05)×104 EU/ml. Overall differences in endotoxin LRV among treatment groups were statistically significant (F=10.480, P=0.001). The hot steam group demonstrated superior removal efficacy (LRV=2.23) compared with the peracetic acid group (t=4.571, P=0.007), the enzymatic detergent group (t=4.636, P=0.006) and the alkaline detergent group (t=5.171, P=0.003).  ③Electron microscopic morphology: Fragments of biofilm were observed in the peracetic acid group, the sodium hypochlorite group and the hot steam group; in the peracetic acid group, bacteria appeared shrivelled but matrix residues remained; in the enzymatic detergent group and the alkaline detergent group, a large number of intact bacteria or cell aggregates were still observed.  Conclusions  Among the five biofilm removal methods, the steam group was significantly superior to other groups in sterilization, endotoxin removal, and physical detachment. Sodium hypochlorite achieved effective bacterial eradication while exhibiting favorable trends in endotoxin clearance and substrate detachment, yielding the second-best results. Although some disinfection methods met the bacterial quality control standards for dialysis water, none of the five single disinfection methods met the dialysis water standards for biofilm endotoxin removal.