AIM:To investigate the preventive effects of low-dose proton-pump inhibitors(PPIs) for upper gastrointestinal bleeding(UGIB) in end-stage renal disease.METHODS:This was a retrospective cohort study that reviewed 544 patients with end-stage renal disease who started dialysis at our center between 2005 and 2013.We examined the incidence of UGIB in 175 patients treated with low-dose PPIs and 369 patients not treated with PPIs(control group).RESULTS:During the study period, 41 patients developed UGIB, a rate of 14.4/1000 person-years.The mean time between the start of dialysis and UGIB events was 26.3 ± 29.6 mo.Bleeding occurred in only two patients in the PPI group(2.5/1000 person-years) and in 39 patients in the control group(19.2/1000 person-years).Kaplan-Meier analysis of cumulative non-bleeding survival showed that the probability of UGIB was significantly lower in the PPI group than in the control group(log-rank test, P < 0.001).Univariate analysis showed that coronary artery disease, PPI use, anti-coagulation, and anti-platelet therapy were associated with UGIB.After adjustments for the potential factors influencing risk of UGIB, PPI use was shown to be significantly beneficial in reducing UGIB compared to the control group(HR = 13.7, 95%CI:1.8-101.6; P = 0.011).CONCLUSION:The use of low-dose PPIs in patients with end-stage renal disease is associated with a low frequency of UGIB. To investigate the preventive effects of low-dose proton-pump inhibitors (PPIs) for upper gastrointestinal bleeding (UGIB) in end-stage renal disease. METHODS: This was a retrospective cohort study that reviewed 544 patients with end-stage renal disease who started dialysis at our center between 2005 and 2013. We examined the incidence of UGIB in 175 patients treated with low-dose PPIs and 369 patients not treated with PPIs (control group). RESULTS: During the study period, 41 patients developed UGIB, a rate of 14.4 / 1000 person-years. The mean time between the start of dialysis and UGIB events was 26.3 ± 29.6 months. There were only two patients in the PPI group (2.5 / 1000 person-years) and in 39 patients in the control group (19.2 / 1000 person-years). Kaplan-Meier analysis of cumulative non-bleeding survival showed that the probability of UGIB was significantly lower in the PPI group than in the control group (log-rank test, P <0.001) .Univariate analysis showed that coronary artery diseas PPI use, anti-coagulation, and anti-platelet therapy were associated with UGIB. After adjustments for the potential factors influencing risk of UGIB, PPI use was shown to be significantly beneficial in reducing UGIB compared to the control group (HR = 13.7 , 95% CI: 1.8-101.6; P = 0.011). CONCLUSION: The use of low-dose PPIs in patients with end-stage renal disease is associated with a low frequency of UGIB.