OBJECTIVE： To evaluate efficacy and saftey of prophylactic use of proton pump inhibitor （PPI） for upper gastrointestinal hemorrhage in patients with chronic kidney disease （CKD） systematically， and to provide evidence-based reference in clinic. METHODS： Retrieved from PubMed， Cochrane library， CNKI， CBM and VIP， randomized controlled trials （RCTs） or cohort study about prophylactic use of PPI （trial group） compared with blank control （control group） for upper gastrointestinal hemorrhage in CKD patients were collected. After literature screening， data extraction and literature quality evaluation with NOS scale， Meta-analysis was performed by using Stata 14.0 software. RESULTS： Fourteen cohort studies were included， involving 13 740 patients in total. The incidence of upper gastrointestinal hemorrhage [RR＝0.16，95％CI（0.07，0.34），P＜0.01] and serum creatinine [WMD＝－62.14，95％CI（－86.98，－37.30），P＜0.01] in trial group were significantly lower than control group； the levels of alkaline phosphatase [WMD＝51.25， 95％CI（10.72，91.78）， P＝0.01] in trial group was significantly higher than control group. There was no statistical significance in the mortality [RR＝1.08，95％CI（0.83，1.40），P＝0.56]， blood calcium concentration [WMD＝－0.14， 95％CI（－0.37，0.08）， P＝0.22]， blood phosphorus concentration [WMD＝0.03， 95％CI（－0.08，0.14）， P＝0.60]， blood magnesium concentration [WMD＝－0.06， 95％CI（－0.22，0.10）， P＝0.44]， the level of parathyroid hormone [WMD＝7.85， 95％CI（－24.88，40.58）， P＝0.64] or glomerular filtration rate [WMD＝3.97， 95％CI（－6.87，14.81），P＝0.47] between 2 groups. CONCLUSIONS： The prophylactic use of PPI can effectively reduce the risk of upper gastrointestinal hemorrhage in CKD patients， without increasing the risk of renal function disorder， aggravating electrolyte disorder or reducing the incidence of death， but with the risk of increasing the level of alkaline phosphatase.