“…In this meta-analysis, we retrieved 40 relevant studies from the three databases; MEDLINE, CENTRAL, and Ichushi-web, as shown in the flowchart ( Fig 1 ). Among these studies, data for analysis were available for six of seven DPP-4 inhibitors: alogliptin (n = 2) [ 29 , 30 ], anagliptin (n = 1) [ 31 ], linagliptin (n = 3) [ 32 – 34 ], saxagliptin (n = 4) [ 35 – 38 ], sitagliptin (n = 26) [ 39 – 64 ], and vildagliptin (n = 4) [ 65 – 68 ], but not for teneligliptin (n = 0). Comparators of DPP-4 inhibitors were α-GIs (n = 9) [ 30 , 31 , 34 , 35 , 41 , 50 , 63 – 65 ], GLP-1 analogues (n = 5) [ 47 , 52 , 58 , 62 , 68 ], metformin (n = 5) [ 29 , 38 – 40 , 56 ], SGLT2 inhibitors (n = 5) [ 36 , 53 , 54 , 60 , 61 ], sulfonylureas (n = 12) [ 32 , 33 , 37 , 43 – 46 , 48 , 57 , 59 , 66 , 67 ], and thiazolidinediones (n = 4) [ 42 , 49 , 51 , 55 ].…”