Table 2.

Summary of main outcomes of all included studies

Author, year	Main outcome	Result
Meier, 2004 [29]	GLP-1 IV lowered mean glucose levels	+
Sokos, 2007 [34]	GLP-1 IV reduced peri-operative glucose levels	+
Müssig, 2008 [31]	GLP-1 IV reduced insulin administration with comparable glycaemic control	+
Deane, 2009 [21]	GLP-1 IV lowered mean post-prandial glucose levels	+
Deane, 2010 [22]	GLP-1 IV lowered mean post-prandial glucose levels	+
Deane, 2011 [23]	GLP-1 IV lowered mean post-prandial glucose levels	+
Galiatsatos, 2014 [24]	GLP-1 IV did not lower mean glucose levels	–
Kohl, 2014 [27]	GLP-1 IV lowered mean glucose levels	+
Miller, 2017 [30]	GLP-1 IV reduced intestinal glucose absorption	+
Kar, 2015 [26]	GIP IV did not lower mean glucose levels	–
Lee, 2013 [28]	GIP IV did not lower mean glucose levels	–
Polderman, 2018 [18]	Liraglutide SC reduced post-operative glucose levels	+
Holmberg, 2014 [25]	Exenatide IV did not lower post-operative cardiac enzymes	–
Besch, 2017 [19]	Exenatide IV did not increase number of patient that spend > 50% in target range	–
Lipš, 2017 [17]	Exenatide IV did not improve left ventricular ejection fraction	–
Garg, 2017 [35]	Saxagliptin PO resulted in similar glucose levels compared with basal bolus insulin	+
Pasquel, 2017 [32]	Sitagliptin PO as adjunct to basal insulin resulted in similar glucose levels compared with bolus insulin	+
Umpierrez, 2014 [33]	Sitagliptin PO resulted in similar glucose levels compared with basal bolus insulin	–
Brackbill, 2012 [20]	Sitagliptin PO did not lower the mean postoperative glucose levels	–

Abbreviations: GIP Gastric inhibitory polypeptide, GLP-1 Glucagon-like peptide-1, IV Intravenously, PO By mouth, SC Subcutaneous

+ = study positive for primary outcome, − = study negative for primary outcome