“…Of these, 19 articles were excluded; two articles were duplicate [ 22 , 23 ], one measured FMD of popliteal artery [ 24 ], two had patients with different kinds of diseases as well as T2D in the groups [ 25 , 26 ], three had T2D patients with peripheral arterial disease or diabetic peripheral neuropathy [ 27 – 29 ], two did not provide precise data [ 30 , 31 ], two additionally treated dietary control for weight loss [ 32 , 33 ], three involved unstructured or unsupervised exercise intervention [ 34 – 36 ], and four were just abstracts with incomplete data [ 37 – 40 ]. Four exercise groups (EX) with additional interventions were included after discussing because there was no difference in the effect of interventions on FMD, our major outcome, between the CON and experimental groups: (i) exercise in the hypoxic environment (16.5% O 2 , 2000 m) [ 41 ]; (ii) endothelin (ET) receptor blockade or a placebo [ 42 ]; (iii) walking meditation [ 43 ]. One AE trial which recorded the time, frequency, and intensity of exercise by a multi-record accelerometer were also included because the subjects visited the laboratory every 1 or 2 weeks [ 44 ].…”