There is limited information regarding the role of left ventricular (LV) twist and the effect of exercise in type 2 diabetes (T2D). The aim of this study was to compare LV twist parameters in patients with T2D versus healthy control subjects and the effects of high-intensity interval exercise (HIIE) and moderate-intensity exercise (MIE) on LV twist in patients with T2D with diastolic dysfunction. This study, which included both prospective and retrospective components, included 47 patients with T2D and diastolic dysfunction and 37 healthy individuals. Patients with T2D were randomized to HIIE (4×4min at 90%-95% of maximal heart rate, three times a week, 120min/wk; n=24) or MIE (210min/wk; n=23) for 12weeks and examined with echocardiography (LV twist by speckle-tracking method) at baseline and posttest. The control subjects received no intervention and were matched according to age, gender, and body mass index to those completing the intervention. In total, 37 subjects completed 12weeks of MIE (n=17) or HIIE (n=20). LV peak untwist rate (UTR) was similar in patients with T2D and control subjects (P˃.05). At baseline, LV peak UTR, relative to total diastolic period, occurred 5.8 percentage points later in patients with T2D compared with control subjects (P=.004). Time to peak UTR was shortened by 6.5 percentage points (P=.002) and 7.7 percentage points (P<.001) after MIE and HIIE, respectively. Time to peak UTR was similar to that in control subjects after exercise interventions. In patients with T2D and diastolic dysfunction, LV peak UTR was similar, but time to peak LV UTR was delayed compared with control subjects. Twelve weeks of endurance exercise normalized the timing of UTR.