Comparison between different types of exercise training in patients with type 2 diabetes mellitus: A systematic review and network metanalysis of randomized CONTROLLED trials

Abstract Aim Aim of the present meta-analysis and network metanalysis (NMA) is the assessment of the effects of physical exercise on glucose control and cardiovascular risk factors in type 2 diabetes. Data synthesis This metanalysis includes all available trials exploring the effects of different exercise modalities in type 2 diabetes, with a duration of ≥ 3 months. The standardized difference in means (SDM) with 95% Confidence Intervals were calculated. Data were analyzed using MetaXL and Rev Man 5.0. Primary endpoint was the effect of exercise versus no exercise on HbA1c and fasting plasma glucose (FPG) at endpoint. Secondary endpoints were body weight and fat, waist circumference, and blood pressure. A comparison of different exercise training modalities (aerobic, resistance and combined) for the same endpoints was also performed, choosing ‘no exercise’ as the reference for indirect comparisons. We included 25 trials fulfilling all inclusion criteria. Physical exercise versus no exercise produced a small, but significant, improvement of HbA1c, body fat, and systolic blood pressure at endpoint (-0.3[-0.1;-0.4]%, -1.44 [-2.22, -0.66]%, and -5.6 [-9.5, -1.6] mmHg, respectively). Combined, supervised aerobic and resistance exercise were associated with a significantly greater reduction of HbA1c (SDM, -0.4[-0.6;-0.3], -0.2[-0.4;-0.1], and -0.2[-0.3;-0.1]%, respectively), but not of FPG, in comparison with no exercise. Conclusions Physical exercise produces small, but detectable, advantages on glycemic control and cardiovascular risk factors and should be suggested in type 2 diabetes. Combined aerobic/resistance training seems to be superior to aerobic training alone, but differences are small and the reliability of supporting evidence limited.