Health status of older adults with Type 2 diabetes mellitus after aerobic or resistance training: A randomised trial

In Asia, more than 100 million people were living with T2DM in 2007 [1]. The prevalence in Singapore is 8.2% in adults aged 18 to 69 years and is expected to rise [2]. It is important to assess the impact of interventions that affect blood glucose control on health status besides clinical outcomes such as glycemic control [3]. Exercise is considered a critical part of therapeutic lifestyle intervention in the treatment of individuals with type 2 diabetes mellitus (T2DM) [4, 5]. Exercise has been shown to improve quality of life in special populations [6, 7]. In patients with T2DM, it is recommended that patients undertake both aerobic training and progressive resistance training [4]. We have recently shown that both types of training improve metabolic control to a similar degree [8]. In a recent study by Reid et al, it appeared that resistance training had more beneficial effects on physical health status than aerobic training [9]. However, the differences in the effects were not statistically significant [9]. Furthermore, they did not attempt to ensure similar volume or duration of exercise in all groups.

The aim of this study was to compare the effects of progressive resistance training (PRT) and aerobic training (AT), of similar volume and duration, on health status in middle-aged patients with T2DM.

Contrary to Reid et al.'s findings, we hypothesised that PRT and AT of similar volume would have similar effects on health status.

We analysed data from 60 subjects with T2DM who participated in a randomised trial of PRT vs AT over an 8-week period [8]. The PRT group undertook nine resistive exercises (three sets of 10 repetitions) at 65% of their assessed one repetitive maximum while the AT group underwent 50 minutes of aerobic training with a target heart rate of 65% of their age-predicted maximum heart rate [8]. The calorie expenditure of both exercise programs was estimated to be 3.5 kcal/kg body weight. More details on the exercise regimes are provided (Additional file 1, Table S1). The main outcomes and the description of the exercise regimen have already been published [8]. In that study, we found that glycosylated haemoglobin (HbA1C) reduced by 0.4(0.6)% and 0.3(0.9)% in PRT and AT group respectively, but there was no significant difference between the groups (-0.1%, 95% CI -0.5 to 0.3) [8]. Systolic blood pressure as well as aerobic fitness in the form of peak oxygen consumption (VO₂peak) favoured the AT group more by 9 mmHg (95% CI 2 to 16) and 5.2 ml/kg (95% CI 0.0 to 10.4) respectively [8]. The PRT group showed a greater reduction in waist circumference by 1.8 cm (95% CI 0.5 to 3.1) [8]. In this secondary analysis, we report on the impact of PRT and AT on health status as measured by the SF-36 questionnaire. All subjects gave written informed consent.

The baseline demographics of the subjects are presented in Table 1. Both groups were not significantly different (p > 0.05).

In this study comparing the impact of PRT and AT on health status in a multi-ethnic Asian population, the PRT group showed significant improvement in physical functioning, general health, vitality and mental health while the AT group demonstrated significant change in general health and vitality after an eight week supervised exercise program. Both groups also showed significant improvement in the mental component summary score. The better outcomes observed in the PRT group could be due to several possibilities: i) the novelty of resistance training, ii) the resistance training increasing subjects ability to perform activities of daily living, and iii) the perception of the exercises being less monotonous than being on a exercise machine for 20 minutes.

To the best of our knowledge, this is the first randomized trial investigating the effect of AT versus PRT on health status in an Asian population. The improvement observed in general health and vitality is in contrast to a study by Hill-Briggs et al. [17] in 149 African Americans that found that despite improvement in clinical outcomes in T2DM, there was no change in SF-36 domains. A possible explanation was that the effect of exercise on health status was short-term rather than long-term as Hill-Briggs et al. [17] studied her subjects before and after a 2-year period while we followed our patients over eight weeks. Another previous study also found that aerobic exercise training did not have any benefit on health status [18]. This study had a small sample size of nine subjects with T2DM and the aerobic sessions ranged from 20 to 45 minutes [18]. In the study by Reid et al, participants had a wider range in baseline HbA1C (6.6 to 9.9%) than our study and the exercise was also gradually increased in intensity and duration over the intervention period (15 minutes increased to 45 minutes for the aerobic group and two sets of up to eight repetitions increased to three sets of up to eight repetitions in the resistance exercise group). They observed a clinically significant improvement in the PCS favouring the resistance group compared to the aerobic group (mean difference of 2.7 points; p = 0.048) [9]. Although we did not observe an improvement in the PCS with either form of exercise in our study, we did find an improvement in the physical functioning domain of the SF-36 (a major component of PCS) in those randomized to PRT, which is in line with the observed benefits of resistance exercise in the PCS observed by Reid et al. It is possible that the larger sample size of at least 50 subjects in each group in the study by Reid et al and the longer intervention period of six months [9] allowed them to detect an effect on PCS that we did not observe.

An important strength of our study is that we made an attempt to match both exercise regimens as closely as possible for volume, frequency and rate of progression which previous studies did not control for. Another strength is that we have conducted a randomized trial design. We acknowledge that, the absence of a control group might limit our ability to assess the true effects of exercise on health status. However, we do not believe that this prevents us from comparing the benefits of AT vs PRT, which was the aim of our study. The small sample size may also have limited our ability to detect important difference in health status between the two types of exercise. Differences between the groups on role emotional exceeded a minimal important difference of 5 points [16] but did not reach statistical significance. In addition, we have used the original SF-36 rather than the SF-36 version 2 in our study as our institution held licence for the former but not the latter. The SF-36 version 2 was introduced to correct deficiencies identified in the original SF-36 and improve measurement properties to increase clarity and sensitivity (e.g. the response categories in mental health and vitality scales were reduced from six to five). Hence, we may have underestimated the effects of AT and PRT on health status. The sustained effect of exercise on health status over time and the combined effect of aerobic and resistance exercise on health status have also not been evaluated in our study and should be explored in future studies. Nevertheless, we believe that we have added new information to the sparse literature available on the impact of different forms of exercise on health status of Asian patients with T2DM.

Both aerobic and progressive resistance training improved general health and vitality subscales in SF-36, as well as the mental component summary score. Although there was no significant difference between the groups, it did appear that progressive resistance training had more beneficial effects as there were significant changes in more domains of the SF-36 than that observed for the aerobic training group.