Fig.1 – PRISMA flow chart.
Study selection
The search strategy identified 600 potentially relevant records. After exclusion of duplicates and screening of titles and abstracts, 20 full-text papers were retrieved for more detailed review. A total of seven and five clinical trials were included in the qualitative synthesis and quantitative analysis, respectively. The process of the study selection and the reasons for exclusions are summarized in Fig. 1.
Study characteristics
The main characteristics of the included studies are presented in Table 1. The studies were published between 2013 and 2016 and included a total 120 patients with T2D (57% females) with an average age of 61.7 ± 6 years.
The duration of the studies ranged from 12 to 16 weeks for the T2D groups. The included studies used different HIIT and MICT protocols. The exercise types, session duration, and intensity varied widely between studies. All exercises were performed post-meal and were monitored by direct supervision or objective measures using heart rate monitors or accelerometers. Moreover, the subjects were instructed to not alter their dietary intake habits and medications throughout the study period.
Three studies on T2D, reported mean adherence rates of 87% and 90% for HIIT and MICT, respectively. In the study by Terada et al., one participant was excluded from fasting glucose and HbA1c analysis due to the discontinuation of medication.
Effects of the interventions
VO2max
Maximal O2 consumption was reported by three studies including 89 participants with T2D [24,25,28] (Fig. 2A). The mean difference in the VO2max was 3.02mL/kg/min (95% confidence interval [CI] 1.42 – 4.61, I2 = 0%), significantly favoring HIIT (p < .001).
Fasting glucose
Baseline and post-intervention fasting glucose were reported by four studies [24–26,30] that included a total of 82 participants with T2D (Fig. 2B). There was no difference in fasting glucose between HIIT and MICT [WMD = 0.11 (95%CI: 0.45 – 0.67, I2 = 0%, p = .70)].
HbA1c
All studies on T2D evaluated HbA1c in a total of 119 participants [24–26,38,30]. There was no difference between interventions HIIT and MICT [WMD = -0.17 (95%CI: 0.36 – 0.02, I2 = 0%, p = .07)] (Fig. 2C).
Blood pressure
Three studies on T2D, with a total of 89 participants, observed no difference in systolic [WMD = 2.92 (95%CI 7.62 – 1.78, I2 = 0%, p = .22)] or diastolic blood pressures [WMD = 2.14 (95%CI 4.37 – 0.09, I2 = 0%, p = .06)] between HIIT and MICT [24,25,28].
Total cholesterol
Four studies on T2D, with a total of 83 participants [24–26,30], observed no difference in total cholesterol concentrations between HIIT and MICT groups [WMD = 0.16 (95%CI 0.68 – 0.35, I2 = 40%, p = .50)].
HDL and LDL cholesterol
Four studies on T2D, with a total of 83 participants [24–26,30], found no difference between the interventions in HDL [WMD=0.07 (95% CI 0.06 – 0.19, I2 =47%, p=.29)] and LDL [WMD = 0.06 (95%CI 0.41 – 0.28, I2 = 67%, p = .71)] cholesterol levels.
Triglycerides
Four studies on T2D, with a total of 83 participants [24–26,30], observed no difference in triglycerides between HIIT and MICT [WMD = 0.14 (95%CI 0.27 – 0.55, I2 = 27%, p = .49)].
BMI
All the studies on T2D evaluated BMI in a total of 120 particitpants [24–26,28,30]. There was no difference in BMI [WMD = 0.62 (95%CI 1.32 – 0.08, I2 = 0%, p = .08)] between HIIT and MICT.
Waist-to-hip ratio
Four studies on T2D, with a total of 83 participants [24–26,30] reported no difference in waist-to-hip ratio between HIIT and MICT [WMD = 0.02 (95%CI 0.02 – 0.05, I2 = 69%, p = .33)].
Waist circumference
Two studies on T2D [26,30], with a total of 31 participants, reported decreased waist circumference after HIIT (102.6 ± 7. 2 vs. 102.2 ± 6.9) [26] (109.8 ± 4.3 vs. 109.6 ± 4.6) [28], and MICT (116 ± 6.7 vs. 114.3 ± 8.9) [27] (108.5 ± 2.1 vs. 107.9 ± 2.3) [28], but no significant difference between groups.
Risk of bias
The final assessment of the risk of bias in the included studies is shown in Table 2. All evaluated papers included a statement on the randomization method; however, an adequate method to generate the random sequence was reported in only one study [26]. Most studies were scored as having a low risk of bias for ‘‘incomplete outcome data”. In the other domains, the unclear risk of bias predominated. Low-quality evidence was judged according to the GRADE for HbA1c and VO2max outcomes and was very low quality for fasting glucose outcome (Table 3).
Discussion
This is systematic review compare the effects of HIIT versus MICT on functional capacity and physiological markers in adults with T2D. The main finding of this meta-analysis was that the VO2max values significantly increased in individuals with T2D who underwent HIIT [WMD = 3.02 (95%IC 1.42 – 4.61), I2 = 0%, p = .0002] in comparison to MICT. In relation to fasting glucose, HbA1c, blood pressure, BMI, total cholesterol, HDL, LDL, triglycerides, and waist- to-hip-ratio, the two exercise modalities induced similar effects.
Previous meta-analysis showed that HIIT and MICT effectively improvement functional capacity, as evidenced by increased VO2max in different populations [16,18,31–34]. However, HIIT is likely to elicit greater increases in VO2max that MICT in patients with cardiometabolic disorders [16,18,32,34] and in healthy individuals [31,33]. The superiority of HIIT for aerobic fitness has important clinical implications given that VO2max is a stronger predictor of cardiovascular risk [35] and its improvement is associated with reduced cardiovascular disease morbidity and mortality and diabetes prevalence. The mechanisms involved in the superiority of HIIT may be due to changes in the stroke volume of the heart induced by increased cardiac contractility [38,39] and skeletal muscle oxidative capacity and changes in glucose transport, which improves mitochondrial function to generate more ATP, thus increasing aerobic capacity.
It is important to consider that the non-difference between interventions on fasting glucose and HbA1c may be related to the small sample size of the primary studies. Moreover, the intake of differences antihyperglycemic drugs may be considered as a confounding factor that mask the effects of exercise on metabolic outcomes, and sex-related differences in the metabolic response [45], as well as the inclusion of active individuals with low levels of physical activity in most of the primary studies [25,26,30]. However, any reduction in HbA1c is likely to reduce the risk of macrovascular and microvascular complications in T2D [46].
A recent meta-analysis also reported no difference between HIIT and MICT on fasting glucose level [17,18] and HbA1c [17] in a group of patients with T2D. On the other hand, the analysis by Liubaoerjijin et al. observed greater improvement in HbA1c with HIIT [WMD = 0.23 (95%CI 0.43 to 0.02), I2 = 0%, p = .03] [17]. These findings could not be confirmed because the systematic review by Jelleyman et al. utilized broad inclusion criteria and the authors included individuals with metabolic syndrome and T2D, which demonstrate different metabolic responses, in the same subgroup. Thus, the feasibility and efficacy for glucose regulation between the two exercise modalities require further investigation.
Both modalities, HIIT and MICT, showed similar effects on systolic and diastolic blood pressures in individuals with T2D. Increased blood pressure is common among patients with diabetes and is considered a strong risk factor for atherosclerotic cardiovascular disease, heart failure, and microvascular complications [49]. However, only in the study of Karsoft et al. the authors reported the use of antihypertensive by the subjects, who were suspended for five day prior to pre- and post-intervention tests. The mechanisms involved in the blood pressure decline mediated by exercise training may be improvements in peripheral vascular structure and function, with increased popliteal artery dispensability [42] and endothelial function [50].
Regarding the lipid profiles of individuals with T2D, the results of this meta-analysis suggest that there is no significant difference between interventions. This is likely because the individuals were within the normal ranges of total cholesterol, HDL, and LDL levels, which did not change greatly after exercise. Moreover, the fact that there was no diet modification or control may attenuate the beneficial effects of exercise on lipid profiles. However, comparing the total cholesterol, HDL, LDL, and triglyceride levels between HIIT and MICT groups showed heterogeneities of 40%, 47%, 67%, and 27% in the I2 test, respectively. This may be due to methodological differences between studies, including exercise duration, frequency, and intensity.
Most individuals with T2D included in the studies were considered overweight (25–29.99 kg/m2) and obese (30kg/m2) [51]. High BMI is associated with an increased risk for developing T2D [52,53] and cardiovascular complications [53,54]. Moreover, there is a relationship between BMI at the time of a diabetes diagnosis and the risk of death [1,55,56]. In relation to the decrease of the BMI, considering the hour of training time, the individuals of the HIIT group presented similar effects to the MICT group, but with a shorter training time. Thus, the tendency of the HIIT to reduce BMI with less training time may be important to public health and requires further investigation.
Waist circumference and waist-to-hip ratio are indicators of central obesity and have also been associated with T2D risk [52,57]. In this meta-analysis it showed similar effect between HIIT and MICT for the waist-to-hip ratio; however, significant heterogeneity was also observed (I2 = 69%), which may be partly due to the lower methodological quality of the included studies. The waist circumference was reported in studies [26,27,30] and did not demonstrate the superiority of some type of exercise; further studies are needed to explore the relationship between central adiposity, diabetes, cardiovascular disease, and increased long-term cardiometabolic risk.
According to the Cochrane risk of bias tool [22], most items (77.8%) were classified as unclear risk of bias due to insufficient information or uncertainty about the potential for bias; the VO2max, fasting glucose, and HbA1c outcomes were considered serious by GRADE because blinding and randomization were not properly performed in most of the studies. This finding demonstrated that need for well-designed future studies.
This meta-analysis has several limitations. First, only five studies with T2D and were included, with small sample size( 20) and most of the studies had methodological limitations. The studies used different HIIT protocols with a variety of exercise modalities, intervals, intensities, volumes, duration, and different ways of determining the intensities of HIIT or MICT. Another challenge was the lack of clear information on some data. With the purpose to provide the best available evidence for clinical practice, our results demonstrate a need for more research with greater methodological rigor and with larger sample sizes in order to strengthen the quality of current evidence and determine which exercise modality most positively affects cardiometabolic markers in individuals with T2D.
Conclusion
High-intensity interval training has the potential to be used as a treatment modality for individuals with T2D. This intervention induces cardiometabolic adaptations similar to those of MICT and provides greater benefits to functional capacity in patients with T2D. However, strong supporting research is further needed.
