Introduction

Type 2 diabetes mellitus is associated with reduced sensitivity of insulin receptors, making glucose uptake difficult [1]. In addition, individuals with type 2 diabetes mellitus have a higher prevalence of cardiovascular disease (CVD) [2], which is considered the main cause of mortality worldwide [3]. Approximately 32.2% of individuals with type 2 diabetes mellitus are affected by CVD, with a higher incidence of strokes and coronary artery disease [2].

Physical inactivity, in turn, is associated with the development and worsening of diseases such as type 2 diabetes mellitus and CVD [4], in addition to being considered one of the main causes of premature mortality worldwide [5, 6]. Several studies have reported a high percentage of physical inactivity in older populations [7,8,9,10]. Population-wide estimates indicate that if physical inactivity were eliminated, on average, life expectancy would increase by 0.68 (range, 0.41. 0.95) years worldwide [11]. And, specifically, in the older adults Brazilian, the fifth largest elderly population in the world [12], if physical inactivity were eliminated, approximately 5.53% of cases CVD, 6.85% of diabetes mellitus 2 and 8.91% of mortality from all causes would be avoided [13].

In this sense, the regular practice of physical activity has numerous health benefits, as it prevents the development of chronic non-communicable diseases (NCDs) and consequently reduces the risk of premature mortality [6]. In addition, regular physical activity is associated with a decrease in CVD-related biomarkers, such as homocysteine [14,15,16].

According to Neves et al. [17], homocysteine is an essential amino acid synthesized in the liver by demethylation, is associated with fasting, or due to methionine overload via transulfurization. At higher levels in the bloodstream, it generates a condition called hyperhomocysteinemia (above 15 μmol / L), directly contributing to the increased incidence of CVD [18].

Hyperhomocysteinemia is considered an independent risk factor for vascular, cerebral, coronary, and peripheral arterial disease [19, 20], and is a predictor of mortality, regardless of the risk factors [21]. According to a meta-analysis of prospective studies, an increase of 5 μmol / L in circulating levels of homocysteine raise the risk of all-cause mortality by 33.6% [22]. Thus, controlling homocysteine levels is important to reduce the risk of thrombogenesis and atherogenesis [21].

Physical inactivity has been linked with higher homocysteine ​​values [15, 23], but the effect of regular physical activity is little explored and is not well understood in the literature. To this end, the current study compared serum levels of homocysteine and cardiovascular risk factors among physically active and inactive older women with type 2 diabetes mellitus. This information is essential to confirm the benefits of regular physical activity on serum levels of homocysteine and cardiovascular risk factors, since hyperhomocysteinemia is independently linked to the development of CVD.

Methods

Study design, participants and eligibility criteria

For the primary and quantitative cross-sectional study, 80 older women with type 2 diabetes mellitus, aged between 60 and 91 years, were assessed for eligibility. The study was conducted in the city of Itajubá, in the state of Minas Gerais, which is located in the southeastern region of Brazil (Latitude 22°25′37″/Longitude 45°27′11″), occupying a territorial area of 294,825 km2. In 2010, the estimated population was 96,869 inhabitants, the demographic density 307.49 inhabitants per km2, with a human development index of 0.787.

The study involved a single meeting, in which the international physical activity questionnaire (IPAQ) was used to classify the participants into physically active and inactive groups. Blood samples were collected to determine the biochemical parameters of interest, and blood pressure was measured. Anthropometric measures, body composition, and information regarding the characteristics of the volunteers were also investigated.

The eligibility criteria adopted were: (I) age ≥ 60 years, (II) lives independently in the community, and (III) not engaged in systematic exercise programs.

The non-inclusion criteria were as follows: (I) a recent history of CVD, malignant neoplasms, or surgical procedures, (II) physical limitations (orthopedic) that interfere with body movement, (III) diagnosis of neurodegenerative diseases, (IV) refusal to perform blood collection, and (V) not interested in participating in the research. For the exclusion criteria, the following were adopted: (I) inadequate responses in the IPAQ, and (II) insufficient data.

Ethical considerations

All individuals in the study signed a free and informed consent form (ICF), accepting that the science of the research and all the benefits and risks would be disclosed during the execution of the study. The study was approved by the Ethics and Research Committee of the Centro Universitário de Itajubá (protocol 139) and is in accordance with the National Health Council.

Outcome measures

Sample characterization

Information regarding age, family income, skin color, education, smoking, and medication (to control blood glucose, cholesterol, and blood pressure) were collected by self-reporting for the purpose of characterizing the sample.

International physical activity questionnaire (IPAQ) long version

Physical activity was subjectively measured by the long version of the IPAQ, an instrument validated in older Brazilian women by Benedetti et al. [24]. The IPAQ was administered in the form of an interview by researchers experienced in the application of this instrument. Participants were asked individually in a separate room as to the frequency and time of moderate, vigorous, and walking activities performed in sessions ≥10 min at work, at home, as a form of transportation or leisure during a typical week.

The questionnaire classifies the subjects into four different levels: very active, active, irregularly active, and sedentary. However, in the present study, we divided the participants into two groups: active (very active and active) and inactive (irregularly active and sedentary).

Thus, the volunteers who performed one session of ≥20 min/session vigorous physical activity on three or more days a week, or one session ≥30 min/session of moderate physical activity and/or walking for at least 5 days a week, or one session of ≥150 min of moderate to vigorous physical activity and/or walking for at least 5 days a week were classified as physically active. Participants who did not reach this level activity were considered physically inactive.

Blood collection

The volunteers were subjected to the collection of blood samples by venipuncture (antecubital vein) to determine the concentrations of homocysteine, glycated hemoglobin, glycemia, and total cholesterol, HDL, LDL and VLDL. About 10 mL of blood was collected from each participant after a minimum fast of 12 h. The samples were separated into test tubes with ethylenediaminetetraacetic acid (EDTA) to assess plasma levels of homocysteine and glycated hemoglobin (HbA1c). The plasma was separated immediately after collection and frozen at minus 20 °C until analysis. To determine the concentrations of total cholesterol, HDL, LDL, VLDL, triglycerides, and glycemia, serum and plasma were separated from the samples. Serum was used to measure total cholesterol, HDL, LDL, VLDL, triglycerides, and fluoridated plasma to measure glucose.

Analysis of homocysteine and glycated hemoglobin

The analysis of homocysteine and HbA1c was measured using high-performance liquid chromatography (HPLC) equipment, which has a high sensitivity and reliability [25,26,27,28]. The range considered normal for the level of HbA1c is between 4% and 5.6%, levels between 5.7% and 6.4% indicate pre-diabetes, and levels of 6.5% or more indicate diabetes. The DPC-Medlab reagent was used, with values between 5.0 and 15.0 μM per liter and analytical sensitivity of 0.5 μM per liter as the normal range, considering the quantitative analysis of the peak area expressed in the chromatograms [25,26,27,28,29].

Analysis of plasma levels of cholesterol, HDL, LDL, VLDL, and glucose

The enzymatic method by end point reaction was used to analyze total cholesterol, HDL, LDL, VLDL, and blood glucose. After collection, samples were screened, and after analysis, no collection was discarded. To analyze the plasma levels of total cholesterol, HDL, LDL, VLDL, and glucose, a spectrophotometer (Biospectro Benfer SP-22®) (reading at 505 nm) and a Cholesterol-PP Kit Cat.460 (Labtest) was used, as well as a PAP Glucose Kit Liquiform (Labtest), centrifuge (Benfer-BMC macro centrifuge), and a water bath (Quimis 021/4®) [30,31,32].

Each sample was subjected to a technical test according to the manufacturer’s procedure, respecting volumes, color reagent sequence, and zeroing the absorbance of white, which is the color reagent. After mixing, the solutions were placed in a water bath for 10 min at 37 °C, as determined by the manufacturer. Then, according to the wavelength of each analyte, the absorbance of the standard and sample of each patient, and the internal control of each analyte were read (total cholesterol wavelength values = 500 nm, HDL = 500 nm, triglycerides = 505 nm and glucose = 505).

After reading the absorbances, the concentration of the samples was calculated by comparing them to the standard. Only when the value of internal control was within the parameters of Levey Jenning were the results of the volunteers released [30,31,32].

Systemic blood pressure

Aneroid equipment (Welch-Allyn, USA) and a cuff compatible with the volunteers’ arm circumference were used to determine blood pressure. The volunteers remained seated in a chair for approximately 5 min in a quiet and peaceful environment. Then, with the forearm and back of the hand resting on a table, and the cuff at heart level, their blood pressure was measured. The collection was carried out in the morning by the same equipment and evaluator.

Cardiovascular risk

The Framingham score, proposed by D’Agostino et al. [33], was used to estimate the risk of CVD (heart failure, coronary artery, cerebrovascular, and peripheral artery disease) within 10 years. To determine this score, specific scores were assigned to age, HDL-cholesterol, total cholesterol, systolic blood pressure (treated or untreated), presence of diabetes mellitus, and smoking. The scores obtained for each of these parameters were then summed to obtain a final score. A higher score represented a greater cardiovascular risk.

Anthropometric and body composition assessment

Anthropometric and body composition measures were recorded according to the instructions of the International Society for the Advancement of Kinanthropometry [34]. Height was measured using a solid PVC stadiometer (Seca® brand), with a scale from 0 to 2.20 m and a resolution of 0.1 cm. Body mass was obtained using a Filizola® scale (Filizola, Brazil) with a capacity of up to 180 kg and fractions of 100 g, with a plate base and a rubber mat [35]. Based on these parameters, the body mass index (BMI) was calculated using the formula: body mass/height2 [35, 36].

Skin folds were determined using a Cescorf® compass. The protocol adopted to verify the fat percentage was the specific equation for older people from Durnin and Womersley with four skin folds (subscapular, tricipital, suprailiac and bicipital). The formula used was as follows: {1.1567–0.0717 * Log10 [Σ (subscapular, tricipital, suprailiaca, bicipital)]} [37, 38]. Meanwhile, lean mass was determined by weight – fat mass. In addition, a semi-rigid measuring tape from the Cescorf® brand was used to assess waist and hip circumference.

Sample size

The sample was calculated using the GPower version 3.1.9.2 (Uiversität Kiel, Germany). The estimation was based on first 20 volunteers homocysteine data (Cohen’s d = 0.93). It was indicated that 21 subjects per group were needed for a power of 90% (1-β), 0.05 alpha value (two-tailed), and an allocation ratio of 1.

Data analysis

The research data were analyzed using descriptive statistical techniques, including mean, standard deviation, and absolute frequency (percentage). The normality of the data was verified by the Shapiro-Wilk test, and the variance and outliers were analyzed. Frequency (histogram), symmetry, or asymmetry were analyzed to establish whether the data were parametric or nonparametric. The characteristics of the participants were compared using the independent t test and chi-square (χ2) test.

To compare the dependent variables between the groups, an one-way ANOVA test was used. The partial eta-square (ηp2) method was used to estimate the size of the effect. The effect size was interpreted as small (ηp2 < 0.035), medium (ηp2 ≥ 0.035 or < 0.140), or large (ηp2 ≥ 0.140) [39]. Statistical analyses were performed using the Statistical Package for the Social Sciences (SPSS) version 20.0, considering the rejection of the null hypothesis of 5% (p < 0.05) for all results.

Table 1 Descriptive characteristics of the participants (n = 59)
Full size table

Results

The allocation diagram of the study participants is shown in Fig. 1. Initially, 80 women (age: 65 ± 4 years) were randomly recruited from the community and assessed for eligibility. Fifteen volunteers were not included because they were not interested in participating (n = 10) or refused to perform blood collection (n = 5). In addition, six older women were excluded due to insufficient data. In this way, 59 volunteers were analyzed, where, following the principles established by IPAQ, 21 participants were allocated to the active group (244.10 ± 99.27 MVPA), and 38 to the inactive group (92.16 ± 19.19 MVPA). The average age of the volunteers was 64.3 years, with an average BMI of 29.47 kg/m2. The majority had a family income >3 minimum wages (86.44%), declared to have a white skin color (84.74%), education >8 years (96.61%), not smoking (89.84%), and using medications to control blood pressure (91.52%), cholesterol (81.35%), and blood glucose (100.0%). There were no significant differences in age, height, body mass, BMI, sociodemographic aspects, and medication use between the active and inactive groups (Table 1).

Fig. 1
figure 1

Participants group allocation diagram.

Full size image

The anthropometric parameters, CVD factors, blood pressure, and diabetes mellitus factors are shown in Table 2. A significant difference was found only for variable waist circumference (F = 2.95, p = 0.013, ηp2 = 0.049).

Table 2 Changes in anthropometric parameters, cardiovascular disease factors, blood pressure, and diabetes factors in the active and inactive groups
Full size table

Figure 2 shows a comparison of homocysteine levels and cardiovascular risk scores between the active and inactive groups. A significant difference was found for homocysteine, with higher values in the inactive group (F = 17.79, p < 0.001, ηp2 = 0.238). The same was observed for the cardiovascular risk score, where the active group had a lower score (F = 15.84, p = p < 0.001, ηp2 = 0.217).

Fig. 2
figure 2

Comparison of serum homocysteine (1A) and cardiovascular risk score (2A) between the active and inactive groups

Full size image

Discussion

In this study, we aimed to compare serum homocysteine levels and cardiovascular risk factors among physically active and inactive older women with type 2 diabetes mellitus. Our main findings indicate that active older women with type 2 diabetes mellitus have lower serum concentrations of homocysteine, lower cardiovascular risk scores, and a lower waist circumference. This corroborates our initial hypothesis, indicating better responses in the parameters of homocysteine and cardiovascular risk in the active individuals with type 2 diabetes mellitus active compared to inactive individuals. Our findings on serum homocysteine levels go in the same direction as the results found in the studies by Buckner et al. [40], Hellgren et al. [41], Unt et al. [42], and Loprinzi and Cardinal [43], who analyzed the relationship between physical activity and homocysteine.

High levels of homocysteine may be associated with increased body adiposity, inadequate nutrition, sedentary behavior, and physical inactivity itself [44]. Higher serum homocysteine levels in women than in men are due to greater muscle mass and creatine phosphate synthesis in men, in addition to the estrogen-reducing effect and variations in the rate of synthesis of homocysteine compared to men [45,46,47,48]. It is important to note that in older adults, high levels of homocysteine are also related to intestinal malabsorption, a slowing down of the metabolism, impaired renal metabolism, and reduced intake of vitamin B12.

For Dankner et al. [49], physical activity can reduce homocysteine independently of B vitamins, corroborating these findings, the study by Alomari et al. [23] suggests that regular physical activity may be an additional treatment strategy to reduce levels of homocysteine, regardless of lifestyle and B vitamins.

Therefore, the encouragement of doctors and health professionals to partake in the regular practice of physical activity/physical exercise for cardiovascular health is of great importance, since the adoption of healthy lifestyle habits, such as increased physical activity and decreased sedentary behaviors, is able to decrease the risk of diabetes type 2 mellitus, stroke, cardiac events, and CVD [50, 51] improving quality of life and decreasing the risk of death, and be considered a public health intervention [52].

Meta-analyses indicated that individuals who practice the equivalent of 150 min per week of moderate-intensity leisure physical activity have a 15% to 20% lower risk of developing coronary artery disease than those who do not [53, 54]. Our findings add to the findings of the meta-analyses, indicating that physically active older adults with diabetes type 2 mellitus by IPAQ had lower cardiovascular risk compared to inactive individuals.

Our study also found a significant difference in waist circumference between the two groups. Studies indicate that waist circumference is a good parameter for tracking CVD [55]. In a study published by the Scottish Health Survey, carried out between the years 1998 and 2008, an increase of 5 to 10 cm in the waist circumference was observed in both sexes between 50 to 70 years [56], an age equivalent to the volunteers in our study, suggesting an increased risk of CVD. Additionally, the simultaneous presence of excessive sedentary time (> 10 h/day) and noncompliance with recommendations for physical activity (<150 min/week of physical activity of moderate to vigorous intensity MVPA) were related to greater waist circumference by Figueiro et al. [57]. On the other hand, Ribeiro et al. [10] observed the performance of 150 min per week in sessions ≥10 min was linked to lower chances of older Brazilian adults having a waist circumference above the recommended parameters.

The study by Loureiro et al. [58] recommends the adoption of simple methods such as waist circumference, as they are low-cost and noninvasive. They can be adopted in clinical practice and in epidemiological studies with older adults and adults, and this could contribute to the screening and early identification of risk factors, thus enabling actions and strategies for the prevention and control of CVD.

For Del Pozo-Cruz et al. [59], in a meta-analysis of objective assessment studies, reallocating sedentary time to light physical activities or MVPA can result in reductions in waist circumference. The equal exchange of sedentary behavior with MVPA can lead to further reductions in fasting glucose, all-cause mortality, and increased HDL-c levels. The replacement of sedentary behavior by light physical activities may be a possible alternative intervention strategy, as it is more viable and less challenging than more strenuous activities, therefore recommended for older adults.

In our study, the other anthropometric parameters related to cardiometabolic risk and obesity in individuals with type 2 diabetes mellitus did not show statistical differences between the active and inactive groups. Individuals in the active group did not perform systematic control of their diet. For Carbone et al. [60], physical activity is important in controlling variables related to obesity; however, healthy eating habits and diet control over a long period are necessary. We emphasize that obesity is a multifactorial chronic disease that is related to genetic factors or disturbances in energy metabolism, related to insulin resistance with its receptors, leading to type 2 diabetes mellitus.

There was also no significant difference in the factors of isolated cardiovascular diseases. These findings are also corroborated by studies that used systematic exercise. In the study by Silva et al. [61] and Silva et al. [62], aerobic and strength activities, respectively, were not able to significantly change cardiometabolic risk factors. In both studies, the weekly frequency of activities was low and the intensity was mild. In another study, when we compared active individuals with type 2 diabetes mellitus with the control, we also found no differences in the main cardiometabolic risk variables [63]. Systolic and diastolic blood pressure also did not show statistical differences between the active and inactive groups.

Based on the above findings and understanding the importance and relevance of the subject addressed in this study for quality of life and decreased mortality in the older population with type 2 diabetes mellitus, we propose that activity programs be controlled to control cardiometabolic risk factors systematized and oriented physics, with adjustment of weekly frequency, volume, and intensity of exercises, respecting biological individuality.

In the present study, we emphasize that the isolated analysis of cardiovascular risk factors can lead to misinterpretations due to the responses of each individual to the stimuli applied. On the other hand, when grouping these factors, we see differences in favor of the active group. This finding has great implications for public health, since the score used in this study uses variables commonly investigated in clinical practice and may be a viable alternative for the early detection of cardiovascular risk.

It is observed, in a qualitative analysis, that physical activity tends to decrease in older adults, it also suggests that the results of biochemical parameters and body composition tend to be better with a control of the diet, since, the high values of glycated hemoglobin may be related to an unbalanced diet, the last being a limitation of the study, since nutritional diet was not analyzed and systematized by a professional in the area of nutrition for the inclusion of people in the groups.

The study also has other limitation that need to be emphasized, this was a cross sectional study, therefore, it is not possible to determine the causal action of the event. However, it is emphasized that the variables investigated in the present study have great applicability in primary and secondary healthcare in most diverse populations. There is a need for studies with nutritional control protocols to analyze the effects of physical activity plus the systematization of the diet in the studied population.

Finally, the study compared serum homocysteine levels and cardiac risk factors between active and inactive elderly people with type 2 diabetes mellitus. It was concluded that active older women with type 2 diabetes mellitus demonstrated a reduction in homocysteine levels and a lower cardiovascular risk score when compared to inactive older women. The analysis of isolated anthropometric and biochemical variables did not show any differences.