Uncomplicated diabetes does not accelerate age-related sarcopenia

Abstract

Background: Diabetes is reported to accelerate sarcopenia (age-related loss of muscle mass and function). We aimed to assess muscle mass and strength in elderly diabetics, elderly non-diabetics, younger diabetics and healthy subjects, and to define correlates of muscle mass and strength in these subjects.

Methods: Sixteen elderly diabetics, 16 younger diabetics, 16 elderly non-diabetics and 18 younger non-diabetics were included. Elderly and diabetic subjects were first evaluated with exercise testing. Isokinetic leg extension and flexion tests were performed using a Cybex 350 dynamometer. Muscle mass was calculated using bioelectric impedance analysis.

Results: Muscle mass was similar between all groups; however, muscle strength was significantly lower in diabetic and non-diabetic elderly subjects compared with younger diabetic subjects and non-diabetics. Muscle strength was positively correlated with albumin, metabolic equivalent and hemoglobin, and inversely correlated with age, HbA1c, functional capacity and CRP. Independent correlates of muscle strength were age and hemoglobin. There was no clinically significant correlate of muscle mass. Presence or duration of diabetes was not associated with muscle mass or strength.

Conclusions: Uncomplicated diabetes does not seem to accelerate aging-related muscle mass or strength loss. Exercise test parameters may be useful markers in the screening of sarcopenia.

• Aging
• diabetes mellitus
• exercise test
• isokinetic muscle strength
• sarcopenia

Introduction

Sarcopenia is defined as age-related loss of muscle mass and function [1]. Due to the structural and functional regression of muscles, mobility decreases, frailty, falls and morbidity increases with aging. Interactions of neurological, environmental, nutritional, hormonal and genetic factors may be important in the development of the sarcopenia [2]. People who have diabetes mellitus tend to have an accelerated aging process. Thus, loss of muscle mass and function may occur earlier with the influence of various factors. Older adults with type 2 diabetes may have a lower muscle mass than older adults without diabetes [3]. Diabetic patients with renal insufficiency show accelerated muscle function loss [4,5]. The mechanisms for the loss of skeletal muscle strength in older adults with diabetes are unclear. Furthermore, studies regarding sarcopenia in patients with uncomplicated diabetes are lacking. We aimed to assess the presence of sarcopenia in elderly diabetic patients and to compare their results with younger diabetic patients and elderly and younger non-diabetic subjects.

Methods

Study population

Sixteen elderly (≥65 years old) diabetics, 16 younger diabetics, 16 elderly non-diabetics and 18 younger non-diabetics (healthy control group) were enrolled to our study. Significant conditions associated with loss of muscle mass like malignancy, chronic infections, long-term immobilization, significant anemia (hemoglobin <11 g/dl) and renal failure (creatinine >1.5 mg/dl or glomerular filtration rate [GFR] <30 ml/min/1.73 m²) were excluded from this study. Diabetic subjects were questioned for the presence of symptomatic peripheral neuropathy and none of them stated having such symptoms. However, we did not perform a fundoscopic examination during this study and did not record their previous fundoscopy findings. Furthermore, individuals with findings of ischemia in electrocardiography or positive results in exercise test were also excluded. The study was conducted according to the guidelines laid down in the Declaration of Helsinki and was approved by the Ethics Committee of Istanbul School of Medicine of Istanbul University with file number 2010/322-48. Written informed consents were obtained from each of the participants.

Measurements

The subjects were initially evaluated with exercise testing and electrocardiography. The exercise test was carried out in accordance with the Bruce protocol using Burdick Quest version 4.2 and GE Case version 6.51 effort devices. Only the younger non-diabetic group did not undergo an exercise test. We recorded metabolic equivalent (MET), which is routinely used in the exercise test, functional capacity (FC) and maximum heart rate (MHR) values. Height, weight and waist circumference (WC) were measured, and body-mass index (BMI, kg/m²) was calculated. Body surface area (BSA) was calculated with the Dubois and Dubois formula (weight[kg]^0.425 × height[m]^0.725 × 0.007184) [6]. After overnight-fasting, glucose, hemoglobin, creatinine, hemoglobin A1C, albumin, C-reactive protein and 25-hydroxy vitamin D levels were determined through blood samples obtained in the morning. Microalbuminuria was calculated as microalbumin/creatinine ratio from morning spot urine sample. GFR was calculated using the short Modification of Diet in Renal Disease (186 × creatinine^−1.154 × age^−0.203) formula for each case and was corrected according to BSA [7].

Fat-free mass (FFM) was measured by means of bioelectrical impedance analysis (BIA) using BC-532 body analysis monitor for personal use. FFM was corrected relative to height squared to determine the corrected fat-free mass (CFFM). CFFM values 2 standard derivations below the mean value of the younger non-diabetic group were defined as sarcopenic muscle mass (SMM). Isokinetic leg extension and flexion tests from both lower extremities were performed using a Cybex 350 isokinetic dynamometer at speeds of 60°/s and 90°/s. The test protocol comprised three trials and four retests for both speeds. The maximum strength value produced was detected as peak torque in Newton-meters (Nm). Average extensor muscle strength (AEMS), average flexor muscle strength (AFMS) and average muscle strength (AMS) measurements were calculated. AMS values which were 2 standard deviations below the mean value of the control group were defined as sarcopenic muscle strength (SMS).

Statistical analysis

Distributions of continuous variables were determined with the Shapiro–Wilk test. For inter-group comparisons, Chi-square and Fischer’s exact tests were used for categorical variables, Student’s t test was used for continuous variables with normal distribution, and the Mann–Whitney U test was used for ordinal variables and continuous variables with skewed distribution. In correlation analysis, Pearson’s test was used for data with normal distribution and Spearman’s test was used for ordinal variables and continuous variables with skewed distribution. Multivariate linear regression analysis was performed for significantly correlated variables in the univariate analysis to determine independent correlates of AMS. In correlation analysis, r values between 0.1 and 0.3 were considered as weak, 0.3 and 0.5 as moderate and above 0.5 were defined as strong correlations. In all analyses, a two-sided p value < 0.05 was accepted as statistically significant.

Results

The mean age of the control group was 28.7 ± 3.9 years (range: 20–35 years). The mean CFFM was 19.6455 ± 1.36911 kg/m² for the control group, and SMM was defined as CFFM <16.90728 kg/m². The mean AMS was 142.986 ± 23.7142 in the control group, and SMS was calculated as <95.5576 Nm. SMM was present only in two subjects, one from the elderly non-diabetics and one from the younger non-diabetics. Furthermore, these were the only two subjects who had a BMI <20 kg/m² in this study. SMS was present in five elderly non-diabetics, three elderly diabetics, one younger diabetic and none of the younger non-diabetics. The elderly non-diabetic subject with SMS had a normal AMS (101.1 Nm).

Height, weight, WC, BMI, muscle mass and CFFM were similar in all three study groups (Table 1). Albumin was higher in the younger diabetics compared with the elderly groups. HbA1c levels were higher in the diabetic groups compared with the elderly non-diabetic group. GFR levels were lower in the diabetic groups compared with the elderly non-diabetic group. MET was higher in the younger diabetics compared with elderly diabetics. FC and maximum heart rate were similar between the groups. The healthy control group had lower WC and BMI, HbA1c, and creatinine and higher hemoglobin, albumin and GFR levels compared with the elderly diabetics (Table 2). The healthy control group had higher hemoglobin and albumin levels compared with the elderly non-diabetics. The healthy control group had lower HbA1c and higher albumin and GFR levels compared with the younger diabetics.

Table 1. Demographic, clinical, bioelectrical impedance analysis, effort test and laboratory data of diabetic and elderly groups.

	Younger DM (n = 16)	Elderly NDM (n = 16)	Elderly DM (n = 16)	p*
Age (year)	57.3 ± 5.1	71.2 ± 5.7	69.4 ± 4.1	<0.001, <0.001, 0.58
WC (cm)	98 ± 6	98 ± 10	106 ± 17	0.9, 0.12, 0.08
Height (cm)	171.6 ± 4.1	170.3 ± 5.8	169.5 ± 5.7	0.79, 0.52, 0.9
Weight (kg)	82 ± 7	79 ± 11	84 ± 8	0.57, 0.9, 0.32
BMI (kg/m^2)	28 ± 2	27 ± 4	29 ± 3	0.83, 0.57, 0.26
Muscle mass (kg)	59 ± 4.9	56 ± 4	58 ± 5	0.17, 0.84, 0.42
CFFM (kg/m^2)	20.1 ± 1.8	19.3 ± 1.5	20.3 ± 1.7	0.46, 0.99, 0.32
C reactive protein (mg/l)	2.1 ± 1.57	2.08 ± 2.69	3.6 ± 3.2	0.4, 0.1, 0.15
HbA1c (%)	7.3 ± 1.8	5.7 ± 0.3	6.8 ± 1	<0.001, 0.53, <0.001
Urine microalbumin (mg/l)	134.9 ± 264.2	16.5 ± 15.1	62.9 ± 134.7	0.09, 0.8, 0.06
Microalbuminuria present (%)	43.8	12.5	30.8	0.1, 0.7, 0.4
25-OH D vitamin (ng/ml)	17.8 ± 5.6	21.7 ± 10.6	20.7 ± 13.5	0.7, 0.9, 1
Hemoglobin (g/dl)	13.9 ± 1	13.8 ± 1.4	13.2 ± 1.2	1, 0.5, 0.6
Creatinine (mg/dl)	0.9 ± 0.1	1 ± 0.1	1.08 ± 0.2	1, 0.2, 0.2
Albumin (g/dl)	4.5 ± 0.2	4.3 ± 0.2	4.3 ± 0.2	0.028, 0.034, 1
GFR (ml/min/1.73 m^2)	58 ± 11	73.8 ± 11.8	49.3 ± 11.2	<0.001, 0.09, <0.001
Functional capacity	1 ± 0	1.1 ± 0.3	1.4 ± 0.5	0.72, 0.12, 0.28
Metabolic equivalent	10 ± 1.7	8.7 ± 2.3	7.3 ± 1.9	0.26, 0.008, 0.26
Maximum heart rate	95.1 ± 5	94.8 ± 6.8	97.5 ± 4.5	0.82, 0.18, 0.39

NDM non-diabetic, DM diabetic, WC waist circumference, BMI body mass index, CFFM fat-free mass corrected for height squared, GFR glomerular filtration rate.

*Comparison between younger DM and elderly NDM, younger DM and elderly DM, and elderly NDM and elderly DM groups.

Bold values indicate statistical significance (p < 0.05).

Table 2. Demographic, clinical, bioelectrical impedance analysis and laboratory data of the control group in comparison with other groups..

	YNDM (n = 18)	YD (n = 16)	END (n = 16)	ED (n = 16)	p*
Age (year)	28.7 ± 3.9	57.3 ± 5.1	71.2 ± 5.7	69.4 ± 4.1	<0.001, <0.001, <0.001
WC (cm)	91.1 ± 8.2	98 ± 6	98 ± 10	106 ± 17	0.2, 0.3, 0.002
BMI (kg/m^2)	25.9 ± 3	28 ± 2	27 ± 4	29 ± 3	0.2, 0.6, 0.02
Muscle mass (kg)	59.7 ± 5.5	59 ± 4.9	56 ± 4	58 ± 5	0.9, 0.2, 0.8
CFFM (kg/m^2)	19.6 ± 1.4	20.1 ± 1.8	19.3 ± 1.5	20.3 ± 1.7	0.8, 0.9, 0.6
C reactive protein (mg/l)	1.6 ± 1	2.1 ± 1.57	2.08 ± 2.69	3.6 ± 3.2	0.9, 0.9, 0.07
HbA1c (%)	5.4 ± 0.3	7.3 ± 1.8	5.7 ± 0.3	6.8 ± 1	<0.001, 0.8, 0.001
Urine microalbumin (mg/l)	16.8 ± 22.3	134.9 ± 264.2	16.5 ± 15.1	62.9 ± 134.7	0.2, 1, 0.9
25-OH D vitamin (ng/ml)	22 ± 8.9	17.8 ± 5.6	21.7 ± 10.6	20.7 ± 13.5	0.6, 1, 1
Hemoglobin (g/dl)	14.9 ± 1	13.9 ± 1	13.8 ± 1.4	13.2 ± 1.2	0.077, 0.048, 0.001
Creatinine (mg/dl)	0.9 ± 0.1	0.9 ± 0.1	1 ± 0.1	1.08 ± 0.2	0.4, 0.4, 0.004
Albumin (g/dl)	4.7 ± 0.2	4.5 ± 0.2	4.3 ± 0.2	4.3 ± 0.2	0.02, <0.001, <0.001
GFR (ml/min/1.73 m^2)	72.3 ± 14.3	58 ± 11	73.8 ± 11.8	49.3 ± 11.2	0.006, 1, <0.001

YND younger non-diabetics, YD younger diabetics, END elderly non-diabetics, ED elderly diabetics, WC waist circumference, BMI body mass index, CFFM fat-free mass corrected for height squared, GFR glomerular filtration rate.

*Comparison between YND and YD, YND and END, and YND and ED groups, respectively.

Bold values indicate statistical significance (p < 0.05).

Duration of diabetes was significantly longer in younger diabetics compared with elderly diabetics (13 ± 8.6 versus 6 ± 4.6 years, p = 0.009). Although albuminuria level and microalbuminuria rate tended to be higher in the younger diabetic group compared with the elderly diabetic and elderly non-diabetic groups, the differences were not significant.

The AMS was significantly lower in the elderly non-diabetic and elderly diabetic subjects compared with younger diabetics (Table 3). In the elderly diabetic and non-diabetic groups, AMS was similar. The healthy control group had higher AMS values compared with the other three groups (Table 4). These results were similar for both flexor and extensor muscle strengths.

Table 3. Laboratory data and muscle strength measurements of diabetic and elderly groups.

Muscle strength	Younger DM	Elderly NDM	Elderly DM	p*
Right extensor (60°/s)	143 ± 33	123 ± 22	128 ± 23	0.07, 0.21, 0.32
Left extensor (60°/s)	146 ± 37	131 ± 23	128 ± 25	0.14, 0.19, 0.76
Right flexor (60°/s)	102 ± 18	90 ± 16	93 ± 11	0.05, 0.1, 0.71
Left flexor (60°/s)	103 ± 17	88 ± 16	89 ± 21	0.02, 0.033, 0.94
Right extensor (90°/s)	120 ± 28	109 ± 20	110 ± 17	0.11, 0.21, 0.56
Left extensor (90°/s)	131 ± 31	113 ± 23	116 ± 21	0.1, 0.25, 0.84
Right flexor (90°/s)	97 ± 18	88 ± 15	87 ± 14	0.12, 0.08, 0.96
Left flexor (90°/s)	95 ± 16	78 ± 16	85 ± 19	0.004, 0.016, 0.69
Average extensor	135 ± 30	119 ± 20	121 ± 20	0.09, 0.29, 0.67
Average flexor	99 ± 16	86 ± 14	89 ± 16	0.02, 0.029, 0.8
General average	117 ± 21	103 ± 16	105 ± 17	0.038, 0.044, 0.67

NDM non-diabetic, DM diabetic.

*Comparison between younger DM and elderly NDM, younger DM and elderly DM, and elderly NDM and elderly DM groups, respectively.

Bold values indicate statistical significance (p < 0.05).

Table 4. Muscle strength measurements of the control group in comparison with other groups.

Muscle strength	YND	YD	END	ED	p*
Right extensor (60°/s)	179 ± 37	143 ± 33	123 ± 22	128 ± 23	0.008, <0.001, <0.001
Left extensor (60°/s)	183 ± 36	146 ± 37	131 ± 23	128 ± 25	0.006, <0.001, <0.001
Right flexor (60°/s)	119 ± 21	102 ± 18	90 ± 16	93 ± 11	0.018, <0.001, <0.001
Left flexor (60°/s)	117 ± 24	103 ± 17	88 ± 16	89 ± 21	0.072, <0.001, 0.002
Right extensor (90°/s)	160 ± 29	120 ± 28	109 ± 20	110 ± 17	<0.001, <0.001, <0.001
Left extensor (90°/s)	166 ± 28	131 ± 31	113 ± 23	116 ± 21	0.002, <0.001, <0.001
Right flexor (90°/s)	112 ± 20	97 ± 18	88 ± 15	87 ± 14	0.040, <0.001, <0.001
Left flexor (90°/sc)	108 ± 20	95 ± 16	78 ± 16	85 ± 19	0.065, 0.001, 0.002
Average extensor	172 ± 31	135 ± 30	119 ± 20	121 ± 20	0.001, <0.001, <0.001
Average flexor	114 ± 20	99 ± 16	86 ± 14	89 ± 16	0.032, <0.001, <0.001
General average	143 ± 23.7	117 ± 21	103 ± 16	105 ± 17	0.003, <0.001, <0.001

YND younger non-diabetics, YD younger diabetics, END elderly non-diabetics, ED elderly diabetics.

*Comparison between YND and YD, YND and END, and YND and ED groups, respectively.

Bold values indicate statistical significance (p < 0.05).

The CFFM was inversely correlated with hemoglobin, MET and GFR levels (Table 5). AMS showed strongly positive correlation with albumin and MET, moderately positive correlation with hemoglobin and inverse correlations with age, HbA1c, FC and CRP (Table 5). Although there was no significant correlation between CFFM and AMS (r = 0.001, p = 0.99), there was a moderate and positive correlation between FFM and AMS (r = 0.38, p = 0.002). There was a weak and non-significant correlation between hemoglobin and CRP levels (r = −0.23, p = 0.07). There was no correlation between duration of diabetes and FFM, CFFM, or AMS in all diabetic subjects (r = 0.09, r = 0.1 and r = 0.02 and p = 0.96, p = 0.6 and p = 0.9, respectively).

Table 5. Correlation of CFFM and AMS with other parameters.

	CFFM		AMS
Variable	r	p	R	p
Age (year)	−0.47	0.36	−0.5	<0.001
Waist circumference (cm)	0.17	<0.001	−0.05	0.211
Body mass index (kg/m^2)	0.23	<0.001	−0.02	0.276
C reactive protein (mg/l)	−0.05	0.216	−0.11	0.048
Hemoglobin A1c (%)	0.18	0.442	−0.04	0.001
Urine microalbumin (mg/l)	0.04	0.436	−0.38	0.203
25-OH D vitamin (ng/ml)	0.14	0.537	−0.05	0.865
Hemoglobin (g/dl)	−0.17	0.008	0.38	<0.001
Albumin (g/dl)	0.12	0.431	0.37	<0.001
GFR-MDRD (ml/min/1.73 m^2)	−0.25	0.04	0.1	0.4
Functional capacity	−0.16	0.8	−0.56	0.002
Metabolic equivalent (kcal/kg/h)	−0.2	0.015	0.5	0.002
Maximum heart rate	0.04	0.3	−0.18	0.550

CFFM Fat-free mass corrected for height squared, AMS average muscle strength, GFR-MDRD glomerular filtration rate calculated by short MDRD formula.

Bold values indicate statistical significance (p < 0.05).

Multivariate linear regression analysis revealed independent correlates of AMS as age (B = −4.6, 95% CI [−2.2]–[−0.8], p < 0.001) and hemoglobin (B = 2, 95% CI [−0.06]–[8.6], p = 0.053).

Discussion

In our study, CFFM was similar in all four groups. However, there were significant differences in muscle strength values, which seemed to result primarily from older age. Presence of diabetes did not seem to affect muscle mass or muscle strength. Furthermore, CFFM was not correlated with muscle strength. Muscle strength was positively correlated with albumin, hemoglobin and MET; however, it was negatively correlated with age, FC and albuminuria. These clinically significant correlations were not present for CFFM.

The findings of our study indicate that isokinetic muscle strength is a better marker than muscle mass or CFFM determined using BIA. Our study also suggests that for subjects with uncomplicated diabetes who have a relatively fine glycemic control, sarcopenia is primarily associated with aging. Correlation of HbA1c with AMS in univariate analysis suggests a potential protective effect of better glycemic control on the development of sarcopenia. However, we cannot extrapolate our results to all diabetic subjects, because diabetic subjects with established complications like significant renal failure or cardiovascular diseases were excluded. This is primarily because of the design of our study, in which we utilized isokinetic muscle strength and excluded subjects with a positive exercise test. As isokinetic muscle strength measurement is an examination requiring effort, its application in diabetic subjects with established complications may have led to undesired results such as myocardial ischemia. Another important point is that the glycemic control of the diabetic subjects in our study was relatively fine, especially in the elderly diabetics. The expected rate of muscle mass and strength loss may have been much higher if elderly diabetic subjects with established renal and/or cardiovascular complications had been selected [4]. In the Health ABC study, low-muscle strength associated with aging was primarily due to low-muscle mass; and age and body fats were inversely associated with muscle strength and quality [8]. Park et al. [9] reported accelerated loss of lower extremity muscle mass and strength in older adults with diabetes compared with non-diabetic older adults. However, their diabetic patients were older (mean 73.5 years), had poorer glycemic control (mean HbA1c 7.9%) and higher rates of comorbidities (23.7% coronary heart disease, 16.6% cancer and 11.5% renal insufficiency) compared with our diabetic patients. Gougeon et al. [10] reported that diabetes was associated with diminished net protein balance and hyperglycemia further potentiated this effect, especially in men. Because diminished net protein balance is associated with muscle loss, this may be an explanation to muscle mass loss in diabetic subjects. However, glycemic control was better in our younger and older diabetics compared with these younger diabetic subjects (mean HbA1c 8.2%).

When sarcopenia rates determined according to CFFM values were compared, there was no significant difference between the groups. However, the rate of sarcopenia determined with this method was very low in this study. The lack of correlation between muscle mass and clinically significant factors for sarcopenia like albumin and FC indicates that utilization of more validated methods to determine muscle mass could provide more reliable information. However, lower isokinetic muscle strength was correlated with clinically significant factors for sarcopenia such as older age, lower albumin, MET and hemoglobin, worse FC, and higher albuminuria. Because of drawbacks of isokinetic muscle strength measurements in diabetic subjects with established complications, simpler methods like handgrip strength measurement may be safer.

Muscle strength of the elderly diabetic subjects tended to be lower than younger diabetics. However, duration of diabetes was shorter, and glycemic control was better in elderly diabetics compared with younger diabetics. Furthermore, the rate of microalbuminuria tended to be higher in younger diabetics. These factors may have attenuated a potentially significant muscle strength difference between elderly and younger diabetic subjects.

The muscle strength was better in younger non-diabetics compared with younger diabetics. However, the mean age of the younger non-diabetic group was significantly lower compared with younger diabetics. This is primarily because we had to include subjects aged under 40 years, which is accepted as the onset age for muscle loss [11]. Thus, our results do not suggest that uncomplicated diabetes causes muscle loss in younger subjects.

In our study, we detected that albumin levels, which are the indicator for nutritional state, were in positive correlation with muscle strength. Albumin levels of all of our subjects were within normal limits and varied between 3.8 and 4.9 g/dl. In a study performed by Visser et al., loss of muscle mass in subjects with albumin levels <3.8 g/dl or between 3.8 and 4.19 g/dl was more pronounced than subjects with an albumin level of ≥4.2 g/dl [12]. Another study by Schalk et al. indicated that even in the subjects with an albumin level of >3.8 g/dl, higher albumin levels were associated with higher baseline handgrip strength and lower handgrip strength loss during follow-up [13]. The relationship between inadequate protein intake and sarcopenia supports the relationship between lower albumin levels and loss of muscle strength [12]. In diabetic subjects, albumin levels may be low due to albuminuria. However, in our study, only three subjects had macroalbuminuria, and the albumin level of each of these subjects was above 4.3 g/dl.

In our study, hemoglobin levels were positively correlated with muscle strength. It is evident that anemia may cause muscle weakness. However, the lowest hemoglobin level was 11.1 g/dl in our study population. Therefore, our findings could not have resulted from muscle weakness associated with anemia. Another explanation for this association may be the fact that chronic inflammation precipitates both sarcopenia and anemia [14]. However, there was no significant correlation between hemoglobin and CRP levels, and CRP levels were generally in the normal range in our study population.

We found a negative correlation between albuminuria level and muscle strength. This finding suggests that microvascular complications of diabetes may be associated with sarcopenia. Microvascular changes may theoretically contribute to sarcopenia [15]. However, to our knowledge, no previous study investigated the effect of microalbuminuria on sarcopenia.

In our study, GFR did not show a clinically significant correlation with muscle mass or muscle strength. GFR values were similar in the elderly diabetic and younger diabetic subjects. GFR values of the two diabetic groups were significantly lower compared with younger and elderly non-diabetics. Pupim et al. [4] showed that in patients with end stage renal disease, diabetes was independently associated with increased muscle mass loss. In subjects with renal failure, factors such as limitation of protein intake, anorexia and chronic inflammation may facilitate development of sarcopenia. Although the mean GFR was lower in our diabetic patients, none of our study population had a GFR <30 ml/min/1.73 m².

The MET, which shows the energy spent during the exercise test and functional capacity, had a strong and significantly correlation with muscle strength. This situation may be associated with the role of physical limitation in the pathogenesis of sarcopenia. This finding may be explained by the fact that functional capacities of sarcopenic subjects may be limited due to muscle weakness. To our knowledge, no previous study investigated the relationship between exercise test parameters and muscle strength. The findings of our study suggest that these variables may be used in adequate people in the screening of sarcopenia.

The limitations of our study include the cross-sectional design, limited sample size and assessment of muscle mass with BIA instead of more validated modalities such as magnetic resonance imaging. The lack of diabetic subjects with complications may also be considered as a limitation.

Conclusions

According to our findings, among the factors determining muscle strength, advanced age appears to be more important than existence of uncomplicated diabetes. Muscle mass determined using BIA seems to be less reliable compared with isokinetic muscle strength measurement to assess the presence of sarcopenia. Albumin, hemoglobin, glomerular filtration rate, microalbuminuria and exercise test parameters (functional capacity, MET) may be useful markers for the determination of sarcopenia.

Declaration of interest

As authors of this article and the institution we work for, we received no grant, consulting fee or honorarium, support for any travel or other purposes, fees for participation in any review activities, payment for writing or reviewing, provision of writing assistance, medicines, equipment or administrative support or other. We have no relevant activities outside the submitted work as board membership, consultancy, employment, expert testimony, grants/grant pending, payment for any lectures, payment for manuscript preparation, patents, royalties, payment for educational activities, stock/stock options, travel/accommodation/meeting expenses unrelated to activities listed or any other. There are no other relationships/conditions/circumstances that present a potential conflict of interest.

Acknowledgements

We thank the sports medicine center staff for their help in obtaining our data. We also thank David Chapman for the editing of the English in the document.