Sarcopenia Is Associated With Lower Skeletal Muscle Capillarization and Exercise Capacity in Older Adults

Abstract

Background:

Skeletal muscle capillary rarefaction limits the transcapillary transport of nutrients and oxygen to muscle and may contribute to sarcopenia and functional impairment in older adults. We tested the hypothesis that skeletal muscle capillarization and exercise capacity (VO_2max) are lower in sarcopenic than in nonsarcopenic older adults and that the degree of sarcopenia is related to lower skeletal muscle capillarization.

Methods:

Body composition, VO_2max, and vastus lateralis capillarization were determined in 76 middle-aged and older men and women (age = 61±1 years, body mass index [BMI] = 30.7±0.5kg/m² [mean ± SEM]). Participants were classified as sarcopenic if appendicular lean mass divided by BMI (ALM_BMI) was less than 0.789 for men or less than 0.512 for women.

Results:

Sarcopenic subjects (ALM_BMI = 0.65±0.04, n = 16) had 20% lower capillary-to-fiber ratio, as well as 13% and 15% lower VO_2max expressed as mL/kg/min or L/min, respectively, compared with sex-, race-, and age-matched participants without sarcopenia (ALM_BMI = 0.81±0.05, n = 16; p < .05). In all 76 subjects, ALM_BMI, thigh muscle cross-sectional area, and VO_2max correlated directly with capillarization (r = .30–.37, p ≤ .05), after accounting for age, sex, and race.

Conclusions:

These findings suggest that low skeletal muscle capillarization is one factor that may contribute to sarcopenia and reduced exercise capacity in older adults by limiting diffusion of substrates, oxygen, hormones, and nutrients. Strategies to prevent the aging-related decline in skeletal muscle capillarization may help to prevent or slow the progression of sarcopenia and its associated functional declines in generally healthy older adults.

Aging and physical inactivity are associated with reduced muscle mass, low physical function, and disability (1,2). Likewise, there is an aging-associated worsening of macrovascular function, measured as arterial stiffening or impaired endothelium-dependent vasodilation (3–5), as well as microvascular dysfunction, assessed as capillary recruitment and muscle capillarization (6,7). These vascular impairments may contribute to declines in muscle mass and function because blood flow and muscle perfusion affect oxygen, hormone, and substrate delivery, as well as nutrient and amino acid uptake into muscle (8–10). Arterial stiffness and other metrics of macrovascular dysfunction are associated with low muscle mass in older adults (11–13); however, few data exist on the relationships between skeletal muscle capillarization, sarcopenia, and age-associated reductions in exercise capacity.

The diffusion and transport of substrates into skeletal muscle are limited by capillary surface area (14). Skeletal muscle capillarization is reduced in older adults compared with that in younger adults (6,15,16), and this aging-associated capillary rarefaction may contribute to lower muscle mass and physical function by limiting transcapillary transport of essential amino acids, nutrients, hormones, and oxygen. Nicklas and colleagues (17) showed that reduced skeletal muscle capillarization and metabolic enzyme activities are associated with lower physical function in older adults, but it is unclear whether these relationships are linked to reduced muscle mass. It is our hypothesis that skeletal muscle capillarization and exercise capacity (VO_2max) are lower in sarcopenic than in nonsarcopenic sedentary older adults and that the degree of sarcopenia is related to lower skeletal muscle capillarization.

Methods

Participants

Adults 45–80 years of age were recruited from the Baltimore, MD regional area to participate in studies examining the metabolic responses to exercise training and weight loss. Participants with complete dual-energy X-ray absorptiometry (DXA), VO_2max, and metabolic data at baseline were considered for this cross-sectional study. All participants for whom adequate muscle biopsy samples were available at the time of immunohistochemical muscle analyses are included in this study. The resultant sample included 32 men and 44 postmenopausal women; 23 participants were African American and 53 were Caucasian. Participants were all weight stable (<2.0kg weight change in past year), sedentary (<20min of aerobic exercise two times per week), and screened by physical examination, medical history, fasting blood chemistry, and graded maximal exercise test to confirm health status. Participants had no history of coronary artery disease, heart failure, peripheral arterial disease, stroke, and liver, kidney, or lung disease. Only participants with no physical or laboratory evidence of these diseases, calculated glomerular filtration rate ≥80mL/min/body surface area, and normal, asymptomatic graded treadmill exercise tests were included in the study. All participants were current nonsmokers; those with a history of smoking had not smoked for ≥2 years. The women in the study had not menstruated for at least 1 year. The research protocols were approved by the Institutional Review Board at the University of Maryland School of Medicine, and all participants provided written informed consent.

Body Composition

Body mass index (BMI) was calculated as weight (kg) divided by height (m²). Total and regional appendicular lean mass (ALM), fat mass, fat-free mass, and percent body fat were measured with DXA (Prodigy or iDXA, LUNAR Radiation, Madison, WI). Sarcopenia was assessed as ALM divided by BMI (ALM_BMI). Participants were classified as sarcopenic if ALM_BMI was less than 0.789 for men or less than 0.512 for women based on the Foundation for the National Institutes of Health Sarcopenia Project criteria (18). Thigh muscle cross-sectional area was determined by a mid-thigh computed tomography scan as previously described (19) using a Siemens Somatom Sensation 64 Scanner (Fairfield, CT) and Medical Image Processing, Analysis and Visualization software (MIPAV v.7.0.0, NIH, Bethesda, MD).

Exercise Capacity (VO_2max)

VO_2max was measured by indirect calorimetry (Quark, Cosmed USA, Chicago, IL) during a maximal graded treadmill exercise test as previously described (20). Briefly, participants walked at a constant speed throughout the test; grade was initially set to 0% and increased every 2 minutes until maximal effort was achieved. VO_2max was verified by standard physiological criteria (respiratory exchange ratio >1.10 or a plateau in VO₂ with an increase in workload).

Skeletal Muscle Capillarization

Percutaneous needle biopsies were obtained from the vastus lateralis, approximately 12–13cm above the patella on the anterolateral aspect of the right thigh using a Bergstrom needle (Stille, Solna, Sweden) as previously described (21). Muscle samples were embedded and rapidly frozen in optimal cutting temperature-tragacanth gum mixture, then stored at −80 °C for histochemical analyses. Samples were sectioned to a thickness of 14 µm on a cryostat, and capillaries were identified using a modified double stain technique as previously described (20). Three indices of capillarization were measured: (i) capillary contacts (the number of capillaries in contact with each muscle fiber), (ii) capillary-to-fiber ratio (the number of whole capillary equivalents in contact with each muscle fiber), and (iii) capillary density (the number of capillaries per mm² of muscle cross-sectional area) on at least 50 fibers for each sample (mean = 67±2 fibers/sample); sampling more fibers does not improve the estimation of capillarization in human muscle (22).

Statistical Analyses

The primary study variables were skeletal muscle capillary-to-fiber ratio, ALM_BMI, and VO_2max; secondary variables included capillary contacts per muscle fiber, capillary density, muscle fiber cross-sectional area, and thigh muscle cross-sectional area. Data are presented as means ± SEM. Statistical analyses were performed using IBM SPSS Statistics v22 (IBM, Armonk, NY). Analysis of variance was used to test for differences in variables between sarcopenic and nonsarcopenic participants matched for sex, race, and age (±5 years). Matches were selected by sorting sarcopenic and nonsarcopenic participants into groups by sex and race, then matching the nonsarcopenic participant closest in age to the sarcopenic participant of the same sex and race. If more than one nonsarcopenic participant of the same age was available, the participant with the closest BMI was selected as the match. Analysis of covariance accounting for sex and BMI was used to test for differences in variables between the groups of nonsarcopenic matched to sarcopenic participants and those not matched. Chi-square was used to test for differences in medication use between sarcopenic and nonsarcopenic participants. Pearson product-moment correlation analyses were used to test for bivariate correlations between variables. Multivariable regression analyses were used to assess associations among variables, accounting for age, sex, and race. A Type I error rate of α = .05 was selected, and two-tailed probabilities are reported for all analyses.

Results

Skeletal Muscle Capillarization in Sarcopenic and Nonsarcopenic participants

Participants were all middle-aged to older men and women with BMI ranging from 23.8–39.5kg/m² (normal weight to Class II obesity). The sarcopenic participants (9 men and 7 women) were compared with sex-, race-, and age-matched nonsarcopenic participants (Table 1). There were no differences in blood pressure, lipoprotein-lipid levels, or plasma glucose levels between sarcopenic and nonsarcopenic participants; however, a larger proportion of the sarcopenic participants were taking medication (statins) for dyslipidemia (p ≤ .01). ALM_BMI was 19% lower and thigh muscle cross-sectional was 17% lower in participants with sarcopenia compared with nonsarcopenic participants (p < .05 for both). Likewise, participants in the sarcopenic group also had significantly lower VO_2max compared with nonsarcopenic participants (p < .05). Vastus lateralis biopsy analyses showed that sarcopenic participants had 18% fewer capillary contacts per fiber and 20% lower capillary-to-fiber ratio (p < .01). There was no difference in capillary density between the groups, as muscle fiber cross-sectional area was lower in sarcopenic participants with low capillary-to-fiber ratio (p < .05).

Table 1.

Participant Characteristics, Body Composition, and Skeletal Muscle Phenotypes

	All participants	Sarcopenic	Nonsarcopenic	p Value
	(n = 76)	(n = 16)	(n = 16)
Sex (male / female)	32 / 44	9 / 7	9 / 7	—
Race (African American / Caucasian)	22 / 54	3 / 13	3 / 13	—
Age (years)	61±1	63±2	62±2	—
BMI (kg/m2)	30.7±0.5	34.0±1.2	33.1±1.3	0.59
Systolic blood pressure (mmHg)	124±1	127±2	124±2	0.36
Diastolic blood pressure (mmHg)	78±1	77±2	78±2	0.67
Medication for hypertension (%)	38.2	37.5	37.5	1.0
Total cholesterol (mmol/L)	4.81±0.10	4.39±0.21	4.72±0.24	0.31
LDL-cholesterol (mmol/L)	2.90±0.09	2.61±0.20	2.80±0.20	0.50
HDL-cholesterol (mmol/L)	1.24±0.04	1.10±0.07	1.19±0.11	0.51
Triglycerides (mmol/L)	1.50±0.06	1.59±0.17	1.61±0.14	0.92
Medication for dyslipidemia (%)	33.8	56.3	12.5	0.01
Fasting plasma glucose (mmol/L)	5.51±0.11	5.60±0.16	5.58±0.09	0.95
G120 (mmol/L)	7.80±0.28	9.13±0.55	8.20±0.62	0.28
Body fat (%)	41.1±1.0	43.4±2.0	39.1±2.2	0.16
ALMBMI	0.74±0.02	0.65±0.04	0.81±0.05	0.024
Capillary contacts (capillaries/fiber)	4.0±0.1	3.7±0.2	4.5±0.2	0.008
Capillary-to-fiber ratio (whole capillary equivalents/fiber)	1.5±0.1	1.3±0.1	1.7±0.1	0.005
Capillary density (capillaries/mm2)	303±7	284±12	292±12	0.65
Muscle fiber cross-sectional area (µm2)	5185±205	4972±414	6309±421	0.032
Mid-thigh muscle cross-sectional area (cm2)	96.6±3.8	86.0±6.1	104.7±6.1	0.047
	(n = 56)	(n = 11)	(n = 11)
VO2max (L/min)	2.08±0.07	1.94±0.11	2.27±0.11	0.045
VO2max (mL/kg/min)	22.9±0.6	20.5±1.0	23.6±1.0	0.039

Notes: Data are presented as means ± SEM with the exceptions of sex, race, and percentage of participants taking medications for hypertension or dyslipidemia.

p Values in bold type denote statistical significance.

ALM = appendicular lean mass; BMI = body mass index; VO_2max = maximal exercise capacity.

To confirm there was no inherent selection bias, we compared matched nonsarcopenic participants with nonsarcopenic participants whom were not matched, adjusting for sex, race, and BMI. These analyses revealed no differences between the matched and unmatched nonsarcopenic participants, respectively, in capillary-to-fiber ratio (1.6±0.1 vs 1.5±0.1 capillaries per fiber), ALM_BMI (0.78±0.19 vs 0.81±0.12), muscle fiber cross-sectional area (5573±400 vs 5213±272 µm²), thigh muscle cross-sectional area (103.9±5.3 vs 106.1±3.4cm²), or VO_2max (24.8±1.1 vs 23.7±0.7ml/kg/min or 2.25±0.11 vs 2.17±0.07L/min).

Association of Skeletal Muscle Capillarization With Muscle Phenotypes and Exercise Capacity

To determine relationships among variables in the larger population, correlation coefficients among capillarization data, ALM_BMI, muscle fiber size, and VO_2max were calculated in the entire group of 76 participants, and correlations between thigh muscle cross-sectional area and capillarization were determined in 56 participants for whom CT scans were available. In bivariate analyses, capillary-to-fiber ratio directly correlated with muscle fiber cross-sectional area (r = .82, p ≤ .001), mid-thigh muscle cross-sectional area (r = .55, p ≤ .001), ALM_BMI (Figure 1: r = .51, p ≤ .001), and VO_2max expressed in mL/kg/min and L/min (r = .37–0.52, p ≤ .01). Age was inversely associated with VO_2max (Table 2), but was not associated with other variables in bivariate analyses. The strength and significance of correlations between capillary-to-fiber ratio and other variables were similar in African American (r = .55–0.77, p ≤ .05 for all) and Caucasian participants (r = .44–.83, p ≤ .01 for all), although capillary-to-fiber ratio was higher in African American than in Caucasian participants (1.6±0.01 vs 1.4±0.01 capillaries per fiber, p ≤ .05).

Figure 1.

Scatterplot depicting the relationship of skeletal muscle capillary-to-fiber ratio with ALM_BMI in sedentary, older, African American and Caucasian men and women. In bivariate correlation analysis, ALM_BMI was directly associated with capillary-to-fiber ratio (r = .51, p ≤ .001). In regression analysis accounting for age and race, there remained a direct association between ALM_BMI and capillary-to-fiber ratio in both men and women (partial r = .36 and .37, respectively, p ≤ .05).

Table 2.

Associations of Capillarization, Skeletal Muscle Phenotypes, and Aerobic Capacity (VO_2max)

	Capillary- to-fiber ratio	Muscle fiber cross-sectional area	Mid-thigh muscle cross- sectional area	ALMBMI	VO2max (mL/kg/min)	VO2max (L/min)	VO2max (mL/ kgFFM/min)
Age	−0.09	−0.04	−0.19	−0.15	−0.39**	−0.45***	−0.49***
Capillary-to-fiber ratio	—	0.69***	0.35*	Men: 0.36*	0.30*	0.35**	0.31*
				Women: 0.37*
ALMBMI	—	—	—	—	0.57***	0.40**	0.22

Notes: Bivariate Pearson correlation coefficients are presented to show the relationships among age and other variables. Partial correlation coefficients are presented to show the relationships among capillary-to-fiber ratio, ALM_BMI, and other variables in regression analyses accounting for age, sex, and race.

ALM = appendicular lean mass; BMI = body mass index; FFM = fat-free mass.

*^,**^,***Statistically significant correlation: *p < .05. **p < .01. ***p < .001.

In regression analyses accounting for age, sex, and race, higher capillary-to-fiber ratio was associated with higher muscle fiber cross-sectional area, thigh muscle cross-sectional area, and VO_2max expressed in mL/kg/min, L/min, or mL/kgFFM/min (Table 2). There was little overlap in ALM_BMI values between men and women; therefore, regression analyses with ALM_BMI as the dependent variable were conducted separately in men and women. After adjusting for age and race, capillary-to-fiber ratio directly correlated with ALM_BMI in both men and women (Table 2). Additionally, ALM_BMI correlated directly with VO_2max expressed in mL/kg/min and L/min, but not expressed in mL/kgFFM/min, after accounting for age, sex, and race (Table 2).

Discussion

The findings of this study support the hypothesis that low skeletal muscle capillarization is associated with the presence and degree of sarcopenia in sedentary older adults. A number of factors contribute to sarcopenia and reduced physical function with aging (23–26), but vascular impairments including microvascular rarefaction remain an understudied area of investigation. Our results show that low muscle cross-sectional area, ALM_BMI, and exercise capacity are all associated with low skeletal muscle capillarization in sedentary older adults. These relationships suggest that reduced capillarization could be one mechanism contributing to sarcopenia in sedentary older adults.

Adequate blood flow and perfusion are critical for anabolism in muscle (8), but these are limited by reduced capillarization. Muscle protein synthesis is reduced in older adults compared with that in young adults after consumption of a mixed meal (10), and this is related to reduced postprandial microvascular blood flow and attenuated vascular responses to feeding in older adults (7). In response to a mixed meal, insulin increases microvascular perfusion and amino acid uptake by muscle to maintain muscle protein synthesis (27). Reduced capillarization in skeletal muscle of older people is associated with lower insulin-stimulated glucose uptake (20,28,29), and it is likely that reduced capillarization also limits the anabolic effect of insulin on muscle protein synthesis. Reduced muscle capillarization also limits oxygen delivery to muscle, and chronic or intermittent hypoxia is associated with muscle atrophy in disease states such as chronic heart failure, obstructive pulmonary disease, and peripheral arterial disease (30). However, it is unclear whether the degree of reduced capillarization in the absence of these diseases would induce sufficient hypoxia to cause muscle atrophy. Considering these previous findings, it is possible that microvascular rarefaction could contribute to reduced muscle mass in sedentary older people by limiting insulin, oxygen, and substrate delivery to muscle.

Low skeletal muscle capillarization in older adults is also associated with functional outcomes including low Short Physical Performance Battery score, slow gait speed, and greater disability (17). Although the present study did not assess these measures of physical function, our finding of a direct relationship between capillarization and exercise capacity is congruous with the previous report. On one hand, reduced capillarization in skeletal muscle could directly affect exercise capacity by limiting oxygen delivery, as indicated by the association of capillary-to-fiber ratio with VO_2max expressed per unit of fat-free mass (mL/kgFFM/min). Reduced capillarization may also impact exercise capacity through effects on muscle mass and sarcopenia in older adults. This is supported by the significant relationship of low capillarization with muscle phenotypes (low muscle fiber size, cross-sectional area, and ALM_BMI) and the direct relationship between ALM_BMI and exercise capacity.

The present study is cross-sectional; thus, cause-and-effect conclusions cannot be drawn. However, patterns of differences in the metrics of capillarization may provide insight to temporal relationships between changes in capillarization and muscle mass. For example, if muscle fiber size decreased prior to the number of capillaries, both capillary contacts and capillary-to-fiber ratio would be similar in sarcopenic and nonsarcopenic adults, and capillary density would appear inflated in those with sarcopenia (ie, there would be a similar number of capillaries with smaller muscle fibers). However, capillary density is similar between groups, whereas capillary contacts and capillary-to-fiber ratio are lower in the sarcopenic participants. This suggests that any decrease in fiber size is either preceded by a reduction in capillaries, or that the capillary rarefaction and muscle fiber size occur simultaneously. Animal studies show that endothelial apoptosis, vascular rarefaction, and impairments in capillary function precede declines in muscle mass (31,32). Conversely, capillary proliferation appears essential to increase muscle mass (33), and increasing capillarization is associated with increases in muscle mass (34) and improvements in muscle phenotypes (eg, a higher number of regenerating fibers and less fibrosis) in animal models (35). We showed that aerobic exercise training–induced increases in muscle capillarization independently contribute to increases in insulin-stimulated glucose uptake (20,36), and we speculate that an increase in capillarization would also improve the delivery of oxygen, amino acids, and other substrates, as well as the anabolic effect of insulin on muscle. A comprehensive study is required to elucidate the temporal relationship between changes in capillarization and muscle mass and the associated effects on protein synthesis in human participants.

The strengths and limitations of this study warrant discussion. A major strength of this study is the careful screening and inclusion of participants generally in good health, with no clinical manifestations of central or peripheral cardiovascular diseases, diabetes, or kidney disease. However, although none of the participants demonstrated symptoms associated with these diseases during screening and baseline testing, we cannot absolutely exclude the presence of asymptomatic vascular disease. Such subclinical disease could impair blood flow and compound the adverse effects of low capillarization on muscle phenotypes and exercise capacity. Another strength is the comparison of matched sarcopenic and nonsarcopenic participants. Although the regression analyses assessed relationships among variables with adjustment based on age, sex, and race, the matching of participants allowed direct comparison of capillarization and exercise capacity between sarcopenic and nonsarcopenic participants of the same sex and race, and similar age. Additional analyses showed that these matched nonsarcopenic participants were representative of the entire nonsarcopenic group, with no differences in capillarization, muscle phenotypes, or exercise capacity between those selected and not selected as matches. In addition, the sarcopenic and nonsarcopenic participants were well matched with respect to blood pressure, lipoprotein-lipid levels, and plasma glucose levels. The only notable difference between the sarcopenic and matched sarcopenic participants was the higher prevalence of statin use in sarcopenic participants. Evidence of the effects of statin use on muscle phenotypes and exercise capacity is largely inconclusive (37), and the incidence of frailty is similar in older adults who do or do not take statins (38). Thus, it seems unlikely higher statin use in the sarcopenic participants significantly affected the results of this study. However, the potential effects of dyslipidemia in these participants prior to statin treatment cannot be quantified in the present study. Finally, although all participants reported being sedentary, objective assessments of physical activity levels were not obtained. Future studies will need to determine whether habitual physical activity levels affect the relationship between skeletal muscle capillarization and sarcopenia in sedentary older adults.

Conclusions

The findings of this study show that low capillarization may affect sarcopenia and exercise capacity in otherwise healthy middle-aged and older adults. This could occur because low capillarization impairs the transcapillary transport of substrates required to maintain muscle mass, metabolism, and physical function. Aerobic exercise training increases skeletal muscle capillarization (20,39,40), and although aerobic exercise does not necessarily increase muscle mass, it may be a valuable adjunct to improve skeletal muscle capillarization. Thus, combinations of exercise, dietary, and/or pharmacologic interventions to increase capillarization in skeletal muscle may help slow the cycle leading to sarcopenia that is associated with reduced fitness, dysmobility, and frailty in sedentary older adults.

Funding

This research was supported by a Paul B. Beeson Patient-Oriented Research Career Development Award in Aging (NIH K23-AG040775 and AFAR, S.J.P.), a VA Career Development Award (S.J.P.), a VA Advanced Research Career Development Award (J.B.B.), NIH R01-AG019310 (A.S.R.) and R01-AG020116 (A.P.G), a VA Senior Research Career Scientist Award (A.S.R.), the University of Maryland Claude D. Pepper Center (NIH P30-AG028747), the Mid-Atlantic Nutrition Obesity Research Center (P30-DK072488), and the Baltimore Veterans Affairs Medical Center Geriatric Research, Education and Clinical Center (GRECC).