Body Cell Mass to Fat-Free Mass Ratio and Extra- to Intracellular Water Ratio Are Related to Maximal Oxygen Uptake Free

Abstract

Fat-free mass (FFM) is a heterogeneous compartment comprising body cell mass (BCM), intracellular water (ICW), extracellular solids, and extracellular water (ECW). The BCM/FFM and ECW/ICW ratios vary among individuals and decrease with age. This study aimed to determine whether BCM/FFM and ECW/ICW ratios are predictors of maximal oxygen uptake (V̇̇O₂peak) independently of age, sex, and objectively measured physical activity (PA). A total of 115 Japanese males and females, aged 55.3 ± 8.0 years (mean ± standard deviation), were included in the study. Anthropometry, explosive leg muscle power, and V̇̇O₂peak were measured, and BCM, FFM, ICW, and ECW were estimated. Step count and PA were objectively measured using a triaxial accelerometer. Blood flow volume was assessed using ultrasonography. BCM and ICW were negatively correlated with age, whereas FFM and ECW were not significantly correlated with age. FFM, ICW/ECW, BCM/FFM, step counts, moderate and vigorous PA, and leg muscle power were positively correlated with V̇̇O₂peak, even after adjusting for age and sex (p < .05). Multiple regression analysis indicated that either BCM/FFM or ECW/ICW, leg power, and objectively measured PA were associated with V̇̇O₂peak independent of age, sex, and FFM. Blood flow volume was significantly correlated with ECW (p < .05), but not with BCM. The BCM/FFM and ECW/ICW ratios were significant predictors of V̇̇O₂peak, independent of age, sex, FFM, leg power, and objectively measured PA.

Fat-free mass (FFM), considered a metabolically active compartment of the human body, is often used to adjust for differences in resting or exercise-state energy metabolism, including maximal oxygen uptake (V̇̇O₂peak), to account for body composition (1–4). As a chemical component, FFM is composed of water, minerals, and proteins that do not expend energy (2), which include body cell mass (BCM), extracellular solids (ECS), and extracellular water (ECW) (5–7). BCM is responsible for almost all metabolic processes (8–11). Although BCM is correlated with FFM, the average BCM/FFM ratio is 0.58 in the Reference man (11), which can vary among individuals depending on age, medical status, and obesity (6,12,13).

V̇̇O₂peak is directly associated with the rate of adenosine triphosphate (ATP) generation, which is typically maintained during endurance exercise (14). Most ATP is synthesized in the mitochondria through oxidative phosphorylation (15). Skeletal muscle tissues require a substantial amount of energy for mechanical work, which is generated by mitochondria (15–17). It is well known that FFM is positively correlated with V̇̇O₂peak in healthy adults (1,3,18,19); however, the physiological association between FFM and V̇̇O₂peak in terms of muscle cell mass and oxygen consumption capacity remains unclear (3). This is because FFM includes the blood volume in the ECW compartment, which significantly influences the left ventricular stroke volume (20,21). Lundby and Montero have published a series of articles highlighting that plasma volume determines V̇̇O₂peak (22). Therefore, FFM is a determinant of stroke volume in healthy humans (3). However, BCM is a component of FFM after the removal of ECW and ECS and is physiologically independent of blood and stroke volumes (10,11).

We hypothesized that BCM, the metabolically active component of FFM, which is more closely related to muscle cell mass and is an independent factor of blood volume, is positively associated with V̇̇O₂peak in healthy middle-aged adults (23). Additionally, we propose that explosive lower limb muscle power is another peripheral factor of V̇̇O₂peak, independent of BCM. To test these hypotheses, we objectively measured physical activity (PA), which is known to be an essential predictor of V̇̇O₂peak (18,24–26).

BCM can only be defined by intracellular water (ICW) or total body potassium (TBK) (27). ICW can be calculated from total body water (TBW) and ECW, where TBW is measured using deuterium dilution, and ECW is measured using sodium bromide dilution (28). TBK can be measured using a whole-body counter (28). These methods are accurate but complicated to conduct regularly. Bioelectrical impedance spectroscopy (BIS) is a noninvasive technique that can be used to estimate ICW and, subsequently, BCM (28). Cell membranes are composed of phospholipid bilayers that operate as capacitors and shield the intracellular chamber from low-frequency currents. Because membranes are receptive to currents at higher frequencies, both ICW and ECW contribute to electrical conductance at higher frequencies (27,29). BIS can split water across intracellular and extracellular layers in skeletal muscle, enabling the measurement of the difference between ICW and ECW (29–31).

Method

Participants

The study recruited participants from a local community at the National Institute of Health and Nutrition, Tokyo, Japan (32–34). A total of 115 participants, consisting of 30 males and 85 females aged 32–69 years (mean ± standard deviation [SD], 55.3 ± 8.0), were included in the study. Participants who underwent the following assessments were included in the study: (a) anthropometric and PA measurements; (b) blood tests; (c) V̇̇O₂peak measurements; and (d) body composition measurements using BIS. This study was reviewed and approved by the Institutional Review Board of the National Institute of Biomedical Innovation, Health, and Nutrition (No. KENEI-102), and all participants provided written informed consent to participate in the study.

Anthropometry and Body Composition

Anthropometric measurements, including weight, height, and waist circumference, were measured for each participant. Body mass index (BMI) was calculated using height and weight, whereas body composition was assessed using the BIS device (SFB7; ImpediMed, Pinkenba, Australia) (35,36). Two injectable electrodes (Red Dot; 3M Health Care, St. Paul, MN) were positioned on the dorsal surfaces of the right hand and foot, close to the metacarpophalangeal and metatarsophalangeal joints, respectively. The electrodes were then positioned at the center of the imaginary line that connects the radius and ulnar bony protrusions on the dorsum of the right wrist and at the center of the front edge of the right ankle, connecting the medial and lateral malleoli. The measurements were taken for 15 minutes while the participants were in a reclining position. The frequency range of the BIS device was 3 kHz–1 MHz. The resistance was calculated at 2 frequencies: infinitely high frequency (R_∞) and zero frequency (R₀), using analytical software combined with the Cole–Cole plot and using the analytical settings as reported in previous studies (30,31). The resistance of the extracellular components (R_e) was equal to R₀, while that of the intracellular components (R_i) was estimated as 1/([1/R_∞] – [1/R₀]). Finally, the equations presented by de Lorenzo et al. (27,29) were used to estimate TBW, ECW, ICW, and FFM from R_e and R_i. The SFB7 system incorporates Lorenzo’s equation, and BCM was calculated using the equations proposed by Wang et al. (11).

Maximal Oxygen Uptake

The maximal oxygen uptake was measured using a maximum-rated exercise test with bicycle dynamometers to determine V̇̇O₂peak (Monark Ergomedic 828E; Varberg, Sweden) (32,33). The test began with a workload of 30–60 W, and the work rate was increased every minute until the required pedaling frequency (60 rpm) became unattainable. Heart rate was monitored throughout the exercise using a WEP-7404 (Nihon Kohden, Tokyo, Japan), and ratings of perceived exertion were recorded. Exhaled air was collected at 30-second intervals during the progressive activity test using Douglas bags. The concentrations of oxygen and carbon dioxide were measured using a mass spectrometer (Arco-1000; Arco System, Ogaki, Japan), and a dry gas volume meter was used to calculate the volume of expired air and convert it to Standard Temperature, Pressure, and Dry. Test operators vocally urged each participant to give their best effort throughout the last part of the exam, and V̇̇O₂peak was accepted if the participant’s peak heart rate was >95% of the age-predicted maximal heart rate (220 – age), with the highest value of V̇̇O₂ during the exercise test being used as V̇̇O₂peak (33).

Physical Activity

A triaxial accelerometer (Actimarker; Panasonic, Osaka, Japan) was used to monitor the intensity of PA (37,38). All participants wore the accelerometer for 28 days. We used data from a minimum of 14 consecutive days during which the accelerometer was worn from waking up until going to bed. The technical and estimated equation features of the accelerometer are described in detail. The accelerometer sampled acceleration at a rate of 20 Hz and had a performance range of 0 to twice the momentum of gravity. The SD of the 3-dimensional vector norm of the composite acceleration was monitored for 1 minute. A study of healthy adults found that the vector norm was significantly associated (R² = 0.86) with V̇̇O₂ during walking and running at 7 paces, ranging from 40 to 160 m min⁻¹, and during 7 common activities, including food preparation, self-care while upright, replacing clothing, cleaning dishes, eating supper, vacuuming, and laundry. The SD of the vector norm at 1-minute intervals was used to calculate the metabolic equivalent (MET) values of PAs. Moderate-intensity physical activity (MPA; 3.0–5.9 METs), vigorous-intensity physical activity (VPA; ≥6.0 METs), and step counts were recorded.

Leg Muscle Power

In the sitting posture, the explosive leg extension power was evaluated using a dynamometer (Anaero Press 3500; Combi Wellness, Tokyo, Japan) (39). Participants were instructed to stretch their bent legs as quickly as possible. After 5 attempts at 15-second intervals, the average of the two highest recorded power outputs (W) was calculated and used as the official measurement. The mathematical calculation of extension power involved dividing the earlier-obtained average power output by the body weight.

Blood Samples

Blood samples were collected from the antecubital vein of the participants who had fasted for at least 10 hours. The collected blood was transferred into tubes without additives or ethylenediaminetetraacetic acid. The tubes were centrifuged at 3 000 rpm for 20 minutes to extract plasma or serum (40). Hemoglobin and hematocrit levels were also measured.

Blood Flow Volume

An ultrasonography instrument (Vivid I; GE Medical System, Chicago, IL) with a high-resolution linear array transducer was used to measure the blood flow volume (41). We obtained 2-dimensional and longitudinal ultrasound images at the proximal 1–2 cm straight portion of the common carotid artery. The mean blood velocity was measured at an insonation angle of <60°, and the blood flow volume was calculated as mean blood velocity × π × (arterial radius)² × 60 (42). On average, 10 measurements were taken per participant.

Statistical Analysis

To establish a prediction equation for V̇̇O₂peak, the sample size was calculated using multiple linear regression analysis with seven predictors with an effect size f² of 0.15, α of 0.05, and power of 0.80, which amounted to a sample size of 103 (G*Power 3.1.9.7, Universität Kiel, Germany). The results are presented as mean ± SD or range (minimum to maximum). Pearson’s correlation coefficients were calculated between V̇̇O₂peak and independent variables. Partial correlation coefficients were obtained between V̇̇O₂peak and the independent variables, with sex alone or both age and sex as the control variables. A multiple linear regression analysis was performed using V̇̇O₂peak as the dependent variable. The statistical significance value was set at p < .05. IBM SPSS Statistics for Windows (IBM Corporation, Armonk, NY) was used for all analyses.

Results

The physical characteristics of the participants and the results of the partial correlation analysis with sex as a control variable are shown in Table 1. Both the absolute (L·min⁻¹) and relative values (mL·kg⁻¹·min⁻¹) of V̇̇O₂peak showed a significant negative correlation with age (p < .001). BCM and ICW also showed a significant negative correlation with age (p < .05), whereas weight, FFM, and ECW were not significantly correlated with age (p > .05). The BCM/FFM and ECW/ICW ratios showed significant negative and positive correlations with age, respectively (both p < .05). Leg muscle power also showed a significant negative correlation with age (p < .001).

The results of the partial correlation analysis with both age and sex as control variables are shown in Table 2. Weight, BMI, and waist circumference were positively correlated with the absolute value of V̇̇O₂peak (L·min⁻¹) but negatively correlated with the relative value of V̇̇O₂peak (mL·kg⁻¹·min⁻¹) after adjusting for age and sex. On the other hand, the absolute values of FFM (kg), BCM (kg), ICW (kg), and leg muscle power (kW) were positively correlated with the absolute value of V̇̇O₂peak (L·min⁻¹¹). Additionally, the relative values of FFM (%), BCM (%), and leg muscle power (W·kg⁻¹) versus body weight were positively correlated with the relative value of V̇̇O₂peak (mL·kg⁻¹·min⁻¹). Notably, the BCM/FFM ratio was positively correlated with both the absolute and relative values of V̇̇O₂peak (L·min⁻¹ and mL·kg⁻¹·min⁻¹, respectively), even after controlling for age and sex (p < .001). Conversely, the ECW/ICW ratio was negatively correlated with both the absolute and relative values of V̇̇O₂peak (L·min⁻¹ and mL·kg⁻¹·min⁻¹, respectively), even after controlling for age and sex (p < .001). Furthermore, the V̇̇O₂peak showed a partial correlation with objectively measured PA.

Blood flow volume (mL·min⁻¹) was found to show a significant positive correlation with V̇̇O₂peak (L·min⁻¹; r = 0.235; p < .05), FFM (r = 0.279; p < .01), and ECF (r = 0.300; p < .01) but a negative correlation with the BCM/FFM ratio (r = 0.184; p < .05). However, no correlation was observed between blood flow volume and BCM (kg; p > .05). Additionally, after adjusting for age and sex, blood flow volume showed a significant positive correlation with ECF (kg; r_p = 0.186; p < .05), but a negative correlation with BCM (%; r_p = −0.237; p < .05).

Table 3 shows the results of multiple linear regression analysis for the absolute value of V̇̇O₂peak (L·min⁻¹). In Model 1, age had a negative association with V̇̇O₂peak (L·min⁻¹), whereas MPA, VPA, BCM, and leg power had positive associations. In Model 2, additional analysis revealed that the BCM/FFM ratio was a significant predictor of V̇̇O₂peak (L·min⁻¹; p < .01), even when FFM was included as an independent variable. Finally, in Model 3, the ECW/ICW ratio was a significant and negative predictor of V̇̇O₂peak (L·min⁻¹; p < .01), even when FFM was included as an independent variable.

Discussion

To the best of our knowledge, this is the first study to examine the association between the BCM/FFM ratio, ECW/ICW ratio, and V̇̇O₂peak. BCM is a fundamental and metabolically active component of FFM related to muscle cell mass. FFM contains ECW, which is associated with blood flow volume, whereas BCM is an independent component of FFM. Köhler et al. (23) demonstrated that BCM was significantly correlated with V̇̇O₂peak, and our findings support this conclusion. Additionally, we found that the BCM/FFM ratio was a significant predictor of both the absolute and relative values of V̇̇O₂peak, even after controlling for age, sex, and PA. We also found a significant association between explosive leg muscle power and V̇̇O₂peak, even after controlling for age, sex, PA, and body composition.

Weight, FFM, and ECW were not significantly correlated with age in our study population. However, V̇̇O₂peak was found to show a significant negative correlation with age, indicating that age-related changes in FFM or ECW could not account for the decline in V̇̇O₂peak. In contrast, BCM and ICW showed significant negative correlations with age. Specifically, the BCM/FFM ratio showed a significant negative correlation, whereas the ECW/ICW ratio showed a significant positive correlation with age. Furthermore, both the BCM/FFM and ECW/ICW ratios were significant predictors of V̇̇O₂peak. These results suggest that individual differences in FFM composition are essential determinants of V̇̇O₂peak.

There has been some debate regarding the significance of the relationship between FFM and V̇̇O₂peak, even after controlling for sex (1,3,18,19). The physiological explanation for this association is thought to be a direct link between the skeletal muscle mass and its ability to utilize oxygen for energy metabolism. However, since FFM includes blood volume, it is linked to central circulatory parameters that determine the V̇̇O₂peak. A previous study found a significant correlation between FFM and blood volume in sedentary adults (3), where FFM may have operated as a factor of stroke volume given the substantial impact of blood volume on the left ventricular stroke volume. Hunt et al. (3,43) found that the volume of a supine stroke at rest was a robust physiological correlate of V̇̇O₂peak. Therefore, instead of having a direct influence on muscle oxygen-consuming capacity, it is probable that FFM is connected to V̇̇O₂peak through this central circulatory process.

As previously mentioned, FFM is a heterogeneous component of the human body (10,11), and its chemical components are distributed between BCM, ECW, and ECS. Blood volume is an ECW component that is physiologically independent of BCM. Our findings on the relationship between blood flow volume and ECW or BCM support this hypothesis. Although the average BCM/FFM ratio of our participants was 0.58, which is consistent with that of the Reference man (11), this ratio varied between participants and was positively associated with V̇̇O₂peak, even after controlling for age and sex.

BIS can differentiate between ICW and ECW (29–31). Muscle fiber cell membranes are composed of phospholipid bilayers that serve as capacitors and protect the intracellular chamber from low-frequency currents; however, they remain receptive to high-frequency currents, leading to electrical conductance in both ICW and ECW at higher frequencies (27,29). Segmental BIS can be used for water splitting across the intracellular and extracellular layers in skeletal muscles. Previous studies have demonstrated that muscle composition varies with age, leading to changes in muscle density and intramuscular adipose tissue (30,44–49). Magnetic resonance imaging (MRI) results reveal that intramuscular fat increases with age, whereas signal intensity decreases with age on diffusion tensor MRI (46). According to a recent study, the mean water T2 values and heterogeneity measured by MRI indices of older individuals were substantially higher than those of younger individuals (44). As people age, their skeletal muscles undergo significant changes in composition, notably fiber degeneration and fibrosis, which may go unreported during basic muscle girth testing. BCM assessed using the BIS is an informative parameter for exercise physiology and metabolism.

Objectively measured PA parameters such as MPA and VPA were significantly associated with VO₂peak, similar to previous studies (24–26,32). Our findings confirm the importance of daily PA in maintaining maximal oxygen capacity in nonathletic middle-aged adults. In 2005, Plaqui and Westerterp (26) developed a nonexercise model for calculating V̇̇O₂peak based on heart rate and accelerometer count. Cao et al. (32) employed a triaxial accelerometer to create a nonexercise model for calculating V̇̇O₂peak using objectively recorded PA in Japanese individuals. Our results are consistent with these previous findings.

The V̇̇O₂peak is one of the most critical indicators of cardiorespiratory fitness. In our study, we found that explosive leg muscle power was a significant predictor of V̇̇O₂peak in the participants. Whether central or peripheral variables limit V̇̇O₂peak remains controversial. The sequence of stages from atmospheric oxygen to mitochondrial utilization can be a possible barrier to oxygen flux, including (a) pulmonary intake capability, (b) blood oxygen-carrying capacity, (c) maximum cardiac output, and (d) skeletal muscle properties (14). According to the literature, oxygen availability rather than skeletal muscle oxygen extraction is the limiting factor for V̇̇O₂peak in physically active individuals. However, the effects of strength training on aerobic capacity and cardiorespiratory fitness remain controversial. In our study, the association between explosive leg muscle power and V̇̇O₂peak may be due to the characteristics of our participants. They were nonathletic middle-aged adults aged 32–69 years, and muscle power may be a determining factor for maintaining the required pedaling frequency during higher-workload cycling exercises.

Limitations

A limitation of the study is the use of the BIS to estimate ICW, ECW, and BCM, as it is a secondary indirect method and may have some determinate bias and indeterminate errors. The gold-standard methods for estimating these parameters are deuterium and sodium dilution techniques and the whole-body counter method. However, despite this limitation, the BIS method is a conventional, noninvasive, and easy-to-use method for estimating ICW, ECW, and BCM in daily practice. It is worth noting that even with the use of a conventional method, a significant relationship was observed between V̇̇O₂peak and FFM composition.

Conclusion

In this cohort, BCM and ICW showed a significant negative correlation with age, whereas FFM and ECW did not show a significant correlation. Our findings revealed that BCM, a metabolically active component of FFM and a representative biomarker of muscle cell mass, was not associated with blood flow volume but had a significant association with V̇̇O₂peak. This association remained statistically significant even after controlling for other influential variables. These results suggest that V̇̇O₂peak is partly determined by BCM and that individual differences in FFM composition (ie, the BCM/FFM or ICW/ECW ratios) are essential determinants of V̇̇O₂peak. Furthermore, objectively measured PA and explosive leg muscle power were significant predictors of V̇̇O₂peak in nonathletic middle-aged adults independent of BCM.

Funding

This study was funded by Health and Labor Sciences Research Grants (200825016 B and 201222028 B) to M.M. and the Japan Society for the Promotion of ScienceKAKENHI (18H03164) to Y.Y. Y.Y has a patent for a physical fitness assessment system (Japanese Patent No. 6709462) that is partly related to this publication.

Conflict of Interest

Y.Y. has a patent for a physical fitness assessment system (Japanese Patent No. 6709462) that is partly related to the publication. No conflicts of interest, financial or otherwise, are declared by the other authors.

Acknowledgments

We wish to thank Mr. Yuki Nishida for the BIS measurements.

References