PET/CT imaging of age- and task-associated differences in muscle activity during fatiguing contractions

Abstract

The study compared positron emission tomography/computed tomography (PET/CT) of [¹⁸F]-2-fluoro-2-deoxy-d-glucose ([¹⁸F]-FDG) uptake by skeletal muscles and the amount of muscle activity as indicated by surface electromyographic (EMG) recordings when young and old men performed fatiguing isometric contractions that required either force or position control. EMG signals were recorded from thigh muscles of six young men (26 ± 6 yr) and six old men (77 ± 6 yr) during fatiguing contractions with the knee extensors. PET/CT scans were performed immediately after task failure. Glucose uptake in 24 leg muscles, quantified as standardized uptake values, was greater for the old men after the force task and differed across tasks for the young men (force, 0.64 ± 0.3 g/ml; position, 0.73 ± 0.3 g/ml), but not the old men (force, 0.84 ± 0.3 g/ml; position, 0.79 ± 0.26 g/ml) (age × task interaction; P < 0.001). In contrast, the rate of increase in EMG amplitude for the agonist muscles was greater for the young men during the two contractions and there was no difference for either group of subjects in the rate of increase in EMG amplitude across the two tasks. The imaging estimates of glucose uptake indicated age- and task-dependent differences in the spatial distribution of [¹⁸F]-FDG uptake by skeletal muscles during fatiguing contractions. The findings demonstrate that PET/CT imaging of [¹⁸F]-FDG uptake, but not surface EMG recordings, detected the modulation of muscle activity across the fatiguing tasks by the young men but not the old men.

modulation of motor unit activity during voluntary contractions varies with both the magnitude of the requisite force and the characteristics of the load against which the limb acts. When subjects performed submaximal, isometric contractions with the elbow flexor muscles, for example, the adjustments in motor unit activity to sustain the same net muscle torque differed when the wrist pulled against a rigid restraint to match a target force compared with maintaining a constant joint angle while supporting a more compliant load (30, 37, 38, 40). The mean discharge rate of motor units in biceps brachii declined more rapidly and was accompanied by greater recruitment of additional motor units when the task required position control. Moreover, position control was associated with an increase in the variability of motor unit discharge times, which suggested a difference between the two tasks in the amount of synaptic noise at the level of the motor neuron (11). Because the amount of antagonist muscle coactivation did not differ between the two tasks in young subjects, the two control strategies required different combinations of synaptic input to sustain the same net muscle torque. Consistent with this interpretation, Baudry et al. (5, 6) observed greater modulation of Ia presynaptic inhibition during position control than during force control.

Despite the different adjustments in motor unit activity during force and position control, the rates of change in concurrently recorded global measures of motor unit activity, as indicated by the amplitude of the surface electromyogram (EMG) for biceps brachii, did not differ during sustained contractions with the two control strategies (30, 40). Due to the limitations of surface EMG recordings (9, 11, 13), the current study estimated the intensity and spatial distribution of muscle activity during fatiguing contractions by measuring the uptake of [¹⁸F]-2-fluoro-2-deoxy-d-glucose ([¹⁸F]-FDG) by skeletal muscles with positron emission tomography (PET) immediately after the contraction ended (16, 21, 24, 32, 35, 45). Because [¹⁸F]-FDG uptake is proportional to exercise intensity, accumulation of the radioactive tracer can be detected after the action and the spatial distribution of signal intensity used to characterize the contributions of selected muscles to the task (15, 24, 33).

To evaluate the sensitivity of PET measurements, the current study also compared surface EMG recordings and [¹⁸F]-FDG signal intensity in young and old adults during force and position control. In old adults, the amount of agonist-antagonist muscle coactivation tends to increase when switching from force to position control, whereas in young adults, modulation of the afferent feedback is preferred (3). The purpose of the current study was to compare the amount of muscle activity as indicated by PET/computed tomography (CT) imaging of [¹⁸F]-FDG uptake by skeletal muscles and surface EMG recordings when young and old men performed fatiguing isometric contractions that required either force or position control.

METHODS

The current study comprised two sets of experiments: 1) endurance-time experiments to establish the relative time to failure for both the force and position tasks when performed with the knee extensor muscles in a supine posture; and 2) muscle-activation experiments during which the task durations were established on the first visit and the two subsequent visits were used to determine the spatial and temporal distribution of muscle activity during two types of fatiguing contractions on the basis of [¹⁸F]-FDG PET/CT imaging and surface EMG recordings. Informed consent was obtained from all participants, who reported being free from cardiovascular and neurological disorders and participating in moderate levels of structured physical activity (2–4×/wk). The experimental procedures were approved by the Institutional Review Board at the University of Colorado Boulder and were in accordance with the Declaration of Helsinki.

Endurance-time experiment.

Endurance time was measured in three young (23 ± 4 yr) and three old (72 ± 4 yr) men. Each subject reported to the laboratory to perform two types of fatiguing contractions with the knee extensors of the left leg. The two tasks required each subject to sustain an isometric contraction with the knee extensors at 25% of maximal voluntary contraction (MVC) force for as long as possible. At this target force, blood flow is impaired but not occluded (41). One fatiguing contraction required the knee extensors to pull against a rigid restraint and to match the force exerted by the leg to the target force that was displayed on a monitor (1% MVC/cm). The other fatiguing contraction required subjects to use the knee extensors to support an equivalent inertial load and to maintain the position of the leg by matching knee angle to the target displayed on a monitor (1°/cm). The two fatiguing contractions are referred to as the force and position tasks, respectively.

Each type of fatiguing contraction was performed on a separate occasion, with 1 wk between the two sessions. The fatiguing contractions were performed with subjects in a supine posture with the trunk-thigh angle at 3.14 rad, the left knee joint angle at 0.78 rad, and the right knee angle at 1.57 rad (Fig. 1). One strap was placed around the waist to stabilize the subject and another strap was wrapped around the ankle to connect the load to the leg. The force exerted by the leg was measured with a load cell (0–500 lb, Noraxon, Scottsdale, AZ) placed in series with the load. The force signal was low-pass filtered (0–5 Hz) and recorded on a computer (1,000 samples/s). Knee joint angle during the position task was measured with a flexible, two-dimensional goniometer sensor (Noraxon) secured to the lateral aspect of the knee joint. The output of the goniometer was recorded, displayed on a monitor, and stored (1,000 samples/s) on a computer. The inertial load (25% MVC force) for the position task was suspended from the ankle at the same location that the restraint was applied during the force task. The force task was terminated when subjects were not able to achieve the target force for 5 s, and the position task was ended when subjects were unable to maintain the knee angle within 0.17 rad of the target value for 5 s.

Fig. 1.

Experimental arrangement for the two fatiguing contractions. As subjects lay in a supine position and the loads for the 2 types of fatiguing contractions, the force (A) and position (B) tasks, were attached to the left leg. A force transducer (a) was in series with the load during both tasks. Surface EMG electrodes (b) were placed over rectus femoris, vastus medialis, vastus lateralis, and the short head of biceps femoris. The position task (B) involved supporting an inertial load (c) that was equivalent to the force exerted against the rigid restraint (A). Knee angle was measured with a goniometer (d) during the position task.

Before and immediately after each fatiguing contraction, subjects performed an MVC with the knee extensor muscles of the left leg. The initial MVC was used to determine the target force for the fatiguing contraction, and the final MVC was used to derive an index of fatigability. The MVC task comprised a 3-s increase in force from zero to maximum with the maximal force held for ∼3 s, and subjects were verbally encouraged to achieve maximal force. Subjects rested for 60 to 90 s between trials. When the peak forces achieved in two of the three trials differed by >5%, additional MVCs were performed until this criterion was met. The greatest force achieved by each subject was taken as the MVC force.

Muscle activation experiment.

Six young (26 ± 6 yr) and six old (77 ± 6 yr) men with similar body mass (young men, 77.3 ± 5.9 kg; old men, 79.0 ± 6.2 kg; P = 0.7) who did not participate in the first experiment were recruited for the second set of experiments and visited the laboratories on three occasions to perform MVCs and fatiguing contractions with the same approach described for the measurement of endurance time. The first visit involved determining the endurance time for the position task (Fig. 1B) with the target force set at 25% MVC. All subjects performed the position task until failure. The other two sessions involved subjects performing the two fatiguing contractions in a randomized order, in a room adjacent to the PET/CT scanner for 90% of endurance time for the position task as determined in the first session. Immediately after the target time was achieved, subjects were placed in the PET/CT scanner to estimate [¹⁸F]-FDG uptake in selected muscles.

EMG signals were recorded during the two fatiguing contractions prior to the PET/CT scans with bipolar surface electrodes (Ag-AgCl; 8-mm diameter, 20-mm distance between electrodes) that were placed over the rectus femoris, vastus medialis, vastus lateralis, and, as a representative antagonist muscle, the short head of biceps femoris (Telemyo 2400 T G2; Noraxon). The electrodes were attached to the skin on the basis of established landmarks between the innervation zone and the end of the tendon. The EMG signals were amplified (×2,000), band-pass filtered (13–1,000 Hz), and recorded on a computer (2,000 samples/s).

The maximal EMG for the knee extensor muscles was calculated as the average amplitude over a 0.5-s interval about the peak MVC force. The maximal EMG for the biceps femoris was calculated as the average value over a 0.5-s interval about the peak-rectified EMG during maximal knee flexor activity. The maximal EMGs were recorded in the same experimental setup prior to the fatiguing contraction in each session. EMG activity during the fatiguing contractions was quantified by averaging the rectified EMG (aEMG) over the first and last 20 s of endurance time and over 20-s intervals centered about the 20, 40, 60, and 80% time points. EMG values were normalized to the aEMG obtained during the MVC. Muscle coactivation was quantified from the EMG measurements (12): (2 × antagonist aEMG/agonist aEMG + antagonist aEMG).

Prior to the two sessions in which PET/CT imaging was to be performed, subjects were required to fast for at least 4 h, to refrain from any kind of strenuous activity for at least 1 day, and to consume water and void the bladder just before the experiment (24). After the MVC had been determined, a buffalo-cap line was placed into the antecubital vein of the right arm to deliver the tracer. A finger stick was used before the injection of [¹⁸F]-FDG to determine the level of plasma glucose concentration. Approximately 2 min after the start of the fatiguing contraction, ∼260 MBq of [¹⁸F]-FDG in 5 ml of saline was infused into the vein and the sustained contractions continued thereafter for total durations of 848 ± 137 s (young men) and 751 ± 83 s (old men). Immediately after the injection of the tracer, the buffalo cap was removed. Once the fatiguing contraction had been sustained for the prescribed duration (90% of endurance time for the position task), subjects were moved into the scanner within 2 min and the acquisition and processing of the PET/CT images was performed following the standard protocol used in the Division of Nuclear Medicine, Department of Radiology, University of Colorado School of Medicine, Denver, CO.

PET scans were performed with a GE Discovery ST scanner (General Electric Medical Systems, Milwaukee, WI). The scanner has 24 PET detector rings of bismuth germinate crystals forming 47 two-dimensional imaging planes with a sampling interval of 3.27 mm each. PET scans were immediately preceded by CT scans for attenuation correction. Both sets of data were acquired consecutively with subjects on the same scanning table and in the same position. The feet and lower legs of each subject were secured to maintain co-registration. The lower limb was scanned from hip to feet in 2-min time frames, with 6 to 7 frames for each subject depending on the height of the individual. The data sets were reconstructed using an iterative method (ordered subsets-expectation maximization) with 21 subsets and 2 iterations with a Gaussian filter. All data sets were corrected for dead-time and random coincidence. The axial and in-plane resolution of the reconstructed images was ∼5 mm full-width at half maximum.

Standardized uptake values (SUV) were calculated for each muscle: SUV = [tissue radioactivity concentration/(injected dose/subject body mass)] (42). Because PET images were acquired immediately after the fatiguing contractions, the SUVs closely reflected the uptake of [¹⁸F]-FDG during the sustained contractions (24). Twenty-four regions of interest (ROI) were identified in the skeletal muscles of the lower limb. In the lower leg section, defined as 30% of the distance from the knee joint to the external malleolus, ROIs were located for tibialis anterior and posterior, gastrocnemius lateralis and medialis, soleus, and peroneus longus and brevis. In the thigh section, defined as 50% of the distance from the femoral head to the knee joint, ROIs were located for the knee extensors (vastus lateralis, vastus intermedius, vastus medialis, and rectus femoris) and flexors (biceps femoris short and long head, semimembranosus, and semitendinosus). In the hip region, which was 30 mm above the femoral head, ROIs were located for adductor magnus; sartorius; gracilis; iliopsoas; tensor fascie latae; quadratus femoris; obturator internus; and gluteus maximus, medius, and minimus. ROIs on the PET images were identified using cylinders with reference to the CT image (Fig. 2). The data were analyzed with the software package Carimas (Cardiac Image Analysis System), developed at the Turku PET Centre and validated by Nesterov et al. (31).

Fig. 2.

Transaxial CT/PET images at a mid-thigh level. Identification of leg muscles on a CT image (A) and the corresponding PET image (B) after one of the fatiguing contractions. Red denotes the greatest signal intensity followed by yellow, green, and blue. Am, adductor magnus; Bfs, biceps femoris short-head; Bfl, biceps femoris long-head; Gr, gracilis; Rf, rectus femoris; St, semitendinosus; Sm, semimembranosus; Sr, sartorius; Vi, vastus intermedius; Vl, vastus lateralis; Vm, vastus medialis.

As a translational reference for the muscle activation data, participants wore an accelerometer (ActiGraph GT3X, Pensacola, FL) mounted at the hip to record accelerations in the vertical and horizontal directions during waking hours for 7 consecutive days. The accelerometers recorded data in 60-s intervals and also counted the number of steps. The data were downloaded onto a Microsoft Excel spreadsheet using ActiLife software. Data recorded on the first and last days were discarded and only data sets for at least 4 complete days (including 1 weekend day) were used in the comparison, consistent with current recommendations (28).

Statistical analysis.

The dependent variables included endurance time for the position task, standardized uptake value for glucose in the selected muscles, and EMG activity of the principal and accessory muscles. Assessments with the Kolmogorov-Smirnov test confirmed that the distributions were normal. A two-factor, repeated-measures ANOVA (task × age) was used to compare MVC forces and endurance times for the force and position tasks (first experiment) between young and old men. Independent t-tests were used to compare MVC forces and endurance times for the position task sustained to failure. A four-factor ANOVA (task × age × time × muscle) with repeated measures on time was used to compare the changes in EMG activity and coactivation ratios. Changes in MVC force and glucose uptake were examined with a two-factor, repeated-measures ANOVA (task × age). A repeated-measures ANOVA was also performed to test the significance of differences in glucose uptake between muscles. After a significant F-test, pairwise differences were identified using paired and unpaired t-tests with Bonferroni corrections as post hoc tests. Stepwise, linear regression analysis using forward selection was performed to examine the associations between physical activity levels (average steps/day) and SUVs. The coefficient of determination (adjusted R²) was used to evaluate the fit of the model and guided the stepwise selection procedure. The significance level was set at P < 0.05. Statistical analyses were performed with SPSS software (SPSS version 17.0). Data are reported as means ± SD within text and tables and displayed as means ± SEM in figures.

RESULTS

The knee extensor MVC forces of the six men who participated in the first experiment were 500 ± 28 N for the young men and 331 ± 29 N for the old men (P = 0.005). The target force for the two fatiguing contractions was 125 ± 7 N for the young men and 83 ± 7 N (P < 0.01) for the old men. Endurance time for the force task (1,105 ± 103 s) was longer than that for the position task (770 ± 94 s; P < 0.001) with no difference between the young and old men (P = 0.5). The decline in MVC force for the two groups of participants was similar (25.7 ± 13.4% MVC; P = 0.2) after the two fatiguing contractions.

The muscle activation experiment compared [¹⁸F]-FDG uptake and EMG activity during two types of fatiguing contractions performed with the knee extensor muscles of the left leg. The MVC force at the beginning of each session was greater for the young men (462 ± 77 N) than for the old men (354 ± 91 N; P < 0.001) (Table 1). The target force for the two fatiguing contractions was 115 ± 19 N for the young men and 89 ± 20 N for the old men. There was no difference in endurance time between groups for the position task (943 ± 153 s vs. 835 ± 92 s, respectively; P = 0.166), which was longer than that in another study when the leg was in a more elevated position (39). Accordingly, the young (848 ± 137 s) and old (751 ± 83 s) men performed the fatiguing contractions prior to the PET/CT imaging for similar durations (P = 0.166). Moreover, the decline in MVC force immediately after the fatiguing contractions was similar for young and old men (24.5 ± 7.5% MVC and 22.9 ± 4.3% MVC; P = 0.247). However, MVC force was decreased to a greater extent after the position task had been performed to 90% of its endurance time compared with the force task for the young men (28.8 ± 2.8% MVC and 20.2 ± 7.7% MVC; P = 0.02) and for the old men (26.1 ± 2.6% MVC and 19.6 ± 2.4% MVC; P = 0.017).

Table 1.

MVC forces (N) for knee extensors of young and old men before and after force and position tasks

	Position task to failure			Force (90%)			Position (90%)
Group	Before	After	% Δ	Before	After	% Δ	Before	After	% Δ
Young	473 ± 85	339 ± 75*	−28 ± 11	466 ± 85	371 ± 127*	−20 ± 8	458 ± 68	326 ± 50*	−29 ± 3
Old	346 ± 115	243 ± 115*	−32 ± 11	384 ± 75	309 ± 66*	−19 ± 3	333 ± 72	246 ± 55*	−26 ± 3

Mean ± SD for MVC forces (N) of young and old men performed before and after force and position tasks during 3 sessions that comprised the second experiment: the position task to failure and the force and positions for a duration equal to 90% of endurance time for the position task.

^*P < 0.05 compared with before MVCs.

EMG amplitude.

The aEMG for the agonist muscles (rectus femoris, vastus lateralis, and vastus medialis) increased similarly during the two types of fatiguing contractions for both groups of subjects (task × muscle × age; P = 0.957), but at a greater rate for the young men (time × age; P = 0.009) (Fig. 3). The aEMG for the knee extensor muscles collapsed across the force and position tasks increased from 16.4 ± 3.9% to 48.4 ± 17.8% MVC for the young men and from 18.1 ± 4.7% to 31.1 ± 10.6% MVC for the old men. The aEMG for the knee extensor muscles at the beginning of the contraction (first 30 s) was similar for the force task (18.1 ± 5.7% MVC) and the position task (16.4 ± 4.9% MVC) for the young and old men. The aEMG for the knee extensor muscles was similar at the start and at 20% of endurance time for the two groups, but greater for the young men at 40, 60, 80, and 100% of endurance time (muscles × time × age; P = 0.009).

Fig. 3.

Change in EMG amplitude during the fatiguing contractions. Mean ± SEM for average EMG amplitude (aEMG, normalized to the peak MVC value) for the agonist muscles (vastus medialis, vastus lateralis, rectus femoris) and the antagonist muscle (short head of biceps femoris) during the 2 fatiguing contractions. The aEMG was averaged over 20-s intervals for each task at 6 time points that correspond to the absolute times at start, 20, 40, 60, 80, and 100% of the prescribed contraction duration (90% of endurance time for the position task). *P < 0.05 between young and old.

The aEMG for the antagonist muscle (short head of biceps femoris) increased similarly during the two fatiguing contractions for both groups of subjects (time × age; P = 0.14; Fig. 3). There were no differences in aEMG for the antagonist muscle between force and position tasks for either group of subjects (task × muscle × age; P = 0.59). An age × time interaction (P = 0.04) indicated that aEMG for the antagonist muscle was greater at each time point for the old men relative to the young men.

The coactivation ratios did not change during the sustained contractions and were comparable between tasks at each time point during the two contractions for the young men (force task start, 0.11 ± 0.05; force task end, 0.09 ± 0.05; position task start, 0.13 ± 0.07; position task end, 0.10 ± 0.04) and the old men (force task start, 0.40 ± 0.15; force task end, 0.34 ± 0.20; position task start, 0.39 ± 0.12; position task end, 0.35 ± 0.20) (task × age × time; P = 0.47). However, there was a significant main effect for age (P < 0.001) due to greater coactivation ratios for the old men in both the force and position tasks.

[¹⁸F]-FDG uptake in skeletal muscles.

Plasma glucose concentration immediately prior to the infusion of the [¹⁸F]-FDG was similar for the young (88 ± 7 mg/dl) and old (92 ± 8 mg/dl) men, which ensured that the measurement of glucose uptake began from comparable baseline conditions for the two groups of participants. SUVs (g/ml) were calculated for three-dimensional volumes of 24 leg muscles that were identified in CT images referenced to a standardized atlas (Table 2). When collapsed across the two fatiguing contractions (force and position tasks), the average SUVs (mean ± SD) for the 24 muscles were significantly greater for the old men (0.82 ± 0.26 g/ml) than for the young men (0.68 ± 0.32 g/ml) (age main effect, P < 0.01). Furthermore, average SUVs differed between the two fatiguing contractions for the young men (force, 0.64 ± 0.3 g/ml; position, 0.73 ± 0.3 g/ml), but not the old men (0.84 ± 0.3 g/ml and 0.79 ± 0.26 g/ml for young and old men, respectively) (age × task interaction, P < 0.001). Thus the amount of [¹⁸F]-FDG uptake was greater for the old men and they did not modulate the uptake across the two tasks (Fig. 4 and Fig. 5).

Table 2.

Glucose uptake in lower limb muscles after fatiguing contractions that required either force or position control

	Force		Position
	Young	Old	Young	Old
Knee extensors	0.86 ± 0.41	1.15 ± 0.34†	1.14 ± 0.41§	1.01 ± 0.40
Vastus lateralis	0.88 ± 0.48	1.04 ± 0.41	1.18 ± 0.51‡	1.02 ± 0.40
Vastus intermedius	1.00 ± 0.59	1.34 ± 0.21	1.23 ± 0.41	1.08 ± 0.47
Vastus medialis	0.84 ± 0.31	1.17 ± 0.29*	1.01 ± 0.31‡	0.94 ± 0.19
Rectus femoris	0.73 ± 0.17	1.17 ± 0.35*	1.15 ± 0.46‡	0.99 ± 0.34
Knee flexors	0.52 ± 0.15	0.69 ± 0.21†	0.60 ± 0.19§	0.64 ± 0.17
Biceps femoris short	0.69 ± 0.17	0.97 ± 0.22*	0.84 ± 0.18‡	0.86 ± 0.17
Biceps femoris long	0.45 ± 0.08	0.67 ± 0.13†	0.51 ± 0.07	0.61 ± 0.10
Semimembranosus	0.47 ± 0.09	0.56 ± 0.05*	0.52 ± 0.09	0.56 ± 0.07
Semitendinosus	0.44 ± 0.09	0.56 ± 0.08*	0.51 ± 0.13‡	0.54 ± 0.10
Hip muscles	0.62 ± 0.30	0.78 ± 0.30†	0.68 ± 0.33§	0.72 ± 0.16
Adductor magnus	0.52 ± 0.09	0.81 ± 0.19†	0.66 ± 0.20‡	0.72 ± 0.12
Sartorius	0.48 ± 0.08	0.67 ± 0.12†	0.58 ± 0.13‡	0.67 ± 0.12
Gracilis	0.48 ± 0.11	0.59 ± 0.08*	0.54 ± 0.11‡	0.56 ± 0.09
Iliopsoas	0.67 ± 0.19	0.95 ± 0.64	0.67 ± 0.19	0.75 ± 0.13
Tensor fasciae latae	0.52 ± 0.07	0.71 ± 0.19*	0.54 ± 0.09	0.63 ± 0.09
Quadratus femoris	0.54 ± 0.08	0.70 ± 0.11*	0.59 ± 0.10	0.74 ± 0.18
Obturator internus	1.33 ± 0.54	1.05 ± 0.15	1.44 ± 0.55	0.85 ± 0.13*
Gluteus maximus	0.46 ± 0.08	0.61 ± 0.09*	0.51 ± 0.10	0.63 ± 0.16
Gluteus medius	0.58 ± 0.11	0.87 ± 0.25*	0.61 ± 0.11	0.75 ± 0.13
Gluteus minimus	0.60 ± 0.14	0.87 ± 0.15†	0.63 ± 0.13	0.93 ± 0.13*
Lower leg muscles	0.60 ± 0.17	0.91 ± 0.24†	0.60 ± 0.16	0.86 ± 0.19†
Tibialis anterior	0.37 ± 0.12	0.75 ± 0.41*	0.40 ± 0.19	0.65 ± 0.26
Tibialis posterior	0.66 ± 0.15	1.15 ± 0.25*	0.66 ± 0.09	1.01 ± 0.16*
Gastrocnemius lateralis	0.63 ± 0.13	0.86 ± 0.13†	0.65 ± 0.10	0.85 ± 0.15*
Gastrocnemius medialis	0.64 ± 0.15	0.81 ± 0.10*	0.63 ± 0.13	0.78 ± 0.12*
Soleus	0.66 ± 0.12	0.99 ± 0.14*	0.65 ± 0.11	0.98 ± 0.12†
Peroneus longus brevis	0.62 ± 0.31	0.78 ± 0.27†	0.63 ± 0.17	0.90 ± 0.13†

SUVs (mean ± SD) for left lower limb muscles of 6 young and 6 old men.

^*P < 0.05 and

^†P < 0.01 between young and old men.

^‡P < 0.05 and

^§P < 0.01 between force and position tasks in young men.

Fig. 4.

Lower-limb PET images after position and force tasks. Longitudinal PET images (ventral, left; dorsal, right) taken after a young and an old man performed the force and position tasks for a similar duration. The high signal intensity (red) at the center of the pelvis resulted from accumulation of [¹⁸F]-FDG in the bladder.

Fig. 5.

Glucose uptake in lower limb muscles after 2 types of fatiguing contractions. Glucose uptake [SUV_bm (g/ml)] collapsed across 24 leg muscles was greater during the position task than the force task for young men, similar for both tasks in old men, and greater for old men than young men when collapsed across tasks. *P < 0.01 between force and position task for the young men and collapsed across tasks between the young and old men.

The group difference in [¹⁸F]-FDG uptake by the 24 muscles in the leg was also apparent for the primary agonist muscles (vastus lateralis, intermedius, and medius; rectus femoris). The SUVs for the force (1.15 ± 0.34 g/ml) and position tasks (1.01 ± 0.40 g/ml) were similar for the old men and the force-task value was greater than that for the young men, but the SUVs for the young men were greater for the position task (1.14 ± 0.41 g/ml) than for the force task (0.86 ± 0.41 g/ml). Similarly, the SUV for the antagonist muscles (short and long heads of biceps femoris, semimembranosus, and semitendinosus) after the force task was greater for the old men (0.69 ± 0.21 g/ml) than for the young men (0.52 ± 0.15 g/ml), and was greater for the position task (0.60 ± 0.19 g/ml) than the force task for the young men.

Furthermore, SUVs for the hip muscles (adductor magnus; sartorius; gracilis; iliopsoas; tensor fasciae latae; quadratus femoris; obturator internus; and gluteus maximus, medius, and minimus) were greater after the force task for the old men (0.78 ± 0.30 g/ml) than for the young men (0.62 ± 0.3 g/ml), and was greater for the position task (0.68 ± 0.33 g/ml) than the force task for the young men. In contrast, SUVs for the lower leg muscles (tibialis anterior and posterior,gastrocnemius lateralis and medialis, soleus, and peroneus longus and brevis) were greater for the old men after the force (0.91 ± 0.24 g/ml) and position (0.86 ± 0.19 g/ml) tasks than for the young men (force, 0.60 ± 0.17 g/ml; position, 0.60 ± 0.16 g/ml) (Fig. 6). Thus the more difficult fatiguing contraction, which required position control, was associated with greater overall levels of [¹⁸F]-FDG uptake (muscle activity) for the young men but not the old men, and the old men used greater amounts of agonist, antagonist, and accessory muscle activity than young men during the force task but not the position task, except for the lower leg muscles.

Fig. 6.

Glucose uptake in 4 lower limb muscle groups after 2 types of fatiguing contractions. Glucose uptake [SUV_bm (g/ml)] across 4 muscle groups for young and old men after completing the force and position tasks. Knee extensors: vastus lateralis; vastus intermedius; vastus medialis; and rectus femoris. Knee flexors: biceps femoris short and long heads; semimembranosus; and semitendinosus. Hip muscles: adductor magnus; sartorius; gracilis; iliopsoas; tensor fascie latae; quadratus femoris; obturator internus; and gluteus maximus, medius, and minimus. Lower leg muscles: tibialis anterior and posterior; gastrocnemius lateralis and medialis; soleus; and peroneus longus and brevis. *P < 0.01 between force and position tasks.

Physical activity.

Young men were more physically active than the old men (9,193 ± 1,829 vs. 4,393 ± 2,518 avg. steps/day for young men and old men, respectively; P = 0.004). Stepwise linear regression analysis using forward selection was adopted to develop a parsimonious model using the SUV values of the leg extensors and flexors, lower leg, and hip muscles and knee extensor MVC force to predict physical activity levels (steps/day) of young and old men. The stepwise procedure converged on a model (R² = 0.73; P < 0.001) that included the SUV values of the lower leg muscles (partial r = −0.72; P < 0.001) and knee extensor MVC force (partial r = 0.55; P = 0.006) (Fig. 7).

Fig. 7.

Prediction of daily levels of physical activity (steps/day) of young and old men. Daily levels of physical activity (steps/day) were strongly predicted by the linear combination of glucose uptake [SUV_bm (g/ml)] in the lower leg muscles of young and old men during 2 types of fatiguing contractions and MVC force. The lower leg muscles included tibialis anterior and posterior, lateral and medial gastrocnemius, soleus, and peroneus longus and brevis. The equation for the predicted physical activity (steps/day) was 9,614 + [−9,525 × SUV_bm (g/ml) in the lower leg muscles] + (11.225 × MVC force).

DISCUSSION

The main findings of the study were that the amount of [¹⁸F]-FDG uptake was greater in the leg muscles of old men than young men after performing the force task and that the young men, but not the old men, modulated [¹⁸F]-FDG uptake across the two fatiguing contractions. In contrast, the rate of increase in EMG amplitude for the agonist muscles was greater in the young men during the two fatiguing contractions, and there was no difference for either group of subjects in the rate of increase in EMG amplitude during the two tasks.

The surface EMG signal comprises the algebraic sum of field potentials associated with the currents that underlie muscle fiber action potentials and provides a temporal measure of muscle activity. Because muscle fiber action potentials correspond to polyphasic waveforms, the summation of the overlapping positive and negative phases of concurrent field potentials reduces the absolute amplitude of the signal (1). This effect is known as amplitude cancellation (14). The reduction in signal amplitude due to cancellation from the summation process increases progressively with contraction force, reaching a value of about 70% during a maximal contraction (7, 23), and increases progressively during sustained low-force contractions (14). Amplitude cancellation reduces the sensitivity of a surface EMG signal to detect modest changes in the associated motor unit activity. For example, Mottram et al. (30) found that sustaining a submaximal isometric contraction with position control involved greater changes in motor unit recruitment and rate coding than when the task required force control, yet there was no difference in the rate of change in the amplitude of the surface EMG signal between the two conditions.

Because an EMG signal is also influenced by the properties of muscle fibers and characteristics of the recording configuration in addition to the activation of motor units by the nervous system (14, 15), the relative rate of change in EMG amplitude during the two types of fatiguing contractions varies across muscles and subjects (20, 25, 27, 36). In contrast to the results in the current study, for example, we have previously found that EMG amplitude for the knee extensors increased more rapidly during the position task when the target force was 20% MVC and knee angle was 1.57 rad (39). Changes in EMG amplitude when measured with bipolar surface electrodes, therefore, provide an unreliable measure of the difference in the control strategy used during these two types of fatiguing contractions.

As an alternative approach to quantifying muscle activity, the current study used PET/CT measurement of [¹⁸F]-FDG uptake by skeletal muscles during the two fatiguing contractions. Pappas et al. (33) have shown a close association between [¹⁸F]-FDG uptake and muscle contraction intensity. A fivefold increase in resistance during elbow flexion, for example, increased [¹⁸F]-FDG uptake in the biceps brachii by a factor of 4.9. Because [¹⁸F]-FDG PET imaging relies on active muscle cells increasing glucose uptake (17, 34) and [¹⁸F]-FDG uptake is closely correlated with exercise intensity (15, 24, 33), it was possible to assess spatial differences in the magnitude of muscle activation across the two tasks and age groups as measured by [¹⁸F]-FDG that had accumulated in the muscles during the fatiguing contractions.

The results of the current study indicating a difference in [¹⁸F]-FDG uptake by the young men across the two types of fatiguing contractions are consistent with previous work showing greater reflex responses elicited by activating group I afferents during position control relative to force control (2, 6, 10, 26). When submaximal (20% MVC force) isometric contractions were sustained with the wrist extensors for as long as possible by requiring either force or position control, Baudry et al. (5) observed a difference in the time course of the change in presynaptic inhibition of homonymous Ia afferent input for extensor carpi radialis during the two types of fatiguing contractions. The task-dependent modulation of synaptic input received by the involved motor neurons underlies the different adjustments in motor unit recruitment and rate coding during the two tasks (3, 30, 40). The current findings indicate that the differences in motor unit activity during force and position control was manifested as greater [¹⁸F]-FDG uptake by skeletal muscles during position control, at least for the young men.

The old men, however, did not exhibit a difference in [¹⁸F]-FDG uptake across the two fatiguing contractions, which is consistent with a lack of modulation in the amount of Ia presynaptic inhibition when subjects were asked to perform brief isometric contractions requiring either force or position control (4). Whereas young adults modulated the strength of Ia presynaptic inhibition across the two load types when performing steady contractions, the old adults tended to vary the amount of antagonist coactivation. Old adults often use greater amounts of antagonist coactivation when performing actions in which postural stability is critical (7, 18, 19, 22, 43, 44), and the preference for this strategy was observed in the current study by greater [¹⁸F]-FDG uptake in the knee extensors, knee flexors, and hip muscles after the force task relative to the young men and the absence of a difference across the two tasks. Thus the old men had greater amounts of [¹⁸F]-FDG uptake after the force task than the young men, but similar uptake values after the two types of fatiguing contractions.

In contrast to [¹⁸F]-FDG uptake values for knee extensors, knee flexors, and hip muscles, those for lower leg muscles were greater for the old men relative to the young men after both types of fatiguing contractions. The combination of the age × task interaction for the uptake values for the lower leg muscles after the fatiguing contractions and knee extensor MVC forces were able to predict physical activity levels of young and old men. The analysis indicated that the least physically active participants had lower knee extensor MVC forces and greater [¹⁸F]-FDG uptake in the lower leg muscles during the two fatiguing contractions. Therefore, the physical activity levels observed in the participants, which was strongly influenced by the amount of ambulation participants performed (29), was not related to any of the age- or task-related differences in [¹⁸F]-FDG uptake during the fatiguing contractions with the knee extensor muscles. Instead, the findings indicate that [¹⁸F]-FDG uptake in the lower leg muscles, which produce most of the joint power for ambulation (19), were greater in participants with lower physical activity levels.

In conclusion, the findings of the current study demonstrate that PET/CT imaging of [¹⁸F]-FDG uptake, but not surface EMG recordings, detected the modulation of muscle activity across the two types of fatiguing contractions by young men and the greater levels of muscle activity among agonist, antagonist, and accessory muscles after the force task for old men relative to young men. The [¹⁸F]-FDG measurements of muscle activity obtained with PET/CT imaging are consistent with age-associated differences in the modulation of motor unit activity and spinal reflexes during tasks that require either force or position control, but provide greater spatial information about the magnitude of the difference in muscle activity between young and old men when performing fatiguing contractions.

GRANTS

The work was supported by National Institute on Aging Grant AG033744 to T. Rudroff.

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: T.R. and W.C.K. conceived and designed research; T.R. and D.E.B. performed experiments; T.R. analyzed data; T.R., K.K.K., D.E.B., W.C.K., and R.M.E. interpreted results of experiments; T.R. prepared figures; T.R. and J.R.G. drafted manuscript; T.R., K.K.K., W.C.K., and R.M.E. edited and revised manuscript; T.R. and R.M.E. approved final version of manuscript.