Aging is associated with a functional decline of the oxidative metabolism due to progressive limitations of both O2 delivery and utilization. Priming exercise (PE) increases the speed of adjustment of oxidative metabolism during successive moderate-intensity transitions. We tested the hypothesis that such improvement is due to a better matching of O2 delivery to utilization within the working muscles. In 21 healthy older adults (65.7 ± 5 yr), we measured contemporaneously noninvasive indexes of the overall speed of adjustment of the oxidative metabolism (i.e., pulmonary V̇o2 kinetics), of the bulk O2 delivery (i.e., cardiac output), and of the rate of muscle deoxygenation (i.e., deoxygenated hemoglobin, HHb) during moderate-intensity step transitions, either with (ModB) or without (ModA) prior PE. The local matching of O2 delivery to utilization was evaluated by the ΔHHb/ΔV̇o2 ratio index. The overall speed of adjustment of the V̇o2 kinetics was significantly increased in ModB compared with ModA (P < 0.05). On the contrary, the kinetics of cardiac output was unaffected by PE. At the muscle level, ModB was associated with a significant reduction of the “overshoot” in the ΔHHb/ΔV̇o2 ratio compared with ModA (P < 0.05), suggesting an improved O2 delivery. Our data are compatible with the hypothesis that, in older adults, PE, prior to moderate-intensity exercise, beneficially affects the speed of adjustment of oxidative metabolism due to an acute improvement of the local matching of O2 delivery to utilization.
- V̇o2 kinetics
- cardiac output kinetics
- near-infrared spectroscopy
the dynamic characteristics of pulmonary O2 uptake (V̇o2) kinetics at the onset of moderate, heavy, or very heavy-intensity exercise transitions, has been extensively investigated (6, 8, 23, 39, 43, 47, 48, 54, 56, 57, 70, 71) as a means to unveil the mechanisms that control/limit oxidative metabolism in vivo (31). The results obtained in these experiments have stirred a scientific debate between those in favor of a central (O2 delivery) and those in support of a peripheral (O2 utilization) limitation of oxidative metabolism. This issue is of particular importance in older adults as aging is characterized by a significant and progressive decline in the speed of adjustment of oxidative metabolism during metabolic transitions (i.e., V̇o2 kinetics) (4, 5, 10, 13, 17, 18, 62), which results in a larger O2 deficit and in a greater reliance on substrate-level phosphorylation to provide ATP for a given activity (21). For the above reason, older adults may experience reduced tolerance to repeated exercise bouts compared with young adults (21). The slower V̇o2 kinetics in older adults is commonly ascribed to a muscle O2 delivery limitation (45, 50, 52, 64) and to a slower speed of adjustment of cellular O2 utilization (16, 49), their relative role remaining controversial.
Priming exercise (PE, also named heavy warm-up) is a powerful experimental tool that has been classically applied to the study of the limitations of oxidative metabolism in young (14, 19, 28, 37, 38) and in older adults (19, 36, 62). By applying PE, several studies have demonstrated an increase in the speed of adjustment of oxidative metabolism, during subsequent moderate-intensity step transitions, in older adults and in young presenting initially slow V̇o2 kinetics, but not in young adults presenting fast kinetics (19, 62). Since PE is believed to produce an acute improvement of oxygen delivery (28) and muscle perfusion (19, 37, 38), the above finding has been interpreted as indirect evidence of a larger role of muscle O2 delivery in the limitation of oxidative metabolism in older compared with young adults. In recent years, however, some studies showed that a single bout of PE can activate pyruvate dehydrogenase in young subjects (37) and older adults (36). These data suggest that enzymes involved in the oxidative metabolism are liable to acute, short-term adaptation, in response to PE, and, therefore, that metabolic inertia may also play a role in the limitation of oxidative metabolism at exercise onset.
In summary, following PE, an increased ability to utilize O2 might parallel a faster adjustment of muscle perfusion. By examining the dynamic coupling between O2 delivery and uptake, both at the systemic level and at the peripheral level, inference can be made on the relative role of O2 delivery and O2 utilization on the regulation of oxidative metabolism during moderate-intensity exercise transitions in older adults. Toward this aim, our study is the first to evaluate, in parallel, pulmonary O2 uptake, cardiovascular/bulk O2 delivery, by noninvasive recording of Q̇ (25, 44, 69), and noninvasive muscular deoxygenation kinetics, by near-infrared spectroscopy (NIRS) (32), during the transition to moderate intensity with or without prior priming exercise. We hypothesized that, in older adults, PE would increase the speed of adjustment of oxidative metabolism during successive moderate-intensity transitions, attributable to a better matching of O2 delivery to O2 utilization.
Fifty-four healthy older males were recruited by local advertisements in the metropolitan area of Verona (Italy). Inclusion criteria were an age of >60 yr and a sedentary lifestyle (2). Exclusion criteria were obesity (body mass index of >30), diabetes or other metabolic disorders, hypertension, heart diseases, and the use of medications affecting the cardiovascular function. All of the subjects underwent a preliminary medical examination, to evaluate possible exclusion criteria and a preliminary cycle-ergometer stress test, to exclude pathological responses to exercise. Twenty-one men were eligible for inclusion (65.7 ± 5.0 yr) and participated in the study after giving written informed consent (Table 1). The reasons for exclusion were pharmacologically treated high blood pressure (23 participants excluded), ischemic heart disease (4 participants excluded), diabetes (2 participants excluded), neurological/orthopedic conditions (4 participants excluded). In conformity to the principles of the Declaration of Helsinki and U.S. Title 45, the study was approved by the Ethical Committee of the Department of Neurological, Neuropsychological, Morphological and Movement Sciences, University of Verona, and subjects were informed of the aims, the procedures, and possible risks involved in the study, and they gave written consent.
All of the tests were carried out under medical supervision and ECG monitoring, in three different, nonconsecutive days at the Exercise Physiology Laboratory of the School of Human Movement Sciences, University of Verona. The subjects were asked to abstain from vigorous physical activity the day before the tests and to have a light meal without coffee 2 h before arriving to the laboratory. All tests were performed at approximately the same time of day. A possible confounding effect of diet on the speed of adjustment of oxidative metabolism was minimized by asking the subjects to consume the same mixed-composition meal each of the three experimental days.
On the occasion of the first appointment, subjects underwent the anthropometric measurements followed by a maximal incremental cycle ergometer test to exhaustion. In the two successive experimental sessions (summarized in Fig. 1), both separated by at least 2 days of recovery, subjects performed one of the following cycle ergometer tests, in randomized order: three-step transitions at moderate-exercise intensity, without PE (ModA), and two-step transitions at moderate-exercise intensity, preceded by a PE at very heavy intensity (ModB).
The incremental test consisted in 3 min at rest, 3 min of warm-up at 30 W followed by a continuous increment, every 1 min, of the workload by 10–15 W, depending on the prospective training status of each subject, until voluntary exhaustion. The latter was defined as the inability to maintain the pedaling frequency (60–80 revolutions/min), despite vigorous encouragement by the experimenters. The test was considered valid when the following criteria were attained: respiratory exchange ratio (R) > 1.1 and heart rate (HR) higher than 90% of the predicted maximum based on age (3). Invalid tests were repeated after 2 days.
For moderate-intensity exercises, the workload was set so as to elicit a V̇o2 equal to the 80% of the first ventilatory threshold (VT1) measured during the incremental exercise (9). For PE, the workload was set at 50% of the difference between V̇o2 max and the V̇o2 at the second ventilatory threshold (VT2). Step transitions, each lasting 6 min, were preceded by 1 min of rest and 1 min of free-cycling and followed by 5 min of recovery. The two ModB sets (i.e., priming exercise + moderate-intensity exercise) were separated by 10 min of recovery (see Fig. 1 for a summary scheme).
During step transitions, the subjects were asked to maintain a pedaling frequency of 30 rpm for free cycling (FC) and between 60 and 80 rpm when the workload was applied (visual display and auditory feedback by a metronome). A FC as opposed to a resting starting point was chosen to limit the muscle pump effect on the NIRS measurements. We chose a low pedaling frequency to minimize the increase in V̇o2 before the onset of the actual moderate exercise phase (pedaling against a 0-W resistance still represents a work rate of ∼1 W per revolution, and pilot tests in our laboratory showed that 30 rpm with no resistance would result in an increase in V̇o2 of ∼150 ml/min above resting) and, therefore, to obtain a larger possible amplitude of the response for the V̇o2 kinetics. A 1-min duration for the FC was chosen to contain the overall duration of the testing sessions and, therefore, favor subject's compliance. Given the small increase in the V̇o2, a steady baseline was established with 1 min of this FC.
Body mass and stature were measured on the first testing day, before the incremental test (analogical scale 761 and portable stadiometer 214; SECA, Hamburg, Germany). Concurrently, subcutaneous skin-fold thickness was measured sequentially, using a pincer-type caliper at six sites (pectoral, biceps, subscapular, abdominal, iliac, and thigh) (skin-fold caliper; Holtain, Crymych, UK) for percent body fat calculation (30).
All of the exercise tests were performed on an electronically braked cycle ergometer (Excalibur Sport, Lode, Groningen, The Netherlands) connected to and operated by a metabolic cart (Quark b2, Cosmed, Rome, Italy) that also allowed continuous, breath-by-breath measures of pulmonary gas exchange and ventilation (at the mouth) and HR. Before each test, the gas analyzers and the turbine flow meter of the system were calibrated, following the manufacturer's instructions, by using a gas mixture of known concentration (FO2: 0.16; FCO2: 0.05; N2 as balance) and a 3.0-liter calibrated syringe.
For measurement of the Q̇, the left arm was bent and suspended at the level of the xyphoid with a sling, while arterial pressure profile was continuously recorded at a fingertip by using a noninvasive photoplethysmographic method (Portapres, FMS, Amsterdam, The Netherlands). Pressure values were corrected for the height difference between the heart and the fingertip, as indicated by the manufacturer. Furthermore, during the last repetition of ModA and the second ModB set, Q̇ was measured at rest and during the last minute of exercise (the exercise duration was prolonged to 7 min), by means of an inert gas rebreathing method (68) (Innocor, Innovision, Odense, Denmark).
Muscle oxygenation was evaluated by means of a frequency-domain multidistance (FDMD) NIRS system (OxiplexTS, ISS, Champaign, IL) that provides a continuous measurement of absolute concentrations (micromolar) of oxyhemoglobin ([O2Hb]) and deoxyhemoglobin ([HHb]) (26). The physical principles and the technology behind the FDMD method have been given in detail elsewhere (33). After shaving, cleaning, and drying of the skin area, the NIRS lightweight plastic probe was longitudinally positioned on the belly of the vastus lateralis muscle 15 cm above the patella and attached to the skin with a biadhesive tape. The position of the probe on the thigh was then pen-marked to check for sliding and to allow repositioning on the following tests days. Finally, the probe was secured with elastic bandages around the thigh. The apparatus was calibrated on each test day after a warm-up of at least 30 min, following the manufacturer's recommendations.
During repetitions 1 and 2 of ModA and the first ModB set, arterialized capillary blood samples were taken from an earlobe at rest, at FC, and at the end of the step exercises at min 0, 1, 3, and 5, for the determination of blood concentration ([La]b, mmol) by means of an electro-enzymatic method (Biosen C_line, EKF Diagnostics, Barleben, Germany).
VT1 and VT2 were individually estimated by applying the Wasserman method (9). V̇o2 max and all maximal variables were calculated as the average of the last 10 s before the end of the exercise.
The V̇o2 series were time aligned with the onset of work rate and interpolated to 1-s intervals (46). Then, the two to three series obtained in the different exercise protocols were overlapped and averaged in 1-s series to obtain a single data file for each condition (i.e., ModA and ModB) for each subject.
Heart stroke volume (SV) was determined on a beat-by-beat basis by means of the Modelflow method (69), applied off-line to the pulse pressure profiles using the Beatscope software (Portapres, FMS, Amsterdam, The Netherlands). Beat-by-beat Q̇ was then calculated as the product of the corresponding SV times the corresponding beat HR. Total peripheral resistances (TPR) were calculated as a ratio of mean arterial pressure (mmHg), as measured beat-by-beat, to Q̇ (ml/s), and expressed in arbitrary medical units (MU; mmHg·ml−1·s).
Since the aortic compliance used by the Modelflow algorithm may substantially differ from that of a given individual, Modelflow Q̇ needs to be corrected with an independent and valid measure. To this aim, each beat-by-beat Q̇ value was multiplied by an individual correction factors [equal to the ratio of the steady-state Q̇ values obtained by the Modelflow algorithm (65) at steady state during the constant loads exercise and the Q̇ values measured by the inert gas rebreathing method].
For both ModA and ModB, the beat-by-beat Q̇ and TPR values obtained from the first exercise transition were time aligned for the exercise onset and interpolated (1 s). The HHb data obtained from the first exercise transition were time-aligned and averaged to a 1-s interval.
Steady-state values of V̇o2, respiratory exchange ratio (R), HR, Q̇, TPR, and HHb during moderate-intensity transitions were calculated by averaging the corresponding data recorded during the last 30 s of each repetition.
Net blood lactate accumulation during square wave exercises was calculated as the difference between the peak [La]b (i.e., the highest value found during the recovery phase) and the [La]b during FC (22).
Net mechanical efficiency (η) during ModA and ModB was calculated as the ratio between the net increment of workload and the corresponding net increase of metabolic power. The latter, in turn, was obtained from the amplitude of the O2 uptake response multiplied by the energy equivalent of O2 in Joules per milliliter of O2, calculated on the basis of the R that was measured at steady state (22).
V̇o2 responses to exercise above baseline, from the onset of workload to the end of the exercise (360 s), were fitted by means of a two-component exponential model, where the first component corresponds to the so-called cardiodynamic (or phase I) and the second component to the metabolic or primary phase (also called phase II) of the kinetics (44, 70). (1) where Y(t) represents the increase in V̇o2 at the onset of exercise, A1 and A2 are the amplitudes of the first and second component, and τ1 and τ2 and TD1 and TD2 correspond to the time constants and time delays of the expression, respectively. H(t − TD1) and H(t −TD2) are the Heaviside functions defined as (2)
The time to reach the 63% of the response, the so called effective time constant (τ′), was calculated as (41): (3) where τ, TD, and a are the time constant, the time delay, and the amplitude of the cardiodynamic (1) and primary (2) phase, respectively. In this model, TD2 corresponds to the duration of phase I.
Additionally, O2 deficit was calculated as the integral of the difference between the individual steady state V̇o2 and the experimentally measured V̇o2 as a function of time of exercise, set equal to 360 s (12).
Q̇, TPR, and HHb data were fitted by a linear function from −60 s through to the physiological TD that follows exercise onset. After that TD, a monoexponential function was fitted up to 360 s for Q̇ and TPR. For the NIRS signal, a shorter fitting window was chosen (i.e., 120 s) to avoid the impact of variations in the HHb signal that are known to occur after about 180–240 s from exercise onset, on the fitting of the initial portion of the transient, while allowing the reaching of a steady state (i.e., four time constants) (26, 27, 55). The parameters of the models (i.e., baseline, amplitude, time constant, τ; time delay, TD) were estimated using a weighted nonlinear least-squares fitting procedure and implemented with Sigmaplot 11 (Systat Software San Jose, CA). For HHb, the TD derived by our exponential modeling is equivalent to the so-called “calculated time delay” (i.e., the first data-point showing a consistent increase in HHb over time after exercise onset) used by other authors (55). For these three variables, the τ′ was calculated as the sum of τ and TD.
For the calculation of the ΔHHb/ΔV̇o2 ratio index (53), the individual second-by-second ΔHHb and V̇o2 data were firstly normalized (from 0%, corresponding to the FC baseline, to 100% reflecting the steady-state response). Then, ΔHHb and V̇o2 data were time-aligned by left-shifting the V̇o2 data, accounting for the duration of the cardiodynamic phase. Finally, a peak value was calculated as the highest data point of the ΔHHb/ΔV̇o2 curve as a function of time, between 0 and 120 s.
The data are presented as the mean and SD. ModA and ModB conditions were compared by paired Student's t-test (Sigmaplot 11, Systat). The significance level was set at P < 0.05 for all comparisons. When a significant difference was detected, Cohen's d effect size was also determined (15). On the basis of variances in the parameters of V̇o2 kinetics measured in our laboratory in older adults (within subject variation ∼10%), using a power of 0.8 and alpha level of 0.05, sample size analysis for t-test, indicated that the minimum number of subjects required to detect a significant difference (i.e., a 15% variation) was six.
The average values of the main variables assessed during the maximal incremental test, along with physical characteristics of the 21 subjects included in the study, are reported in Table 1. During the incremental test, V̇o2 (ml/min) increased as a linear function of power output (W) (y = 10.5x + 543). A plateau in V̇o2O2, was attained, at volitional exhaustion, in 50% of the subjects, while a “peak” V̇o2 (i.e., HR > 95% HRmax, R >1.1) (3) was reached in the remaining 50%.
The priming exercise was performed at an average workload of 153 ± 27 W and elicited a V̇o2 of 2,311 ± 387 ml/min (98 ± 8.2% V̇o2 max), a heart rate of 147 ± 12 bpm (94.7 ± 5.4% HRmax), and a net blood lactate accumulation of 7.0 ± 1.7 mmol/l, confirming a high-intensity exercise session.
The main physiological variables measured at FC and at steady state during the moderate-intensity exercise, either with or without priming (ModB, ModA), are summarized in Table 2. No difference was detected between ModB and ModA in power output, V̇o2, Q̇, TPR, or HHb. On the contrary, HR was significantly higher, and R was lower for ModB compared with ModA. Furthermore, the net blood lactate accumulation was 1.42 ± 0.86 mmol/l for ModA exercise (indirectly confirming an exercise of moderate intensity) and had a negative value (−1.65 ± 0.60 mmol/l, significantly different from ModA) for ModB exercise (suggesting a reutilization of this metabolite).
Net mechanical efficiency was 0.141 ± 0.03 in ModA, and it was unaffected by PE (0.146 ± 0.03 for ModB).
O2 deficit was 774 ± 182 ml for ModA step transitions, and it was significantly reduced in ModB (673 ± 224 ml). The Cohen's d effect size was 0.49 (i.e., medium).
Pulmonary V̇o2 Kinetics
Average data regarding the V̇o2 kinetics are presented in Table 3, and Fig. 2 reports the experimental data of a typical subject. While no significant differences were detected in τ2 between ModA and ModB step transitions, on the contrary, phase I duration and τ′ became significantly shorter in ModB compared with ModA (Cohen's d effect size for τ′ was 0.44, rated as medium).
The speed of adjustment of Q̇ and TPR during moderate-intensity step transitions was significantly faster compared with pulmonary V̇o2 kinetics (Table 3). Furthermore, no difference was detected in the speed of adjustment of either Q̇ or TPR following priming exercise compared with ModA.
Muscle Deoxygenation Kinetics
The speed of adjustment of muscle O2 extraction during moderate intensity step transitions was significantly faster compared with τ2 of pulmonary V̇o2 kinetics. Priming produced a significant shortening of the HHb TD but a significant increase of the HHb τ (Table 3). The net effect of these opposite modifications was an apparently unchanged τ′ between ModA and ModB step transitions.
ΔHHb/ΔV̇o2 Ratio Index
During ModA exercise step transitions, the ΔHHb/ΔV̇o2 ratio (Fig. 3) displayed an “overshoot” (relative to the steady state ratio of 1.0, representing the balance of delivery to utilization) that was nearly abolished during the exercise preceded by PE. Also, the individual peak value of the ΔHHb/ΔV̇o2 ratio was significantly larger for ModA than for ModB exercise (2.0 ± 0.4; 1.7 ± 0.4). Cohen's d effect size was 0.75 (i.e., large). These data suggest a larger mismatch between local O2 delivery and utilization in ModA compared with ModB exercise.
The present study was the first to evaluate, in parallel, the kinetics of V̇o2, Q̇ (bulk O2 delivery), TPR, and muscular deoxygenation (HHb) of older adults at the onset of moderate-intensity cycling exercise and the effect of a very high-intensity “priming” exercise. By examining the dynamic coupling between O2 delivery and uptake, both at the systemic level and at the local (muscle) level, the purpose of the present study was to examine the relative role of O2 delivery and O2 utilization in the limitation of oxidative metabolism, during moderate-intensity exercise transitions, in older adults.
The main findings were 1) PE increased the overall speed of adjustment of pulmonary V̇o2 (i.e., significantly shorter τ′ and reduced oxygen deficit), 2) PE did not modify the speed of adjustment of bulk O2 delivery (i.e., Q̇) and TPR; 3) at the muscle level, PE produced significant yet opposite changes in the HHb kinetics variables (i.e., shorter TD and longer τ), without modifications in τ′; and 4) at the muscular level, PE brought about a reduction in the “overshoot” of the ΔHHb/ΔV̇o2 ratio, suggesting a better matching of O2 delivery to O2 utilization during the exercise on-transition. Our data are compatible with the hypothesis that, in older adults, priming exercise, prior to moderate-intensity exercise, may beneficially affect oxidative metabolism due to acute improvement of the local matching of O2 delivery to O2 utilization.
In agreement with the literature, our data confirm slower V̇o2 kinetics in older compared with young adults (18). Compared with previous work in older adults (19, 62), our study found shorter τ2 values. Such difference could be related to slightly higher V̇o2 values in our subjects compared with those reported in the quoted papers.
Whereas other studies documented that a priming exercise increases the speed of adjustment of oxidative metabolism, during moderate-intensity step transitions in older adults, due to a reduction of τ2, our data show unchanged τ2 and significantly shorter phase I duration. Two considerations can be made on this issue: 1) As the V̇o2 kinetics was relatively fast, the potential speeding with PE manipulation may not be detectable. Indeed, as observed in young adults, when τ2 is close to 20 s (i.e., to the time constant of the splitting of phosphocreatine), any improvement of the local oxygen delivery may hardly affect the rate of adjustment of V̇o2 kinetics (42). 2) A modeling study (7) suggested that modifications of phase I duration, even in the presence of no changes in muscle V̇o2 kinetics, may modify the phase II τ2. On the basis of this model, a reduction in the duration of phase I, in the presence of an unchanged rate of muscle O2 utilization, would result in a longer phase II τ2. This model could lead to the speculation that our “apparently” unchanged τ2 of phase II V̇o2, in the presence of a shorter phase I duration, is, indeed, indicative of a faster adjustment of muscle O2 utilization in moderate-intensity step transitions, when preceded by very heavy-intensity warm-up. In support of this view (i.e., accelerated oxidative metabolism after PE), our study documented a significant reduction of the total O2 deficit (approximately −12%).
Blood lactate concentration was elevated, prior to ModB, as a consequence of the preceding heavy-intensity exercise bout. During ModA exercise, there was a net blood lactate accumulation. On the contrary, the negative net lactate concentration balance in ModB (−1.65 ± 0.6 mmol) reflects a greater lactate oxidation (see Refs. 11 and 29 for reviews on the topic). These data are consistent with faster activation of mitochondrial oxidative phosphorylation during the transition to moderate-intensity exercise after priming (40, 67) and, as a result, with a reduced reliance on anaerobic glycolysis and substrate level phosphorylation for ATP production (further confirmed by a reduced R).
In summary, the different and independent approaches converge to suggest that the response of muscular O2 uptake is faster after PE.
Q̇ Kinetics (Index of Bulk O2 Delivery)
Q̇ was used as an index of bulk O2 delivery response based on the assumption that CaO2 remains unchanged in the exercise transition in healthy subjects who exercise in normoxia at moderate intensity (they operate in the flat portion of the O2 dissociation curve and no increase in the physiological shunt is expected). Under these conditions, the kinetics of O2 delivery (i.e., the product of Q̇ and CaO2) would be well represented by that of Q̇ (1, 44). Previous studies (19, 62) have considered HR as a surrogate of the rate of adjustment of Q̇ and presumably O2 delivery. This approach is based on the assumption that stroke volume changes little after the first heartbeats of the transition, especially during moderate-intensity dynamic exercise. The above-cited studies documented either no change (19) or a slowing of HR on-kinetics (62) as a result of priming exercise, the latter being reflective of changes in the autonomic control of HR (i.e., reduced parasympathetic control and increased sympathetic tone). Yet, these authors suggested that priming exercise increased muscle perfusion and O2 delivery to the working muscles, based on the observation that the HR was elevated before and throughout the moderate-intensity exercise, when it was preceded by the priming exercise. However, solely on the basis of the HR, it cannot be excluded that opposite changes in SV may occur following priming exercise, leading to small, if any, changes in Q̇ (25). Indeed, our HR data confirm the findings of previous studies (i.e., an increased HR at baseline and throughout exercise). However, we detected no changes in either baseline or steady state values and in the speed of adjustment of Q̇ following PE. Therefore, in agreement with the results of studies on young individuals (25), our experimental data do not support the hypothesis that the bulk O2 delivery is elevated or accelerated during the exercise transition after a very heavy-intensity warm-up.
In our study, as in several other investigations, the HHb signal has been used as a surrogate of O2 extraction, which reflects the balance between O2 utilization and O2 delivery in the region of NIRS interrogation (21, 32). The NIRS parameters have been traditionally interpreted as follows: the TD preceding the increase in HHb has been typically interpreted as evidence of a tight coupling between muscular O2 uptake and local O2 delivery during this interval (20, 32); the fast increase in HHb concentration after the TD has been interpreted as suggestive of a slower microvascular perfusion response relative to the O2 uptake (26).
In agreement with previous studies on older adults (18, 19, 36), our data confirm that priming exercise produces, on the subsequent moderate-intensity step transitions, a shorter TD and a slower τ. A shorter TD may indicate that the mismatch between local muscle O2 utilization and O2 delivery occurs earlier following PE, with the muscle O2 consumption in excess of local O2 delivery, possibly because of a faster activation of mitochondrial enzymes and/or of PDH following heavy-intensity warm-up (35, 36).
A greater τ HHb following PE may suggest that, once the matching of O2 delivery and utilization is broken (i.e., after the TD), the adjustment of muscle oxygen consumption proceeds at a slower rate compared with the adjustment of O2 delivery. Therefore, despite the possibly beneficial effect of PE on O2 extraction, in the first few seconds of exercise, O2 delivery becomes relatively faster than extraction (compared to ModA) as the exercise continues. Finally, in agreement with the majority of studies, our data indicate that the overall effect of the opposite changes in the HHb kinetic parameters is an unchanged τ′ (18, 19, 36).
In summary, NIRS data can be interpreted to mean that an increased O2 delivery is associated with an increased extraction following PE compared with control moderate exercise conditions, in healthy older adults. Yet, the relative role of these effects in the improvement of oxidative metabolism remains somewhat elusive. To elucidate this point, we calculated the ΔHHb/ΔV̇o2 ratio (18, 53). This index combines NIRS data with breath-by-breath V̇o2 measures, providing an indication of the rate of adjustment for O2 extraction compared with the rate of adjustment for V̇o2 and thus allowing indirect inference on the matching of microvascular blood flow distribution to local O2 utilization (20). The effect of heavy priming exercise is a reduction of the peak value of ΔHHb/ΔV̇o2O2 as a function of time. This evidence suggests that PE may cause a faster adjustment of O2 delivery that is in excess of the (possibly also improved) speed of the muscle O2 utilization. This evidence may also suggest a possible limitation in O2 extraction (related to the fundamental metabolic control of oxidative phosphorylation), in older adults that may be masked under flow-limited (i.e., no PE, ModA) conditions, becoming evident only under flow-enhanced conditions (PE, ModB).
The present finding of an O2 delivery constraint to V̇o2 kinetics in older adults (ameliorated with PE) is supported by other observations of altered physiological responses to exercise with aging. An impaired vasodilation in active muscles has been documented in healthy sedentary (59, 61) and endurance-trained older adults (60). The potential mechanisms of the impaired local hemodynamic response are controversial and remain to be clearly determined in humans. Recent studies on animals suggest that the effects of aging on microvascular regulation are heterogeneous and involve age-related modifications of both endothelium and smooth muscle-mediated vasomotor responses (50–52). Furthermore, the accumulation of reactive oxygen species and the low-grade inflammation associated with aging and with the increase in proinflammatory agents, can negatively affect the nitric-oxide-mediated vasomotor signal (63) and alter the structure of microvascular endothelium (58) impairing blood flow. Priming exercise seems to restore the dynamic blood flow response in the exercising muscles of older adults. The physiological mechanisms underlying this effect remain to be fully understood. A previous study showed that even a single bout of exercise is able to induce an acute (i.e., 12–24 h) and sustained (i.e., still present after 2 days) improvement in both endothelium and flow-mediated vasodilatation (34). In agreement with the above explanation, we observed a reduction of the peak value of ΔHHb/ΔV̇o2 and throughout the exercise transition from ∼20 to 120 s, suggesting a better matching between local O2 delivery and consumption, at least in the portion of muscle under the NIRS investigation.
Recent studies have shown that in older adults the energy cost of high-intensity muscle contractions is smaller than in young adults, possibly due to an increased contribution of more efficient, type I fibers to the force development with aging (in turn, caused by a selective atrophy of type II muscle fibers). The above phenomenon may be responsible for the superior fatigue resistance that has been observed in older compared with young adults, despite a reduced aerobic fitness (66).
In ModA, our subjects showed a net mechanical efficiency not different from that reported in the literature for cycling exercise at workloads below 100 W, i.e., ∼0.14–0.15 (24). Therefore, during voluntary dynamic exercise in the moderate-intensity domain, our data do not support a superior efficiency of muscle contraction in older compared with young adults. This observation may be ascribed to the fact that, during dynamic exercise of moderate intensity, mainly type I muscle fibers are recruited in young and older adults alike.
In young adults, an increased efficiency has been shown in heavy-intensity exercise following PE, due to a reduction of the contribution of the slow component of V̇o2 kinetics (42). On the contrary, in our study, which evaluated exercise in the moderate-intensity domain, net mechanical efficiency was unaffected by PE. This is not surprising since the subjects performed a moderate-intensity exercise in which 1) a slow component of V̇o2 is absent, 2) highly efficient type I motor units are likely to be recruited irrespective of age (since a low percentage of the maximal mechanical power is required), and 3) the recruitment pattern should not be altered by a previous bout of high-intensity exercise.
In conclusion, on the basis of our data, we cannot rule out a possible contribution of an improved efficiency of muscle contraction to the increased fatigue resistance that has been observed in older adults in heavy-intensity exercises. Yet, our evidence does not support either a superior efficiency of muscle contraction in older compared with young adults in the moderate-intensity domain or changes in muscle efficiency following PE.
Perspectives and Significance
Our study has provided evidence that, in healthy older adults, prior heavy-intensity exercise beneficially affects the speed of adjustment of oxidative metabolism, during moderate-intensity exercise transitions, by removing the relative insufficiency of the local O2 delivery and by improving the matching between O2 delivery and utilization. A faster provision of ATP from the aerobic metabolism reduces the contribution of anaerobic ATP sources, with a lesser disturbance of cell homeostasis and reduced muscle fatigue during the repeated metabolic transitions of moderate-intensity exercise that characterize everyday life. Therefore, our study underlines that future interventions aimed at increasing exercise tolerance and, therefore, functional independence in older adults should focus on microvascular blood flow regulation as opposed to bulk O2 delivery. Furthermore, protocols for exercise prescription in older adults should include heavy-intensity priming exercise to allow the muscles to fully exploit their oxidative capacity and to increase exercise tolerance, in turn, increasing the training load that can be tolerated in a training session, and therefore, the benefit derived by exercise prescription.
The study was financially supported by the Grant PRIN 2007, allocated to Dr. Carlo Capelli by the Italian Ministry for Research and University.
No conflicts of interest, financial or otherwise, are declared by the author(s).
Author contributions: G.D.R., A.A., and C.P. performed experiments; G.D.R., S.P., and C.P. analyzed data; G.D.R. drafted manuscript; S.P. and C.C. conception and design of research; S.P. and C.C. interpreted results of experiments; S.P. prepared figures; S.P. and C.C. edited and revised manuscript; S.P. and C.C. approved final version of manuscript.
We heartily thank the subjects who volunteered for the study.
- Copyright © 2012 the American Physiological Society