Oxygen delivery and the restoration of the muscle energetic balance following exercise: implications for delayed muscle recovery in patients with COPD

Abstract

Patients with chronic obstructive pulmonary disease (COPD) experience a delayed recovery from skeletal muscle fatigue following exhaustive exercise that likely contributes to their progressive loss of mobility. As this phenomenon is not well understood, this study sought to examine postexercise peripheral oxygen (O₂) transport and muscle metabolism dynamics in patients with COPD, two important determinants of muscle recovery. Twenty-four subjects, 12 nonhypoxemic patients with COPD and 12 healthy subjects with a sedentary lifestyle, performed dynamic plantar flexion exercise at 40% of the maximal work rate (WR_max) with phosphorus magnetic resonance spectroscopy (³¹P-MRS), near-infrared spectroscopy (NIRS), and vascular Doppler ultrasound assessments. The mean response time of limb blood flow at the offset of exercise was significantly prolonged in patients with COPD (controls: 56 ± 27 s; COPD: 120 ± 87 s; P < 0.05). In contrast, the postexercise time constant for capillary blood flow was not significantly different between groups (controls: 49 ± 23 s; COPD: 51 ± 21 s; P > 0.05). The initial postexercise convective O₂ delivery (controls: 0.15 ± 0.06 l/min; COPD: 0.15 ± 0.06 l/min) and the corresponding oxidative adenosine triphosphate (ATP) demand (controls: 14 ± 6 mM/min; COPD: 14 ± 6 mM/min) in the calf were not significantly different between controls and patients with COPD (P > 0.05). The phosphocreatine resynthesis time constant (controls: 46 ± 20 s; COPD: 49 ± 21 s), peak mitochondrial phosphorylation rate, and initial proton efflux were also not significantly different between groups (P > 0.05). Therefore, despite perturbed peripheral hemodynamics, intracellular O₂ availability, proton efflux, and aerobic metabolism recovery in the skeletal muscle of nonhypoxemic patients with COPD are preserved following plantar flexion exercise and thus are unlikely to contribute to the delayed recovery from exercise in this population.

exercise intolerance is a frequent complaint, and an important predictor of mortality (55), in patients with COPD. This attenuated exercise capacity, perhaps initiated by a chronic obstructive pulmonary disease (COPD)-driven downregulation of oxidative capacity in skeletal muscle (49), can be so debilitating that patients progressively become unable to perform activities of daily living. Interestingly, this downward spiral appears to be aggravated by a prolonged recovery of functional capacity following exercise (48). For instance, after performance of a knee-extension exercise to the point of exhaustion, both postexercise maximum voluntary contraction and electrically evoked quadriceps twitch force were significantly depressed in patients with COPD compared with healthy controls and remained so even following ~1 h of rest (41). As daily life is characterized by repetitive activities, this delayed recovery is likely an important contributor to the poor exercise tolerance, the dimished physical activity and ultimately the loss of mobility experienced by patients with COPD. Therefore, elucidating the prevailing physiological determinants of impaired muscle recovery from exercise in this population is an important and clinically relevant endeavor.

Immediately after exercise, the rate of skeletal muscle recovery is influenced by mechanisms intrinsic to contractile function (e.g., Ca²⁺ handling), the restoration of energetic balance, and O₂ delivery to muscle. With regard to contractile function, the effects of exhaustive exercise on Ca²⁺ handling have yet to be investigated as, currently, the effects of COPD on the contractile apparatus during exercise are still unclear in humans (9, 37). In contrast, a prolonged recovery of the muscle energy stores (i.e., phosphocreatine concentration [PCr]) and pH after exercise has already been documented in the lower limb of hypoxemic (75) and nonhypoxemic patients with COPD (51, 57, 62, 65), likely as a consequence of the detrimental effects of both emphysema per se (49) and muscle disuse (37, 51) on muscle oxidative capacity. Finally, studies focused on the adequacy of postexercise O₂ supply to the skeletal muscle in patients with COPD are relatively scarce and conflicting. For instance, the pulmonary O₂ consumption (V̇o₂) time constant at the offset of a cycling exercise has been correlated with microvascular reoxygenation kinetics measured by near-infrared spectroscopy (NIRS) in the quadriceps of patients with COPD (56), suggestive of O₂ supply-limited muscle oxidative metabolism. This interpretation has, however, been challenged by the recent evidence that pulmonary and muscle V̇o₂ may be dissociated during recovery from exercise (34), making inferences specific to muscle oxidative metabolism from pulmonary V̇o₂ postexercise rather tenuous. More recently, slower microvascular deoxygenation recovery kinetics, measured by NIRS, have been reported following neuromuscular electrical stimulation in patients with COPD compared with sedentary controls (2). This finding was interpreted as evidence that impeded recovery of muscle functional capacity is caused by metabolic abnormalities. However, as the NIRS-derived deoxyhemoglobin (HHb) signal reflects the balance between O₂ utilization and supply, in the absence of any direct measurements of limb blood flow or oxidative metabolism, an inadequate hemodynamic response should not be ruled out and could account for the slower kinetics in patients with COPD.

Therefore, it is still unclear whether the interaction between postexercise muscle metabolism and peripheral hemodynamics is altered in patients with COPD compared with healthy sedentary controls. Accordingly, this study sought to examine the dynamics of peripheral O₂ transport and metabolism following the cessation of plantar flexion exercise. We hypothesized that if contractile dysfunction is the predominant mechanism delaying muscle recovery in patients with COPD, then 1) postexercise limb and capillary hemodynamics would not be different between controls and patients; 2) as a result, both the convective and diffusive components of O₂ transport would adequately match the metabolic demand both in controls and patients; and 3) postexercise PCr and pH kinetics would be similar between controls and patients with COPD.

METHODS

Subjects.

After written informed consent was obtained, 12 nonhypoxemic patients with COPD and 12 age-matched sedentary subjects participated in this study. The patients were recruited based on spirometric evidence of COPD, while the controls subjects were recruited based on no evidence of regular physical activity above that required for activities of daily living (assessed by both questionnaire and accelerometry). Exclusion criteria for the study included overt cardiovascular disease, diabetes, obesity, neuromuscular disease, and known cancer. All subjects performed standard pulmonary function tests during an initial visit to the laboratory. The study was approved by the Human Research Protection Programs of both the University of Utah and the Salt Lake City Veterans Affairs Medical Center.

Exercise protocol.

After subjects were familiarized with the equipment, the individual maximal work rate (WR_max) was determined by having them perform incremental dynamic plantar flexion exercise until exhaustion. On two separate occasions designed to be balanced, subjects performed constant-load submaximal plantar flexion at ~40% of WR_max (frequency of 1 Hz). The exercise protocol was performed on two separate days, once in the whole body MRI system (TimTrio, 2.9T; Siemens Medical Systems, Erlangen, Germany) to assess metabolism and then repeated on a separate day in the laboratory to assess limb and capillary blood flow as well as microvascular oxygenation using Doppler ultrasound imaging and NIRS, respectively. Specifically, the exercise protocol entailed 1 min of rest and 4 min of the constant-load submaximal plantar flexion, followed by 5 min of recovery. Before initiation of the hemodynamics protocol, blood samples were collected to assess blood lipids, fasting glucose, and metabolism and to perform a complete blood cell count. All experimental trials were performed by the participants in a thermoneutral environment following an overnight fast, after they had refrained from any physical activity and smoking for 12 h.

Popliteal blood flow.

Measurements of popliteal artery blood velocity and vessel diameter were performed in the popliteal fossa of the exercising leg proximal to the branching of the medial inferior genicular artery with a Logic 7 Doppler ultrasound system (General Electric Medical Systems, Milwaukee, WI). The ultrasound system was equipped with a linear array transducer operating at an imaging frequency of 9 MHz. Vessel diameter was determined at a perpendicular angle along the central axis of the scanned area. Blood velocity was measured using the same transducer with a frequency of 5 MHz. All blood velocity measurements were obtained with the probes appropriately positioned to maintain an insonation angle of 60° or less. The sample volume was maximized according to vessel size and centered within the vessel. Arterial diameter was measured offline every 12 s using automated edge-detection software (Medical Imaging Applications, Coralville, IA), and mean velocity (V_mean) (angle corrected and intensity-weighted area under the curve) was automatically calculated beat by beat (Logic 7). With the use of arterial diameter and V_mean, blood flow in the popliteal artery was calculated as blood flow = V_mean × π (vessel diameter/2)² × 60, where blood flow is in milliliters per minute. Arterial O₂ saturation ($\text{S}{\text{a}}_{{\text{O}}_2}$) was monitored at baseline with a finger probe oximeter (OxiMax N-600x; Nellcor, Pleasanton, CA). Mean arterial pressure, heart rate, stroke volume, and cardiac output were determined with a Finometer (Finapres Medical Systems, Amsterdam, The Netherlands). Leg vascular conductance was calculated as popliteal artery blood flow divided by mean arterial pressure. Arterial O₂ content ($\text{C}{\text{a}}_{{\text{O}}_2}$) was calculated, utilizing baseline $\text{S}{\text{a}}_{{\text{O}}_2}$ and hemoglobin (Hb), as the sum of bound O₂ (1.34 × Hb × $\text{S}{\text{a}}_{{\text{O}}_2}$) and dissolved O₂ (0.003 × Po₂), based on a normal Hb association curve (64). O₂ delivery was then calculated as the product of $\text{C}{\text{a}}_{{\text{O}}_2}$ and popliteal artery blood flow.

Microvascular oxygenation and capillary blood flow.

Microvascular oxygenation was assessed using the NIRS technique, which provides continuous, noninvasive measurements of oxygenated (HbO₂), deoxygenated (HHb), and total (Hb_tot) Hb levels as well as a “tissue” oxygenation index (TOI, i.e., HbO₂/Hb_tot). Because of identical spectral characteristics, Hb and myoglobin (Mb) are not separated using NIRS. However, although still somewhat contentious (14), the signal is usually considered to be derived mainly from Hb (30). Changes in microvascular oxygenation of the calf muscle were continuously monitored at 2 Hz using a near infrared frequency resolved spectroscopy oximeter (Oxiplex TS; ISS, Champaign, IL). The probe was positioned at the level of the largest circumference of the calf and secured with velcro straps and biadhesive tape. This NIRS device uses intensity-modulated light, and the probe consisted of eight infrared light sources (4 emitting at 690 nm and 4 emitting at 830 nm) and one detection channel (interoptode distance = 1.5 to 4.5 cm) including a selected light detector (photomultiplier tube), thus providing a measurement of absorption and the scattering coefficient of the tissues. Measurement of adipose tissue thickness under the NIRS sample site was performed with a Logic 7 Doppler ultrasound system (General Electric Medical Systems). Microvascular Po₂ was derived from the tissue oxygen index (4), assuming that the near infrared spectroscopy signal mainly originates from Hb (30) and then computed from the HbO₂ dissociation curve (64).

The estimated capillary blood flow response following the offset of exercise was calculated from a modified version of the method proposed by Ferreira et al. (12, 18) using the kinetics of muscle O₂ consumption and the HHb data, as previously described (38). Specifically, the PCr resynthesis rate, measured by ³¹P-MRS, which is derived almost exclusively from oxidative phosphorylation (58), was used as an index of muscle O₂ consumption. Then, as the HHb response determined by NIRS is considered to reflect muscle capillary O₂ extraction (i.e., $\text{C}{\text{a}}_{{\text{O}}_2}$−$\text{C}{\text{v}}_{{\text{O}}_2}$) (30), and based upon the Fick equation, the temporal characteristics of capillary blood flow were estimated using the PCr resynthesis rate-to-HHb ratio.

³¹P-MRS.

³¹P-MRS was performed using a clinical 2.9T MRI system (Tim-Trio; Siemens Medical Solutions) operating at 49.9 MHz for ³¹P-resonance. ³¹P-MRS data were acquired with a dual-tuned ³¹P-proton (¹H) surface coil with linear polarization (Rapid Biomedical, Rimpar, Germany) positioned under the calf at its maximum diameter. The ³¹P-single-loop coil diameter was 125 mm surrounding a 110-mm ¹H-coil loop. The centering of the coil around the leg was confirmed by T₁-weighted ¹H-localizing images, and the coil was repositioned if the coil was not actually centered on the calf, as determined by the thickness of the gastrocnemius muscle. After a three-plane scout proton image, advanced localized volume shimming was performed. Before each experiment, two fully relaxed spectra were acquired at rest with three averages per spectrum and a repetition time of 30 s. Then, MRS data acquisition was performed throughout the rest-exercise-recovery protocol using a free-induction-decay pulse sequence with a 2.56-ms adiabatic-half-passage excitation radiofrequency pulse and the following parameters: repetition time (TR) = 2 s; receiver bandwidth = 5 kHz; 1,024 data points; and 3 averages per spectrum. Saturation factors were quantified by the comparison between fully relaxed (TR = 30 s) and partially relaxed spectra (TR = 2 s).

As previously described (36), relative concentrations of PCr ([PCr]), inorganic phosphate ([P_i]), and ATP ([ATP]) were obtained by a time-domain fitting routine using the Advanced Method for Accurate, Robust, and Efficient Spectral (AMARES) fitting algorithm (71) incorporated into the CSIAPO software (40). Intracellular pH was calculated from the chemical shift difference between the P_i and PCr signals. The free cytosolic ADP concentration ([ADP]) was calculated from [PCr] and pH using the creatine kinase equilibrium constant (K_CK = 1.66 × 10⁹ M) and assuming that phosphocreatine represents 85% of total creatine content (23). The resting concentrations were calculated from the average peak areas of the three relaxed spectra (TR = 30 s; n = 3) recorded at rest, assuming an 8.2 mM [ATP]. When P_i splitting was evident, the pH corresponding to each P_i pool was calculated separately as pH₁ and pH₂ on the basis of the chemical shift of each peak relative to PCr. The overall muscle pH was then calculated as pH = pH₁ (area P_i1/total P_i area) + pH₂ (area P_i2/total P_i area).

Leg volume.

Leg volume was calculated based on lower leg circumference (3 sites: distal, middle, and proximal), leg length, and skinfold measurements (25). This method has recently been confirmed to provide a valid estimate for muscle volume across a spectrum of individuals with normal muscle mass to severe muscle atrophy (39).

Assessment of physical activity.

Physical activity level (PAL) was assessed using both a subjective PAL recall questionnaire and objective accelerometer data. The PAL questionnaire included items regarding the average type, frequency, intensity, and duration of physical activity in any given week. After receiving standardized operating instructions, subjects wore an accelerometer (GT1M; Actigraph, Pensacola, FL) quantifying both the number of steps per day and intensity of movement for seven continuous days, with adherence automatically recorded by the device. According to the manufacturer specification, thresholds for sedentary, light, lifestyle, moderate, vigorous, and very vigorous activity were defined as <99, 100–759, 760–1,951, 1,952–5,724, and >5,725 counts/min, respectively.

Data analysis.

The limb and capillary blood flow, vascular conductance, PCr, HHb, and TOI recovery kinetics were determined by fitting the time-dependent changes during the recovery period to a monoexponential curve described by the following equation:

$$Y\left(t\right)=Y_{end}+Y_{amp}\left(1-e^{-(t-TD/\tau )}\right)$$ (1)

where Y_end is the level of variable measured at end-of-exercise and Y_res refers to the amplitude of the blood flow response, PCr resynthesized, or the resaturation during the recovery. Unlike the other variables, there is no time delay (TD) in the resynthesis of PCr, and therefore, TD was fixed to 0 for PCr kinetics. Then, the initial rate of PCr resynthesis from ³¹P-MRS (Vi_PCr) was calculated from the derivative of Eq. 1 at time 0:

$$V{i}_{PCr}=k·\Delta \left[PCr\right]$$ (2)

in which Δ[PCr] represents the amount of PCr resynthesized during the recovery and the rate constant k = 1/τ (27).

The peak rate of oxidative ATP synthesis from ³¹P-MRS (V_max in mM/min) was calculated using the initial rate of PCr synthesis (Vi_PCr) during the recovery period and [ADP] obtained at the end of exercise as previously described (68):

$$V_{max}=V{i}_{PCr}\left(1+\left(K_m/{\left[\text{ADP}\right]}_{end}{}^{2.2}\right)\right)$$ (3)

in which K_m (the [ADP] at half the highest oxidation rate) is 30 µM in skeletal muscle (27).

During the recovery period, PCr is regenerated throughout the CK reaction as the consequence of oxidative ATP production in mitochondria. Thus H⁺_efflux can be calculated from the rates of proton production from the CK reaction (H⁺_CK, in mM/min) and mitochondrial ATP production (H⁺_Ox, in mM/min) on one side and the rate of pH changes on the other side. At this time, ATP production is exclusively aerobic, and lactate production is considered negligible:

$$V_{eff}={\beta }_{total}·dpH/dt+\gamma ·ViPCr+m·{ATP}_{ox}$$ (4)

To improve precision, we use a modified version of this calculation (29) in which the total proton disappearance (i.e., ∫Edt) is estimated cumulatively from the start of recovery and then fitted to an exponential function to obtain the initial efflux recovery rate.

Model variables were determined with an iterative process by minimizing the sum of squared residuals between the fitted function and the observed values. Goodness of fit was assessed by visual inspection of the residual plot and the frequency plot distribution of the residuals, χ²-values, and the coefficient of determination (r²), which was calculated as follows:

$$r^2=1-\left({SS}_{reg}/{SS}_{tot}\right)$$ (5)

where SS_reg is the sum of squares of the residuals from the fit and SS_tot is the sum of squares of the residuals from the mean.

Statistical analysis.

The assessment of differences between COPD and controls was performed with either paired t-tests or nonparametric Wilcoxon tests, where appropriate (Statsoft, version 5.5; Statistica, Tulsa, OK). Statistical significance was accepted at P < 0.05. Results are presented as means ± SD in tables and means ± SE in the figures for clarity.

RESULTS

Subject characteristics.

Subject characteristics are presented in Table 1. Patients with COPD exhibited reduced pulmonary function relative to the healthy sedentary controls and blood gas characteristics consistent with COPD. Despite considerable effort in terms of seeking out sedentary controls, there was still a significant difference in physical activity between groups (Table 1). However, based on the number of steps per day and the time spent in the different intensity domains measured by accelerometry, all subjects can still be confidently defined as sedentary to low active (69). This similar sedentary lifestyle between groups was further confirmed by the lack of a significant difference in plantar flexion peak power output and limb muscle volume between groups. Adipose tissue thickness was not significantly different between groups (controls: 0.88 ± 0.21 cm; COPD: 0.73 ± 0.25 cm, P > 0.05). One patient and one control subject were current smokers. Two patients were studied while using supplemental O₂ (resting $\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_{\mathrm{2}}}$: 94.5 ± 2.1%).

Table 1.

Subject characteristics

	Controls	COPD
Sample size	12	12
Age, yr	68 ± 6	66 ± 6
Anthropometric characteristics
Height, cm	173 ± 9	171 ± 9
Weight, kg	76 ± 13	77 ± 16
BMI	25 ± 3	26 ± 5
Limb muscle volume, dl	21 ± 5	23 ± 4
Functional characteristics
Steps per day	6,079 ± 1,775	3,021 ± 747*
Sedentary physical activity, min	1,244 ± 61	1,303 ± 43*
Light physical activity, min	104 ± 37	84 ± 43
Lifestyle physical activity, min	56 ± 25	31 ± 28*
Moderate physical activity, min	24 ± 15	9 ± 14*
Vigorous and very vigorous, min	0 ± 0	0 ± 0
Gait speed, m/s	1.5 ± 0.2	1.2 ± 0.2*
Plantar flexion maximal work rate, W	9 ± 3	7 ± 2
Pulmonary function
FVC, liter	4.8 ± 1.4	3.2 ± 0.7*
FVC, %pred	117 ± 22	80 ± 16*
FEV1, liter	3.4 ± 0.8	1.6 ± 0.5*
FEV1, %pred	114 ± 18	53 ± 16*
FEV1/FVC%	73 ± 8	51 ± 9*
Blood characteristics
Arterial blood saturation, %	94 ± 1	94 ± 2
Glucose, mg/dl	84 ± 13	91 ± 8
Cholesterol, mg/dl	202 ± 39	198 ± 28
Triglycerides, mg/dl	151 ± 81	87 ± 25
HDL, mg/dl	51 ± 13	71 ± 34
LDL, mg/dl	130 ± 32	108 ± 25
WBCs, K/μl	5.3 ± 1.4	7.4 ± 3.0
RBCs, M/μl	4.9 ± 0.5	4.7 ± 0.3
Hemoglobin, g/dl	15 ± 2	14 ± 1
Hematocrit, %	45 ± 4	44 ± 3
Neutrophil, K/μl	2.9 ± 1.1	4.8 ± 3.4
Lymphocyte, K/μl	1.7 ± 0.5	1.8 ± 0.8
Monocyte, K/μl	0.5 ± 0.2	0.6 ± 0.2
Bicarbonate, mM/l	25 ± 1	28 ± 3*
Potassium, mM/l	4.0 ± 0.3	3.8 ± 0.3

Data are expressed as means ± SD. COPD, chronic obstructive pulmonary disease; BMI, body mass index; FVC, forced vital capacity; FEV₁, forced expiratory volume in 1 s; HDL, high-density lipoprotein; LDL, low-density lipoprotein; WBCs, white blood cells; RBCs, red blood cells.

^*P < 0.05, significantly different from controls.

Baseline and plantar flexion exercise.

Intracellular metabolite concentrations and pH, in addition to tissue and microvascular oxygenation indexes (Hb_tot, HHb, and TOI) at rest and during the last 30 s of exercise in both controls and patients with COPD are summarized in Table 2. Apart from the resting concentrations of P_i and phosphodiester (PDE), which were lower in COPD (P < 0.05), pH, phosphorylated compounds ([PCr], [P_i], and [ADP]), as well as microvascular oxygenation were not significantly different between groups (P > 0.05) at baseline or at the end of exercise.

Table 2.

Metabolic and microvascular oxygenation responses at rest and during steady-state submaximal plantar flexion exercise in control and COPD subjects

	Controls	COPD
Baseline
Phosphorylated compounds and pH
PCr, mM	34 ± 5	36 ± 8
Pi, mM	1.9 ± 1.0	1.1 ± 0.4*
ADP, μM	8.2 ± 0.3	8.0 ± 0.7
pH	6.97 ± 0.02	6.96 ± 0.04
PDE, mM	2.0 ± 1.4	0.6 ± 0.6*
Microvascular oxygenation
Oxygenation index, %	62 ± 4	60 ± 7
Hb total, µM	55 ± 20	56 ± 27
HHb, µM	21 ± 8	23 ± 13
End exercise
Phosphorylated compounds and pH
PCr, mM	22 ± 6	21 ± 6
Pi, mM	10 ± 4	10 ± 4
ADP, μM	43 ± 21	47 ± 47
pH	6.99 ± 0.06	6.93 ± 0.09
Microvascular oxygenation
Oxygenation index, %	60 ± 4	55 ± 9
Hb total, µM	58 ± 22	58 ± 29
HHb, µM	24 ± 10	28 ± 20

Values are expressed as means ± SD. PCr, phosphocreatine; P_i, inorganic phosphate; PME, phosphomonoester; ADP, adenosine diphosphate; PDE, phosphodiester; Hb, total hemoglobin; HHb, deoxyhemoglobin.

^*P < 0.05, significantly different from controls.

Recovery period: peripheral hemodynamics.

Limb blood flow, vascular conductance, and capillary blood flow dynamics during the recovery period in controls and COPD are displayed in Fig. 1. While end-exercise limb blood flow was not significantly different between groups (controls: 741 ± 216 ml/min; COPD: 704 ± 253 ml/min; P > 0.05), the mean response time at the offset of exercise was significantly prolonged in patients with COPD (controls: ~56 s; COPD: ~120 s; Fig. 1; P < 0.05). Similarly, end-exercise vascular conductance was not significantly different between groups (controls: 7.5 ± 1.4 ml·min⁻¹·mmHg⁻¹; COPD: 6.0 ± 2.5 ml·min⁻¹·mmHg⁻¹; P > 0.05). However, the mean response time at the offset of exercise was significantly prolonged in patients with COPD (controls: ~61 s; COPD: ~127 s; Fig. 1; P < 0.05). In contrast, the time constant for capillary blood flow at the offset of exercise was not significantly different between groups (Fig. 1; P > 0.05).

Fig. 1.

The recovery kinetics of limb blood flow (A), vascular conductance (B) and capillary blood flow (C) following dynamic plantar flexion exercise in controls and patients with chronic obstructive pulmonary disease (COPD). Insets: illustrations of the mean response time. Both for limb blood flow and vascular conductance, the mean response time was significantly slower in patients with COPD compared with controls (P < 0.05). Values are presented as means ± SE; a.u., arbitrary units.

Convective O₂ delivery and O₂ diffusional conductance.

The convective O₂ delivery dynamics during the recovery period in controls and patients with COPD is displayed in Fig. 2. The initial postexercise convective O₂ delivery and the corresponding oxidative ATP demand were not significantly different between controls and patients with COPD (P > 0.05, Fig. 2). The relationship between microvascular Po₂ and initial postexercise PCr resynthesis rate, an index of O₂ utilization, during the recovery from plantar flexion exercise in controls and patients with COPD is documented in Fig. 3. The slope of each line from the origin, which reflects O₂ diffusional conductance, was not significantly different between groups (controls: 0.47 ± 0.25 mM·mmHg⁻¹·min⁻¹; COPD: 0.55 ± 0.26 mM·mmHg⁻¹·min⁻¹; P > 0.05).

Fig. 2.

Recovery kinetics of convective O₂ delivery following dynamic plantar flexion exercise in controls and patients with COPD. Inset: illustration of the immediate postexercise convective O₂ delivery and the corresponding oxidative ATP demand in both groups. Neither convective O₂ delivery nor oxidative ATP demand were significantly different between controls and patients with COPD (P > 0.05) indicative of a similar matching of O₂ supply and demand in both groups. Values are presented as means ± SE.

Fig. 3.

Relationship between microvascular partial pressure of O₂ (Po₂) and initial postexercise phosphocreatine (PCr) resynthesis rate, an index of O₂ utilization, during the recovery from plantar flexion exercise in controls and patients with COPD. The slope of the lines from the origin reflects O₂ diffusional conductance according to Fick’s law. Values are presented as means ± SE.

Microvascular oxygenation offset kinetics.

Both the postexercise TOI time constant (controls: 48 ± 74 s; COPD: 53 ± 56 s; P > 0.05) and mean response time (controls: 63 ± 78 s; COPD: 62 ± 59 s; P > 0.05) were not significantly different between groups. Similarly, the HHb recovery time constant (controls: 39 ± 22 s; COPD: 57 ± 46 s; P > 0.05) and mean response time (controls: 75 ± 63 s; COPD: 76 ± 78 s; P > 0.05) were not significantly different between groups.

Metabolic offset kinetics.

Changes in pH and [PCr] dynamics during the recovery period in controls and patients with COPD are displayed in Fig. 4. Table 3 documents mitochondrial function and proton handling assessed via postexercise metabolic kinetics in both groups. None of these variables were different between controls and patients with COPD.

Fig. 4.

Recovery kinetics of phosphocreatine (A) and pH (B) following dynamic plantar flexion exercise in controls and patients with COPD.

Table 3.

Mitochondrial function and proton handling assessed via postexercise metabolic kinetics in controls and patients with COPD

	Controls	COPD
Mitochondrial function
PCr recovery time constant, s	46 ± 20	49 ± 21
IC 95, s	16 ± 12	15 ± 9
Peak mitochondrial phosphorylation rate, mM/min	23 ± 10	24 ± 10
Proton (H+) handling
Initial H+ efflux, mM/min	4.1 ± 3.0	4.0 ± 3.8

Values are expressed as means ± SD. IC 95, 95% confidence interval.

DISCUSSION

Patients with COPD experience a delayed recovery from skeletal muscle fatigue in the first minutes following exhaustive exercise that likely contributes to their progressive loss of mobility. As this delayed postexercise recovery is not well understood, this study sought to examine the interaction between postexercise muscle metabolism and peripheral hemodynamics following plantar flexion exercise in patients with COPD and healthy sedentary controls. The main findings of this study were that 1) while end-exercise limb blood flow was not significantly different between groups, postexercise vascular conductance and limb blood flow kinetics, but not capillary dynamics, were slower in patients with COPD compared with controls; 2) despite these altered hemodynamics, convective O₂ delivery and O₂ diffusional conductance appeared to appropriately match muscle metabolic demand both in controls and patients with COPD; and 3) the metabolic recovery and mitochondrial capacity in patients with COPD were not significantly different to the controls. Therefore, in the face of perturbed peripheral hemodynamics, both intracellular O₂ availability and metabolic recovery in the skeletal muscle of nonhypoxemic patients with COPD are actually preserved following exercise and thus are unlikely to contribute to the delayed functional recovery from exercise exhibited by this population.

Prolonged postexercise peripheral hemodynamics in patients with COPD.

For a given metabolic demand at the end of the exercise (Fig. 2), blood flow and vascular conductance were not significantly different between sedentary controls and patients with COPD. In contrast, the kinetics of postexercise limb blood flow and vascular conductance were significantly slower in patients with COPD compared with controls (Fig. 1). Similarly, two studies (19, 61) reported no difference in exercise-induced blood flow and vascular conductance during steady-state exercise in patients with COPD compared with healthy age-matched controls. Our results not only extend these findings to the calf muscle but also reveal that, despite no significant difference in the steady-state blood flow response, postexercise hemodynamics are actually prolonged in patients with moderate to very severe COPD. This novel finding confirms the importance of studying not only steady-state response but also postexercise hemodynamics as it provides a sensitive and unique window to examine the regulation of blood flow.

Interestingly, the prolonged recovery dynamics at the macrocirculatory level did not translate into slower capillary hemodynamics within the working muscle, as the mean response time of capillary blood flow in patients with COPD was not significantly different from that of the controls (Fig. 2). While initially somewhat puzzling, this dissociation between macro- and microcirculatory dynamics likely indicates inefficient blood flow redistribution following exercise. Indeed, neither the PCr resynthesis nor pH kinetics were significantly different between groups (Fig. 4), thereby indicating that prolonged limb blood flow recovery did not stem from an elevated ATP demand or metabolic acidosis within the working muscle. Instead, these findings suggest a significant portion of postexercise blood flow was directed toward other areas of the lower limb, with low metabolic activity, in patients with COPD.

Another possible explanation for the different hemodynamics in the conduit artery and capillaries of the working muscle of patients with COPD might stem from substantial heterogeneity in O₂ delivery to O₂ utilization matching in the exercising muscles (31–33). Specifically, animal studies have revealed that fast-twitch glycolytic fibers rely, to a greater extent, on adjustments in fractional O₂ extraction, rather than O₂ delivery, to cope with the changes in skeletal muscle O₂ demand (33, 50). According to these findings and given the well-established shift in muscle fiber type toward type II glycolytic fibers with COPD (15), it is possible that the region investigated by the NIRS (medial gastrocnemius muscle) might have relied more on alterations in O₂ extraction rather than blood flow to match muscle O₂ demand compared with other muscles of the lower limb composed of a greater proportion of slow-twitch oxidative fibers.

The potentially abnormal regulation of lower limb hemodynamics in patients with COPD is consistent with disease-related vascular dysfunction in this population. For instance, COPD has been associated with augmented sympathetic activity (20), impaired endothelium-dependent and -independent dilation (8, 10, 22, 53), and alterations in the concentration of circulating vasoactive substances such as nitric oxide (22, 52). Individually or in combination such phenomena have the potential to impair vascular control and peripheral hemodynamics following exercise. Also relevant to the present findings is the previously documented evidence of an attenuated contribution of the muscle metaboreflex to hemodynamic control in the calf muscles of patients with COPD (60). Specifically, the sympathetically mediated vasoconstriction of the calf vasculature following stimulation of the metabosensitive afferents was attributed to the metabolic products of handgrip exercise (60). Such maladaptations can lead to the inefficient coupling between blood flow and metabolic demand, especially at times when metabolic demand varies rapidly, such as the onset and offset of exercise. Combined with the current results, these findings suggest a possible overperfusion of muscle tissue with low metabolic demand after exercise in the lower limb. Although such alterations in blood flow distribution may be of little consequence during a small muscle exercise, this has the potential to result in greater perfusion-metabolism mismatch and/or blood pressure dysregulation at the onset and offset of whole body high-intensity exercise, with the potential to compromise exercise capacity and recovery.

Matching of peripheral O₂ supply and utilization in the leg of patients with COPD.

An important finding of this study was that convective O₂ delivery to the plantar flexor muscles appeared to be preserved at the offset of exercise in patients with COPD and appeared to match muscle metabolic demand (Fig. 2). In addition, as illustrated in Fig. 3, skeletal muscle O₂ diffusional conductance was also not significantly different between sedentary controls and patients with COPD. As a result of this preserved O₂ transport system, both the HHb recovery time constant and mean response time were not significantly different between groups, implying that muscle O₂ extraction in the plantar flexor muscles of patients with COPD was also not significantly different from that of the controls. In agreement with these findings, Richardson et al. (59) documented a similar relationship between peak muscle O₂ consumption and O₂ delivery, assessed directly across the exercising muscle, in patients with COPD and controls during single leg knee-extensor exercise (59). It should, however, be noted that in the present study the postexercise HHb response was quite heterogeneous in both groups, which may allude to different determinants of muscle aerobic capacity. However, a discussion of these factors and the existence of different muscle phenotypes in patients with COPD are beyond the scope of the present study. Interestingly, the present functional findings were also supported by morphometric analysis of skeletal muscle fiber capillarization (59). Indeed, according to a series of recent studies, capillary density, capillary-to-fiber ratio, number of capillaries around a fiber, and capillary-to-fiber cross-sectional area are similar between healthy sedentary controls and patients with COPD (11, 14, 59, 73). While not unanimous (24), perhaps due to contrasting levels of physical activity between groups in this latter study, the majority of findings suggest that the size of the capillary-fiber interface, an important site of functional resistance to O₂ flux, is likely maintained in patients with COPD. Therefore, in combination, these morphometric and functional findings lend strong support to the concept that the capacity to transport O₂ in the periphery is actually well preserved in patients with COPD, especially when employing a small muscle mass exercise paradigm that does not heavily tax central hemodynamics or the pulmonary system (59). This conclusion should, however, be put in perspective with the results from prior studies employing high-intensity cycling exercise, which have consistently documented that O₂ transport to limb locomotor muscles was compromised in patients with COPD due to impaired central and peripheral hemodynamics (3, 6, 7, 42, 43, 54), likely caused indirectly by abnormal respiratory mechanics and gas exchange (5, 13, 66, 72).

Metabolic recovery and mitochondrial function in the skeletal muscle of patients with COPD.

Confirming the results from previous work by our group (37), but with a larger sample size, this study has documented that the PCr recovery time constant and the peak rate of mitochondrial phosphorylation of the plantar flexor muscles were not significantly different between sedentary controls and patients with COPD (Table 3). Given the strong dependence of this measurement on the level of muscle activity, these results confirm that despite significant group differences in overall physical activity, measured by accelerometry (due to the extreme inactivity of COPD patients), both groups actually exhibited characteristics of a sedentary lifestyle and thus the same level of muscle disuse. Such metabolic findings add to the growing evidence suggesting that skeletal muscle mitochondrial capacity assessed in vivo is actually preserved with COPD when the level of physical activity is not significantly different between patients and controls (37, 65) and that therefore much of the decline in mitochondrial capacity reported in these patients can likely be attributed to muscle disuse. For instance, 8 wk of supervised endurance and strength training in patients with COPD restored mitochondrial phosphorylation capacity in the quadriceps to that of the controls (51). Furthermore, Shields et al. (65) recently examined the PCr recovery kinetics in the biceps brachial and quadriceps femoris muscles in patients with COPD, with the premise that the upper extremities are less affected by the disease-related reduction in physical activity. This approach allowed the effects of disease and deconditioning on skeletal muscle mitochondrial phosphorylation capacity to be parsed out. Interestingly, using such approach, the authors reported a slower PCr recovery halftime in the quadriceps but not in the biceps brachial, thereby providing additional evidence that muscle disuse may be the predominant factor accounting for the impaired mitochondrial capacity in the skeletal muscle of patients with COPD (65).

It has previously been reported that the initial postexercise H⁺ efflux rate in the calf muscle was similar between hypoxemic patients with COPD and age-matched controls (67). However, given the contrasting difference in end-exercise pH in this prior study (6.93 in controls and 6.68 in COPD), it is unclear whether this result should, in fact, be interpreted as evidence of impaired H⁺ clearance. In the current study, for a given metabolic state and pH (Table 2), the initial H⁺ efflux rate at the offset of the exercise was not significantly different between controls and patients with COPD (Table 3), which translated into similar pH recovery kinetics between groups. This is an important finding as excessive accumulation of lactate and exaggerated metabolic acidosis have previously been reported in patients with COPD during both cycling (42) and plantar flexion exercise (35, 57, 75). In light of the present results, this metabolic acidosis appears not to be related to impaired H⁺ transport.

One potential explanation in the present study is that preserved limb and capillary blood flow in patients with COPD (Fig. 1) contributed to preserved H⁺ efflux. During the initial phase of recovery from exercise, H⁺ transport is also regulated by the monocarboxylate transporters 1 and 4 (MCT1 and MCT4) (17) and the sodium-H⁺ antiporter (28). Interestingly, and somewhat in contrast with our functional measurement of H⁺ efflux, a lower MCT4 expression, but not MCT1, has been documented in the vastus lateralis of patients with COPD (16). Considering the severe physical impairment of the patients recruited in the study of Green et al. (26) ($\mathrm{S}{\mathrm{a}}_{{\mathrm{O}}_{\mathrm{2}}}$: ~90%; V̇o_2peak: ~8.6 ml·min⁻¹·kg⁻¹) and the role of physical activity in modulating MCT4 content, here again, it is possible that extreme inactivity in these patients may help to explain the discrepancy between studies. Future studies examining both H⁺ efflux rate and skeletal muscle MCT content at different stages of the disease across a wide range of exercise intensities, while also controlling for the level of physical activity, will help clarify these divergent results.

Experimental considerations and limitations.

A known limitation of the NIRS technique is that the signal originates from both Hb and Mb, and the relative contribution of Mb to the overall NIRS signal has been a matter of contention (47). Although this may influence our estimation of microvascular Po₂ due to different O₂ affinity between Hb and Mb, it is unlikely that this issue actually confounded our interpretation of the data. Indeed, most experimental models have demonstrated that the contribution of Mb to the NIRS signal is actually quite minimal, such that the NIRS signal qualitatively reflects blood O₂ saturation (30, 44–46, 63, 74). In addition, it has previously been documented that the Mb content in the skeletal muscle of patients with COPD was not significantly different from controls (59). Therefore, the contribution of Mb to the NIRS signal was likely not different between groups in the present study. Along the same lines, adipose tissue thickness, which has the potential to complicate NIRS interpretation, was not significantly different between groups and, therefore, also did not likely confound our interpretation of the data.

This study did not assess the recovery of functional capacity following exercise and, therefore, could not directly investigate the link between muscle recovery and exercise tolerance. However, future studies using an all-out test during small muscle mass exercise to explore the relationships between critical power (and W’) and the determinants of muscle recovery would provide some valuable information on the factors contributing to exercise intolerance in patients with COPD.

Perspective and implications of the delayed skeletal muscle fatigue recovery following exercise.

Overall, the results of this study do not support the hypothesis that the persistent force deficit following exercise experienced by patients with COPD is due to inadequate intracellular O₂ availability or prolonged metabolic disturbance and acidosis. Possible alternative explanations for the delayed recovery in skeletal muscle fatigue may involve a reduction in Ca²⁺ sensitivity of the contractile apparatus (21), depressed Ca²⁺-ATPase activity (70), or a reduction in Ca²⁺ release that stem from an impaired coupling between the dihydropyridine receptor and the ryanodine receptor, which releases Ca²⁺ from the sarcoplasmic reticulum (1). However, whether impaired Ca²⁺ handling in the skeletal muscle of patients with COPD actually contributes to the persistent postcontractile depression will require further investigations.

Conclusions.

In summary, with the use of an integrative approach, this study revealed altered postexercise peripheral hemodynamics in the plantar flexor of nonhypoxemic patients with COPD, suggesting inefficient blood flow redistribution within the lower limb. Nevertheless, convective O₂ delivery and O₂ diffusional conductance within the exercising muscle appeared to be appropriately matched with metabolic demand and were associated with preserved aerobic metabolism recovery and proton handling in the calf muscle of these patients. Together, these findings suggest that the persistent muscle force deficit following exercise previously documented in patients with COPD is not likely a consequence of lower intracellular O₂ availability or metabolic abnormalities but instead may be linked to impaired Ca²⁺ handling within the contractile apparatus, at least, in the majority of nonhypoxemic COPD patients, as in this study.

GRANTS

This work was funded in part by a grant from the Flight Attendant Medical Research Institute (FAMRI); National Heart, Lung, and Blood Institute Grants P01-HL-091830 and K99HL125756; Veterans Affairs (VA) Merit Awards E6910-R and E1697-R; VA SPiRe Award E1433-P; and VA Senior Career Scientist Award E9275-L.

DISCLOSURES

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

AUTHOR CONTRIBUTIONS

G.L., J.D.T., Y.L.F., E.-K.J., and R.S.R. conceived and designed research; G.L., C.R.H., J.D.T., O.-S.K., M.J.R., and R.M.B. performed experiments; G.L., C.R.H., J.D.T., and Y.L.F. analyzed data; G.L. and C.R.H. interpreted results of experiments; G.L. prepared figures; G.L. drafted manuscript; G.L., C.R.H., J.D.T., O.-S.K., M.J.R., R.M.B., Y.L.F., E.-K.J., and R.S.R. edited and revised manuscript; G.L., C.R.H., J.D.T., O.-S.K., M.J.R., R.M.B., Y.L.F., E.-K.J., and R.S.R. approved final version of manuscript.