INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Impaired exercise performance is a frequently occurring problem in patients with chronic obstructive pulmonary disease (COPD) (1). The two dominant symptoms limiting exercise tolerance in COPD are dyspnea and a sensation of fatigue in leg muscles (1). Recently, it has been shown that both skeletal and respiratory muscle weakness contribute to the severity of these symptoms (2) and to reduced exercise tolerance, independent of lung function impairment (2, 3).

The cause of muscle weakness in COPD patients is incompletely understood. It has been shown that the development of muscle weakness is frequently associated with loss of body mass and in particular loss of muscle mass (4, 5). However, in this group of patients, apart from muscle wasting, also changes in muscle energy metabolism have to be considered as a factor contributing to the impairment of muscle function.

Experimental data on muscle energy metabolism in patients with stable COPD are scarce. In two recent studies, the activity of glycolytic and oxidative enzymes was investigated in resting quadriceps femoris muscle of patients with stable COPD. Jakobsson and associates (6) found indications for an augmented glycolytic and a decreased aerobic capacity. The latter phenomenon was also observed in the study performed by Maltais and colleagues (7).

The metabolic consequences of the observed changes in enzyme capacities in COPD patients are as yet unclear. Therefore, the first aim of this study was to investigate possible changes in high-energy phosphate concentrations in resting muscle of patients with stable COPD in comparison with healthy age-matched control subjects. We hypothesized that, next to measuring concentrations of adenine nucleotides, analysis of concentrations of their degradation products might be useful in detecting subtle alterations in muscular energy status. In particular, inosine monophosphate (IMP), a deamination product of adenosine monophosphate (AMP), is of interest in this context. IMP has been primarily studied in healthy volunteers, where an increased IMP formation was universally found during exercise (8). This increased IMP formation during exercise is generally believed to reflect an imbalance between adenosine triphosphate (ATP) utilization and resynthesis. Under resting conditions, muscle IMP levels are undetectable or very low in healthy subjects.

The second aim of this study was to examine, within the COPD population, the relationship between muscle energy status and lung function parameters, blood gases, and nutritional status.

Because it is probable that activity level is reduced in COPD patients compared to healthy control subjects, and since theoretically the anterior tibialis muscle is less influenced by activity level than the quadriceps femoris muscle, we chose to examine the former muscle.

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Subjects

All patients had COPD according to the criteria of the American Thoracic Society (9) and an FEV₁ expressed as a percentage of predicted (FEV₁% pred) < 50%. Only male subjects were included. Patients who demonstrated an increase in FEV₁ of > 10% of the predicted value after inhalation of a bronchodilator (terbutaline, 500 µg) were excluded. Other exclusion criteria were a history of cardiac insufficiency, distal arteriopathy, malignancy, endocrine, hepatic or renal disease, or use of anticoagulant drugs. All patients were stable at the time of the study and did not have any infection or exacerbation of their disease at least 6 wk prior to the investigation. The patients were compared with a group of healthy age-matched volunteers.

Written informed consent was obtained, and the study was approved by the Medical Ethical Board of the University Hospital Maastricht.

Muscle Biopsy Analysis

Muscle biopsies were obtained under resting conditions between 9:00 and 10:00 A.M., after an overnight fast, while the subjects were breathing room air. As described elsewhere (10), the biopsies were obtained from the anterior tibialis muscle, under local anesthesia, using a conchotome. The biopsies were immediately frozen in liquid nitrogen and stored at 80° C until analysis.

The muscle samples were freeze-dried. After freeze-drying, adherent blood and connective tissue were removed and the samples were divided into two portions. One portion was used for determination of adenine nucleotides and related compounds, phosphocreatine (PC), creatine (C), and glucose; the other part was used for determination of muscle glycogen content.

ATP, adenosine diphosphate (ADP), AMP, and IMP were determined with a high performance liquid chromatographic technique using a modified method after Wynants and van Belle (11), as described by Van der Vusse and coworkers (12). In short, the freeze-dried tissue was extracted at 15° C in a mixture of perchloric acid (3.0 M) and dithiothreitol (5 mM). After the tissue was ground in the extraction fluid with a glass rod and after subsequent centrifugation at 4° C at 1,200 × g for 5 min, the supernatant was neutralized with KHCO₃. Aliquots of the neutralized supernatant were injected on a reversed-phase column (Lichrosorb RP-18; Merck, Darmstadt, Germany). Stepwise gradient elution using two solvents was applied to separate the compounds of interest. Solvent A was an aqueous buffer of NH₄(H₂PO₄) (150 mM, pH 6.0); solvent B consisted of a 1:1 (vol/vol) mixture of acetonitrile and methanol. Flow rate amounted to 0.8 ml/ min. Peaks were detected at 254 nm and were identified by comparing retention times with known standards. LiChroCART 4-4 (Merck) was used as guard column. The detection level for IMP was 0.06 mmol/kg dry weight. Total adenine nucleotides (TAN) was computed adding ATP, ADP, and AMP contents. PC and C were measured fluorometrically (13). Total muscular creatine (CTOT) was computed adding PC and C contents. Free glucose was assayed as described elsewhere (13).

The second part of the freeze-dried tissue was used for glycogen determination. To this end, the tissue specimen was kept at 100° C for 3 h after addition of 1.0 ml of 1 M HCl in order to hydrolyze glycogen. Subsequently, the samples were neutralized with TRIS (0.12 M)- KOH (2.1 M) saturated with KCl. The glucose residues were measured fluorometrically as described elsewhere (13). The values obtained were corrected for the amount of free glucose already present at the time of tissue sampling.

Pulmonary Function Tests

FEV₁ and FVC were measured, using the pneumotachograph of a constant-volume plethysmograph (Masterlab; Jaeger, Wurzburg, Germany), until three reproducible recordings were obtained. Highest values were used for analysis. DL_CO was measured by the single-breath carbon monoxide method (Masterlab Transfer; Jaeger). Total lung capacity (TLC) and residual volume (RV) were measured by body plethysmography (Masterlab Body). All values were expressed as a percentage of reference values (14).

Blood was drawn by puncture of the radial artery while subjects were breathing room air. Pa_O2 and Pa_CO2 were measured using a blood gas analyzer (ABL 330; Radiometer, Copenhagen, Denmark).

Functional Capacity

Inspiratory muscle strength was assessed by maximal inspiratory mouth pressure (PI_max) according to the method described by Black and Hyatt (15). Exercise performance was evaluated with a 12-min walking test, performed in a level enclosed corridor according to the methods described by McGavin and colleagues (16). All tests were performed in the afternoon, and no encouragement was given. As learning effects have been noticed to occur quickly with repeated walk tests, the patients performed one practice test. The patients were not encouraged during the test.

Statistical Analysis

Statistical analysis was performed using Student's t test for unpaired measurements. In case the normality hypothesis was not fulfilled, the Mann-Whitney U test was used. Results were expressed as mean (SD). Frequency data were compared using the chi-square test. Linear regression analysis was used to study relationships between parameters. Significance was determined at the 5% level. The statistical analyses were performed using the SPSS for Windows Statistical package (17).

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Characteristics of COPD Patients and Control Subjects

Thirty-four male patients and eight male volunteers were included in the study. Anthropometric and pulmonary function data are listed in Table 1. Patients had a severe airflow obstruction (FEV₁% pred: 32 [10] %), marked air trapping (RV: 205 [74] %), moderate hyperinflation (TLC: 126 [20] %), reduced DL_CO (62 [29] %), and slightly reduced values of arterial oxygen tension (Pa_O2: 9.1 [1.3] kPa) in the presence of normocapnia (Pa_CO2: 5.6 [0.6] kPa). In the control group, all lung function parameters were in the normal range. The body mass index (BMI) was lower in the patient group (mean difference: 2.4; 95% confidence interval [95% CI]: 0.5-4.4 kg/m²).

All patients used inhalation therapy: 2-agonists (n = 34), anticholinergics (n = 29), and steroids (n = 29). Twenty-six patients used theophylline, and 17 patients used acetylcysteine continuously. Furthermore, 17 patients used oral prednisolone as a maintenance treatment (mean daily dose: 7.6 [4.4] mg; mean duration of therapy: 3.4 [5.3] yr), while three patients used oral betamethasone.

Muscle Metabolites in COPD Patients and Control Subjects

The values of the muscle metabolites are summarized in Table 2. No significant differences were found between the total COPD group and the control group in ATP, ADP, AMP, or TAN. However, the ATP/ADP ratio was significantly lower in the patient group compared with the control group (mean difference: 0.5; 95% CI: 0.02-0.98). Muscle content of PC, C, and CTOT was not significantly different between both groups, while the PC/C ratio was lower in the patient group (mean difference: 0.19; 95% CI: 0.02-0.36).

IMP could be detected in 25 of 34 patients compared with one of eight control subjects (² = 10.23, p = 0.0014). The mean IMP level in the 25 patients with detectable IMP was 0.18 (0.14), ranging from 0.06 to 0.59 mmol/kg dry weight. The IMP level of the one control subject with detectable IMP was 0.06 mmol/kg dry weight. The lower level of detection of IMP with the present analytic technique was found to be 0.06 mmol/kg dry weight. Muscle contents of glycogen, glucose, and lactate were not significantly different between patients and control subjects.

Characteristics of IMP+ and IMP Patient Subgroups

Based on the presence or absence of detectable levels of muscle IMP, the patient group was subdivided into two groups, i.e., an IMP-positive (IMP+) and an IMP-negative (IMP) group (Tables 3 and 4). No differences in airflow obstruction or resting arterial blood gases were found between both patient subgroups (Table 3), but DL_CO was significantly lower in the IMP+ subgroup (median difference: 15%; p = 0.049). No differences in BMI or PI_max were found between both groups. A 12-min walking test was performed by 15 of 25 IMP+ and eight of nine IMP patients, and the 12-min walking distance tended to be lower (p = 0.09) in the IMP+ group. The patient subgroups did not differ in use of maintenance treatment of oral steroids (14 of 25 in the IMP+ and six of eight in the IMP patient subgroup; ² = 0.31, p = NS). Furthermore, no differences were found in daily dose of steroids or in duration of therapy. Also regarding other drug therapy, no differences were found between both subgroups (data not shown).

Muscle Metabolites in IMP+ and IMP Patient Subgroups

Muscle metabolites were compared in both patient subgroups (Table 4). No differences in ATP, ADP, AMP, or TAN were found between the two subgroups. The ATP/ADP ratio was lower in the IMP+ compared with the IMP subgroup (mean difference: 0.38; 95% CI: 0.02-0.73). Although PC, C, and CTOT did not differ, the PC/C ratio was significantly lower in the IMP+ patients (mean difference: 0.28; 95% CI: 0.13-0.43). Muscle glycogen (mean difference: 50; 95% CI: 2-97 mmol glycosyl units/kg dry weight) and glucose (mean difference: 3.0; 95% CI: 1.0-4.9 mmol/kg dry weight) were higher in the latter group. However, no significant differences in muscle glycogen were found comparing both patient subgroups separately with the healthy control subjects.

In the IMP+ patients, a significant negative correlation was found between IMP levels and the ATP/ADP ratio (r = 0.63, p = 0.001) (Figure 1). Correlations between IMP and PC, PC/C ratio, Pa_O2, and DL_CO did not reach the level of significance.

View larger version (6K): [in this window] [in a new window]   Figure 1.   The relationship between ATP/ADP ratio and IMP levels (mmol/kg dry weight muscle) in 25 COPD patients with detectable IMP in resting muscle (r = 0.63, p = 0.001).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

In the present study, elevated muscular IMP levels and decreased ATP/ADP and PC/C ratios were found in patients with stable COPD compared with healthy control subjects under resting conditions. ATP, ADP, AMP, and TAN did not differ between patients and control subjects. Within the patient group, two subgroups could be distinguished: a subgroup with muscular IMP content  0.06 mmol/kg dry weight (IMP+ patients) and a subgroup with an IMP content < 0.06 mmol/kg dry weight (i.e., below the detection level) (IMP patients). This division seemed relevant since only one of eight control subjects also had a measurable IMP content. In the IMP+ patients, lower ATP/ADP and PC/C ratios were found compared with IMP patients. Clinically, IMP+ patients differed from IMP patients in that they had a significantly lower DL_CO.

IMP and ammonia (NH₃) are produced during deamination of AMP (AMP IMP + NH₃). A proportion of the IMP produced is degraded to inosine, which can be further catabolized to hypoxanthine and eventually to uric acid. However, IMP can also be reaminated via the purine nucleotide cycle, eventually resulting in ATP resynthesis.

AMP is mainly formed by the myokinase reaction, which also produces ATP (2ADP AMP + ATP). AMP formation occurs in times of metabolic stress, when the mitochondrial process for rephosphorylating ADP is unable to maintain a high ATP/ADP level. The preservation of a high ATP/ADP ratio is imperative for adequate cellular function. To ensure an ongoing myokinase reaction in times of metabolic stress, AMP is deaminated to IMP, as outlined above.

In healthy subjects, IMP content in resting muscle is very low. Elevated IMP levels are only found during high intensity exercise (75% O₂max) (8). Several studies in healthy subjects indicated that special conditions, such as hypoxemia (18) or initial shortage of muscular glycogen, induce enhanced IMP accumulation during exercise (19). It is generally believed that, in these acute situations, IMP formation reflects an imbalance between ATP utilization and resynthesis.

In the present study, we hypothesized, based on the previously observed changes in aerobic and glycolytic capacity (6, 7), that in patients with stable COPD alterations might occur in muscular high-energy phosphate concentrations. Furthermore, we assumed that IMP measurements would enable us to detect subtle changes in muscular energy balance. Indeed, in this study, elevated IMP levels were found that were associated with lower ATP/ADP ratios. The finding of elevated IMP content in resting muscle, is to our knowledge, unique. The observed negative relationship between IMP levels and ATP/ADP ratios seems to be in line with other studies where a direct stimulation of AMP deaminase by low ATP/ADP ratios was suggested (20, 21). Furthermore, this negative relationship suggests that the elevated IMP levels are associated with an imbalance in ATP utilization and resynthesis. Because of the observed relationship between IMP content and ATP/ADP ratio, other explanations for the enhanced IMP content, such as disturbances in pathways catabolizing IMP (such as degradation of IMP to inosine and further) or disturbances in the purine nucleotide cycle (where IMP is reaminated to AMP), seem less relevant.

An imbalance between ATP utilization and resynthesis can be caused by enhanced ATP utilization, decreased ATP resynthesis, or both. Earlier studies by our group and others have shown that a significant proportion of COPD patients have increased resting energy expenditure (22). The cause of this abnormality is incompletely understood. This increased resting energy expenditure might reflect enhanced ATP utilization in rest. In COPD patients, a decreased ATP resynthesis may occur because of hypoxemia or a decreased substrate availability. In our study, resting Pa_O2 did not differ in IMP+ compared with IMP patients. However, IMP+ patients had a lower DL_CO, and a decreased DL_CO is frequently associated with desaturation during exercise (23). Therefore, intermittent hypoxemia might play a role in the observed enhanced IMP content in COPD patients. Regarding the influence of substrate availability on IMP formation, several studies suggested increased IMP formation during exercise in glycogen-depleted muscle (19, 24). However, during resting conditions, IMP levels were not elevated in glycogen-depleted muscle of healthy volunteers (24). In any case, because in our study glycogen was even elevated in resting muscle of IMP+ patients compared with IMP patients, it was clear that glycogen depletion did not play a role in the elevated IMP levels in COPD patients. The observed higher muscle glycogen and free glucose content in IMP+ patients compared with IMP patients suggest differences in glucose handling of the skeletal muscle cells between both patient subgroups, the nature of which is presently unknown.

Because medication, physical activity, and nutritional status may influence muscle function and metabolism, special attention has been paid to these variables. More than half of the patients used oral corticosteroids as a maintenance treatment. However, no differences were found between the two subgroups of patients in relative number of patients using corticosteroids, daily dose of corticosteroids, or duration of treatment. Also, no differences were found between the two patient subgroups in use of other maintenance treatment. All patients were fully ambulatory, and there were no apparent differences in training status between both patient subgroups. Nevertheless, 12-min walking distance tended to be lower in the IMP+ patients, suggesting a functional difference between IMP+ and IMP patients. Regarding nutritional status, the BMI was significantly lower in the COPD patients compared with the control subjects. Furthermore, there was a trend toward a lower BMI in IMP+ patients compared with IMP patients, but information on body composition would be essential to fully appreciate these data.

Biopsies were obtained from the tibialis muscle using a conchotome. However, in all the studies quoted here, biopsies were taken from the quadriceps femoris muscle, which has a different fiber type distribution compared with the anterior tibialis muscle (40% and 70% type I fibers, respectively) (25). In healthy subjects during exercise, higher IMP formation has been found in type II fibers compared with type I fibers (8). Therefore, if quadriceps femoris muscle would have been used in this study, results might have been different.

In two other studies, high-energy phosphate compounds were investigated in patients with stable COPD (26, 27). In accordance with our study, Möller and associates (26) found a decreased ATP/ADP ratio in COPD patients. However, in contrast with our study, both Möller and associates and Jakobsson and Jorfeldt (27) found decreased ATP and glycogen content in resting muscle in COPD patients. Differences in patient selection regarding resting Pa_O2, nutritional status, and use of corticosteroids might have attributed to these discrepancies, as it is known that ATP levels might be decreased in acute hypoxemia (28), while glycogen concentration might be decreased due to malnutrition (29) and increased during use of corticosteroids (30). Further cause for the discrepancies between the studies may include the fact that both Jakobsson and Jorfeldt and Möller and associates used quadriceps femoris muscle instead of anterior tibialis muscle.

In summary, in patients with stable COPD elevated muscular IMP content was found under resting conditions. IMP levels were negatively related to the ATP/ADP ratio, suggesting an imbalance between ATP utilization and resynthesis. The fact that these abnormalities were already found under resting conditions suggests that muscular energy status in COPD patients might even be more compromised during exercise. The cause and consequences of these disturbances in muscle energy metabolism in COPD patients need further exploration.