139 - 147: Prevention of Post-Stroke Disuse Muscle Atrophy with a Free Radical Scavenger

Motor weakness is one of the most common symptoms of stroke which largely disturbs activities of daily living in stroke patients. Even with optimal therapy, 15-30% of stroke patients are permanently disabled and 20% require institutional care at 3 months after onset [1]. Such a long-term functional disability is caused mainly by damage to the motor neuron tract at the frontal cortex, the corona radiata, the internal capsule or the pons which leads to hemiparetic weakness of the contralateral upper and lower limbs. The impaired function of the motor tract may be restored afterwards by the compensatory development of the motor neuron tract as discussed by several authors in this book. Yet, even after the compensatory development of the motor neuron tract, motor dysfunction may remain unaltered, if disuse muscle atrophy is developed in the paretic limb associated with muscle weakness and/or articular contracture during the acute and subacute phases. Disuse muscle atrophy occurs most likely in the paretic lower limb of elderly stroke patients who are bedridden during the acute phase of stroke. The antigravity muscles of the lower limbs, such as the soleus, gastrocnemius, and vastus lateralis, are most commonly affected by immobilization [2,3]. Stroke is primarily a central nervous system disease. However, we should be reminded that secondary muscle changes play an important role in the persistence of lower-limb disability in the chronic phase. The muscle changes may partly be reversible and may be treatable with rehabilitation. Yet, delayed treatment of muscle changes likely fails to restore normal muscle structure and muscle strength eventually leading to articular contracture. Therefore, myoprotective drug therapy should be initiated in the early phase of stroke. Myoprotection is considered easier and clinically more effective as compared with neuroprotection. Unfortunately, however, little attention has been paid to myoprotective drug therapy in the past history of acute stroke management.

During normal ageing, muscle tissue gradually changes in its distribution with advancing age; fast-twitch muscle fibers (MHC type IIa and IIb) decrease and mitochondria-rich slow-twitch muscle fibers (MHC type I) increase. This change in fiber type distribution results from motor unit denervation and subsequent reinnervation from adjacent intact muscle fibers. The changes result in reduction of muscle strength. According to the study of Kostka [4], muscle strength decreases by 18% between 30 and 60 years of age, and further decreases by another 20% between 60 and 90 years of age. Thus, elderly subjects are in a preparatory condition for motor weakness even prior to stroke. Once stroke occurs in elderly subjects, it likely causes severe hemiparesis which forces elderly patients to stay bedridden for 1-2 weeks or longer. Immobilization due to the bedridden state likely causes disuse muscle atrophy in the paretic leg as well as the nonparetic leg (fig. 1).

Long-term muscle disuse due to immobilization or chronic bed rest is known to induce muscle atrophy and muscle strength reduction particularly in legs [5,6]. Disuse atrophy of lower-limb muscles was confirmed to occur following 35 days of bed rest in healthy young subjects with a mean age of 24 years [7]. Disuse atrophy occurs more rapidly in older subjects. In the study of Kortebein et al. [8], lower-limb disuse atrophy was observed following 10 days of bed rest in healthy subjects with a mean age of 67 years. In the study of Paddon-Jones [9], elderly patients showed the same amount of lean leg muscle mass after 3 days of hospitalization as healthy older subjects experienced after 10 days of inactivity, and a 3-fold greater loss of lean leg muscle mass than a younger cohort confined to bed for 28 days. Thus, elderly subjects may develop disuse muscle atrophy even a few days after bed rest immobilization. Early development of disuse atrophy was confirmed to occur in rats with hind limb unloading [10]. The soleus muscle mass began to decrease at 4 days after hind limb immobilization.

Recent advances in cellular and molecular biology have provided quite a few pieces of evidence for understanding the pathophysiology of disuse atrophy. Disuse atrophy of skeletal muscle occurs due to both a decrease in muscle protein synthesis and an increase in the rate of proteolysis [11,12]. It has not yet been fully clarified how a decrease in muscle protein synthesis and an increase in proteolysis are initiated. However, experimental studies indicated a close relationship between oxidative stress and disuse muscle atrophy [13,14,15]. More specifically, atrophic muscles of rats showed elevated levels of thiobarbituric acid reactive substances and oxidized glutathione, which are both markers of oxidative stress. Furthermore, an increase in the formation of superoxide radical anions (O₂^-) was indicated by elevated Cu/Zn-containing superoxide dismutase activity in the cytoplasm of atrophic muscle cells, and 12 days of disuse atrophy showed a 2.2-fold rise in xanthine oxidase levels relative to controls. Histochemical studies using transmission electron microscopy have revealed elevated levels of reactive oxygen species such as H₂O₂ in atrophic muscles. Metal compounds such as iron have also been shown to increase oxidative stress in atrophic muscles, and the administration of the iron-chelating agent, deferoxamine, suppressed the increase in thiobarbituric acid reactive substances and oxidized glutathione in the atrophic muscles of rats [16]. In more recent years, the role of iron in oxidative stress has attracted close attention in the field of physiology [17,18].

Another factor that may be involved in the development of muscle atrophy is elevation of the intracellular calcium ion (Ca²⁺) levels during muscle inactivity [13,14,15]. Ca²⁺ levels in the cytoplasm of atrophic muscle cells were approximately 4-fold higher than levels in normal muscle cells. This accumulation of intracellular Ca²⁺ during muscular disuse may occur due to ionic disturbances of the cell membrane, which retards cellular removal of Ca²⁺. This resulting elevation of cytosolic Ca²⁺ may lead to the activation of calcium-dependent proteases that mediate, thus leading to the breakdown of muscle tissue.

In addition to inducing muscle atrophy during disuse, oxidative stress was shown to be enhanced following remobilization in rats [19]. Oxidative stress markers increased rapidly when rabbits with leg immobilization began to move the immobilized leg. This suggests that the production of free radicals occurs continuously in patients who start moving paretic limbs. In order to prevent disuse muscle atrophy, therefore, free radicals need to be suppressed continuously for a considerably long duration.

Edaravone, a free radical scavenger, was shown to have neuroprotective action and to ameliorate functional outcome of acute stroke patients in a placebo-controlled double-blind study performed in Japan [20]. Edaravone is now widely used clinically for the treatment of acute ischemic stroke in Japan. This is the only drug in the world which has been approved as a neuroprotective drug and permitted for clinical use in the management of acute stroke. During the last 2 decades, more than 30 substances with neuroprotective action were subjected to phase III clinical trials to confirm the efficacy in improving the clinical outcome of acute stroke patients [21,22,23,24,25]. However, all of them including free radical scavengers, except for edaravone, failed to show clinical effectiveness. Questions are raised here. Why could only edaravone prove clinical usefulness? Are free radical scavengers superior to the other types of neuroprotective agents, such as NMDA receptor antagonists or Ca²⁺ channel blockers? If free radical scavengers are superior to the other types of neuroprotective drugs, then why did the other free radical scavengers, such as tirilazad [21] or NXY-059 [22], fail to show clinical effectiveness? In order to find the answers to these questions, we should pay attention to the design of the phase III clinical trial of edaravone which quite differs from that of all the other neuroprotective drugs. In the edaravone trial, ischemic stroke patients admitted within 72 h after onset were enrolled into the study. They were randomly allocated to two groups. In one group, edaravone was intravenously administered twice daily for 14 days, and in the other group, placebo was given in a similar manner for 14 days. Thus, the time limit of initial drug administration and the duration of drug administration are enormously longer in the edaravone trial as compared with the other phase III trials of neuroprotective drugs. In almost all the phase III trials of neuroprotectants, the drugs were initially administered within 6-12 h after stroke onset and were continued for 1-3 days. For instance, in the clinical trials testing the efficacy of a free radical scavenger, tirilazad, the drug was first given within 4-12 h after onset and was continued for 3-4 days [21]. Likewise in the SAINT II study for evaluating the efficacy of a free radical scavenger, NXY-059, the drug was initially administered within 6 h after onset and was continued for 3 days [22]. Such an early and short-term administration of drugs is reasonable for testing the clinical usefulness of neuroprotective drugs, since irreversible changes likely occur within 6 h in the majority of ischemic neuronal cells and the process of neuronal cell death may be completed within several days after onset. In most phase III clinical trials of other types of neuroprotective agents, the drugs were given in a similar manner as in the SAINT II study or tirilazad studies. The results in these studies were all unsuccessful. In the edaravone phase III trial, the initial administration of the drug was considerably late; the average time for the first drug administration was 37.3 h after onset. Thus, the initiation of edaravone administration was unfavorably late for testing the clinical efficacy of neuroprotective drugs. Presumably, in the edaravone trial, neuronal cells in the majority of ischemic areas may have had irreversible changes prior to edaravone administration. Edaravone was given consecutively for 14 days in all the patients. Such a long-term administration of a free radical scavenger may be meaningless from the viewpoint of neuronal protection, since the process of neuronal cell death is likely completed within several days after onset. On the other hand, delayed and long-term administration of a free radical scavenger for 14 days is meaningful, if the target is myoprotection, since the process of disuse muscle atrophy begins at several days after immobilization and is enforced at the time of remobilization. The results of the phase III clinical trial of edaravone suggest that the beneficial effects of the drug may be attributable to myoprotective effects rather than to neuroprotective effects.

In order to test the hypothesis that long-term administration of a free radical scavenger may prevent disuse muscle atrophy of lower limbs and improve locomotor function in stroke patients, we performed a pilot study (Muscle Atrophy Restraint with Vigilant Edaravone Long-Term Use after Stroke: MARVELOUS) in patients with ischemic stroke [26]. The study was an open-label, multicentered, randomized, controlled pilot study performed in Japan. Written consent was obtained from all patients prior to registration. The study protocol was permitted by the local ethics committees of all the institutes that participated. Acute ischemic stroke patients with hemiparesis were randomly allocated to receive edaravone for 3 days (short-term group) or 10-14 days (long-term group). These two durations were determined on the basis of the assumption that 3-day administration likely exerts neuroprotective effects and 10- to 14-day administration may provide both neuro- and myoprotective effects. Inclusion criteria were: (1) age 20-79 years, (2) definitive paresis or paralysis of the leg of the diseased side, (3) admission within 24 h after onset of stroke, (4) no neurological deficits prior to stroke, (5) no thrombolytic therapy and (6) well-maintained consciousness level. Edaravone (30 mg, twice daily) was administered intravenously during the first 3 days of admission in the short-term group and during the first 10-14 days of admission in the long-term group. All the patients were admitted to an acute stroke center and underwent bedside rehabilitation therapy followed by out-of-bed rehabilitation therapy. In order to evaluate changes in femoral muscle volume, the circumference of the thigh was measured at 15, 10 and 5 cm above the upper end of the patella bilaterally in a position with legs stretched on the floor. The circumference of the lower leg was also measured at its largest part bilaterally in the same manner to evaluate crural muscle volume. The first measurement of the leg circumference was performed within 4 days after admission, and the values were regarded to represent the baseline leg muscle volume. The second and third measurements were performed at 3 weeks and 3 months after stroke onset, respectively. The reduction of leg circumference at 3 weeks and 3 months after stroke onset as compared with the baseline value was expressed as percentage reduction and was cited to represent the degree of muscle atrophy. The primary endpoints of the study included the degree of leg muscle atrophy and the severity of leg locomotor dysfunction at 3 months after stroke onset. The severity of locomotor dysfunction was evaluated with maximum walking speed (MWS) over a distance of 10 m. Data are expressed as percentages for categorical variables or means for continuous variables. A Pearson χ² or Fisher exact test was used for the statistical comparison of qualitative or categorical variables. An unpaired Student t test was also used to compare differences in the degree of disuse muscle atrophy and MWS between the two groups. A p value of <0.05 was considered to be statistically significant.

A total of 47 patients were enrolled in the study, and 3-month follow-up was completed in 41 patients (21 in the short-term group and 20 in the long-term group). As shown in table 1, there was no significant difference in age, gender, and severity of neurological symptoms on admission. Edaravone was administered for 3 days in all the patients in the short-term group and for 13.6 ± 1.2 days in the long-term group. Bedside rehabilitation was performed within 3 days of admission in 67% of patients in the short-term group and in 70% of patients in the long-term group. By the end of 3 weeks after stroke onset, rehabilitation therapy was performed for 16 ± 5 days in the short-term group and for 16 ± 7 days in the long-term group. There was no significant difference in the state of rehabilitation between the two groups.

Table 2 shows the state of gait disability and grade of muscle atrophy at 3 weeks and 3 months after stroke onset. Changes in femoral volume measured at 15 cm above the patella are demonstrated as representative markers of leg disuse atrophy. At 3 weeks after onset, 48% of patients in the short-term group and 50% of patients in the long-term group had gait disability either requiring a wheelchair or remaining in a bedridden state. Disuse muscle atrophy was more or less developed bilaterally in almost all the patients in both groups. The grade of muscle atrophy was 5.0 ± 3.4% in the paretic leg and 3.7 ± 4.4% in the nonparetic leg in the short-term group and 4.4 ± 4.1% in the paretic leg and 2.0 ± 3.8% in the nonparetic leg in the long-term group. There was no significant difference in the grade of atrophy between the two groups. At 3 months after stroke, muscle atrophy in the short-term group became more remarkable as compared with 3 weeks after stroke, whereas the atrophy in the long-term group became less remarkable as compared with 3 weeks after stroke. The grade of atrophy was 8.3 ± 5.2% in the paretic leg and 5.7 ± 6.4% in the nonparetic leg in the short-term group and 3.6 ± 5.9% in the paretic leg and 1.5 ± 6.0% in the nonparetic leg in the long-term group. The grade of atrophy in the long-term group was significantly less prominent both for the paretic leg (p < 0.01) and nonparetic leg (p < 0.05) as compared with that in the short-term group. The MWS over the 10-meter distance was 53.6 ± 54.8 cm/s in the short-term group and 97.9 ± 67.3 cm/s in the long-term group. The MWS was significantly faster in the long-term group than in the short-term group (p < 0.05). There was a significant negative correlation between the grade of paretic leg atrophy and MWS (r = -0.87, p < 0.001).

Thus, the pilot study suggests that a free radical scavenger may prevent the development of leg disuse atrophy and ameliorate locomotor function in hemiparetic stroke patients, if administered for a considerably long time. It remains unclear why anti-atrophic effects were not remarkable at 3 weeks after stroke and became prominent at 3 months after stroke. This may be explained by the delayed involvement of apoptosis during the course of disuse muscle atrophy, which eventually leads to muscular cell shrinkage and/or death [17,18]. As a result, the clinical manifestation of muscle atrophy may not become apparent until changes in the muscular intracellular environment have taken place, which may occur over a period of several weeks. Edaravone is a drug with neuroprotective action. Therefore, the possibility that the prevention of muscle atrophy is mediated partly by neuroprotective action cannot be excluded completely. It seems, however, obvious that the prevention of disuse muscle atrophy is closely connected with improvement of functional outcome.

Leg motor weakness and resulting locomotor dysfunction in stroke patients are attributable to both ischemic brain damage and disuse muscle atrophy. In the brain, neurons are unable to survive for extended periods of time under ischemic conditions, and the protection of such vulnerable brain tissue with drug therapy is difficult in a clinical setting. In contrast, muscle is a robust tissue, and the process of disuse muscle atrophy takes place gradually following stroke, which may provide a larger therapeutic window of opportunity. Therefore, agents that offer myoprotection may provide an easier and more effective treatment option to ameliorate the functional disability of stroke patients. Based on the results of the MARVELOUS study, the use of antioxidant therapy for as long as possible appears to be warranted to provide the maximal level of both neuroprotection and myoprotection. Larger, randomized controlled clinical trials are required to confirm the beneficial effects of free radical scavengers in the management of stroke.