Muscle fatigue has been defined as a temporary loss in force or torque-generating ability due to recent muscle contraction.1 The mechanisms underlying muscle fatigue are numerous and may have their origins anywhere from the central nervous system (CNS) to cellular-level cross-bridge cycling.2,3 The major proposed mechanisms of fatigue production are: (1) inadequate excitation of motoneurons, (2) poor action potential transmission along axonal branch points, (3) failure of the action potential to invade the synaptic bouton or failure to trigger transmitter release at the myoneural junction, (4) failure of the post-junctional membrane to be depolarized adequately, (5) failure of the action potential to propagate the full length of the sarcolemma or the action potential has too low of an amplitude, (6) failure of the action potential to invade the T-tubule system and sarcoplasmic reticulum system or failure of the action potential to trigger calcium release, (7) contractile system of the muscle, and (8) mitochondria, signifying metabolic events that sustain contraction.2

Until recently, muscle fatigue was thought to be caused by lactic acid accumulation or a lowered pH within the muscle due to a rising hydrogen ion concentration ([H+]) during anaerobic metabolism.46 Recent literature, however, has shown that at the cross-bridge level, [H+] may not be a major causative factor of muscle fatigue.718 Inorganic phosphate accumulation, on the other hand, has been implicated as a significant cause of muscle fatigue, presumably in its role at the cross-bridge level.7,1921

During the study of muscle fatigue, a phenomenon known as low-frequency fatigue (LFF) was observed. It was first described by Edwards and colleagues22 and is characterized by a proportionately greater loss of force in response to low- versus high-frequency muscle stimulation. This form of fatigue is long-lasting, taking hours or even days to subside, and may play a significant role in the decline in the force-generating capabilities of skeletal muscle. This update will focus on reviewing the characteristics, possible mechanisms, and clinical implications of LFF.

Definition and Characteristics

During LFF, the contractile responses to low-frequency stimulation are diminished.23, 24 The main features of LFF are: (1) the forces at low frequencies of stimulation are the most severely affected, (2) recovery of force is slow, taking hours or days, and (3) the effect persists in the absence of gross metabolic or electrical disturbance of the muscle.23 These features are in contrast to those of high-frequency fatigue, which is characterized by loss of force at high frequencies of stimulation that is rapidly (within seconds) reversed by reducing the frequency.25, 26, 27

Low-frequency fatigue has been induced using voluntary contractions28 and both high-frequency (100 Hz)2931 and low-frequency (10–40 Hz)29, 30 electrical stimulation. Low-frequency fatigue can result from virtually any frequency of muscle stimulation, although, as stated above, the characteristic force declines are noticeable only when tested at low frequencies.23 The observed force decrements in human skeletal muscles are typically greater than 50% of predetermined force values for frequencies between 10 and 30 Hz.23 Interestingly, although these frequencies are similar to discharge rates observed from active motor units during everyday activities, the force decrements resulting from LFF during volitional muscle activation have not been reported.23

Low-frequency fatigue is typically measured by recording the torque responses to different frequencies of electrical stimulation (large surface electrodes are used to stimulate the muscle over the motor point of a muscle or placed near the motor nerve).3235 A commonly used index of LFF is the change in the ratio of the force production with 20-Hz stimulation to that of 50-Hz or 80-Hz stimulation.22, 23, 29 Any decrease in this ratio, compared before and after exposure to a fatiguing protocol, is commonly interpreted as LFF.36

Because LFF takes so long to subside, LFF may not have an ionic or metabolic basis like other forms of fatigue that subside more quickly.23 It has been suggested that LFF may be caused by muscle fiber damage or impairment in the excitation-contraction (EC) coupling mechanism of muscle activation.23 In support of muscle fiber damage, it has been noted that LFF is a prominent characteristic following eccentric exercises or isometric exercises at a long muscle length and may be seen during the repair process if the muscle fiber was damaged during these types of exertions.23, 28, 37, 38

In contrast, impairments in EC coupling that have been suggested to play a role in LFF include a reduction of Ca2+ release from the sarcoplasmic reticulum (SR),22, 24, 37, 39 a decrease in the calcium sensitivity of troponin,24, 37, 40 poor conduction of the action potential in the T tubules,22 and a reduced Ca2+ reuptake by the SR.41


Westerblad and colleagues31 investigated the sites of LFF and showed that, in isolated mouse muscle fibers, after 30 minutes of recovery (when LFF is present but metabolic recovery is largely complete22, 41), Ca2+ sensitivity of troponin is indistinguishable from control conditions. They also found that the failure of Ca2+ release from the SR was uniform across the muscle fiber, indicating proper conduction of the action potential in the T tubules.31 MacIntosh and Rassier42 determined, therefore, that the primary mechanism of LFF is either a decreased Ca2+ release from the SR or faster uptake of Ca2+ by the SR during contraction. Attenuated Ca2+ transient is thought to reduce the amount of Ca2+ that can bind to troponin, thus limiting the amount of force that can be produced.39, 43

Interestingly, Chin and colleagues30 have shown that decreased SR release of Ca2+ has at least 2 components: (1) a metabolic component, which, in the presence of glucose, recovers within 1 hour, and (2) a component dependent on the elevation of the time integral of the concentration of calcium in the interstitial space of a muscle cell ([Ca2+]i-time integral), which recovers more slowly. The [Ca2+]i-time integral depends on both the initial release of Ca2+ by the SR and how long the Ca2+ stays in the intracellular milieu. Because the recovery of the metabolic component of decreased SR release of Ca2+ takes place so quickly, it is unlikely that it is a cause of low-frequency fatigue. In contrast, an elevated [Ca2+]i-time integral associated with repeated tetanic stimulation has been shown to result in prolonged reduction in Ca2+ release and LFF.29 By increasing the time integral, Ca2+ remains in the intracellular space longer, which elicits LFF, as elevated intracellular Ca2+ inhibits SR Ca2+ release, producing a smaller Ca2+ transient.

Fryer and colleagues44 proposed another mechanism for the decreased release of Ca2+ from the SR that deals with a precipitate that may form within the SR lumen of human skeletal muscle under fatigued conditions. According to this theory, the increased inorganic phosphate (Pi) that is produced during fatigue is transported into the SR where it forms an insoluble precipitate with Ca2+. If one assumes that the rate of leak of Ca2+ from the SR is proportional to the amount of free Ca2+ in the SR lumen, a decrease in the free Ca2+ due to precipitate formation would cause a decrease in release of Ca2+ from the SR and LFF.44 Additionally, there may be a slower formation of a second more insoluble Ca2+-Pi species that may contribute to a much longer-lasting form of depressed SR Ca2+ release, which may contribute to the long duration of LFF.31, 44

Westerblad and colleagues31 explained why a decrease in [Ca2+] affects muscle forces at low frequencies of stimulation, but not at high frequencies, by noting the shape of the [Ca2+]i-tension relationship (Figure). Because [Ca2+]i at high frequencies is on the horizontal part of the curve, moderate falls in [Ca2+]i have no effect on muscle tension. At low frequencies, [Ca2+]i is on the steep part of the curve, so falls in [Ca2+]i produce large changes in tension.31

Figure 1.

A hypothetical [Ca2+]i-tension curve is shown. Because [Ca2+]i at high frequencies is on the horizontal part of the curve, moderate falls in [Ca2+]i have no effect on muscle tension. At low frequencies, [Ca2+]i is on the steep part of the curve, so falls in [Ca2+]i produce large changes in tension.

Clinical Implications

The primary clinical implication for LFF is the reduction in muscle forces in response to low-frequency activation. The observed motor unit discharge rates during voluntary skeletal muscle activation rarely exceed 30 Hz, making volitional activation possibly susceptible to the effects of LFF.25 Although force decrements resulting from LFF during volitional muscle activation have not been reported,23 LFF may result in the need for higher levels of activation by the CNS. This need for increased CNS drive may cause patients to experience a greater sense of effort during repetitive activities such as walking and stair climbing and may limit patient performance both in the physical therapy clinic and when performing activities of daily living. An appreciation of the effects, causes, and clinical manifestations of LFF may help clinicians to identify the factors limiting a patient’s performance and to design the most effective treatment program for each patient.

Functional electrical stimulation (FES), which is the use of electrical stimulation to produce functional movements in patients with a damaged CNS, typically uses frequencies in the 20- to 30-Hz range. Thus, the appearance of LFF during application of FES may be an important factor when determining the most effective stimulation protocol to use during FES.45 Binder-Macleod and Russ45 investigated the effects of activation frequency and force on the production of LFF during brief, intermittent stimulation trains and found that lower frequencies of stimulation produced greater LFF within 2 minutes of recovery, which appeared to be related to the average force-time integral produced by the repetitive contractions.45 After 2 minutes of recovery, however, there was no correlation between the force produced during the contractions used to produce fatigue and the degree of LFF.45 Additionally, there was more LFF after 30 minutes of recovery than after 2 minutes of recovery, which was possibly a result of the proposed nonmetabolic component of LFF production.45 Noting the time course of LFF may help clinicians to understand the variability in the responses to electrical stimulation that patients display and to design effective stimulation strategies to overcome LFF.

There is good evidence that once LFF is induced, switching to higher frequencies of stimulation or using stimulation patterns that have a short-duration, high-frequency burst at the start of the stimulation train will augment muscle performance in healthy and paralyzed muscle.4654 This force augmentation can be clinically applicable during FES. According to Kebaetse and Binder-Macleod,46 the use of variable-frequency trains, where the stimulation pattern of a short-duration (<500 milliseconds), constant-frequency train (CFT) was altered by including a 2- or 3-pulse high-frequency burst at the onset of the CFT, produced greater isometric forces than using CFTs of similar train frequency. Variable-frequency trains also produced greater excursions, power, and peak forces than CFTs of similar frequencies during nonisometric contractions, especially when muscles were fatigued.50, 52, 53 Recently, Kebaetse and colleagues54 also demonstrated that switching to higher stimulation frequencies as the human quadriceps femoris muscle became fatigued improved the ability of subjects with spinal cord injuries to perform repetitive, nonisometric contractions compared with using low frequencies (20 or 33 Hz) or a high frequency (66 Hz) alone. The use of high-frequency stimulation was needed, in part, to overcome the effects of LFF during repetitive activation.

In summary, the occurrence of LFF may markedly affect the central drive and sense of effort experienced by patients during voluntarily contractions or the activation pattern needed to produce targeted levels of force during electrically elicited contractions. By understanding the causes and clinical manifestations of LFF, clinicians can better understand the factors limiting their patients’ performance and thus design more effective treatments.


  • Both authors provided concept/idea/project design and writing. Dr Binder-Macleod provided fund procurement and facilities/equipment.

  • This study was supported by National Institutes of Health grant HD-36379 to Dr Binder-Macleod.

  • Received August 10, 2005.
  • Accepted March 7, 2006.


View Abstract