Dating back to the early work of Fechner (14) and Scripture et al. (32), there has been evidence that chronic training or motor practice of one limb can have an effect on the untrained contralateral limb — a phenomenon commonly known as the cross-education effect. The idea can be conceptualized broadly as bilateral training effects after unilateral motor practice. Repetitive “training” of one limb affects homologous muscles in both limbs — whether the motor training is guised as skilled, ballistic, strength, learning, or some combination of these. The type of motor activity being “transferred” to the nontraining side (e.g., cross-education of strength or skill) and how training is executed (i.e., a single session or multiple sessions over weeks) are important because these can influence the presentation and mechanisms of the effect. Here we focus the discussion on cross-education after chronic strength training (i.e., multiple sessions over weeks), which is the paradigm used in the two exemplar clinical studies.
Cross-education is well recognized in kinesiology, motor control, and neurobiology disciplines and has long been argued for its potential clinical applications, but it may not be so well known in other relevant fields such as biomedical engineering and rehabilitation. Some view cross-education of strength as little more than an interesting, yet trivial, side effect of unilateral training. For the healthy neurologically intact person, there is no apparent relevance of cross-education of strength as a training paradigm because the common goal would be to enhance the function of both limbs symmetrically. Chronic strength training of one limb would result simply in functional or morphological asymmetry between limbs, which usually is undesirable unless the goal of practice is to develop a specific unilateral skill such as throwing, kicking, or paddling. From the perspective of rehabilitation, the relevance of cross-education emerges as a way to benefit recovery of function after unilateral injury or neurological damage. The ultimate goal is to determine the best way to exploit these transfer effects to restore symmetry.
Here we operationally define symmetry in the neuromuscular system as a relatively balanced function between homologous limbs. This use of the term symmetry in the human motor system is not meant to imply that both sides of the body are identical, but rather operate within a range of normal lateralization of function. Neurological damage or injury would disturb homeostasis by creating abnormal asymmetry between sides, and the rehabilitation goal would be to attempt to restore that previous symmetry or, in the case of severe neurological damage, to restore some degree of bilateral functional capacity.
The formalized hypothesis presented here is that cross-education of strength — an adaptation after unilateral training often involving an asymmetrical outcome — should be best applied to clinical presentations with asymmetries, such as neurological damage after stroke or unilateral orthopedic injury. In this context, cross-education can be viewed as a clinically relevant therapy to aid in restoring function after neurological damage or injury where one limb is functionally disabled for a persistent or finite period and symmetrical training of both limbs is not possible. Central to the discussion is that underlying cross-education mechanisms should be viewed as evolutionarily conserved bilateral circuits that may underlie some of the adaptations to training. Although they are not used to the same extent as by quadrupeds (36), they afford a neurobiological substrate for restoration of impaired motor function. The fundamental point of this brief review is to increase awareness of these circuits and demonstrate the efficacy of cross-education in two pathological conditions affecting adults: orthopedic — where the injury is peripheral and affects musculoskeletal structures (23) — and neurologic, specifically stroke — where the injury affects the central nervous system (8).
CROSS-EDUCATION AND IMMOBILIZATION
The cross-education effect after unilateral strength training averages about 50% of that observed for the trained side (3,27), but the magnitude varies greatly depending on the characteristics of the task, the duration of training, and the particular training paradigm (10,22). Training the dominant side with novel manual strength tasks may enhance transfer effects (11), whereas training the nondominant upper limb may involve little or no transfer (4,19). In rare instances, the strength transfer effect can exceed improvements observed for the trained side (11,17). There is consensus in the literature that the effect is controlled by neural mechanisms, and although specific loci of adaptation remain unclear, several candidate mechanisms have been clearly identified (3). Although there is ample support for the contribution of cortical mechanisms to cross-education (13,18,21,31) with convincing data supporting plasticity of interhemispheric circuitry (18), there also is support for the contribution of spinal mechanisms (7). Logically, the neural mechanisms that contribute to cross-education reside at multiple sites with different temporal courses for plasticity, and the challenge is to isolate specific mediators.
Many cross-education articles focus on mechanisms and behavioral effects and emphasize the importance of continued work in the area because of its clinical implications. A renewed interest in cross-education as a viable clinical approach was spurred on by immobilization studies in healthy participants that demonstrate profound sparing effects in the immobilized limb of participants who concurrently trained the opposite limb (12,13,24,31). The short-term preservation effects observed in the immobilization studies suggest the use of cross-education as a therapeutic approach to prevent loss of muscle function during periods of disuse.
CLINICAL MODELS OF CROSS-EDUCATION
Unilateral orthopedic injury and poststroke hemiparesis represent viable models to test the clinical utility of cross-education because, in each case, there is an injury that temporarily or permanently results in an interlimb asymmetry. The utility of cross-education emerges because one limb is rendered unable to engage in training or movement and chronic training of the less affected or intact side could be used preventatively to offset maladaptation or therapeutically to restore lost function caused by damage.
The Case for Orthopedic Injury
Unilateral orthopedic injury presents a disuse-induced asymmetry as a consequence of necessary treatment (i.e., casting or immobilization). Aside from the pathology of the injury itself, the prolonged disuse results in functional strength loss and atrophy and heavily impacts the nervous system (33,35). After injury, there are severe decrements in muscle size, strength, and muscle activation (33,35), and efforts to restore symmetry typically involve targeted rehabilitation of the injured limb only after the immobilization period. Therefore, unilateral orthopedic injury is a potential model to apply cross-education as a therapeutic intervention.
As one example of a common orthopedic injury, distal radius fracture presents with lasting deficits and poor long-term outcomes in high-risk groups (2,16,34). Fractured limb strength at 3 months postfracture is approximately 50% of the nonfractured side (15,23) and deficits of approximately 12% or more persist even up to a year after the initial injury (2,34). Wrist fracture represents a logical extension of the preliminary work with wrist cast immobilization models in healthy participants (12,13). The proposition is that cross-education could be used during the disuse period to offset muscle wasting.
To test the hypothesis, Magnus et al. (23) implemented intense handgrip strength training of the nonfractured limb (progressing up to five sets of eight maximal effort, 3-s handgrip contractions, 3 d wk-1) during recovery from distal radius fracture in women older than 50 yr (mean age, 63 yr). The majority of patients were right handed, and the side of fracture was evenly distributed between dominant and nondominant sides. Training began within 1-wk after fracture, in addition to standard rehabilitation, and the outcomes were compared with those of patients receiving only standard rehabilitation. Standard rehabilitation exercises included active and passive range of motion exercises for the fractured limb only initiated at 9 wk after fracture. The cross-education intervention was shown effective for improving recovery of the fractured limb, where handgrip strength and active range of motion at 12 wk postfracture were significantly greater compared with those of the nontraining control group. During the interval from 9 to 12 wk after fracture, the cross-education group experienced a 34% greater increase in fractured limb strength compared with that in the control group (Fig. 1A) and near complete recovery of range of motion indicative of a more rapid recovery for the fractured limb. This time interval seems particularly important because, alarmingly, the nontraining control group experienced no improvement in fractured limb strength or range of motion up to 3 months postinjury. The cross-education strength training intervention continued up to 6 months, but there was no additional benefit compared with control.
The study marks the first “proof-of-principle” example of cross-education applied in an orthopedic clinical setting involving limb fractures. Without measures of brain activation or other markers of neural plasticity, there was no conclusion regarding mechanisms of the sparing effects, but the study demonstrates the potential to offset some of the functional loss after unilateral fracture. Because the intervention did not involve targeted strength training (over and above standard rehabilitation exercises) of the fractured limb after cast removal, the impact of cross-education on subsequent strength training of the fractured limb also remains open for study. Importantly, Magnus et al. (23) demonstrate bilateral strength preservation effects for the cross-education group because the nontraining control group experienced bilateral decrements at 3 months postfracture, indicative of global functional decline (Fig. 1B). The lack of improvement in fractured limb strength of the control group at 3 months after injury (just 45% of nonfractured limb strength) may suggest a heightened risk of reinjury. We propose the wrist fracture as an excellent clinical model to apply cross-education to augment existing therapy, but the results are promising for a wide range of orthopedic injuries involving persistent immobilization. For example, promising recent evidence has emerged in support of cross-education during rehabilitation from knee reconstructive surgery (29). Optimal rehabilitation strategies may involve a cross-education intervention applied during the immobilization period followed by subsequent training of both limbs after immobilization.
The Case for Stroke
Stroke presents a typical asymmetry of hemiparesis producing more affected (MA) and less affected (LA) sides (39). Common clinical efforts attempt to modify the activity in the MA limb by stimulating, training, or treating it directly. An alternate and complimentary perspective is to instead attempt to access interlimb neural circuits (37,38). This allows influencing the MA limb by using pathways from the LA side.
After stroke, weakness of leg flexor and arm extensor muscles is commonly combined with excessive antagonist muscle activity. This presents clinically as hyperexcitable reflex pathways (6, 25) and leads to reduced movement at a given joint with decreased ability to generate purposeful torques and functional movement (5,6). There is a particular dysfunction in reciprocal inhibition between functional antagonists at the wrist (1) and ankle (8).
The overall goal of neurologic rehabilitation is to restore or promote more functional movement. Resistance training improves muscular strength and functional ability in poststroke hemiparesis without deleterious results (26,30), including functional relearning to improve arm movement (9).
Unfortunately, implementation of motor retraining may be compromised for those in critical need of training; namely, those with significant poststroke hemiparesis. In this group, substantial weakness in the MA limb can prevent sufficient activation to gain a training effect (28). A protocol that could “boost” the ability to recruit targeted motor output on the MA side would be of great benefit.
Unilateral plantarflexion (20) or dorsiflexion (7) resistance training produces bilateral strength gains in neurologically intact participants. These changes in strength can be associated with changes in muscle activation and reflex excitability (7). Despite these observations, the applicability and translational implications of the cross-education effect have not been much explored in neurologic rehabilitation. As outlined above, Farthing and colleagues (12,13,24) provided an important translational context for this effect when they showed that training the opposite arm could offset muscle wasting associated with immobilization during limb casting or immobilization.
Such also is applicable in poststroke hemiparesis. In a recent study (8), 19 participants with chronic poststroke weakness performed high-intensity maximal isometric dorsiflexion resistance training (five sets of five maximal 2-s voluntary contractions three times per week for 6 wk) of the LA leg. Voluntary dorsiflexion strength in the trained (LA) leg increased by approximately 34%. In addition, the untrained MA leg showed an approximately 31% increase in strength (schematically illustrated in Fig. 2). Four participants who were unable to generate measurable dorsiflexion torque in the MA leg before training was able to meaningfully activate the untrained dorsiflexors after the intervention. Also, muscle activation was significantly increased bilaterally and reciprocal inhibition between the ankle flexor and extensor muscles was shifted in the direction of that found in the LA leg.
This study showed for the first time that significant gains in voluntary strength and muscle activation in the untrained MA leg can be achieved by training the opposite LA limb after stroke. This proof-of-principle study demonstrates the residual plasticity that exists poststroke and is a significant step toward poststroke rehabilitation, specifically in cases of severe hemiparesis, where training the MA limb is not initially possible. We suggest that the clinical sequelae presenting as hemiparesis after stroke should be viewed as physiologically complimentary to the approach taken above with limb casting in orthopedic injury. The net functional effect in both limbs (and expressed most clearly in the MA limb) is a reduction of the sublime coordination normally present and an overall weakening. With this approach, cross-education of the LA limb could be used to initially “boost” muscle activation and strength in the MA limb. In this way, the MA limb then can be functionally recruited and activated in a targeted training paradigm focused on both limbs.
THE RESTORING SYMMETRY HYPOTHESIS
As with all aspects of motor output and coordination, which make use of our intrinsic quadrupedal neural circuitry (36–38), the cross-education effect may have an evolutionary basis. Considering quadrupedal locomotion with all four limbs directly engaged in locomotor patterning and the production of propulsive force, injury to one limb would have a dramatic effect on functional ability of the animal. The remaining three intact limbs are subjected to increased loading and forced use. This state may tap into the circuitry at play in the cross-education effect to result in a more rapid recovery of function in the damaged or injured limb. As such, ongoing activation serves as the training stimulus leading to enhanced restoration of function after injury but that is completely invisible in the intact and uninjured organism.
This conceptual approach is useful in translating basic science observations related to mechanisms of quadrupedal and bipedal motor control networks and targeted plasticity to human neurologic and orthopedic rehabilitation. Stroke LA limb training to improve MA limb motor output could be used when the MA limb is too weak to be initially applied in training. Before enrollment in constraint-induced movement therapy, for example, in the case of orthopedic injury, training the uninjured limb could be used to offset decrements related to disuse atrophy and strength loss.
We advance the hypothesis here that the cross-education effect is best applied to restore function when there is an asymmetrical deficit. As argued above, two examples of such deficits are limb casting after orthopedic injury and hemiparesis after stroke. Although differing in underlying etiology and pathophysiology, both of these clinical examples present with a functional asymmetry.
In the orthopedic condition, peripheral injury and subsequent immobilization lead to neural maladaptations. In the neurologic condition, central nervous system injury causes behavioral manifestations with similarities to those observed after immobilization. Although this approach and conceptualization do not resolve the question of whether the behavioral manifestations result from an intact nervous system acting on an injured musculoskeletal system or an injured nervous system acting on an intact musculoskeletal system, the common goal is to use persistent neural mechanisms to retain or restore peripheral and behavioral function.
As indicated in Figure 3, we advance the “restoring symmetry hypothesis” to account for and apply the cross-education effect in a clinical setting. The top panel in the figure shows Da Vinci’s Vitruvian Man, long a hallmark of symmetry, with casted limbs on one side of the body. The casted limb is meant to stand in for both literal mechanical immobilization of the limb as in orthopedic injury and also for functional neurological reduced use as occurs in hemiparesis after stroke. The middle panels show the asymmetry that arises in terms of muscle girth and function because of the immobilization and how these could be attenuated by cross-education.
The bottom right panel of Figure 3 illustrates the restoration of symmetry as a result of the cross-education intervention. Clearly, this is an oversimplification and we do not mean to suggest that any intervention can restore completely to baseline the deleterious effects of injury. We believe it helpful and instructive as an illustration, though, to show clearly the effects that an asymmetric intervention could have in rehabilitation, especially because the intervention is opposite to the logical approach of training the injured or more affected limb directly.
In sum, we suggest that, as indicated in Figure 3, an asymmetrical clinical presentation could be best offset by applying a clinical intervention with an asymmetry in the same direction. This is summarized as:
- Training one side only (as in cross-education) yields a bilateral, often asymmetrical, increase in strength and
- Limb immobilization (imposed by orthopedic injury or neurological insult) gives rise to an asymmetrical decrease in strength and skill and leads to
- The Restoring Symmetry Hypothesis — an asymmetrical intervention should be applied to offset asymmetrical deficits.
We caution that this approach would not be used alone or in isolation but rather as part of a larger multidisciplinary rehabilitation program aimed at eventual full integration and bilateral training. We suggest cross-education interventions would best be used during the period after injury when one limb is rendered unable to function (temporarily or chronically) and then continued as part of treatment engaging both limbs independently once some functional ability has been restored.
Importantly, the approach to use activity related to the opposite limb to modulate output of the crossed limb may offer previously untapped avenues for rehabilitation. In this way, musculoskeletal and neurologic rehabilitation can move beyond the concept of deficit compensation and toward attempting to tap into the intrinsic biology of the nervous system to facilitate motor relearning and functional activation. Thus, we propose that, in such cases of asymmetrical activation and limb function, rehabilitative interventions that involve an asymmetrical approach (e.g., cross-education) can be usefully applied.
Dr. Jonathan Farthing’s research was supported by funding from the Royal University Hospital Foundation, Saskatchewan, and the Saskatchewan Health Research Foundation. Dr. E. Paul Zehr’s research was supported by funding from the Natural Sciences and Engineering Research Council of Canada and the Heart and Stroke Foundation, British Columbia and Yukon. The authors acknowledge Dr. Charlene Magnus and Dr. Katie Dragert for their outstanding contributions to the clinical cross-education studies in wrist fracture and stroke, respectively, which formed the basis for this review. The authors also acknowledge the very helpful comments from the reviewers that were used to carefully hone their message. The authors thank Andi Zehr for help with the artwork.
1. Artieda J, Quesada P, Obeso JA. Reciprocal inhibition between forearm muscles in spastic hemiplegia. Neurology
. 1991; 41: 286–9.
2. Brogren E, Hofer M, Petranek M, Dahlin LB, Atroshi I. Fractures of the distal radius in women aged 50 to 75 years: natural course of patient-reported outcome, wrist motion and grip strength between 1 year and 2–4 years after fracture. J. Hand Surg. (Eur)
. 2011; 36: 568–76.
3. Carroll TJ, Herbert RD, Munn J, Lee M, Gandevia SC. Contralateral effects of unilateral strength training: evidence and possible mechanisms. J. Appl. Physiol
. 2006; 101: 1514–22.
4. Davies J, Parker DF, Rutherford OM, Jones DA. Changes in strength and cross sectional area of the elbow flexors as a result of isometric strength training. Eur. J. Appl. Physiol. Occup. Physiol
. 1988; 57 (6): 667–70.
5. Dewald JP, Pope PS, Given JD, Buchanan TS, Rymer WZ. Abnormal muscle coactivation patterns during isometric torque generation at the elbow and shoulder in hemiparetic subjects. Brain
. 1995; 118: 495–510.
6. Dobkin BH. The Clinical Science of Neurologic Rehabilitation
. New York (NY): Oxford University Press; 2003.
7. Dragert K, Zehr EP. Bilateral neuromuscular plasticity from unilateral training of the ankle dorsiflexors. Exp. Brain Res
. 2011; 208: 217–27.
8. Dragert K, Zehr EP. High-intensity unilateral dorsiflexor resistance training results in bilateral neuromuscular plasticity after stroke
. Exp. Brain Res
. 2013; 225 (1): 93–104.
9. Ellis MD, Holubar BG, Acosta AM, Beer RF, Dewald JPA. Modifiability of abnormal isometric elbow and shoulder joint torque coupling after stroke
. Muscle Nerve
. 2005; 32: 170–8.
10. Farthing JP. Cross-education of strength depends on limb dominance: implications for theory and application. Exerc. Sport Sci. Rev
. 2009; 37 (4): 179–87.
11. Farthing JP, Chilibeck PD, Binsted G. Cross-education of arm muscular strength is unidirectional in right-handed individuals. Med. Sci. Sports Exerc
. 2005; 37 (9): 1594–1600.
12. Farthing JP, Krentz JR, Magnus CRA. Strength training the free limb attenuates strength loss during unilateral immobilization
. J. Appl. Physiol
. 2009; 106: 830–6.
13. Farthing JP, Krentz JR, Magnus CRA, et al.
Changes in functional magnetic resonance imaging cortical activation with cross education to an immobilized limb. Med. Sci. Sports Exerc
. 2011; 43 (8): 1394–1405.
14. Fechner G. Beabachtungen, Wiche Zu Beweisen Achenem, Dans Durch Die Usbung Der Allder Der Einem Seite Die Der Andern Zugleich Mit [Using one side at the same time as the other side]. Berlin De K-Suchs Ges D Wies Math Phys
. 1858; X: 70.
15. Földhazy Z, Törnkvist H, Elmstedt E, Andersson G, Hagsten B, Ahrengart L. Long-term outcome of nonsurgically treated distal radius fractures. J. Hand Surg
. 2007; 32A: 1374–84.
16. Handoll HHG, Madhok R, Howe TE. Rehabilitation for distal radial fractures in adults. Cochrane Database Syst. Rev
. 2006, Issue 3: CD003324.
17. Hortobágyi T, Lambert NJ, Hill JP. Greater cross education following training with muscle lengthening than shortening. Med. Sci. Sports Exerc
. 1997; 29 (1): 107–12.
18. Hortobágyi T, Richardson SP, Lomarev M, et al.
Interhemispheric plasticity in humans. Med. Sci. Sports Exerc
. 2011; 43 (7): 1188–99.
19. Housh DJ, Housh TJ, Johnson GO, Chu W. Hypertrophic response to unilateral concentric isokinetic resistance training. J. Appl. Physiol
. 1992; 73: 65–70.
20. Lagerquist O, Zehr EP, Docherty D. Increased spinal reflex excitability is not associated with neural plasticity underlying the cross-education effect. J. Appl. Physiol
. 2006; 100: 83–90.
21. Lee M, Hinder MR, Gandevia SC, Carroll TJ. The ipsilateral motor cortex contributes to cross-limb transfer of performance gains after ballistic motor practice. J. Physiol
. 2010; 588 (Pt 1): 201–12.
22. Lee M, Carroll TJ. Cross education: possible mechanisms for the contralateral effects of unilateral resistance training. Sports Med
. 2007; 37: 1–14.
23. Magnus CR, Arnold CM, Johnston G, Dal-Bello Haas V, Basran J, Krentz JR, Farthing JP. Cross-education for improving strength and mobility after distal radius fractures: a randomized controlled trial. Arch. Phys. Med. Rehabil
. 2013; 94: 1247–55.
24. Magnus CRA, Barss TS, Lanovaz JL, Farthing JP. Effects of cross-education on the muscle after a period of unilateral limb immobilization
using a shoulder sling and swathe. J. Appl. Physiol
. 2010; 109: 1887–94.
25. Mirbagheri MM, Settle K. Neuromuscular properties of different spastic human joints vary systematically. In: Engineering in Medicine and Biology Society (EMBC), 2010 Annual International Conference of the IEEE
26. Morris SL, Dodd KJ, Morris ME. Outcomes of progressive resistance strength training following stroke
: a systematic review. Clin. Rehabil
. 2004; 18: 27–39.
27. Munn J, Herbert RD, Gandevia SC. Contralateral effects of unilateral resistance training: a meta-analysis. J. Appl. Physiol
. 2004; 96: 1861–6.
28. Olsen TS. Improvement of function and motor impairment after stroke
. Neurorehabil. Neural. Repair
. 1989; 3: 187–92.
29. Papandreou M, Billis E, Papathanasiou G, Spyropoulos P, Papaioannou N. Cross-exercise on quadriceps deficit after ACL reconstruction. J. Knee Surg
. 2013; 26 (1): 51–8.
30. Patten C, Lexell J, Brown HE. Weakness and strength training in persons with poststroke hemiplegia: rationale, method, and efficacy. J. Rehabil. Res. Dev
. 2004; 41: 293–312.
31. Pearce AJ, Hendy A, Bowen WA, Kidgell DJ. Corticospinal adaptations and strength maintenance in the immobilized arm following 3 weeks unilateral strength training. Scand. J. Med. Sci. Sports
. 2012. doi: 10.1111/j.1600-0838.2012.01453.x.
32. Scripture E, Brown EM. On the education of muscular control and power. Stud. Yale Psychol. Lab
. 1894; 2: 5.
33. Stevens JE, Pathare NC, Tillman SM, et al.
Relative contributions of muscle activation and muscle size to plantar flexor torque during rehabilitation after immobilization
. J. Orthop. Res
. 2006; 24 (8): 1729–36.
34. Trumble TE, Schmitt SR, Vedder NB. Factors affecting functional outcome of displaced intra-articular distal radius fractures. J. Hand Surg
. 1994; 19A: 325–40.
35. Vandenborne K, Elliott MA, Walter GA, et al.
Longitudinal study of skeletal muscle adaptations during immobilization
and rehabilitation. Muscle Nerve
. 1998; 21 (8): 1006–12.
36. Zehr EP, Hundza SR, Vasudevan EV. The quadrupedal nature of human bipedal locomotion. Exerc. Sport Sci. Rev
. 2009; 37: 102–108.
37. Zehr EP, Loadman PM. Persistence of locomotor-related interlimb reflex networks during walking after stroke
. Clin. Neurophysiol
. 2012; 123: 796–807.
38. Zehr EP, Loadman PM, Hundza SR. Neural control of rhythmic arm cycling after stroke
. J. Neurophysiol
. 2012; 108: 891–905.
39. Zehr EP. Evidence-based risk assessment and recommendations for physical activity clearance: stroke
and spinal cord injury. Appl. Physiol. Nutr. Metab
. 2011; 36 (Suppl 1): S214–31.