The mechanisms underlying the so-called crossover effect—when a unilateral intervention results in bilateral changes—are still unclear, but clinical applications related to lower extremity strengthening, fatigue, and stretching are already being explored by rehabilitation specialists.
By Cary Groner
When LER reported last year on the effects of foam rolling, we noted that even if only one leg was rolled, the other experienced a similar decrease in soreness.1 That a unilateral intervention should have bilateral effects hinted at a neural component to the procedure’s efficacy that wasn’t yet fully understood.
This phenomenon—often called the crossover effect—was first described in the literature more than a hundred years ago and has intrigued researchers and clinicians ever since. The exploration and development of practical applications has begun only relatively recently, but a growing body of research into the neurological basis of crossover effects is shedding light on aspects that include muscle strengthening, fatigue, and stretching. And, though researchers are still elucidating the underlying mechanisms, clinical applications are already being suggested, particularly in situations requiring rehabilitation for unilateral conditions such as a broken leg or poststroke hemiplegia.
The need for effective interventions in such cases is clear, as even relatively brief periods of immobilization can have devastating consequences for the lower extremities. For example, one study reported that immobilization of the leg resulted in a 47% decrease in quadriceps maximum voluntary contraction (MVC) after just three weeks.2
Yale University researchers first reported in 1894 that unilateral strength training of a single limb increased strength in the untrained contralateral limb.3 Now, 122 years later, experts agree that neural adaptations are at the core of the phenomenon, and that cortical, spinal, and peripheral mechanisms are all likely involved.4 Researchers have found crossover effects associated with training protocols that include isometric, dynamic, electrically stimulated, and even imagined muscle contractions,4 and eccentric contractions reportedly produce triple the contralateral strength gains than do concentric or isometric ones.5
Although strength transfer appears more predictable with some approaches than others, the degree of strength gained in the untrained limb is usually proportional to that gained in the trained limb. Training variations such as speed, intensity, and contraction type contribute to the wide range of strength transfer reported,4 though a 2006 meta-analysis found that the average reported strength gain in the untrained limb was 7.6%, a number that, according to some experts, reflects roughly a third of the gains in the trained limb.6
Crossover effects likely arise out of the neuromuscular system’s propensity for symmetry. Human motor systems operate within a range of normal lateralization of function, and mechanical or neurological injuries disturb this homeostasis.7 The goal of therapeutic interventions, then, is to take advantage of this natural tendency to restore symmetry to the greatest degree possible. In mechanical injuries, actual symmetry is a reasonable goal; in the case of a stroke, however, this may simply mean restoring whatever function the patient can achieve in the affected limb.7
“The interconnectivity of the body is phenomenal,” said David Behm, PhD, a university research professor in the School of Human Kinetics and Recreation at the Memorial University of Newfoundland (MUN) in St. John’s, Canada. “Whether you’re looking at muscle fatigue, resistance training, pain, or stretching, they’re all interconnected, and what we do on one side of the body influences the rest of it. This can be beneficial, in terms of trying to open up and excite new pathways on both sides of the body; it can also be troublesome in that it can inhibit pathways on both sides.”
Researchers at Deakin University in Melbourne, Australia—including Ashlee Hendy, PhD, a lecturer in motor learning in the School of Exercise and Nutrition Sciences—are among the leaders in studying crossover effects, particularly regarding strength training. In a 2012 review article, for example, Hendy and her colleagues analyzed the state of knowledge about the phenomenon and reported that muscular mechanisms likely don’t play a major role, given that research to date hasn’t identified significant peripheral muscle adaptations in the untrained limb.4 At the neurological level, moreover, spinal mechanisms may be less important than cortical ones, though the relative degree of involvement appears partly dependent on whether upper or lower extremities are evaluated.
At the cortical level, Hendy reported, complex interhemispheric connections and ipsilateral corticospinal fibers from the primary motor cortex provide pathways for stimulation of the inactive opposite muscle during unilateral contraction. This corticospinal activity, referred to as “motor irradiation,” seems to contribute to strength transfer following unilateral training.
The Deakin team employs a couple of particularly helpful technologies, Hendy told LER. “We apply anodal transcranial direct current stimulation [a-tDCS] to the brain during training to try to enhance the magnitude of crosstransfer by making the neurons in the untrained area of the brain more excitable,” she said. “We also utilize transcranial magnetic stimulation [TMS] to measure physiologically whether the a-tDCS has had an effect.”
In one study, researchers used TMS to demonstrate that corticospinal mechanisms underpinned the maintenance of strength and muscle thickness in an immobilized limb after unilateral training of the free limb.8 In another, participants undertook leg-press strength training for eight weeks; the trained leg showed a strength increase of 29% while the untrained one increased 20.4%.9
The lead author of the second paper, Chris Latella, a doctoral candidate at Deakin, said the team trained the dominant leg because they’d begun to notice that strength transfer seemed to work better from dominant to nondominant limbs, though this hasn’t yet been verified or fully explained. In addition to the strength gains, the team reported a decrease in corticospinal inhibition in both legs (by 17.3 ms for the trained leg and 20.8 ms for the untrained leg).
“I’ve seen two related nervous-system effects with this kind of training—both increased excitability and decreased inhibition,” Latella said. The former is the readiness of the neuronal pathway to send excitatory stimulation to the muscle, causing a contraction; the latter is the degree of the pause between such signals, which reduces the overall neural drive to the muscle.
Deakin researchers have also noticed that increasing the demands of a training regimen seems to increase the crossover effect.10
“We believe that higher-load and more cognitively demanding strength training, such as using a metronome to control movement, facilitates the magnitude of transfer,” Hendy said. “It’s more demanding on the nervous system, both in motor output to control the movement and with afferent feedback, particularly when your muscles are lengthening. So we’re trying to make the movement as skillful as possible. It’s all about learning the activation, learning to relax your antagonists, and applying force using motor unit recruitment in the optimal manner.”
The researchers continue to refine their knowledge of what works and what doesn’t. For example, in a 2015 paper, Hendy reported that crossover strength gains in the untrained limb were more likely to be maintained at 48 hours when strength training was combined with a-tDCS, versus strength training with sham a-tDCS—a finding with implications for improving rehab outcomes after unilateral injury.11 An interesting aspect of the study was that the authors used a-tDCS to stimulate the side of the brain associated with the untrained limb.
“To get a crossover effect, you want to maximize activity in that hemisphere,” Hendy explained.
She acknowledged that much of the Deakin research has been conducted in the upper limbs, and that extrapolating their conclusions to the lower extremities should be done with caution. Even so, she thinks the potential for crossover effects in the legs might be even greater than in the arms.
“There may be more skill-based learning effects because we have the ability to apply complex, multijoint movements in the legs,” she said. “I also think those effects might be facilitated more at the spinal level. In any case, I see no harm in utilizing these types of training to try to maintain neural connections when someone is partially immobilized. I can’t think of any potential detrimental effects other than exacerbating the bilateral difference.”
Studies in the lower extremities bolster Hendy’s position. For example, in 2009, Norwegian researchers concluded that enhanced neural drive to the contralateral agonist muscles contributed to cross-education of soleus strength over a four-week training regimen.12
In 2015, Italian investigators reported in Gait & Posture that crossover effects occurred in ankle dorsiflexors, both for peak torque and muscle work, and in fact the untrained leg gained more than the trained leg, a counterintuitive finding yet to be explained.13 (A similar paper by the same group, also published last year, found the gains were similar in both trained and untrained legs.14) Finally, echoing one of Hendy’s conclusions, researchers at the University of Central Florida in Orlando reported this year that four weeks of unilateral strength training resulted in increases in strength and size of the trained muscles, but crossover of only leg-press strength—not muscle size, activation, or hormonal response—to those on the untrained side.15
The flip side of strengthening muscles is fatiguing them, and David Behm of MUN, as well as other researchers, have investigated crossover fatigue effects extensively. According to Behm, such effects are typically easier to measure in the lower body than in the upper body.
“We have reflexes that allow us to walk automatically, without thinking about it,” he said. “There are more interconnections in the lower body, which facilitates the crossover effect.”
The research about crossover fatigue has been equivocal, however; some studies support the idea and some don’t.16 One of Behm’s coauthors, Jalal Aboodarda, PhD, a MUN colleague who will begin postdoc work in kinesiology at the University of Calgary in Canada this fall, told LER that several investigations currently under peer review have shown no such effect.
“I believe the acute deterioration of muscle performance, which has been observed in only fifty percent of studies, does not necessarily mean that unilateral fatiguing exercise provides longitudinal inhibitory neuro- physiological effects,” Aboodarda said.
In a 2014 study coauthored by Behm and Aboodarda,16 a unilateral fatigue protocol of knee extension exercises showed moderate effects on the contralateral limb (23.7% to 34.6% decreases in MVC in the first 100 ms); by contrast, a 2015 paper by Behm and other colleagues reported no significant changes in force production in the nonfatigued limb.17 For that matter, Aboodarda, Behm, and colleagues published a paper the same year showing that upper-body fatigue, achieved through elbow-flexion exercises, affected electromyographic responses in the knee extensors. However, this did not decrease their strength, supporting the idea of a centrally mediated fatigue mechanism, the overall effects of which remain unclear.18
Other investigators have reported, for example, that a unilateral fatigue protocol induces crossover fatigue during single-leg landings;19 that fatigue-induced anticipatory postural adjustments occur in both fatigued and nonfatigued muscles;20 that two bouts of fatiguing exercise rather than one were needed to produce a crossover effect;21 and that crossover fatigue affected postural control after both stimulated and voluntary contractions.22
What to make of it all?
“The neurons that control whether your muscles are going to activate or not are interconnected with a lot of synapses,” Behm said. “Some are excitatory, but some are inhibitory, so that if you’re getting really tired the system will help prevent you from having a catastrophic event. Bilateral effects make sense because the body would want to slow you down on both sides. This may have to do with the brain; if it’s been pushing your left leg for the last three minutes and then you try to push the right leg, the brain’s ability to concentrate may have been diminished, and if you have less ability to concentrate you’re more likely to have crossover fatigue.”
In terms of practical applications, Behm said that athletes may want to allow time between training different parts of the body.
“If I’m doing an upper-body training program, then going for a run, the run is probably going to be affected by the upper-body workout,” he said. “Nonlocal effects are more apparent with prolonged, high-intensity contractions, so if you want to minimize those effects, you shouldn’t push yourself to total fatigue and failure.”
Israel Halperin, MSc, a colleague of Behm and Aboodarda’s at MUN and coauthor of several of their papers, is now completing his PhD as a scholar at Edith Cowan University in Canberra, Australia, in conjunction with the Australian Institute of Sports. In a recent literature review, he concluded fatigue crossover effects are more prevalent in the lower limb muscles studied (mainly quadriceps) than in the upper (mainly elbow flexors)—76% of outcome measures versus 32%, respectively—and appear dependent on the muscle group studied.23 Moreover, nonlocal fatigue effects appear to be associated with four different but connected pathways: neurological, biochemical, biomechanical, and psychological.
In a 2014 paper, Halperin and his MUN colleagues reported nonlocal fatigue effects in the knee extensors regardless of whether the opposite knee or the elbow flexors were exercised, but also found the elbows weren’t similarly affected by first working the knee.24 Perhaps inevitably, other papers by the team have shown elbow flexors are affected by knee extensor fatigue,25,26 but Halperin stands by his conclusion that the greater effects remain in the legs. In any case, he echoed Behm’s perspective about practical applications of the research.
“Given that the legs are most susceptible to nonlocal muscle fatigue, that should affect exercise programs,” Halperin said. “Lower body exercise should come before upper body exercise to minimize the fatigue-related crossover effects.”
Researchers and clinicians are investigating the potential utility of crossover effects in patients with neurological deficits, as well.
In a 2016 proof-of-concept study in multiple sclerosis (MS) patients, for example, Italian researchers reported crossover strengthening effects in the ankle dorsiflexors.27 Increasing attention is being paid to the potential for treating the weakness associated with poststroke hemiparesis, as well.
“Once you’ve had a stroke, the system tries to find alternative pathways to allow you to function,” said David Behm. “So if you can work and coordinate the good side, it should help get the other side to improve those pathways.”
Research is beginning to support this view. For example, in a 2013 paper from the University of Victoria in Canada, investigators reported unilateral high-intensity resistance training of the dorsiflexors on the less-affected side led to significant gains in muscle activation contralaterally. This included four patients who were unable to generate any dorsiflexion before the intervention and could do so afterward, demonstrating residual neuroplasticity years post-stroke.28 In a related paper,7 coauthor E. Paul Zehr, PhD, professor and director of the Rehabilitation Neuroscience Laboratory at the University of Victoria, wrote that “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.”
“Of course, the best way to train something is directly,” Zehr told LER. “But we’ve had people in our studies who couldn’t produce measurable force in the [paretic] limb at the beginning. Then, after five weeks of contralateral training, they can activate that limb and start to train it. To see someone move their ankle even a little when they couldn’t do it before is amazing, and the participants themselves get very excited. But I don’t think of this approach as existing on its own; it’s got to be part of a bigger picture of therapeutic and rehabilitative intervention.”
As researchers and clinicians refine their understanding of the neurological underpinnings of crossover effects, a variety of patients stand to gain. Those needing rehabilitation after injuries or stroke, or those with other conditions such as MS, may find new, more efficient pathways to strength and independence. After all, as a clinician once told a friend of mine: “The whole body’s connected: Take a look!”
Cary Groner is a freelance writer in the San Francisco Bay Area.
- Groner C. The mechanistic mysteries of foam rolling. LER 2015;7(10):20-25.
- Hortobagyi T, Dempsey L, Fraser D, et al. Changes in muscle strength, muscle fibre size and myofibrillar gene expression after immobilization and retraining in humans. J Physiol 2000;524(1):293-304.
- Scripture EW, Smith TL, Brown EM. On the education of muscular control and power. Studies Yale Psychol Lab 1894;2:114-119.
- Hendy AM, Spittle M, Kidgell DJ. Cross education and immobilization: mechanisms and implications for injury rehabilitation. J Sci Med Sport 2012;15(2):94-101.
- Hortobagyi T, Lambert NJ, Hill JP. Greater cross education following training with muscle lengthening than shortening. Med Sci Sports Exerc 1997;29(1):107-112.
- Carroll TJ, Herbert RD, Munn J, et al. Contralateral effects of unilateral strength training: evidence and possible mechanisms. J Appl Physiol 2006;101(5):1514-1522.
- Farthing JP, Zehr EP. Restoring symmetry: clinical applications of cross-education. Exerc Sport Sci Rev 2014;42(2):70-75.
- Pearce AJ, Hendy A, Bowen WA, et al. Corticospinal adaptations and strength maintenance and the immobilized arm following three weeks unilateral strength training. Scand J Med Sci Sports 2013;23(6):740-748.
- Latella C, Kidgell DJ, Pearce AJ. Reduction in corticospinal inhibition in the trained and untrained limb following unilateral leg strength training. Eur J Appl Physiol 2012;112(8):3097-3107.
- Leung M, Rantalainen T, Teo WP, Kidgell D. Motor cortex excitability is not differentially modulated following skill and strength training. Neuroscience 2015;305:99-108.
- Hendy AM, Teo WP, Kidgell DJ. Anodal transcranial direct current stimulation prolongs the cross-education of strength and corticomotor plasticity. Med Sci Sports Exerc 2015;47(9):1788-1797.
- Fimland MS, Helgerud J, Solstad GM, et al. Neural adaptations underlying cross-education after unilateral strength training. Eur J Appl Physiol 2009;107(6):723-730.
- Manca A, Pisanu F, Ortu E, et al. A comprehensive assessment of the cross-training effective ankle torso flexors of healthy subjects: a randomized controlled study. Gait Posture 2015;42(1):1-6.
- Manca A, Solinas G, Dragone D, et al. Characterization of ankle dorsiflexors performance in healthy subjects following maximal-intensity isokinetic resistance training. J Electromyogr Kinesiol 2015;25(5):773-781.
- Beyer KS, Fukuda DH, Boone CH, et al. Short-term unilateral resistance training results in cross education of strength without changes in muscle size, activation, or endocrine response. J Strength Cond Res 2016;30(5):1213-23.
- Kawamoto JE, Aboodarda SJ, Behm DG. Effect of differing intensities of fatiguing dynamic contractions on contralateral homologous muscle performance. J Sport Sci Med 2014;13(4):836-845.
- Arora S, Budden S, Byrne JM, Behm DG. Effect of unilateral knee extensor fatigue on force and balance of the contralateral limb. Eur J Apl Physiol 2015;115(10):2177-2187.
- Aboodarda SJ et al. Elbow flexor fatigue modulates central excitability of the knee extensors. Apppl Physiol Nutr Metab 2015;40(9):1-7.
- McLean SG, Samorezov JE. Fatigue-induced ACL injury risk stems from a degradation in central control. Med Sci Sports Exerc 2009;41(8):1661-1672.
- Strang AJ, Berg WP, Hieronymus M. Fatigue-induced early onset of anticipatory postural adjustments in non-fatigued muscles: support for a centrally mediated adaptation. Exp Brain Res 2009;197(3):245-254.
- Doix AC, Lefevre F, Colson SS. Time course of the crossover effect of fatigue on the contralateral muscle after unilateral exercise. PLOS One 2013;8(5):e64910.
- Paillard T, Chaubet V, Maitre J, et al. Disturbance of contralateral unipedal postural control after stimulated and voluntary contractions of the ipsilateral limb. Neurosci Res 2010;68(4):301-306.
- Halperin I, Chapman DW, Behm DG. Non-local muscle fatigue: effects and possible mechanisms. Eur J Appl Physiol 2015;115(10):2031-2048.
- Halperin I, Copithorne D, Behm DG. Unilateral isometric muscle fatigue decreases force production and activation of contralateral knee extensors but not elbow flexors. Appl Physiol Nutr Metab 2014;39(12):1338-1344.
- Halperin I, Aboodarda SJ, Behm DG. Knee extension fatigue attenuates repeated force production of the elbow flexors. Eur J Sport Sci 2014;14(8):823-829.
- Sambaher N, Aboodarda SJ, Behm DG. Bilateral knee extensor fatigue modulates force and responsiveness of the corticospinal pathway in the non-fatigued, dominant elbow flexors. Front Hum Neurosci 2016;10:18.
- Manca A, Cabboi MP, Ortu E, et al. Effect of contralateral strength training on muscle weakness in people with multiple sclerosis: proof-of-concept case series. Phys Ther 2016;96(6):828-838.
- 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.