Assessing Inter-limb Asymmetries: Should We Bother?

In this blog I wanted to discuss the topic of reporting inter-limb asymmetries for performance and rehabilitation. This post essentially builds on a previous article led by Thomas Dos’Santos where he outlined the potential options available for practitioners and researchers regarding the quantification, monitoring, and interpretation of interlimb asymmetries titled ‘Assessing Interlimb Asymmetries: Are We Heading in the Right Direction?’ (Dos’Santos et al., 2021). Like the article, this post will continue to review the use of inter-limb asymmetries in performance and rehabilitation and outline potential issues when interpreting such data. However, I believe that relying solely on inter-limb asymmetry data for performance and rehabilitation decisions may be misguided and even harmful in some cases.

An inter-limb asymmetry is a difference in performance of function of one limb with respect to the other (Bishop et al., 2016), such as right vs. left, dominant vs. non-dominant, or injured vs. non-injured. These differences may be categorised into the following (Hart et al., 2014; Maloney, 2019):

It is suggested asymmetries are training- and competition-history specific; thus, athletes may develop an inter-limb asymmetry in part due to handedness, previous injury, or execution of repetitive unilateral and asymmetrical movements (Graham-Smith et al., 2015). For example, prolonged practice in a unilateral skill such as soccer kicking leads to an asymmetrical profile, whereby the stance leg becomes stronger eccentrically and the kicking leg becomes stronger concentrically (Rahnama et al., 2005). Similarly, basketball athletes benefit from repeatedly performing unilateral jumping tasks, specifically taking off on one limb more frequently than the other when performing a lay-up, a drill which is commonplace in basketball training and competition. 

Why Assess Inter-limb Asymmetry, Should We Bother?

Inter-limb asymmetries have become a topic of interest over recent years, with people such as Chris Bishop leading the research and shining light on the subject. Typically, research on the subject has focused on the following areas:

1. Impact of inter-limb on athletic performance

As recently outlined, evidence trying to link asymmetry and athletic performance is questionable (Afonso et al., 2022). Such studies demonstrating associations (correlation) between asymmetry in one task and performance another does not always equate to cause and effect. Furthermore, the effects of training interventions on inter-limb asymmetry show small reductions in asymmetries compared to a control group (Bettariga et al., 2022). Lastly, whether asymmetries contribute to reductions in performance and/or increase the likelihood of injury is unknown, and highly speculative. That said, if an athlete does possess a significant neuromuscular deficit in function in one limb, such as eccentric quadriceps strength, one would assume that this deficit could transfer and mechanically impact tasks such as landing and deceleration which does involve high levels of eccentric strength for braking.

2. Screening and profiling of inter-limb asymmetry in relation to injury risk

Predictive findings for injury have been proposed as loss of >25° internal rotation, and a loss of 20° internal rotation combined with a loss of 5% in total range of movement, doubles the risk of injury in professional baseball pitchers (Herrington et al., 2018). Higher injury rates in collegiate athletes were associated with knee flexor or hip extensor imbalances of 15% (Knapik et al., 1991). Athletes who did not meet functional criteria (10-15% difference between limbs) during isokinetic strength and hop tests before returning to professional sport had a four times greater risk of sustaining an ACL graft rupture compared with those met return-to-sport criteria (Kyritsis et al., 2016).

3. Monitoring of limb function and inter-limb asymmetry during the injury rehabilitation process

Recent work has recommended rehabilitation of anterior cruciate ligament reconstruction may be complete upon reaching limb symmetry measures of >90% in both isokinetic knee-extension peak torque and single-leg CMJ height in multidirectional field-sport athletes (O’Malley et al., 2018). A recent clinical practice guideline (Kotsifaki et al., 2023) outlined >80% symmetry quadriceps strength as criteria to return to running during ACLR. Further recommendations for return-to-sport included 100% symmetry in isokinetic quadriceps and hamstring peak torque for returning to high-demand pivoting sports, while CMJ and DJ jump height and concentric and eccentric impulse should achieve >90% symmetry. Finally, restoration of >90% symmetry of vertical ground reaction force during high-speed running is recommended. Outcome measures should be interpreted with caution, as a recent study showed symmetry in triple hop distance may mask asymmetries in knee function after ACLR and may not be the most appropriate for discharge criterion during rehabilitation (Kotsifaki et al., 2020). Another source of uncertainty is the use of pre-injury values or the injured limb in some asymmetry calculations. It may be the case that pre-injury levels were sub-optimal in the first place, or towards end-stage rehabilitation the injured limb may display superior scores to the un-injured limb, complicating calculation of asymmetry.

Thresholds and Considerations

Although the validity and test-retest reliability of different testing protocols to determine inter-limb asymmetry has been examined, there are clear discrepancies in the methodologies used. Numerous studies have proposed many equations to calculate inter-limb asymmetry, revealing great disparity between inter-limb asymmetry thresholds (Bishop et al., 2016). Therefore, certain equations may elevate or reduce inter-limb asymmetry based on the numerator(s) and denominator(s) chosen. It is difficult to completely justify which equation should be used over another when quantifying inter-limb asymmetry. However, difficulties exist with values inserted into the equation; with researchers reporting using right and left, stronger and weaker, and self-reported preferred and non-preferred as the numerators. Additionally, selection of a denominator (right or left, stronger or weaker, preferred or non-preferred) will also impact the resultant inter-limb asymmetry threshold, as shown in Table 2 below. Using different values in the calculation of inter-limb asymmetry could elevate or reduce inter-limb asymmetry, potentially making it difficult for practitioners to determine the use of inter-limb asymmetry in strength assessment and rehabilitation, influencing the interpretation of an athlete’s asymmetry. Table 2 outlines some common equations used within the literature; yet it should be known that a consistent approach to calculating inter-limb differences is advised to best influence the monitoring process for both injured and non-injured athletes.

Previously, an imbalance of 10-15% has been considered a potentially problematic asymmetry. However, recent research has pointed out that using an arbitrary threshold is problematic due to interlimb asymmetry being task- and metric-specific (Bishop, 2021), with specific equations as highlighted in the Table 2 also inflating asymmetry imbalance measures. Figure 1 shows asymmetry data for two soccer players across strength, power and COD metrics. Not only is the asymmetry imbalance different to the arbitrary threshold of 10-15%, but it is different depending on the test (strength quality) being assessed. 

Furthermore, Figure 2 shows asymmetry data for Player 2 for five variables in a CMJ, with the variation plotted as a dark blue line. This demonstrates the metric-specific nature of asymmetry; thus, any threshold should be interpreted with caution. It may be the case that the asymmetry value of any metric should be interpreted in comparison to its variation (coefficient of variation [CV%]) to determine whether a “real” asymmetry exists (Exell et al., 2012). Time to takeoff and RSImod are exhibiting asymmetries greater than the CV%, whereas peak force, countermovement depth and jump height asymmetries are within the variation of each measure. It has been suggested that an “asymmetry might only be meaningful if the percentage value is greater than the test variability score” (Bishop et al., 2020b; Exell et al., 2012). In our example, despite peak force, countermovement depth and jump height displaying variability close to the actual asymmetry value, it may not actually be problematic in terms of athletic performance and risk of any injury. Due to the asymmetry being so low (<5%) across these variables, it may be questionable to use such an approach when interpreting an individual’s inter-limb asymmetry. On the contrary, as time to take-off (TTT) and RSImod are displaying inter-limb asymmetries of 30 and 25%, respectively, the reliability and thus, use-ability of these metrics may be questioned. Here poses the question if the variability in each measure is too high, practitioners should be weary of using this data to inform decisions regarding monitoring and rehabilitation. 

Figure 3 shows an example of 505 CoD time asymmetry over three trials within the same testing session for 16 soccer players. Values above 0 indicate 505 times faster turning on the right limb and below 0 faster when turning on the left limb. The direction of asymmetry is consistent for some athletes, yet the direction of asymmetry varies for athletes 8, 10, 14, 15 and 16. Given the inherent variability of asymmetry (Bishop et al., 2020a, 2020b), this raises the question of how to interpret such data within a single session. This is also consistent with that exploring the consistency of asymmetry favouring the same limb between separate test sessions, finding direction of asymmetry to be just as variable as the magnitude of asymmetry in healthy athletes (Bishop et al., 2020a). These findings suggest that an individual approach should be adopted when monitoring inter-limb asymmetries. Therefore, it is advised for practitioners to determine the reliability of inter-limb asymmetries between sessions to ensure that the magnitude (expressed as a percentage difference) and the direction (which limb shows greater performance or function) remains consistent before categorising athletes as having asymmetries. To obtain a comprehensive assessment of reliability, a combination of statistical tests, such as paired-samples t-tests or nonparametric equivalents, within-subject variation measures such as the CV%), typical error, or standard error of measurement, and retest correlation measures like intraclass correlation coefficient should be employed. For further reading on assessing reliability in sport science, readers are directed to the excellent work of Atkinson and Nevill (Atkinson and Nevill, 1998).

The authors acknowledge the need for an individualised approach to monitoring inter-limb asymmetries and other variables in strength and conditioning. However, it is challenging to apply this approach without benchmark data related to the specific metric, task, and population being studied. While recent work (Bishop, 2021) has proposed inter-limb asymmetry thresholds, they fail to discuss existing thresholds used in literature (Aldukhail et al., 2013; Dos’Santos et al., 2018, 2017; Graham-Smith et al., 2015; Lockie et al., 2014, 2013). These include thresholds based on population mean + smallest worthwhile change (SWC = 0.2 × between subject SD), mean + (0.5 × between-subject SD), or population mean + (1.0 × between-subject SD). Assuming a normal distribution of the metric, these thresholds could identify approximately 42%, 31%, and 16% of athletes as having “small to moderate” or “high or extreme” asymmetries. These benchmarks have been used to assess inter-limb asymmetry in athletic populations (Aldukhail et al., 2013; Dos’Santos et al., 2018, 2017; Graham-Smith et al., 2015; Thomas et al., 2017), highlighting the task- and metric-dependent nature of asymmetry.

While acknowledging that it is not a perfect solution, using the mean + SWC and mean + SD for asymmetry thresholds enables practitioners to establish specific benchmarks, criteria, and normative data for their athlete population across various metrics and tests. These thresholds allow practitioners to classify athletes as having “small to moderate” or “extreme or high” asymmetries. The mean + SWC threshold is a more sensitive approach, identifying a greater proportion of athletes (~42%) with “small to moderate” inter-limb asymmetries, while the mean + (1.0 × between-subject SD) threshold is a more conservative approach, identifying a smaller proportion (~16%) of athletes with “high or extreme” inter-limb asymmetries. It is important to note that neither threshold is superior, and both should be used together to provide descriptors for the magnitude of the asymmetry. This will aid in the interpretation and classification of inter-limb asymmetry. The concept of interpreting data relative to the population mean and using the SWC to monitor changes in fitness and training load data is common practice in sports science and strength and conditioning (McGuigan, 2017). Therefore, athletes who display inter-limb asymmetries greater than the population mean + SWC and their individual variability (CV%) can potentially be interpreted as showing “meaningful” or “greater” asymmetry in the context of the population and metric for a specific test.

Recently, a study (Bishop, 2021) has proposed methods to interpret inter-limb asymmetry values in relation to an individual’s variability (Exell et al., 2012). Although this is helpful, the authors suggest that including asymmetry threshold lines (mean + SWC and mean + SD) is also useful for interpreting asymmetries specific to a population during a particular test. For instance, Figure 4 illustrates inter-limb asymmetry values for time to completion during 505 testing, CV% values, and asymmetry thresholds calculated on the population mean + SWC and mean + SD. The asymmetry thresholds of 1.3% (small to moderate) and 4.7% (high/extreme) were specific to this population and metric during this test. Out of the 9 athletes who displayed asymmetries greater than their individual CV% and exceeded the small-to-moderate asymmetry threshold, only 4 exceeded the high-asymmetry threshold. It may be the case that a more conservative approach in diagnosing inter-limb asymmetries could be used going forward to classify a smaller number of athletes.

The inter-limb asymmetry percentage is a simple way to indicate differences in performance between an athlete’s right/left and dominant / non-dominant limbs. However, there are some issues with this approach. For instance, the percentage figure can be misleading if the denominator changes. This means that the raw score may be a more accurate method to use when assessing an athlete’s physical profile. Coaches and practitioners should also consider the absolute components, such as the numerator and denominator, when interpreting the athlete’s performance (Bishop et al., 2023). An athlete may display asymmetry, but their weaker-performing limb may still outperform the rest of the team or exceed benchmark data. Similarly, an athlete may be symmetrical but have poor function or performance, which would warrant greater attention and support from strength and conditioning practitioners. As shown in Figure 5, Player 1 exhibited close to 0% asymmetry in 505 CoD time with completion times of 2.95 and 2.94 seconds when turning of the left and right limbs, respectively. Although Player 2 demonstrated greater asymmetry (9.8% – favouring the right limb), completion times were much faster on both left (2.8 seconds) and right (2.55 seconds) limbs.

“These findings suggest that an individual approach should be adopted when monitoring inter-limb asymmetries. Therefore, it is advised for practitioners to determine the reliability of inter-limb asymmetries between sessions to ensure that the magnitude (expressed as a percentage difference) and the direction (which limb shows greater performance or function) remains consistent before categorising athletes as having asymmetries.”

THOMAS (2023)

It is crucial for coaches to interpret the absolute components of an athlete’s physical profile, as they can indicate whether the athlete is weak or has poor function. In the above example, we have one athlete who is symmetrical but slow, and one who is potentially asymmetrical but fast. If we were to improve Athlete 1 in their COD ability, would they develop an inter-limb asymmetry? If we were to try to reduce inter-limb asymmetries in Athlete 2, would they still maintain high COD performance? Arguably, Athlete 1 (symmetrical but poor function) warrants greater attention and strength and conditioning support. Figure 6 illustrates four quadrants to assist in the physical profiling of athletes (Dos’Santos et al., 2021). The optimal scenario for practitioners is athletes who are equally strong and displaying high function and performance from both limbs, which should be viewed as the aim for strength and conditioning and physiotherapist practitioners (Figure 6 — top right). Conversely, the worst case and “red flag” scenario for practitioners is identifying a weak or low function and asymmetrical athlete (Figure 6 — bottom left). In an ideal scenario coaches should aim for athletes who are equally proficient and display high function and performance from both limbs. It is important to note that an athlete’s magnitude of asymmetry may not change over time, but the absolute components may increase. This can still be viewed as positive adaptation as the athlete has improved their function or performance in both limbs proportionately.

On the other hand, athletes may maintain similar asymmetry values over time, but the absolute components may have reduced proportionately. This is problematic as it can affect athletic performance and potentially increase the risk of injury. Injury rehabilitation monitoring should also consider inter-limb asymmetries (Bishop et al., 2020b; Kotsifaki et al., 2023). Athletes may display reductions in inter-limb asymmetries during rehabilitation due to reductions in strength and performance of the contralateral limb. This is problematic, especially in the context of ACL injury, as the contralateral limb is typically at greater risk of injury after an ACL injury (King et al., 2021a, 2021b). Therefore, coaches and practitioners should inspect the raw values when monitoring inter-limb asymmetries to ensure that athletes are not reducing their imbalance through reductions in dominant limb strength, performance, or function while maintaining strength, performance, or function in the non-dominant limb.

Summary

The impact of inter-limb asymmetries on performance and injury risk is a debatable topic, but athletes generally benefit from having equally strong and functional limbs. Asymmetry presents an opportunity for improvement and may require more attention to improve the weaker limb. However, it’s crucial to remember that the stronger limb should not be overlooked, and practitioners should still aim to enhance performance and function in both limbs. Consequently, to improve inter-limb asymmetry profiling and research, the following points should be considered:

• Asymmetry thresholds should be created based on population mean + SWC (0.2 × between-subject SD) and population mean + (1.0 × between-subject SD) to classify athletes with “small to moderate” and “extreme or high” asymmetries. The mean + SWC threshold is more sensitive, while the mean + SD threshold is more conservative.
• Practitioners can use the thresholds to create benchmarks, criteria, and normative data specific to their athlete population for a range of metrics and tests.
• Athletes who display asymmetries exceeding their individual CV% and population asymmetry can be classified as displaying “small to moderate” or “high or extreme” asymmetries.
• Practitioners should establish between-session reliability for inter-limb asymmetry metrics and use a combination of statistical tests for a holistic overview of reliability.
• Practitioners should evaluate the absolute components and percentage value of the inter-limb asymmetry metric to assist in physical profiling and training prescription.

References

Afonso, J., Peña, J., Sá, M., Virgile, A., García-de-Alcaraz, A., Bishop, C., 2022. Why sports should embrace bilateral asymmetry: A narrative review. Symmetry 14, 1993.

Aldukhail, A., Jones, P., Gillard, H., Graham-Smith, P., 2013. Clinical diagnosis of strength and power asymmetry. Biol. Sport 15, 33–38.

Atkinson, G., Nevill, A.M., 1998. Statistical methods for assessing measurement error (reliability) in variables relevant to sports medicine. Sports Med. 26, 217–238.

Bettariga, F., Turner, A., Maloney, S., Maestroni, L., Jarvis, P., Bishop, C., 2022. The effects of training interventions on interlimb asymmetries: a systematic review with meta-analysis. Strength Cond. J. 44, 69–86.

Bishop, C., 2021. Interlimb asymmetries: Are thresholds a usable concept? Strength Cond. J. 43, 32–36.

Bishop, C., Read, P., Chavda, S., Jarvis, P., Brazier, J., Bromley, T., Turner, A., 2020a. Magnitude or Direction? Seasonal Variation of Interlimb Asymmetry in Elite Academy Soccer Players. J. Strength Cond. Res.

Bishop, C., Read, P., Chavda, S., Turner, A., 2016. Asymmetries of the Lower Limb: The Calculation Conundrum in Strength Training and Conditioning. Strength Cond. J. 38, 27–32.

Bishop, C., Shrier, I., Jordan, M., 2023. Ratio data: Understanding pitfalls and knowing when to standardise, Symmetry. MDPI.

Bishop, C., Turner, A.N., Gonzalo-Skok, O., Read, P., 2020b. Inter-limb asymmetry during rehabilitation understanding formulas and monitoring the” magnitude” and” direction”. Aspetar Sports Med. J. 9, 18–22.

Dos’Santos, T., Thomas, C., Jones, P.A., 2021. Assessing interlimb asymmetries: Are we heading in the right direction? Strength Cond. J. 43, 91–100.

Dos’Santos, T., Thomas, C., Jones, P.A., Comfort, P., 2018. Asymmetries In Isometric Force-Time Charcteristics Are Not Detrimental To Change Of Direction Speed. J. Strength Cond. Res. 32, 520–527. https://doi.org/10.1519/JSC.0000000000002327

Dos’Santos, T., Thomas, C., Jones, P.A., Comfort, P., 2017. Asymmetries in single and triple hop are not detrimental to change of direction speed. J. Trainology 6, 35–41.

Exell, T.A., Irwin, G., Gittoes, M.J., Kerwin, D.G., 2012. Implications of intra-limb variability on asymmetry analyses. J. Sports Sci. 30, 403–409.

Graham-Smith, P., Al-Dukhail, A., Jones, P., 2015. Agreement between attributes associated with bilateral jump asymmetry. ISBS-Conf. Proc. Arch. 33.

Hart, N.H., Nimphius, S., Spiteri, T., Newton, R.U., 2014. Leg strength and lean mass symmetry influences kicking performance in Australian Football. J. Sports Sci. Med. 13, 157.

Herrington, L.C., Munro, A.G., Jones, P.A., 2018. Assessment of factors associated with injury risk, in: Performance Assessment in Strength and Conditioning. Routledge, pp. 53–95.

King, E., Richter, C., Daniels, K.A., Franklyn-Miller, A., Falvey, E., Myer, G.D., Jackson, M., Moran, R., Strike, S., 2021a. Biomechanical but not strength or performance measures differentiate male athletes who experience ACL reinjury on return to level 1 sports. Am. J. Sports Med. 49, 918–927.

King, E., Richter, C., Daniels, K.A., Franklyn-Miller, A., Falvey, E., Myer, G.D., Jackson, M., Moran, R., Strike, S., 2021b. Can biomechanical testing after anterior cruciate ligament reconstruction identify athletes at risk for subsequent ACL injury to the contralateral uninjured limb? Am. J. Sports Med. 49, 609–619.

Knapik, J.J., Bauman, C.L., Jones, B.H., Harris, J.M., Vaughan, L., 1991. Preseason strength and flexibility imbalances associated with athletic injuries in female collegiate athletes. Am. J. Sports Med. 19, 76–81.

Kotsifaki, A., Korakakis, V., Whiteley, R., Van Rossom, S., Jonkers, I., 2020. Measuring only hop distance during single leg hop testing is insufficient to detect deficits in knee function after ACL reconstruction: a systematic review and meta-analysis. Br. J. Sports Med. 54, 139–153.

Kotsifaki, R., Korakakis, V., King, E., Barbosa, O., Maree, D., Pantouveris, M., Bjerregaard, A., Luomajoki, J., Wilhelmsen, J., Whiteley, R., 2023. Aspetar clinical practice guideline on rehabilitation after anterior cruciate ligament reconstruction. Br. J. Sports Med.

Kyritsis, P., Bahr, R., Landreau, P., Miladi, R., Witvrouw, E., 2016. Likelihood of ACL graft rupture: not meeting six clinical discharge criteria before return to sport is associated with a four times greater risk of rupture. Br. J. Sports Med. 50, 946–951.

Lockie, R.G., Callaghan, S.J., Berry, S.P., Cooke, E.R.A., Jordan, C.A., Luczo, T.M., Jeffriess, M.D., 2014. Relationship Between Unilateral Jumping Ability and Asymmetry on Multidirectional Speed in Team-Sport Athletes. J. Strength Cond. Res. 28, 3557–3566.

Lockie, R.G., Schultz, A.B., Callaghan, S.J., Jeffriess, M.D., 2013. The effects of isokinetic knee extensor and flexor strength on dynamic stability as measured by functional reaching. Isokinet. Exerc. Sci. 21, 301–309.

Maloney, S.J., 2019. The relationship between asymmetry and athletic performance: A critical review. J. Strength Cond. Res. 33, 2579–2593.

McGuigan, M., 2017. Monitoring training and performance in athletes. Human Kinetics.

O’Malley, E., Richter, C., King, E., Strike, S., Moran, K., Franklyn-Miller, A., Moran, R., 2018. Countermovement jump and isokinetic dynamometry as measures of rehabilitation status after anterior cruciate ligament reconstruction. J. Athl. Train. 53, 687–695.

Rahnama, N., Lees, A., Bambaecichi, E., 2005. A comparison of muscle strength and flexibility between the preferred and non-preferred leg in English soccer players. Ergonomics 48, 1568–1575.

Thomas, C., Comfort, P., Dos’Santos, T., Jones, P.A., 2017. Determining Bilateral Strength Imbalances in Youth Basketball Athletes. Int. J. Sports Med. 38, 683–690. https://doi.org/10.1055/s-0043-112340