Lateral Asymmetries and Their Predictive Ability for Maximal Incremental Cycle Ergometer Performance in Road Cyclists

Abstract

This study aimed to (1) determine and compare the magnitude and direction of asymmetry in lower limbs neuromuscular properties, range of motion, strength and muscle electrical activity (EMG) in well-trained male road cyclist across categories (elite, under-23 and junior); (2) establish test- and age-specific asymmetry thresholds for these variables to enable individualized classification; and (3) examine the relationship between these lateral asymmetries and performance in a maximal incremental cycle ergometer test. Fifty-five well-trained road cyclists were assessed through tensiomyography (TMG), active knee extension test (AKE), leg press and EMG of vastus lateralis (VL-EMG) during a maximal incremental cycling test. Junior cyclists showed lower asymmetry in VM than elite cyclists, but greater asymmetry in AKE. No significant differences were found in strength or VL-EMG during the maximal incremental cycle ergometer test. The magnitude and direction of lateral asymmetry differs between tests (TMG: 11.3–21.3%; AKE: 2.3%; leg-press: 9.8–31.9%; VL-EMG: 20.8–22.7%). Multiple linear regression revealed a significant predictive model for maximal incremental cycling ergometer performance based on lateral asymmetry in AKE, leg press and VL and rectus femoris contraction time (R²_a = 0.23). These reference data can support trainers in monitoring and managing lateral asymmetry throughout the cyclists’ season.

1. Introduction

Road cycling performance is highly related to physical [1], psychological [2] and tactical factors [3]. Physical performance-related factors in road cycling have traditionally been linked to the athlete’s physiological profile. Professional cyclists typically display very high aerobic capacity at both maximal and submaximal levels, as measured through maximal oxygen consumption (VO₂max) [1]. However, evidence suggests that good cycling economy and efficiency can compensate for a relatively low VO₂max [4]. In fact, cycling economy and lactate threshold may be more critical determinants of endurance performance than VO₂max itself [5].

Furthermore, heart rate (HR) monitoring has long been used to relate exercise intensity during competition to maximum and submaximal laboratory reference values [1]. More recently, with the advent of portable outdoor power measurement devices, power output and its relationship with laboratory parameters have become key variables to consider in both training and competitive performance [6].

Beyond dispute, the endurance profile of cyclists—measured through physiological or mechanical variables—plays a fundamental role in cycling performance [1]. However, research in this area seems to have overlooked other highly trainable capacities, such as flexibility and strength. Although the literature on this topic is limited, evidence suggests that replacing a portion of endurance training with resistance training can be beneficial for improving time trial performance and maximum power output [7].

Lateral preference refers to the dominance of one side of the body in performing motor actions [8]. Limb dominance in sport has been examined in various ways depending on the motor characteristics of each discipline. In acyclic sports, lateral dominance is typically identified by determining which limb achieves higher values in sport-specific tasks [9], with lateral asymmetries often arising from sport-specific adaptations [10,11,12]. Although cyclical sports are generally assumed to follow a bilaterally symmetrical pattern, asymmetry has been observed in several of these sports regardless of age, gender, or competitive level [10]. Specifically, in cycling, the dominant leg appears to generate greater force in athletes who exhibit pedaling asymmetries [13,14]. However, to our knowledge, it remains unclear whether the leg that dominates in force application during pedaling also demonstrates dominance across other performance-related variables.

There is no clear consensus regarding the relationship between bilateral symmetry and athletic performance, nor to what extent asymmetry influences injury risk [8,10,15,16,17]. However, scientific literature increasingly supports the idea that lateral asymmetry is both individual and sport-specific, and should therefore be addressed using task- and sport-specific thresholds [10,17,18]. Since the 1990s, these asymmetry thresholds have generally been treated as fixed values, typically ranging between 10% and 15%, regardless of the type of test, the population assessed, or the sporting discipline. This generic approach overlooks the influence of task specificity and individual characteristics on asymmetry. In response to these limitations, more recent proposals, such as that of Dos’Santos et al. [19], have emphasized the need to establish asymmetry thresholds that are specific to the task, the variable measured, and the characteristics of the athlete population under study. This perspective offers greater ecological validity and may enhance the relevance of asymmetry assessments in applied sports contexts; however, asymmetry thresholds have yet to be specifically explored in the context of cycling.

Conversely, research into asymmetries in cycling dates back over half a century [20,21], with early studies primarily examining their impact on performance through kinetic analyses of force application on the crank and pedals during pedaling [22]. Yet, no clear consensus has been reached in the scientific community. Some studies have associated more pronounced asymmetries in effective force with improved performance in a 4 km time trial [23], while others reported that pedaling asymmetries tend to decrease as exercise intensity increases [13]. This ongoing uncertainty highlights the need for further, more comprehensive exploration of this topic.

Pedaling asymmetries have also been studied using kinematic analysis [24,25] and muscle activity measured by electromyography (EMG) [26,27]. Carpes et al. [27] reported that cyclists vastus lateralis (VL) muscle activity increases significantly as the exercise intensity rises, without significant inter-limb differences. Furthermore, although asymmetries were not the focus of this study, Bini et al. [26] concluded that the VL was the only muscle among those analyzed that showed selective activation during a 40 km laboratory time trial. Their findings also reveal a strong relationship between VL and rectus femoris (RF) activation and the power output during the trial. These results suggest that EMG may serve as a valuable tool for detecting functional pedaling asymmetries.

While most research has focused on pedaling analysis, other relevant components of cycling performance remain underexplored. In this regard, Yanci and Los Arcos [28] examined inter-limb asymmetries in relation to vertical jump performance, reporting that cyclists exhibited both lower jumping capacity and greater asymmetries compared to runners. These findings likely reflect sport-specific demands, as cycling does not require reactive strength. Similarly, Pimentel et al. [29] observed greater asymmetries in bone mineral density and lean mass among cyclists compared to non-cyclists. However, it remains unclear whether such asymmetries are also evident in other performance-related tests more directly associated with cycling.

Nevertheless, to the best of our knowledge, asymmetries in competitive cyclists and their impact on performance have scarcely been addressed from a multifactorial perspective that included key components of physical preparation, such as muscle strength, range of motion (ROM), or resting muscle contractile capacity. In contrast, this multifaceted approach has been addressed in other sports such as volleyball [30], soccer [31], or canoeing [32] where different sport-specific capacities—such as concentric strength, ROM, muscle contractile properties assessed via tensiomyography (TMG), or muscle electrical activity during exercise—have been considered to identify athletes’ asymmetries.

Based on the existing evidence and gaps in the literature, we hypothesized that (1) the magnitude of lateral asymmetry would not differ significantly between categories (junior, under-23 and elite cyclists) and (2) that the percentage of lateral asymmetry observed in neuromuscular, ROM, strength and muscle electrical activity variables would demonstrate moderate explanatory power for cyclists’ performance during a maximal incremental cycle ergometer test.

Considering the paucity of research in this field among cyclists, the aim of this study is to (1) determine the magnitude and direction of potential asymmetries in neuromuscular properties, ROM, strength and muscle electrical activity asymmetries of the main muscles involved in pedaling in well-trained male road cyclists across categories (junior, under-23 and elite); (2) establish test- and category-specific asymmetry thresholds for neuromuscular properties, ROM, strength and muscle electrical activity, enabling individualized classification; and (3) examine the relationship between cyclists’ asymmetries in neuromuscular properties, ROM, strength and muscular electrical activity test with and performance in maximal incremental cycle ergometer test. By addressing these aims, this study seeks to provide novel and relevant contributions to scientific literature, offering a comprehensive and multifactorial approach to the analysis of asymmetries in cyclists. It proposes specific thresholds for their classification and develops a predictive model of cycling performance that can serve as a practical tool for coaches and sports performance professionals.

2. Materials and Methods

2.1. Study Design

An exploratory cross-sectional comparative study design was conducted to determine the magnitude and direction of lateral asymmetry in cyclists based on their competitive category (elite, under-23, junior). In addition, a predictive research design was employed to examine the influence of lower limbs lateral asymmetry in contractile properties, ROM, strength and muscular electrical activity on performance in maximal incremental cycle ergometer test (see Figure 1). All assessments were performed at the beginning of the competition season, following a day of active recovery.

2.2. Participants

An a priori sample size analysis was carried out using G*Power v.3.1. for Windows (Heinrich-Heine-Universität Düsseldorf, Düsseldorf, Germany). The analysis considered the number of measurements and the grouping of cyclists, with parameters set at an alpha level of 0.05, statistical power of 0.95 and an effect size of 0.50. The result indicated that a minimum of 54 cyclists was required.

The sample was composed of fifty-five well-trained road cyclists from elite (18), under-23 (27) and junior (10) categories, all actively competing in national and international amateur and professional cycling events and classified as tier 3 or 4 according to Mckay et al.’s [34] classification based on training volume and performance metrics (see

Table 1. Descriptive statistics of road cyclists’ characteristics.

	Age  (years)	Height  (cm)	Weight  (kg)	Body Fat  (%)	Muscle Mass  (kg)	BMI	Water  (%)	VO2max (mL/kg/min)
Junior (n = 10)	16.33 ± 0.50	170.63 ± 4.15	62.22 ± 5.44	8.87 ± 4.58	53.68 ± 2.67	21.41 ± 2.11	66.00 ± 4.03	65.58 ± 6.12
Under-23 (n = 27)	19.47 ± 1.22	175.49 ± 7.21	65.64 ± 7.27	7.04 ± 3.01	57.74 ± 5.04	20.81 ± 2.39	66.03 ± 5.32	69.11 ± 6.51
Elite (n = 18)	23.50 ± 2.22	179.69 ± 8.14	69.75 ± 7.75	7.81 ± 3.87	61.35 ± 7.59	21.57 ± 2.10	65.65 ± 3.14	66.91 ± 6.01
Total (n = 55)	20.31 ± 2.90	175.97 ± 7.24	66.13 ± 6.83	7.56 ± 3.53	58.32 ± 6.29	21.17 ± 2.24	65.90 ± 4.40	68.0 ± 6.61

Body composition was analyzed through bioelectrical impedance Tanita MC-780MA (Tanita Corporation, Tokyo, Japan). VO₂max was obtained through Gas Exchange (Ergostik, Geratherm Respiratory GmbH, Bad Kissingen, Germany) during a maximal incremental cycling test. BMI (body mass index).

). Inclusion criteria required cyclists to be healthy, injury-free at the time of data collection and to have successfully passed a medical screening compromising a submaximal exercise test and an ultrasound assessment to exclude any abnormalities in adaptation to exercise. Only cyclists with a minimum sporting level of tier 3 were eligible for inclusion [34]. Those not meeting these criteria were excluded from the study.

All cyclists included in the sample have signed an informed consent informing about the assessment protocol, risk and benefits, which followed the principles of the Declaration of Helsinki (DoH)—Ethical Principles for Medical Research Involving Human Participants (1964) and its latest amendments from the 75th General Assembly of the World Medical Association (WMA) in Finland, on 19 October 2024. This investigation was approved by the Institutional Review Board from the local University (01-170123) and the cycling club.

2.3. Procedure

All tests were conducted during a recovery session, following the same protocol order for all participants: TMG, ROM, strength and EMG during a maximal incremental cycling test. To determine inter-limb asymmetries, the formula proposed by Bishop et al. [33] was applied

$$Asymmetry\left(\%\right)={\displaystyle \frac{DL-NDL}{DL\times 100}}\times IF(DL<NDL, 1,-1)$$

where DL is dominant leg, and NDL non-dominant leg, self-reported for each cyclist. This formula implements the Excel IF function that enables to monitor both the magnitude and direction of lateral asymmetry, while avoiding issues related to the absolute magnitude of variation. The resulting values are expressed as percentage (%).

2.3.1. Neuromuscular Properties Assessment

TMG was used to assess contractile properties of biceps femoris (BF), RF, VL and vastus medialis (VM) in both legs. Radial muscle belly displacement (Dm) was measured under isometric conditions using electrical stimulation, following the protocol described by García-García et al. [35,36]. A digital displacement transducer (GK 30, Panoptik d.o.o., Ljubljana, Slovenija) was positioned perpendicular to the thickest part of the muscle belly, in accordance with anatomical guidelines from Perotto et al. [37] (BF: at the midpoint of a line between the fibula head and the ischial tuberosity; RF: on the anterior aspect of the thigh, midway between the superior border of the patella and the anterior superior iliac spine; VL: over the lateral aspect of the thigh, one handbreadth above the patella; VM: four fingerbreadths proximal to the superior-medial angle of the patella.). Two self-adhesive electrodes (5 × 5 cm, Lessa®, AB Medica Group SA, Barcelona, Spain) were placed symmetrically at 5 cm from the digital transducer. Progressive electrical stimulation was applied in 10 mA increments up to a maximal output of 110 mA (EMF-FURLAN & Co. d.o.o., Ljubljana, Slovenia). The following parameters were recorded: Dm (in mm); contraction time (Tc, in ms), defined as the time from 10% to 90% of Dm and radial displacement velocity (Vrd, in mm·s⁻¹), calculated as Dm80/Tc, where Dm80 represents the displacement during Tc. The curve with the highest Dm was selected for analysis. All measurements were performed by two experienced evaluators with extensive expertise in the field.

2.3.2. ROM Assessment

After a protocolized 20 min warm-up consisting of five general mobility exercises (4 sets of 10 repetition each), four isometric strength exercises (4 sets of 20 s each) and five minutes of continuous running at an intensity of 4–6 on the modified Borg’s Rate of Perceived Exertion (RPE) scale [38], cyclists performed the active knee extension (AKE) test to assess hip-knee ROM. The AKE was conducted according to the protocol described by Gajdosik and Lusin [39], with participants lying supine on a stretcher and starting at a 90° hip and knee flexion. A digital goniometer (Baseline Absolute Axis 360°, Fabrication Enterprises, Inc., White Plains, New York, USA) was used to measure knee extension ROM. Three attempts were performed on each limb, supervised by two expert evaluators—one responsible for recording the knee extension angle and the other for preventing compensatory movements during the measurement. The best attempt for each cyclist was retained for further analysis.

2.3.3. Strength Assessment

After a specific warm-up focused on position adjustment, cyclists performed a maximal repetition semi-squat using a horizontal leg press machine (RS-1403 Leg Press ROC-IT line; HOIST, Poway, CA, USA). The test was performed unilaterally, using one leg and the cyclist’s own body weight. Athletes were instructed to execute the concentric phase of the movement as explosive as possible while maintaining control during the eccentric phase. The test ended when the cyclist could no longer complete repetitions or when a deterioration in technical execution was observed by consensus among evaluators (i.e., knee valgus, hyperlordosis, oscillatory movements). A Linear Encoder (Chronojump Boscosystem, Barcelona, Spain) paired with Chronojump software (version 1.7.0 for Windows; Chronojump Boscosystem) was used to measure average speed (Vavg), average power (Pavg) and the number of repetitions (reps) performed. Previous studies have confirmed the validity and reliability of this method for measuring movement speed and estimating power, reporting intraclass correlation coefficient (ICC) between 0.95 and 0.988 [40].

2.3.4. Muscular Electrical Activity Assessment

During the maximal incremental cycling test on a cycle ergometer, surface EMG signals were recorded using the FREEEMG 1000 system (BTS Bioengineering, Garbagnate Milanese, MI, Italy). Self-adhesive hydrogel Ag/AgCl electrodes (40 mm, Meditrace-Kendall, Covidien IIc, Mansfield, MA, USA) were applied to shaved and alcohol-cleaned skin. Electrodes were placed on the vastus lateralis of both legs, aligned with muscle fibers in accordance with the SENIAM recommendations.

The system operates in differential mode with a high input impedance of 100 MΩ and a Common Mode Rejection Ratio (CMRR) greater than 110 dB at 50–60 Hz. The EMG signal was acquired at a sampling rate of 1000 H, with the amplifier gain was set to a 3.0 mV range, which is recommended for typical clinical and sports applications. Wireless EMG sensors transmitted the signal via Bluetooth to the receiving unit. Muscle activity was quantified using the Root Mean Square (RMS) of the EMG signal, calculated in moving windows of 100 ms, as follows

$$RMS=\sqrt{{\displaystyle \frac{1}{N}}\sum _{i=1}^N{x}_i^N}$$

where $x_i$ represents the EMG signal value at sample $i$ within the window of $N$ samples [41]. Electrodes were positioned according to the anatomical guidelines described by Perotto et al. [37], with an interelectrode distance of 1 cm. After the test, electrode placement was visually inspected to confirm that no displacement had occurred during the protocol.

2.3.5. Maximal Incremental Cycle Ergometer Test

Prior to starting the maximal incremental test, cyclists completed a standardized 20 min specific warm-up, pedaling on their own bicycle mounted on rollers, with intensity progressively increasing up to their individual functional threshold power.

The maximal incremental cycling test was carried out on an electromagnetic brake cycle ergometer (Cardgirus W3+, Sabadell, Barcelona, Spain). Each cyclist individually adjusted the cycle ergometer setup to replicate their own bicycle position. The test followed a stepwise protocol starting at 100 W and increasing by 5 W every 15 s, while maintaining a constant pedaling cadence above 60 revolutions per minute (rev·min⁻¹). The test concluded when the cyclist reached volitional exhaustion, was unable to maintain cadence or if any medical issues arose, monitored by the cardiologist F-R, D.

Gas Exchange was recorded using an Ergostik CardioPart analyzer (Ergostik, Geratherm Respiratory GmbH, Bad Kissingen, Germany) to determine VO₂max as well as ventilatory thresholds 1 and 2 (VT1 and VT2, respectively). The gas analyzer was calibrated two minutes prior to each test. Blood lactate concentration ([La]) was measured via capillary earlobe puncture at VT1, VT2 and three minutes post-test, using electro-enzymatic reactive strips (Lactate Scout 4, EKF diagnostic, SensLab GmbH, Leipzig, Germany). The lactate analyzer was calibrated before each test using a control solution. HR was recorded via electrocardiogram with a Mortara WAM™ syste, (Mortara Wireless Acquisition Module, Mortara Instrument, Milwaukee, WI, USA).

To ensure maximal effort, all test met the criteria established by Birds and Davison [42]: (1) a post-exercise blood lactate concentration of ≥8 mMol·L⁻¹ within 3–5 min; (2) a maximal HR within ±10 bpm of the age-predicted maximum HR estimated using Whaley et al.’s [43] formula; and (3) a subjective RPE between 9 and 10 on the modified Borg scale [38]. During the maximal incremental cycling test, the following performance variables were obtained: power output relative to body weight (W/kg) at VO₂max, VT₁ and VT₂.

2.3.6. Statistical Analysis

Relative reliability was assessed using intraclass correlation coefficient (ICC) based on single measurements, two-way mixed effects models, and absolute agreement. Absolute reproducibility was evaluated using the coefficient of variation (CV).

The influence of the inter-limb asymmetry (DL vs. NDL) across performance categories (junior, under-23, elite), was analyzed using one-way ANOVA. Prior to this, assumptions of multivariate normality and homogeneity of variances and covariances were tested. Normality was confirmed using the Kolmogorov–Smirnov and Lilliefors tests, while homoscedastic was verified via the Box M test. Post hoc comparisons were performed using the Turkey test HSD. Effect sizes for the ANOVA were reported as partial eta square (η_p²), interpreted as small (0.02), moderate (0.06), or large (0.14) [44].

Kappa coefficients assessed the level of agreement on the direction of asymmetry across variables (i.e., TMG, ROM, leg press), interpreted as slight (0.00–0.20), fair (0.21–0.40), moderate (0.41–0.60), substantial (0.61–0.80), near perfect (0.81–0.99), or perfect (1.00) [45].

Cyclists were classified as “asymmetrical” or “symmetrical” based on specific asymmetry thresholds following Dos’Santos et al. [19], calculated as: % Asymmetry + (0.2 × SD), where % of Asymmetry and SD represent the sample’s average asymmetry percentage and standard deviation. This classification was applied individually to each variable from TMG, upper and lower limb ROM, and leg press assessments.

A stepwise multiple linear regression was performed to develop a parsimonious model predicting performance—defined as W/kg in the maximal incremental cycle ergometer test—based on cyclists’ asymmetry classification through asymmetry thresholds for TMG, AKE and leg press variables (predictor variables). Model assumptions were verified through residual analysis and Durbin–Watson statistics (d = 1.857), indicating no significant autocorrelation. Multicollinearity was assessed via the Variation Inflation Factor (VIF), with values ranging from 1.015 to 1.074, confirming the absence of multicollinearity among the predictor variables.

All statistical analyses were performed using IBM SPSS Statistic version 25.0 (SPSS Inc., Chicago, IL, USA).

3. Results

Relative and absolute reliability were calculated for each test using ICC and CV, respectively. For TMG, ICC across all assessed muscles ranged from 0.94 and 0.99 for Dm, 0.95 to 0.99 for Tc, and 0.92 to 0.98 for Vrd. Corresponding CV values were 2.4%, 2.5% and 2.9%, respectively. For the AKE test, ICC was 0.93 with a CV of 2.9% and for leg press test, ICC was 0.95 with a CV of 5.3%.

As shown in

Table 2. Descriptive statistics and lateral asymmetry (magnitude and direction) for TMG parameters.

		Category	Mean (cm) ± SD		% Asymmetry	Asymmetry	Asymmetry Direction	Threshold
			Dominant	Non-Dominant	(Mean ± SD)	IC 95%
BF	Tc	Junior	36.03 ± 10.24	36.68 ± 14.74	22.66 ± 21.56	6.09–39.24	NDL	26.97
		Under-23	39.68 ± 10.76	36.07 ± 9.78	19.62 ± 16.58	13.06–26.18	DL	22.94
		Elite	41.44 ± 10.13	40.93 ±11.22	23.30 ± 21.90	12.75–33.86	DL	27.68
		Total	39.69 ± 10.43	37.85 ± 11.19	21.39 ± 19.08	16.23–26.55	DL	25.21
	Dm	Junior	7.60 ± 2.00	6.90 ± 1.85	22.77 ± 23.75	4.51–41.03	DL	27.52
		Under-23	7.65 ± 2.25	7.12 ± 2.14	20.16 ± 14.13	14.57–25.75	DL	22.99
		Elite	8.37 ± 2.00	8.08 ± 3.14	21.68 ± 18.04	12.98–30.37	DL	25.98
		Total	7.89 ± 2.12	7.42 ± 2.50	21.11 ± 17.01	16.51–25.71	DL	24.51
	Vrd	Junior	173.54 ± 48.12	163.61 ± 56.00	16.72 ± 12.59	7.04–26.39	DL	19.24
		Under-23	157.77 ± 38.51	161.44 ± 41.84	20.48 ± 12.49	15.54–25.42	NDL	22.98
		Elite	167.21 ± 38.39	156.67 ± 46.42	19.05 ± 12.76	12.90–25.20	DL	21.60
		Total	163.61 ± 39.83	160.15 ± 45.09	19.37 ± 12.44	16.01–22.73	DL	21.86
RF	Tc	Junior	27.26 ± 3.86	26.09 ± 3.93	15.88 ± 8.94	9.00–22.76	DL	17.67
		Under-23	27.71 ± 6.95	28.08 ± 6.18	15.94 ± 10.34	11.85–20.04	NDL	18.01
		Elite	29.76 ± 4.60	28.84 ± 5.25	11.37 ± 10.68	6.22–16.52	DL	13.51
		Total	28.35 ± 5.80	28.01 ± 5.54	14.35 ± 10.30	11.57–17.14	DL	16.41
	Dm	Junior	7.34 ± 2.00	6.72 ± 1.61	20.59 ± 16.09	8.22–32.97	DL	23.81
		Under-23	7.06 ± 2.02	7.15 ± 2.09	19.40 ± 11.08	15.02–23.79	NDL	21.62
		Elite	7.15 ± 2.42	7.22 ± 2.18	21.39 ± 12.96	15.14–27.64	NDL	23.98
		Total	7.14 ± 2.12	7.10 ± 2.02	20.28 ± 12.42	16.92–23.64	DL	22.76
	Vrd	Junior	216.36 ± 53.12	206.89 ± 44.44	16.25 ± 12.52	6.62–25.88	DL	18.75
		Under-23	210.96 ± 61.62	209.68 ± 64.38	17.55 ± 12.59	12.57–22.53	DL	20.07
		Elite	194.61 ± 74.55	199.44 ± 45.52	20.39 ± 14.54	13.38–27.40	NDL	23.30
		Total	206.20 ± 64.61	205.69 ± 54.78	18.32 ± 13.13	14.77–21.87	DL	20.95
VL	Tc	Junior	24.12 ± 3.15	20.72 ± 1.35	14.70 ± 10.13	6.91–22.49	DL	16.73
		Under-23	21.95 ± 2.52	21.51 ± 2.36	7.74 ± 6.18	5.29–10.19	DL	8.98
		Elite	25.72 ± 9.02	22.03 ± 2.60	15.72 ± 17.25	7.40 ± 24.04	DL	19.17
		Total	23.61 ± 5.88	21.56 ± 2.33	11.63 ± 12.16	8.35–14.92	DL	14.06
	Dm	Junior	5.62 ± 1.45	5.32 ± 1.16	13.53 ± 10.33	5.59–21.47	DL	15.60
		Under-23	5.33 ± 1.46	5.49 ± 1.37	18.21 ± 12.31	13.34–23.08	NDL	20.67
		Elite	5.21 ± 1.27	4.76 ± 1.63	21.47 ± 12.42	15.49–27.46	DL	23.95
		Total	5.34 ± 1.38	5.21 ± 1.45	18.57 ± 12.14	15.29–21.86	DL	21.00
	Vrd	Junior	190.66 ± 56.65	204.95 ± 38.13	14.56 ± 15.39	2.73–26.39	NDL	17.64
		Under-23	193.85 ± 46.96	203.50 ± 44.41	15.91 ± 13.86	10.43–21.40	NDL	18.68
		Elite	172.41 ± 53.99	172.99 ± 57.74	23.45 ± 14.18	16.61–30.28	NDL	26.29
		Total	185.92 ± 51.07	193.20 ± 49.96	18.29 ± 14.46	14.38–22.20	NDL	21.18
VM	Tc	Junior	23.97 ± 3.97	25.48 ± 3.61	12.81 ± 7.31	7.19–18.43	NDL	14.27
		Under-23	24.63 ± 4.13	24.90 ± 3.05	7.99 ± 6.15	5.55–10.42	NDL	9.22
		Elite	27.42 ± 9.49	28.85 ± 9.66	15.41 ± 18.79	6.35–24.47	NDL	19.17
		Total	25.48 ± 6.53	26.36 ± 6.39	11.34 ± 12.48	7.97–14.71	NDL	13.84
	Dm	Junior	7.20 ± 0.93	7.63 ± 1.06	10.46 ± 7.74 *	4.50–16.41	NDL	12.01
		Under-23	8.25 ± 1.81	8.56 ± 1.01	15.48 ± 10.77	11.22–19.75	NDL	17.63
		Elite	7.84 ± 2.84	8.44 ± 2.09	23.32 ± 18.61	14.34–32.29	NDL	27.04
		Total	7.94 ± 2.13	8.37 ± 1.49	17.37 ± 14.23	13.52–21.21	NDL	20.22
	Vrd	Junior	243.63 ± 35.06	245.11 ± 58.45	13.83 ± 14.39	2.77–24.90	NDL	16.71
		Under-23	277.11 ± 83.47	278.51 ± 46.29	16.32 ± 13.30	11.05–21.58	NDL	18.98
		Elite	248.30 ± 117.25	254.07 ± 95.52	20.08 ± 12.50	14.05–26.10	NDL	22.58
		Total	261.68 ± 91.41	264.60 ± 69.12	17.21 ± 13.16	13.65–20.78	NDL	19.84

BF (biceps femoris), RF (rectus femoris), VL (vastus lateralis), VM (vastus medialis), Tc (contration time in ms), Dm (maximum radial muscle belly displacement in mm), Vrd (radial displacement velocity in mm·s−1); SD (standard deviation); IC (interval confidence; DL (dominant limb); NDL (non-dominant limb); * (p < 0.05).

, only junior cyclists exhibited a significantly lower percentage of asymmetry in VM Dm compared to elite cyclists (10.46% vs. 23.32%; F = 3.199; p = 0.049). None of the other TMG parameters differed significantly across categories (junior, under-23 and elite). Regarding the evaluated muscles, asymmetry percentages were slightly lower in monoarticular muscles (VL and VM) compared to biarticular muscles (BF and RF), with the BF showing the highest percentage of asymmetry across all parameters (Tc: 21.39%; Dm: 21.11%; Vrd: 19.37%). Among TMG variables, the Tc exhibited the lowest percentage of asymmetry (11.34–21.39%).

The direction of the asymmetry varied by parameter and muscle, showing no clear pattern. Agreement on asymmetry direction between limbs for TMG parameters was generally mild to moderate (see

Table 3. Kappa values and descriptive levels of agreement between asymmetry scores.

			Kappa Coefficient	Level of Agreement
TMG	BF	Tc—Dm	0.58	Moderate
		Tc—Vrd	0.12	Slight
		Dm—Vrd	0.59	Moderate
	RF	Tc—Dm	0.50	Moderate
		Tc—Vrd	0.12	Slight
		Dm—Vrd	0.59	Moderate
	VL	Tc—Dm	0.57	Moderate
		Tc—Vrd	0.15	Slight
		Dm—Vrd	0.56	Moderate
	VM	Tc—Dm	0.31	Fair
		Tc—Vrd	0.08	Slight
		Dm—Vrd	0.71	Substantial
Leg Press		Vavg—Pavg	0.97	Near perfect
		Vavg—Reps	0.36	Fair
		Pavg—Reps	0.39	Fair
EMG VL		VT1—VT2	0.82	Near perfect
		VT1—VO2max	0.87	Near perfect
		VT1—TEST	0.90	Near perfect
		VT2—VO2max	0.77	Substantial
		VT2—TEST	0.77	Substantial
		VO2max—TEST	0.95	Near perfect

TMG (tensiomyography); BF (biceps femoris); RF (rectus femoris); VL (vastus lateralis); VM (vastus medialis); Tc (contraction time in ms); Dm (maximum radial muscle belly displacement, in mm); Vrd (radial displacement velocity, in mm·s⁻¹), Vavg (average speed in m·s⁻¹); Pavg (average power in w); Reps (number of repetitions); EMG (electromyography); VT1 (first ventilatory threshold); VT2 (second ventilatory threshold); VO₂max (maximum oxygen consumption); TEST (RMS of the entire maximal cycle ergometer incremental test).

). Using the specific asymmetry threshold to classify cyclists as “asymmetric”, between 29.0% and 41.1% were classified as asymmetric for BF contractile properties, 34.54% and 43.6% for RF, 27.5% and 40.0% for VL, and 23.63% and 34.5% for VM. The RF muscle has the highest proportion of cyclists classified as asymmetric, while VM had the lowest.

Regarding ROM, the AKE test showed a low percentage of asymmetry in across categories, ranging between 1.7% and 3.6% (see

Table 4. Descriptive statistics and lateral asymmetry (magnitude and direction) for AKE.

	Category	Mean (cm) ± SD		% Asymmetry	Asymmetry	Asymmetry Direction	Threshold
		Dominant Limb	Non-Dominant Limb	(Mean ± SD)	IC 95%
AKE	Junior	152.11 ± 9.04	153.56 ± 12.87	3.6 ± 2.3	2.30–4.90	NDL	4.06
	Under-23	165.10 ± 6.11	166.67 ± 5.39	2.3 ± 1.8	1.55–3.05	NDL	2.66
	Elite	162.77 ± 10.12	162.52 ± 10.63	1.7 ± 1.9	0.75–2.59	DL	2.08
	Total	162.16 ± 9.23	163.10 ± 9.87	2.3 ± 2.0	1.93–3.11	NDL	2.70

AKE (active knee extension); SD (standard deviation); IC (interval confidence); DL (dominant limb); NDL (non-dominant limb).

). When analyzed by category, junior cyclists exhibited significantly greater asymmetry in AKE compared to elite cyclists (p = 0.019; ηp² = 0.104). The direction of the asymmetry was predominantly towards the NDL, indicating that the NDL had a greater ROM than the DL. Using the specific asymmetry thresholds to classify cyclists as “asymmetric”, 33.3% of the cyclists were classified as asymmetric.

Regarding force production in the horizontal leg press, no significant differences in asymmetry were observed between the DL and NDL across the three categories. As shown in

Table 5. Descriptive statistics and lateral asymmetry (magnitude and direction) for leg press.

		Category	Mean (cm) ± SD		% Asymmetry	Asymmetry	Asymmetry Direction	Threshold
			Dominant	Non-Dominant	(Mean ± SD)	IC 95%
Leg  press	Vavg	Junior	0.364 ± 0.04	0.386 ± 0.04	8.36 ± 6.38	3.46–13.27	NDL	9.64
		Under-23	0.392 ± 0.10	0.367 ± 0.10	10.34 ± 8.31	6.34–14.34	DL	12.00
		Elite	0.423 ± 0.93	0.404 ± 0.07	10.23 ± 6.86	5.86–14.59	DL	11.60
		Total	0.395 ± 0.89	0.382 ± 0.08	9.86 ± 7.36	7.51–12.22	DL	11.33
	Pavg	Junior	224.01 ± 35.29	236.91 ± 37.41	8.86 ± 6.73	3.68–14.03	NDL	10.21
		Under-23	226.82 ± 107.28	234.22 ± 75.44	13.17 ± 12.92	6.94–19.40	NDL	15.75
		Elite	286.18 ± 77.69	271.34 ± 62.56	10.23 ± 7.21	5.64–14.81	DL	11.67
		Total	263.00 ± 88.28	245.96 ± 65.59	11.32 ± 10.22	8.05–14.58	DL	13.36
	Reps	Junior	20.78 ± 14.19	26.56 ± 17.37	33.42 ± 19.38	18.75–48.09	NDL	37.30
		Under-23	17.74 ± 9.38	16.42 ± 10.44	29.85 ± 20.05	20.18–39.52	DL	33.86
		Elite	18.50 ± 12.42	16.92 ± 11.21	34.27 ± 29.56	15.48–53.05	DL	40.18
		Total	18.65 ± 11.26	18.85 ± 12.86	31.98 ± 22.61	24.75–39.38	NDL	36.50

Vavg (average speed in m·s⁻¹); Pavg (average power in w); Reps (number of repetitions); SD (standard deviation); IC (interval confidence); DL (dominant limb); NDL (non-dominant limb).

, the percentages of asymmetry for Vavg and Pavg were similar, at 9.86% and 11.32%, respectively. However, asymmetry was notably higher for the maximum number of repetitions performed, reaching 31.98%. The direction of the asymmetry varied depending on the parameter used. Agreement between the direction of the asymmetries for the maximum number of repetitions and the other two parameters (velocity and power) was only fair (see Table 3). In contrast, there was a near-perfect agreement between Vavg and Pavg, both indicating that the DL produced greater speed and power in the horizontal leg press. Applying the specific asymmetry thresholds for classification, 42.5% of the cyclists were classified as asymmetric based on Vavg, 32.5% based on Pavg, and 35.0% based on the maximum Reps.

Regarding VL electrical activity (VL-EMG) during the maximal incremental cycle ergometer test, no significant differences in asymmetry between the DL and NDL across categories were observed. As shown in

Table 6. Descriptive statistics and lateral asymmetry (magnitude and direction) for VL-EMG.

		Category	Mean (cm) ± SD		% Asymmetry	Asymmetry	Asymmetry Direction	Threshold
			Dominant	Non-Dominant	(Mean ± SD)	IC 95%
VL-EMG	VT1	Junior	0.164 ± 0.024	0.174 ± 0.026	17.37 ± 10.28	10.11–30.47	NDL	19.43
		Under-23	0.150 ± 0.032	0.162 ± 0.039	19.85 ± 15.68	9.88–29.81	NDL	22.99
		Elite	0.187 ± 0.060	0.155 ± 0.052	22.36 ± 16.83	11.05–33.68	DL	25.73
		Total	0.165 ± 0.045	0.162 ± 0.042	20.87 ± 14.74	15.37–26.38	DL	23.82
	VT2	Junior	0.200 ± 0.038	0.217 ± 0.036	20.95 ± 14.76	7.29–34.61	NDL	23.90
		Under-23	0.163 ± 0.037	0.185 ± 0.046	20.06 ± 14.88	11.82–28.30	NDL	23.04
		Elite	0.209 ± 0.080	0.73 ± 0.060	22.68 ± 13.70	14.40–30.96	NDL	25.42
		Total	0.185 ± 0.058	0.187 ± 0.051	21.21 ± 14.04	16.39–26.04	NDL	24.02
	VO2max	Junior	0.241 ± 0.092	0.235 ± 0.073	22.15 ± 12.52	10.57–33.73	DL	24.65
		Under-23	0.183 ± 0.062	0.200 ± 0.056	22.37 ± 19.13	11.78–32.97	NDL	26.20
		Elite	0.277 ± 0.095	0.203 ± 0.050	23.50 ± 15.27	14.26–32.73	DL	26.55
		Total	0.226 ± 0.089	0.208 ± 0.057	22.75 ± 16.16	17.19–28.30	DL	25.98
	TEST	Junior	0.183 ± 0.037	0.194 ± 0.044	20.29 ± 11.00	10.11–30.47	NDL	22.49
		Under-23	0.143 ± 0.033	0.169 ± 0.072	19.85 ± 15.68	9.88–29.81	NDL	22.99
		Elite	0.204 ± 0.082	0.159 ± 0.041	22.36 ± 16.83	11.05–33.68	DL	25.73
		Total	0.172 ± 0.060	0.171 ± 0.056	20.87 ± 14.74	15.37–26.38	DL	23.82

VT₁ (first ventilatory threshold); VT₂ (second ventilatory threshold); VO₂max (maximum oxygen consumption); TEST (RMS of the entire maximal cycle ergometer incremental test); SD (standard deviation); IC (interval confidence); DL (dominant limb); NDL (non-dominant limb).

, the percentages of asymmetry were very similar across the physiological milestones—VT₁, VT₂, and VO₂max—ranging from approximately 20.8% to 22.7%, with only a slight increase as exercise intensity rose. In terms of asymmetry direction, there was a clear trend toward the DL. Indeed, the level of agreement in the directions of asymmetries across the physiological milestones was substantial to near perfect (see Table 3), supporting this trend. This indicates that the VL on the DL exhibited greater electrical activity than the NDL. When applying the specific asymmetry threshold to classify cyclists, 40.0% of the cyclists were classified as asymmetric at VT₁, 31.4% at VT₂, and 34.2% at VO₂max. Considering the RMS of the VL across the entire test, 26.6% of cyclists were classified as asymmetric.

The multiple linear regression analysis revealed a significant predictive model for maximal incremental cycle ergometer test performance, using W/kg at VO₂max as the dependent variable. The predictor included lateral asymmetry in ROM measured by AKE (β = −0.0241; t= −1.1636; p = 0.111), TMG variables Tc of the RF (β = −0.0252; t = −1.690; p = 0.100), and Tc of the VL (β = −0.286; t = −1.954; p = 0.050), as well as leg press Vavg (β = −0.311; t = −2.147; p = 0.039) (1):

W/kg at VO₂max = 6.932–0.060 AKE—0.013 Tc of RF—0.013 Tc of VL—0.022 Vavg

(1)

This model is significant, but it explains only a small proportion of the variability in maximal incremental cycle ergometer performance (F = 3.850; p < 0.01; R²_a = 0.23).

4. Discussion

The main findings of this study indicate that junior cyclists show a significantly lower percentage of asymmetry in the Dm of the VM compared to elite cyclists, as well a greater asymmetry in the AKE than elite cyclists, with no differences observed in horizontal leg press performance or VL-EMG during the maximal incremental cycle ergometer test. The magnitude and direction of lateral asymmetry differed between tests, ranging from 11.3% to 21.3% in the contractile properties of the muscles evaluated, 2.3% in ROM (favoring the NDL), between 9.8% and 31.9% in force production and between 20.8% and 22.7% in VL-EMG, with asymmetry in the latter favoring the DL.

Moreover, the percentage of cyclists classified as asymmetric based on the individualized asymmetry threshold in each test varied between 23.6% and 43.6% for TMG, 33.3% for ROM, between 35.0 and 42.5% for force production, and between 31.4% and 40% for the VL-EMG at the physiological milestones of the maximal incremental cycle ergometer test. Finally, 23% of the variability in cyclist’s maximal cycle ergometer incremental test performance could be explained by lateral asymmetry in ROM, contractile properties (Tc of RF and VL), and force production.

To our knowledge, scientific literature has paid little attention to describing lower limb lateral asymmetry in cycling or its effect based on category. Instead, most studies have explored age-related differences in power output and the ability to sustain efforts. In this regard, younger cyclists have consistently shown lower peak power output that adults during maximal incremental tests to exhaustion [46,47,48]. Similarly, this lower power output in younger cyclists is also evident at 2 and 4 mMol L⁻¹ [La] [46], as well at respiratory compensatory threshold and ventilatory threshold [47]. However, our results did not show significant differences in VL-EMG asymmetry between categories during the maximal incremental cycle ergometer test. Therefore, VL-EMG asymmetry does not appear to modulate power production across different ages categories in cyclists. Nonetheless, cyclists’ power output has also been analyzed during constant workload tests, both short- and long-term, with older cyclists generally demonstrating greater performance in long-term efforts [46,49].

Similarly, in terms of force production during strength exercises, our findings reveal no significant inter-limb asymmetry differences in horizontal leg press. These results are consistent with those reported by Alejo et al. [47], who also found no significant differences in strength exercises such as squat, hip thrust, or split squat across junior, under-23 and professional categories. Notwithstanding, our findings also show that younger cyclists exhibit lower lateral asymmetry in muscle tone (Dm) in the VM compared to older (elite) cyclists. Interestingly, this pattern contrasts with findings in volleyball players, where senior athletes display lower levels of lateral asymmetry in rectus femoris and BF muscles [30]. However, this discrepancy should be interpreted with caution, given the mechanical and energetic differences between the two sports. It is possible that the asymmetries observed in cycling are related to the specificity of the discipline, where the motor action is continuous, repetitive and difficult to consciously perceive. Prolonged practice under these conditions could accentuate such asymmetries over time. In contrast, in sports like volleyball, asymmetries may diminish with experience, as players tend to use both legs more evenly to maximize jump height during key game actions.

Furthermore, junior cyclists have shown greater inter-limb asymmetry in the AKE test compared to elite cyclists. To our knowledge, no previous studies have addressed age-related differences in ROM differences in road cyclists, although this has been explored in other sports. For example, Iglesias Caamaño et al. [30] reported that junior volleyball players show a higher percentage of lateral asymmetry in the AKE test than senior players. These findings align with ours; in fact, the percentage of asymmetry is higher in cyclists (3.6 ± 2.3%) than in volleyball players (2.9 ± 2.4%). Despite differences between categories, the overall asymmetry percentage in the AKE test is lower than in other tests performed. However, junior cyclists showed lower ROM values in the AKE test compared to other male athletes such as throwers, runners, jumpers [50] or soccer players [51], while under-23 and elite cyclists presented higher ROM values. To our knowledge, the impact of reduced flexibility on cycling performance has not been explored. Nonetheless, the observed inter-category differences in ROM suggest potential effects on performance that warrant further investigation in future studies.

Overall, it remains unclear whether the percentage of lateral asymmetry differs across categories. The underlying aim of this analysis was to explore whether training specialization over the years may promote the development of greater lateral asymmetry. Similarly, it sought to examine to what extent lateral asymmetry may influence athletic performance—an important question that remains unanswered and requires further investigation.

The magnitude of the asymmetry observed in the different tests cannot, to the best of our knowledge, be directly compared with similar samples. For instance, García-García et al. [52] found no significant inter-limb differences in the contractile properties (Tc and Dm) of the VM, VL, RF and BF muscles of professional cyclists, either during the preparatory or the competitive periods. However, lateral asymmetry was not specifically analyzed in this study. In contrast, Yanci and Los Arcos [28] conducted an asymmetry analysis comparing the elastic explosive strength of both limbs by measuring power output during a countermovement jump, and reported significantly greater asymmetry values (14.6%, p < 0.01) compared to the strength asymmetry power values we observed in the leg press test (11.32 ± 10.22%). Nevertheless, due to the differences in testing protocols, these asymmetry percentages cannot be directly compared.

Altogether, our results show complex patterns of asymmetry in competitive cyclists that vary depending on the parameter evaluated and the exercise protocol. When analyzing muscle electrical activity using EMG, the findings of Carpes et al. [27] demonstrated that asymmetries in muscle activation (VL, medial gastrocnemius and BF) remained stable across different exercise intensities. This behavior contrasts markedly with our findings, where we observed a progressive increase in the VL electrical asymmetry during the incremental test, rising from 20.87 ± 14.74% to 22.75 ± 16.16%. These results also differ from torque and force asymmetry patterns reported in other studies. For example, Carpes et al. [13] showed oscillation in crank torque asymmetry during a 40 km time trial at various stages and intensities of peak oxygen uptake (1st stage at 64.72% of VO₂peak: 8.91 ± 0.7%; 2nd stage at 61.88% of VO₂peak: 13.51 ± 4.17%; 3rd stage at 60.75% of VO₂peak: 17.28 ± 5.11%; 4th stage at 71.78% VO₂peak: 0.32 ± 2.92%). Similarly, Bini and Hume [23] reported effective strength asymmetry during a 4 km time trial using strain gauge instrumented pedals, showing an increasing tendency up to 3.5 km (0.5 km: 36 ± 33%; 2 km: 45 ± 26%; 3.5 km: 54 ± 34%) followed by a decrease at 4 km (39 ± 31%) [23]. It is important to note that the protocols used by Carpes et al. [13] and Bini and Hume [23] involved 40 km and 4 km free time-trial, respectively, whereas our study employed a maximal incremental cycle ergometer test. These differences in performance assessment could partly explain the contrasting asymmetry response to increasing intensity and fatigue. However, the potential influence of accumulated fatigue was addressed in the study by Farrell and Neira [53] who evaluated lateral asymmetry at different powers levels (60, 70, 80, and 90% of VO₂max) using a cycle ergometer with instrumented cranks. Their results showed that power production asymmetries did not vary significantly between intensities, with average asymmetry index ranging from −0.8% to 2.5%.

On the contrary, and in line with the asymmetry values observed in our cyclists, Bini and Hume [14] conducted an incremental cycle ergometer test with cyclists and triathletes, comparing two measurement systems (instrumented pedals and crank system). Their results showed smaller asymmetries at lower intensities (6 ± 17% with the crank system and 11 ± 28% with pedals), with asymmetry increasing at higher intensities (27 ± 18% with the crank system and 51 ± 36% with pedals). Additionally, our values are still significantly higher than those observed for kinematic parameters such as cadence, where the asymmetry index typically ranges between 5% and 8% [54]. This highlights how the magnitude of the asymmetry varies considerably depending on both the parameter evaluated [30] and on exercise intensity [27].

However, in other disciplines, the asymmetries obtained through TMG in our cyclists are lower in the Dm (21.11 ± 17.01%) and Tc (21.39 ± 19.08%) of the BF compared to values reported in male canoeists (Dm: 24.08 ± 15.43%; Tc: 23.64 ± 17.93%) [32], and volleyball players (Dm: 26.38 ± 20.29%) [30]. Similarly, the asymmetry in the AKE test was lower in cyclists (2.3 ± 2.0%) than in canoeists (5.81 ± 4.14%) [32], although similar to volleyball players (2.10 ± 1.84%) [30]. This difference likely reflects sport-specific demands, as canoeists adopt a completely asymmetrical paddling position—paddling only on one side of the boat from a kneeling position—whereas cyclists pedal in a predominantly symmetrical posture. Nonetheless, in the leg press test, asymmetry in Vavg was slightly greater in cyclists (9.86 ± 7.36%) than in canoeists (8.49 ± 6.28%).

Regarding the direction of the asymmetries, specific patterns emerged depending on the type of test performed. As highlighted in the literature, both the nature of the test and the applied load are key determinants of asymmetry direction [8,30]. For example, in our study, VL electromyographic activity showed a clear tendency toward the DL, whereas the categorical analysis suggested that at lower performance levels, there was a predominance of the NDL. Similarly, other parameters—such as BF muscle tone (Dm)—revealed that junior cyclists tended to show asymmetry favoring the DL, while in elite cyclists this shifted toward the NDL. This variability may reflect long-term training adaptations, where chronic exposure to asymmetrical loads in cycling, or strength training conducted without specific compensatory interventions, could result in adaptations that either strengthen or weaken a given limb, ultimately altering the direction of the asymmetry. Although these findings are encouraging, studies involving larger samples are required to confirm these patterns.

Finally, the asymmetry thresholds identified in this study must be interpreted within the context of sport-specificity demands. This notion is consistent with Loturco et al. [55], who highlight the importance of establishing functional criteria tailored to the unique requirements of each discipline. Our results demonstrate that asymmetry thresholds in competitive cyclists exhibit distinctive characteristics compared to those reported in other sports, thereby reinforcing the necessity of applying sport-specific thresholds for meaningful assessment and practical application.

Moreover, we develop a significant explanatory model for cyclists’ performance using W/kg achieved at VO₂max during a maximal cycle ergometer incremental test. Although this model only explains 23% of the variability through ROM, contractile properties (Tc of RF and VL) and force production lateral asymmetry, its practical utility is limited by this relatively low explanatory power. This underscores the complex, multifactorial nature of road cycling performance, which likely cannot be fully captured by isolated physiological and neuromuscular variables alone.

Traditional predictive models, such as those relying primarily on VO₂max, demonstrated stronger correlations with cycling economy/efficiency (r = −0.71) [4] and with time trial laboratory performance (r = 0.93) [56]. Furthermore, biomechanical models like Olds et al.’s [57] integrate environmental, physiological and biomechanical variables (i.e., body mass, frontal area, fractional VO₂max, etc.) to explain over 80% of time-trial performance variability, highlighting the importance of multifactorial approaches. However, the complexity and instrumentation requirements of such models limit their use in routine athlete monitoring.

Similarly, more recently, Leo et al. [58] proposed a novel model using a compound score approach that incorporates post-fatigue power metrics (5 min maximal power output after 2000 kJ) to predict podium success in under-23 racing (R² = 0.55). However, this model depends on accumulated race data and complex field testing. In contrast, our laboratory model offers a practical tool for standardized physiological and neuromuscular monitoring throughout the season without reliance on competitive history.

Nevertheless, future research should aim to develop multifactorial models that incorporate biomechanical, physiological, and performance data to improve predictive accuracy and practical relevance in cycling performance assessment. Overall, given the need for a comprehensive, multifactorial explanatory model for road cyclist performance, it appears that lateral asymmetry should also be considered alongside physical, physiological and mechanical performance indicators. Clearly, further research is still needed to develop a highly predictive and practical model.

The results of this study provide partial confirmation of our hypotheses. While no significant differences in the magnitude of lateral asymmetry were observed between categories, as we had proposed, the contribution of asymmetry variables (ROM, contractile properties, and force production) to maximal cycle ergometer incremental test showed only moderate explanatory power (R²_a = 0.23). This highlights the complex and multifactorial nature of cycling performance, where asymmetry may represent one of several relevant factors.

5. Study Limitations

One of the main limitations of this study is the lack of World Tour cyclists (Tier 5) [34], which limited our ability to address performance differences across varying competitive levels. While the primary focus of this study was on the influence of lateral asymmetry on athletic performance, exploring injury incidence would also have been valuable. However, the fortunate lack of a sufficiently large group of cyclists with common injuries prevented us from investigating this aspect. Additionally, assessing the behavior of lateral asymmetry throughout the competition season using a longitudinal design would provide important insights. Finally, it could be interesting for future studies to integrate systematic injury surveillance alongside asymmetry monitoring to determine whether specific asymmetry patterns could serve as predictors of injury risk, particularly for overuse injuries common in cycling. These limitations highlight key challenges and opportunities for future research on lateral asymmetry in well-trained cyclists.

6. Practical Applications

The findings of this study provide valuable tools for physical trainers to assess and monitor lateral asymmetry in road cyclists. By establishing test-specific asymmetry thresholds—such as TMG, ROM, force and EMG—trainers can design individualized training and prevention strategies. For example, an asymmetry in the AKE test exceeding the threshold established for the elite category (i.e., greater than 2.08%) would indicate the need for a specific intervention. In such cases, applying proprioceptive neuromuscular facilitation stretching techniques to the hamstring muscles of the limb with reduced mobility would be recommended. Alternatively, it may also be appropriate to strengthen the musculature of the limb with greater mobility through unilateral concentric exercises, with the aim of balancing the asymmetry by targeting either the limb with greater or lesser range of motion.

Furthermore, specific asymmetry thresholds make it possible to create cyclist-specific asymmetry profiles based on performance categories. Early detection of asymmetries can guide targeted interventions to address muscular imbalances or mobility restrictions that may negatively impact performance. In brief, our findings reinforce the value of incorporating lateral asymmetry assessments into comprehensive athlete monitoring strategies, in combination with other physiological, mechanical and performance indicators.

7. Conclusions

In conclusion, junior cyclists present a lower percentage of asymmetry in Dm of VM compared to elite cyclists, and a greater asymmetry in the AKE than elite cyclists. The magnitude of lateral asymmetry observed in this study ranged from 11.3% to 21.3% in contractile properties, 2.3% in ROM, 9.8% to 31.9% in force production, and 20.8% to 22.7% in VL electrical activity. Similarly, the percentage of cyclists classified as asymmetric varied between 23.6% and 43.6% in TMG parameters, 33.3% in ROM, 35.0% to 42.5% in force production, and 31.4% to 40% in VL EMG at physiological milestones. Finally, lateral asymmetry in ROM, contractile properties (Tc of RF and VL), and force production (Vavg) counted for 23% of the variance in cycling performance (W/kg at VO₂max).

These findings provide valuable refence values for lateral asymmetry across different assessments (TMG, ROM, force, EMG) in well-trained cyclists. They may be particularly useful for physical trainers and rehabilitation specialists aiming to monitor, interpret, and address asymmetries in competitive cycling.