Anthropometric,muscle power and kinematic characteristics of visually impaired sprinters and their guides: Case studies

Abstract

This study aimed to characterize the anthropometric, muscle power, and kinematic profiles of visually impaired sprinters (AVI) and their guides (GA) during the 100-meter dash. Three male guide pairs participated, all with national and international experience: Pair 1: AVI (T12); Pair 2: AVI (T11); Pair 3: (T11). Assessments included: (a) anthropometry, (b) lower-limb muscle power via countermovement jump (CMJ), and (c) kinematic analysis during an official 100-meter race using a global positioning system (GPS). GAs generally exhibited greater height, body mass, and longer lower limbs compared to AVIs. Except for Pair 3, GAs demonstrated superior CMJ performance and lower-limb power output. Peak running velocity (PV) was comparable across pairs, but AVI in Pair 1 and 2 achieved PV faster than their GAs. Pair 3 displayed pronounced asymmetries in instantaneous velocity and acceleration between AVI and GA during the race. The synchronization is influenced by experience duration, anthropometric compatibility, and muscle power disparities.

Introduction

The Paralympic Summer Games represent the premier international sporting event for athletes with different disabilities, featuring comprehensive classification systems across para-sport disciplines. ¹ Among these, the 100-m sprint stands out as one of the most decisive events for medal attainment, combining extreme competitive demands with high spectator and media appeal, similarly to its Olympic counterpart. ²

It is already known, that sprint performance is fundamentally linked to neuromuscular power and strength capabilities, ³ with ground reaction forces demonstrating biomechanical similarities between sprinting and vertical jumps.^4,5 However, for Paralympic athletes there are other factors to consider. Visually Impaired Sprinters (AVIs) classified in T11/T12 classes must race while tethered to guide athletes (GAs) via a 40-50 cm rope, creating unique biomechanical challenges that distinguish these events from able-bodied sprints.

Biomechanical analyses reveal GAs typically exhibit higher stride frequency, but reduced stride length compared to VIAs, resulting in prolonged ground contact phases and shorter flight times.^6,7 Previous evidence also confirm that GAs generally achieves higher velocities than their VIA partners.^8,9 This can be related to the fact that postural stability, and proprioception is affected by vision in VIAs, which result in abnormal gait and biomechanics. ¹⁰

These differences may also have important biomechanical and performance-related consequences. Mismatches in stride frequency, step length, or acceleration capacity can disrupt the temporal coordination of ground contact, leading to inefficient force transmission through the tether system and affect performance. Over repeated efforts, these mechanical inconsistencies may also elevate neuromuscular strain and fatigue, increasing the risk of overuse injuries for both the athlete and the guide. ¹¹

Another factor that directly influences the performance of the pair is the anthropometric disparities. For AVIs, the mandatory tether system introduces additional complexity, as anthropometric disparities between athlete-guide pairs may amplify biomechanical mismatches, potentially compromising race synchronization and performance outcomes. Despite the relevance of these factors to AVIs performance, no studies have quantitatively examined the synchrony between pairs during competitive 100 m dash running using comprehensive kinematic analysis of speed and acceleration profiles. Understanding how anthropometric and power differentials affect AVI-GA synchrony is crucial for optimizing training paradigms in Paralympic sprinting. ¹² This study therefore aims to: (1) characterize anthropometric, muscle power, and kinematic profiles of elite AVI-GA pairs, and (2) identify possibles relationships between these variables and 100 m performance outcomes.

Materials and methods

Participants

Three pairs of male athletes who trained and competed at national and international level took part in the study. All the methodological procedures used in the study were approved by the Ethics Committee for Research with Human of University (register number. 6.074.333).

Pair 1: AVI (T12, 22 years, 74.7 kg and 180 cm); GA (40 years, 84.9 kg and 175 cm). They have been training for 4 years and their best performance was 3^rd place in the world youth tournament - Gymnasiade held in France in 2022.

Pair 2: AVI (T11, 37 years, 78.7 kg and 174 cm); GA (36 years, 83.0 kg and 181 cm). They has been training for 8 years, with his best performances being 2 Paralympic gold medals, 1 gold in the London 2012 edition; 1 gold in the Rio 2016 edition in the 4x100 T11/T13 relay; 3 silver medals in the 100 m, 200 m and 400 m in the Rio 2016 edition; 1 bronze in the 100 m in the London 2012 edition; and gold in the 100 m in the Parapan American Games in Santiago - CL, 2023.

Pair 3: AVI (T11, 21 years, 57.2 kg and 167 cm); GA: (20 years, 64.7 kg and 175 cm). They have been training for 3 years and their best performances were winning the gold medal at the Brazilian under-20 championships in 2022 and the gold medal at the Brazilian University Games - JUBs in 2022.

Parameters assessed and instruments used

Anthropometric parameters

The anthropometric parameters assessed were height (cm), body mass (kg), wingspan (cm), right upper limb length (RULL), left upper limb length (LULL), right lower limb length (RLLL) and left lower limb length (LLLL). A stadiometer, scales and segmometers were used. All procedures followed the guidelines of the International Society for the Advancement of Kinanthropometry (ISAK).

Countermovement jump (CMJ) assessment

The lower-limb neuromuscular function was accessed through countermovement jump (CMJ) height accessed by Optojump Next^® (Microgate^®, Italy) device. The athletes performed three attempts of CMJ starting from a standing position with the trunk erect and knees extended, utilizing the stretch-shortening cycle (i.e., knee flexion and extension) and maintaining extended knees during flight. ¹³ During the test, the athlete was instructed not to use upper limbs to assist with propulsion and to remain with his knees extended during the flight. Statistical analysis considered the mean of the attempts.

Kinetic and kinematic variables

The 100 m data were collected using a portable 10-Hz GPS unit (Catapult Sports^®, Melbourne, Australia), previously validated for the assessment of instantaneous velocity during high-speed, field-based running. Each athlete wore the GPS device in a manufacturer-designed harness positioned between the scapulae to ensure optimal satellite reception and data accuracy. Prior to data acquisition, a satellite lock was established for a minimum of 15 min to optimize signal stability. Raw GPS data were imported into an automated processing pipeline for variable standardization. Individual race performance was automatically identified from the time distance series based on odometer discontinuities, temporal inconsistencies, or the reinitialization of distance values following substantial progression. Each valid performance was segmented from the starting point to 400 m into fixed spatial intervals of 2 m. For each segment, mean running speed was calculated using a time-weighted average based on the sampling interval (Δt), thereby minimizing biases associated with irregular GPS sampling.

Finally, instantaneous running speed (Rs) was derived from Doppler-based velocity estimates, and peak running velocity (PV) was defined as the highest instantaneous speed achieved during the sprint. In addition, the time to reach peak running velocity (TP) was defined as the instant at which maximal running speed was achieved. Acceleration (Acc) was calculated from the temporal variation of horizontal velocity variation of horizontal velocity, which was automatically calculated by Catapult’s proprietary software (OpenField™). Thus, acceleration corresponds to the first derivative of velocity with respect to time and is expressed in meters per second squared (m·s^-2).

The asymmetry in speed and instantaneous acceleration between the athletes was calculated by the difference between the athletes’ performance in each variable, where positive values indicated a predominance of GA over AVI while negative values indicated a superiority of AVI over GA.

The peak muscle power output of the lower limbs (PPO) was calculated using the equation proposed by, ¹⁴ where:

$Po\text{wer}\hspace{0.17em}\left(W\right)=61,9\hspace{0.17em}*\hspace{0.17em}\text{height}\hspace{0.17em}\text{jump}\hspace{0.17em}\left(\text{cm}\right)+36,0\hspace{0.17em}*\hspace{0.17em}\text{body}\hspace{0.17em}\text{mass}\hspace{0.17em}\left(\text{kg}\right)\hspace{0.17em}–\hspace{0.17em}1822.$

Data collection procedures

Data collection took place over one week during the National Championship, in final events (each pair competing in a single race), which served as a qualifier for the 2024 Parapan American Games.

The assessments were carried out in stages. One hour before the race, the pairs underwent anthropometric assessments in the laboratory. After the anthropometric assessment, the athletes performed a standardized warm-up lasting six (6) minutes, consisting of exercises that they regularly perform in their training routines. Subsequently, for muscle power assessment, three (3) countermovement jump (CMJ) attempts were performed, with a one-minute interval between attempts. For kinematic data collection, GPS units were fitted using a vest to both the guide athlete (GA) and the visual impairment sprinters (AVI) thirty minutes before the competition and were removed only after the race. Kinematic data were collected during the 100-m sprint.

Data analysis

Given the very small sample size and the paired nature of the dataset (three athlete–guide dyads), conventional inferential statistics were deemed inappropriate, as group-level analyses would violate key statistical assumptions and potentially obscure meaningful individual responses. Therefore, a descriptive and case-based analytical approach was adopted. Data were presented at the individual and pair levels, allowing direct comparison between each athlete and their respective guide. Differences between pairs were examined using absolute and relative values, as well as graphical visualization of velocity, peak running velocity, and acceleration profiles across the 100-m sprint. Accordingly, results should be interpreted as pair-specific performance profiles rather than generalized population estimates. To do this, we used Microsoft Excel spreadsheets. The graphs were building using de GraphPad Prism® software.

Results

Table 1 shows individual values and relative differences (%) between GA and AVI regarding anthropometric characteristics. The highest relative differences were found for RULL at 12.4% (Pair 2) and RLLL at 10% (Pair 3), while in the other parameters analyzed, the differences were less than 10%.

Table 1.

Individual anthropometric characteristics values and relative differences between guide-athletes and visual impairment sprinters.

Pairs		RULL (cm)	%	LULL (cm)	%	RLLL (cm)	%	LLLL (cm)	%	Arm span (cm)	%
Pair 1	GA	76.5	−5.2	75.3	−4.6	95.0	+0.1	94.5	−0.3	174	+6.3
	AVI	72.5		71.8		95.1		94.2		185
Pair 2	GA	83.7	−8.6	84.0	−8.6	93.2	−5.2	93.0	−6.4	180	−2.2
	AVI	76.5		76.8		88.3		87.8		176
Pair 3	GA	76.5	−6.5	75.3	−4.4	95.0	−10.0	94.5	−9.7	174	−2.6
	AVI	71.5		72.0		85.5		85.3		170

Legend: GA: Guide Athlete; AVI: Visual Impairment Sprinter; RULL: Right Upper Limb Length; LULL: Left Upper Limb Length; RLLL: Right Lower Limb Length; LLLL: Left Lower Limb Length; %: Relative difference between GA and AVI; Arm Span: Full reach from fingertip to fingertip.

Table 2 shows the individual CMJ and peak lower limb power (PPO) values and percentage differences (%) between GA and AVI. Except for Pair 3 (CMJ +19.8%; PPO -0.6%), the GA showed higher jump height and PPO than AVI being the highest difference observed in Pair 1 (CMJ -20.8%; PPO –31.6%).

Table 2.

Individual data and percentage differences of jump height and peak muscle power between guide-athletes and visual impairment sprinters.

Pairs		CMJ (cm)	%	PPO (W)	%
Pair 1	GA	54.8	−20.8	5129.1	−31.6
	AVI	43.4		3505.7
Pair 2	GA	56.4	−8.5	5141.0	−9.9
	AVI	51.6		4633.2
Pair 3	GA	36.3	+19.8	3991.9	−0.6
	AVI	43.5		3966.9

Legend: GA: Guide Athlete; AVI: Visual Impairment Sprinter; CMJ (cm): Countermovement Jump Height; PPO (W): Peak Power Output; %: Relative difference of GA compared to AVI.

We found that AVI from Pairs 2 and 3 have reached higher than their respective GA (Table 3). Pair 3 had the lowest PV and the greatest differences between AVI-GA (1.55%), which may have influenced the synergy of the race and consequently the longer time taken.

Table 3.

Individual data 100-m time, peak running velocity, time to peak running velocity, percentage differences between guide-athletes and visual impairment sprinters.

Pairs		PV (m/s)	%	Time to PV (s)	%	Other Athlete’s speed at GA PV (m/s)	Trial time (s)
Pair 1	GA	9.64	−1.14	7.30	−9.59	9.52	11.96
	AVI	9.53		6.60		9.55
Pair 2	GA	10.13	+1.28	7.98	−8.52	10.00	11.26
	AVI	10.26		7.30		10.04
Pair 3	GA	9.03	+1.55	4.20	+21,42	8.42	12.24
	AVI	9.17		5.10		8.98

Legend: GA: Guide Athlete; AVI: Visual Impairment Sprinter; PV: Peak Running Velocity.

Table 3 shows the individual values and percentage differences for the maximum speed reached during the race, the time taken to reach the maximum speed, the speed at which the partner was at the time their pair reached the maximum speed and the trial time.

Figure 1 shows the individual running speeds, the differences between the speeds and the accelerations of the AVI and GA. GPS data revealed that, although Pair 1 did not reach the best performance (11.96 s), it showed the least variation in running speed (Figure 1(b)) among the pairs (-1.14%), indicating greater symmetry between the athletes. Pair 3, on the other hand, not only had the worst performance (12.24 s) but also showed the greatest variations in running speed (1.55%) and accelerations, indicating a high degree of asymmetry between the athletes.

Figure 1.

Running speed, difference in speed between AVI and GA and accelerations during the 100 m race for Pair 1 (Figures a, b, c), Pair 2 (Figure d, e, f) and Pair 3 (Figure g, h, i).

Discussion

The aim of this study was to evaluate the anthropometric, muscle power, and kinematic characteristics of three pairs of visual impairment sprinters (AVIs) and their respective guides (GAs) during 100-meter sprint races. The main finding here was that for Pair 1, the lower variation in speed and acceleration throughout the race may be associated with anthropometric similarity between the athlete and the guide, in both, upper (RULL: -5.2%; LULL: -4.6%) and lower limb (RLLL: 0.1%; LLLL: -0.3%) as well as higher power values in favor of the GA (-31.6%).

Another important result found here was that although Pair 2 had the best performance in the race (11s26), they showed greater variation in speed (1.28%) and acceleration (Figure 1) during the race when compared to Pair 1 (-1.14%). We believe this is maybe due to a greater anthropometrics difference in the upper (Pair 2: RULL: -8.6% and LULL: -8.6%; Pair 1: RULL: -5.2% and LULL: -4.6%) and lower limbs (Pair 2: RLLL: -5.2%; LLLL: -5.4%; Pair 1: RLLL: 0.1%; LLLL: -0.3%) and a smaller difference in power parameters between AVI and GA (Pair 2: -9.9%; Par 1: -31.6%) which, added to their extended experience in high level competition (eight years) could help them adjust their pace during the race.

Finally, the greatest variations in speed and acceleration during the test were observed in Pair 3 (Figure 1). We believe that this is since this pair has the greatest anthropometric differences, especially in the lower limbs (RLLL: -10,0%; LLLL: -9,7%), and the fact that GA’s PPO is very close to the values observed for AVI (0.6%), which would require GA to make a very high effort during the test. Additionally, the shortest training time (3 years) could contribute to the pair’s asynchronism during the race.

Maximum running speed is the combination of stride length and stride frequency. In athletics speed events, stride length is influenced by the length of the lower limb, strength, mobility, technique and other factors.^15,16 According to Pereira et al. ² guide athletes should have greater power than AVI athletes to avoid decrease in performance during the intermediate and final phases of the race. In this sense, pairs with greater differences in length of their lower limbs tend to have a greater mismatch in stride length and in the number of strides needed to complete the race distance.

This shows that, although anthropometric differences may exist between the athlete and the guide, the functional capacity of the athlete–guide system can be crucial for the pair’s performance. We believe that the GA should have greater power in order to have sufficient physiological and mechanical margin to adapt to the demands of the race. This means that the guide is able to instantaneously reduce or increase power output without entering premature fatigue, synchronize the pace with the AVI, compensate for small variations in speed, stride, or balance, and respond to course demands such as curves, accelerations, decelerations, and changes in pace. This ability to adjust is crucial during the intermediate and final phases of the race, when fatigue in the AVI tends to increase. Thus, if the GA had similar or lower power, they would likely be unable to compensate for these variations without compromising their own performance.

It can be reinforced looking at the differences in speed between the Pairs during the race (Figure 1). In this analysis a greater oscillation was found in the pair with the lower performance and jump height, power and peak running velocity. These greater oscillations are possibly due to the need for greater adjustments between the athletes to maintain synchronization according to their performance levels. These differences may impact in the time to reach the peak running velocity synchronism. Although there are no studies that can point to clearer parameters regarding the differences or ideal minimum/maximum zones of linear speed that pairs of sprinters can reach during a real race, this study provides important insights into these differences (Table 3).

The speed adjustments made by both athletes (GA and AVI) in the race also explain the increases and decreases in speed in order to remain synchronized (Figure 1). However, adjustments at unwanted levels or beyond what is necessary to comply with the rules could change the mechanics of the race, causing a decrease in flight time, kinematic patterns and a natural decrease in speed.⁽⁸⁾ In this context, adjustments in speed refer to intentional accelerations or decelerations imposed by the GA to maintain synchronization with the AVI or to comply with competition rules.

Regarding acceleration behavior during the race (Figure 1), it can also be seen that Pair 3 needed more adjustments, especially after reaching the maximum speed peak. This could be explained by the fact that they reached the peak running speed early (Table 3), with a concomitant lower jump height in GA and muscle power performance similar to the AVI (Table 2). Furthermore, they are less experienced in age and training as a pair. Considering jumping performance,^4,17 state that the characteristics of the mechanics vertical jump tests are also presented during the speed run, which may explain the greater need for acceleration adjustments in Pair 3.

It is therefore important to observe the individual differences in acceleration, peak running velocity and the time that both athletes achieve, as they will be decisive in the synergistic relationship between the guide and the visual impairment sprinters. In this sense, ¹² state that synchronization between the visual impairment sprinters and the guide is an important principle in sprint performance and especially during the initial acceleration phase, to obtain a better running performance and better race speed. This makes it necessary for the differences in individual physical attributes of lower limb power and speed to be in favor of the guide athlete. In a study by Pereira et al., ² the authors observed that the average differences between the visually impaired sprinters and their guides in the sprint tests amounted to 10%, varying between 1-24%. Therefore, substantial differences in sprinting speed (in favor of the guides) between the peers were observed.

The study has both strengths and limitation. One limitation is that we only evaluated three pairs during a single race may limit the generalization of the results. In this sense, considering that a 100 m race in classes T11 and T12 will always be run with the participation of AVI and GA, we suggest that further studies investigate the behavior of kinematic parameters in a greater number of races and athletes, to identify whether our findings are replicable, or whether they were the result of chance. Despite this limitation, our study was a pioneer in integrating anthropometric measurements, lower limb muscle power and the use of a GPS monitoring system in real competition situations. We believe that our study has created insight to better understand the relationship between AVI and GA during races and thus better plan training routines.

Conclusion

Our findings demonstrate that AVI-GA synchronization in 100m races may depends on: (1) anthropometric compatibility (smaller limb length differences = better synchrony, e.g., Pair 1); (2) power differentials (GAs with superior CMJ/PPO showed more stable dynamics); and (3) joint experience (Pair 2’s 8-year partnership overcame anthropometric mismatches for fastest time). Pair 3’s large anthropometric gaps + balanced power may have contributed to pronounced asymmetries and slowest time.