Comparative Kinematic Analysis of Patellar vs. Hamstring Autografts in ACL Reconstruction on Side-Hop Test Performance

Abstract

This study aimed to analyze the biomechanical differences in knees that underwent reconstruction using four-strand hamstring or bone–patellar tendon–bone autografts during the side-hop test. This case–control study included 46 participants: 18 controls, and 28 individuals who underwent reconstruction with hamstring (n = 15) or patellar (n = 13) tendons. The Theia Markerless system and Visual3D provided information on knee kinematics, namely time of contact with the ground, maximum varus and valgus angles, and flexion during maximum valgus and varus. Additionally, we assessed the activity levels of the participants pre- and post-surgery using the Tegner Activity Scale. Data analysis was conducted using ANOVA and Tukey’s post hoc test. Significant differences were observed in terms of contact time with the ground between the control and autograft groups (p = 0.025, g = −1.13; p = 0.014, g = −1.09), but not between the autograft groups. Other variables did not demonstrate statistically significant differences; however, there was a slight tendency to greater valgus in patellar autografts and greater varus with hamstring tendons. The absence of significant differences between the groups may indicate that both autografts allow knee kinematics restoration by engaging neuromuscular and proprioceptive mechanisms that compensate for anatomical alterations.

1. Introduction

The Anterior Cruciate Ligament (ACL) is a crucial stabilizer of the knee, limiting the forward translation of the tibia in relation to the femur and preventing excessive internal rotation of the knee [1,2]. It consists of two bundles: the anteromedial bundle, which functions during flexion to restrain anterior tibial translation, and the posterolateral bundle, which provides rotational stability during extension [2,3]. ACL tears account for approximately 20% to 50% of all knee injuries [3,4]. These injuries affect both athletes and the general population, and have shown an increasing tendency across the years [5], particularly due to a rise in sports participation [4,6,7].

While conservative treatment may be appropriate for cases of partial tears [3], it is typically reserved for less active patients or those with significant comorbidities due to the increased joint instability, greater difficulty with certain activities, such as jumping or descending stairs, lower rates of return to sport, and an increased risk of developing osteoarthritis [8]. Nevertheless, the preferred treatment is surgical reconstruction (ACLR) [7,8], although there is no gold standard for which is the best graft choice. The two most common techniques involve the use of four-strand hamstring tendon (HT) and bone–patellar tendon–bone (PT), each with its own advantages and disadvantages [7]. PT grafts are associated with faster incorporation, greater stability, better return to pre-injury levels, and decreased incidence of graft failure (1.9% vs. 4.9%). Nonetheless, they are more likely to have an extension deficit greater than 3 degrees and more anterior knee pain when kneeling [1,7,9]. Due to the greater stiffness of the PT autograft and the rigid fixation that bone-to-bone healing allows, it was expected that knees reconstructed with this graft would show less valgus, as well as less general laxity, as the ones reconstructed with HT autografts [10]. HT grafts are associated with reduced extensor strength deficits, a lower probability of osteoarthritis development, decreased incidence of a postoperative range of motion complications requiring re-intervention, fewer donor-site morbidities, and improved cosmetic outcomes. However, it predisposes for a greater valgus knee alignment on initial contact during single-hop landing and greater flexion and internal rotation deficits, especially if both the gracilis and semitendinosus were harvested [1,6,7,11,12,13]. The choice of autograft should be shaped to the patient’s necessities and expectations, as well as the surgeon’s experience and preference, to achieve the best possible outcomes [13,14].

Even though ACLR restores knee stability, improves functional outcomes, and prevents further articular damage—specifically cartilage injury, meniscus tears, and osteoarthritis [15]—a significant number of patients still fail to regain their pre-lesion levels of movement [5], with only 65% of athletes returning to pre-injury levels of participation [6,16,17]. Two factors that may contribute to these unsatisfactory outcomes are the lack of standardized return-to-sport guidelines and inadequate assessments to address physical and psychological impairments [5,18,19]. Return-to-sport criteria following ACLR include the time since surgery and subjective measures such as pain and instability, the absence of effusion, range of motion normalization, quadriceps strength, and performance on the side-hop test [5,18,20].

The side-hop test is a pivotal test in return-to-sports evaluation because not only is it capable of analyzing muscle strength and endurance but it can also give information on proprioceptive and neuromuscular adjustments that only a dynamic test can provide [17,19,21]. Its unique ability to challenge dynamic knee stability in the frontal and transverse planes makes it particularly suited for assessing biomechanical parameters such as knee valgus/varus angles and joint movements. It introduces lateral displacement and rapid changes in direction, which place greater demands on the knee’s ability to stabilize, mimicking the type of biomechanical challenges encountered during sports-specific activities, thus providing more ecologically valid insights into functional joint stability post-reconstruction [22,23]. However, it is still a subjective measurement, and its combination with motion capture systems can be helpful to provide more objective data that could integrate rehabilitation and return-to-sports protocols, potentially improving clinical outcomes.

Previous studies [24,25] have shown that the association of force plates and motion capture methods can identify risk patterns for ACL injury, such as asymmetry in the number of hops, peak knee flexion, vertical impact force, flight and contact time, and jump distance between legs. Even though marker-based methods are the gold standard for this type of analysis, it introduces a subjective variable, being highly dependent on the correct position of the sensor, which is influenced by the researcher’s experience and the morphology of the individual’s landmarks. In addition, the signal can be affected by skin and movement artifacts or not recognized by the system [26]. Recent markerless technologies eliminate these disadvantages by creating a 3D representation of the skeleton through images captured by various cameras, providing information on joint angles of the entire body in a more objective manner, being less time-consuming and easier to use [26]. Although markerless systems have demonstrated strong correlations with marker-based systems for sagittal plane kinematic analysis, their correlations in the frontal and transverse planes have been comparatively weaker. Therefore, the interpretation of data from planes of motion involving smaller angular displacements should be exercised with caution [27,28]. While not being exempt from faults, it is a promising technology that has already been validated many times, including against marker-based systems [29,30,31].

This study aimed to examine potential differences in knee kinematics between ACLR using HT or PT autografts during the side-hop test. Additionally, it sought to determine whether the choice of autograft influences knee functionality beyond 12 months postoperatively. It was hypothesized that (i) operated limbs would show longer periods of contact with the ground between hops and (ii) HT autografts would show greater valgus knee alignment during the contact of the limb with the ground, (iii) as well as greater flexion deficits.

The novelty of this study lies in its contribution of new data comparing HT and PT autografts at 12 months postoperatively within a single clinical setting, utilizing a functional performance test. Additionally, this study introduces the application of markerless motion capture technology to the investigation of knee kinematics following ACLR. The implementation of standardized and automated data analysis protocols further enhances the reproducibility and comparability of the results, supporting their potential utility as reference values in future research.

2. Materials and Methods

2.1. Participants

Two groups were recruited for this study. The first group consisted of patients who underwent ACLR and formed the case group. The second group comprised healthy individuals who served as the control group.

The case participants were recruited from the list of patients who underwent ACLR at ULS São João, a tertiary referral center in Porto (Portugal), between the 1st of July of 2020 and the 1st of July of 2023. Participants were included in this group if they met the following criteria: (i) aged between 18 and 50 years; (ii) had undergone ACLR with HT or PT autograft; (iii) the surgery had been performed at least 12 months prior; (iv) had no musculoskeletal injuries in the previous 6 months; and (v) were able to walk without the use of crutches, canes, or other walking aids. Excluded were those that underwent ACLR with quadriceps autograft or those who had concomitant extra-lateral tenodesis techniques or meniscus sutures. Patients with meniscus lesions were included in this study only if no repair or meniscectomy had been performed. All participants followed a rehabilitation protocol established by physiatrists from the institution, who monitored the patient’s progress and adjusted the protocol according to their individual needs. Participants’ contact information was obtained from their medical records, and contact was initiated via phone call. During this call, the project was explained, and if the participant agreed, a session was scheduled for data collection. Information regarding the type of autograft, concomitant lesions, and time since ACLR was extracted from the participants’ medical records.

Control participants were recruited from a convenience sample and included in this group if they met the following criteria: (i) age between 18 and 50 years; (ii) no history of previous lower limb surgeries; (iii) absence of pain or knee effusion; and (iv) no history of neuromuscular diseases. This case–control study was reviewed and approved by the ULS São João Center Review Board. This study was conducted under protocol number 162/2024, which was approved on 3 October 2024.

2.2. Data Collection Equipment

Participant anthropometric measurements included body mass and height, which were measured using an InBody 230 (Biospace, Seoul, Republic of Korea) and a Seca 206 stadiometer (seca, Hamburg, Germany), respectively.

A motion capture system was employed to record full-body movements during side hopping. The system consisted of eight Miqus Video Cameras (Qualisys AB, Gothenburg, Sweden), operating at a sampling frequency of 100 Hz and a resolution of 720 p. Prior to data collection, a 2.0 × 2.0 × 2.0 m performance volume was calibrated with an error < 0.50 mm. Two 60 × 40 cm force platforms (Bertec Inc., Columbus, OH, USA) positioned side by side and fixed to the ground were used to record ground reaction forces at a 2000 Hz sampling frequency and in synchrony with the motion capture system. The Qualisys Track Manager version 2024.2 (Qualisys AB, Gothenburg, Sweden) software was employed to collect both video and force data.

After arriving at the laboratory, participants were given an informed consent form that provided additional details about the planned experimental procedures. If they agreed to participate, they signed the form and were asked to change into more comfortable garments (i.e., shorts and a T-shirt) while remaining barefoot. The Tegner Activity Scale (TAS) was applied to the participants in the case group prior to the beginning of the experimental procedures [32].

A researcher provided a verbal explanation of the side-hop exercise, followed by the demonstration of the movement: stand on one foot (right or left) on a force platform and then hop side to side, alternating between platforms while using the same foot, without performing double taps, for a duration of 15 s. The participant was allowed to try the movement once with each foot. After a rest period of 3 min, data collection was initiated. Three repetitions of side-to-side hopping were recorded for each foot.

2.3. Data Processing

Data processing was conducted using the Qualisys Functional Assessment v.2.4.0 plugin (Qualisys AB, Gothenburg, Sweden). This included using Theia Markerless v2022.1.0.2309 p20 (Theia Markerless, Kingston, ON, Canada) to compute a six-degrees-of-freedom biomechanical model from the video recordings and to integrate it with the ground reaction forces. The resulting motion and force data were then imported into Visual3D v2021.11.3 (HAS-Motion, Kingston, ON, Canada), where further data processing and extraction were performed.

For this study, the following variables were measured during limb contact with the ground: the duration of contact with the ground (CTG), maximum knee varus and valgus angles, and flexion at maximum varus and valgus. In the control group, the lower limbs (left and right) were randomly selected for analysis to mitigate potential effects of limb dominance. This was conducted using a computer-generated random sequence in R v2.4.2 (R Core Team 2024, Vienna, Austria), resulting in the representation of the group by a single limb. In the case group, only the operated limb was analyzed, regardless of its laterality.

2.4. Statistical Analysis

The statistical analysis was conducted using R v.2.4.2 (R Core Team 2024, Vienna, Austria). The normality of the data distribution was assessed using the Shapiro–Wilk test. In groups where the sample distribution was non-normal, the data were transformed using the logarithmic function. This transformation aimed to reduce skewness and achieve a more symmetrical distribution, thereby approximating a normal distribution.

For the comparison of descriptive variables between groups, either Fisher’s Exact Test or the Wilcoxon Rank Sum Test was applied, depending on whether the variable was categorical or numerical. An ANOVA was performed to assess the presence of interactions between groups, followed by Tukey’s post hoc test. The effect size was calculated using Hedges’ g (g) and interpreted as small (0.2), medium (0.5), large (0.8), or very large (>1) [33]. All statistical tests were performed considering an α = 0.05. A post hoc power analysis was conducted to evaluate the statistical power of the tests based on the observed effect sizes, significance level (α = 0.05), and sample size.

3. Results

3.1. Recruited Groups

A total of 46 volunteers (11 females and 35 males) participated in this study, divided into a case group (n = 28) and a control group (n = 18). Summary information regarding these groups can be found in

Table 1. Descriptive statistics of the case and control groups.

Parameter	Case Group n = 28	Control Group n = 18	p-Value
Gender			0.29
Female	5 (18%)	6 (33%)
Male	23 (82%)	12 (67%)
Age (years)	32.5 (18.25)	31.50 (12.50)	0.12
Body mass (kg)	82.0 (20.83)	75.0 (21.45)	0.04 *
Height (m)	1.77 (0.09)	1.71 (0.12)	0.16
Limb dominance			0.99
Right	22 (79%)	15 (83%)
Left	6 (21%)	3 (17%)

Continuous variables are presented as median (interquartile range). Categorical variables are presented as frequency (percentage). An asterisk (*) denotes a significant difference (p < 0.05).

.

No significant differences were observed in age (p = 0.12), height (p = 0.16), and gender (p = 0.29) between groups; however, the case group exhibited a greater body mass (p = 0.04). Both groups were predominantly right-limb dominant (p = 0.99). The case group was further divided into two subgroups based on the type of autograft selected for the ACLR: HT (n = 15) or PT (n = 13). The specific characteristics of the participants in each subgroup can be found in

Table 2. Descriptive statistics of the autograft subgroups of the case group.

Parameter	HT n = 15	PT n = 13	p-Value
Gender			0.64
Female	2 (13%)	3 (23%)
Male	13 (87%)	10 (77%)
Age (years)	37.00 (15.00)	29.00 (9.50)	0.09
TAS variation	0.00 (−4.50)	−2.00 (−3.00)	0.88
Body mass (kg)	84.30 (21.60)	75.00 (23.60)	0.49
Height (m)	1.77 (0.08)	1.75 (0.16)	0.98
Limb dominance			0.65
Right	11 (73%)	11 (85%)
Left	4 (27%)	2 (15%)
Injured Limb			0.04 *
Right	2 (13%)	7 (54%)
Left	13 (87%)	6 (46%)

Continuous variables are presented as median (interquartile range). Categorical variables are presented as frequency (percentage). An asterisk (*) denotes a significant difference (p < 0.05). TAS: Tegner Activity Scale.

.

As can be verified in Table 2, no significant differences were observed between HT and PT autograft participants regarding age (p = 0.09), body mass (p = 0.49), height (p = 0.98), limb dominance (p = 0.65), and TAS variation (p = 0.88). A significant difference was found (p = 0.04) in the injured leg distribution, with the HT group showing more cases on the left limb (87%), while the PT group has more participants with lesions in the right limb (54%).

The levels of physical activity, as measured by the TAS, did not exhibit a significant variation between post- and pre-surgery (post–pre) (p = 0.88). Nevertheless, the PT group demonstrated a median decrease of 2 levels (−3.00), while the median for the HT group was 0 (4.50). These findings can be observed in Table 2.

3.2. Side-Hop Kinematics

Differences in the logarithm of CTG were observed among the experimental groups (F(2,43) = 5.785, p = 0.006), with the surgical limbs exhibiting a longer contact time in HT (p = 0.025, g = −1.13) and PT (p = 0.014, g = −1.09) compared to the control group. No differences were found between HT and PT (p = 1). These results are presented in

Table 3. Comparison of CTG, maximum valgus, flexion in maximum valgus, maximum varus, and flexion in maximum varus in the control and case groups and in the autograft subgroups.

Parameter	Units	Control Group	Autograft Group
			HT	PT
CTG	S	−0.57 ± 0.11 [−0.68; −0.46]	−0.46 ± 0.19 a [−0.65; −0.27]	−0.46 ± 0.20 a [−0.66; −0.26]
Max Valgus	Deg	3.88 ± 3.17 [0.71; 7.05]	2.27 ± 3.89 [−1.62; 6.16]	4.55 ± 3.34 [1.21; 7.89]
Max Varus	Deg	0.68 ± 3.07 [−2.39; 3.75]	2.54 ± 3.40 [−0.86; 5.94]	0.97 ± 2.75 [−1.78; 3.72]
Flexion in Max Valgus	Deg	31.80 ± 8.51 [23.29; 40.31]	30.90 ± 6.24 [24.66; 37.14]	34.90 ± 5.45 [29.45; 40.35]
Flexion in Max Varus	Deg	27.70 ± 6.64 [21.06; 34.34]	27.10 ± 8.47 [18.63; 35.57]	25.80 ± 8.45 [17.35; 34.25]

The p-value given for CTG refers to the analysis of the log (CTG). CTG: contact time with the ground. Max: maximum. Deg: degree. A superscript letter indicates significant differences (p < 0.05) in relation to (a) the control group, (b) the HT subgroup, and (c) the PT subgroup.

.

As shown in Table 3, the analysis revealed no significant differences in maximum varus among the groups (F(2,43) = 0.543, p = 0.585). Although no statistical significance was identified, participants in the HT group demonstrated a mean maximum varus that was slightly greater than that of the PT and control groups. Similarly, no significant differences were observed when comparing maximum valgus across the groups (F(2,43) = 0.795, p = 0.458). However, participants with PT autografts appeared to achieve a greater mean valgus during ground contact (4.55°), followed by healthy controls (3.88°), and finally the HT participants (2.77°).

Flexion did not differ significantly between groups, whether measured at maximum varus (F(2,43) = 0.086, p = 0.915) or maximum valgus (F(2,43) = 1.135, p = 0.331). The median values for flexion at maximum varus were very similar across all groups. However, the PT group demonstrated a slightly greater mean flexion during maximum valgus (34.9°) compared to the HT (30.9°) and control (31.8°) groups. These results are presented in Table 3.

3.3. Post Hoc Power Analysis

The post hoc power analysis revealed limited statistical power in the comparisons between the HT and PT groups. Moderate effect sizes were observed for maximum valgus (g = 0.60) and flexion at maximum valgus (g = 0.65), with corresponding power values of 0.34 and 0.38, respectively. Other variables, including CTG (g = 0.10, power = 0.06) and flexion at maximum varus (g = 0.15, power = 0.07), showed small effect sizes and low power. These results reflect the constraints of the current sample size (n = 28, α = 0.05) in detecting potential differences between groups. This information is detailed in

Table 4. Post hoc power analysis between HT and PT groups.

Parameter	Effect Size (g)	Power (1-β)
CTG	0.10	0.06
Max Valgus	0.60	0.34
Max Varus	0.48	0.23
Flexion in Max Valgus	0.65	0.38
Flexion in Max Varus	0.15	0.07

Sample size was 28, and α error probability was 0.05. CTG: contact time with the ground. Max: maximum.

.

4. Discussion

The primary objective of ACLR is to restore the patient’s knee function after an ACL tear. To achieve this, two main surgical techniques are available, but with somewhat conflicting results reported in the scientific literature [1,14,15,34], particularly regarding the ability to return to pre-injury performance and return to sports [35]. However, most of these studies were conducted within 12 months post-surgery [12,36], a period during which the participants may still be adapting and undergoing physical therapy. Furthermore, during this period, patients may still experience symptoms of kinesiophobia, which can diminish their mobility and confidence in the operated limb [37].

Despite the gold standard for clinical motion capture being marker-based optical cameras, this study utilized a markerless motion capture system. This approach eliminated errors associated with poor marker placement or skin movement artifacts, while still ensuring a high level of reproducibility and accuracy. Additionally, this method is not significantly affected by the participant’s clothing [38], promoting a more comfortable and natural range of motion.

This study only identified significant differences in contact time with the ground between the case and control groups. This suggests that, regardless of the ACLR technique employed, patients require longer periods of ground contact. This extended contact time may be attributed to several factors, including adaptive mechanisms that necessitate more time to control movement or a decline in muscle function regarding myoelectric activation and force production following surgery [39]. After an ACL injury and subsequent reconstruction, the disruption of mechanoreceptors within the ligament and associated structures can impair joint position sense and neuromuscular control, leading to compensatory strategies during dynamic tasks [40,41]. Quadriceps inhibition is also frequently reported post-ACLR and can contribute to prolonged stabilization phases during landing [42]. Additionally, altered pre-activation and co-contraction patterns, particularly between the quadriceps and hamstrings, have been observed and are associated with diminished dynamic knee stability [43]. These neuromuscular adaptations may prompt individuals to adopt more cautious landing strategies, as evidenced by increased contact times, in an effort to mitigate mechanical load and perceived instability. Consequently, these findings underline the importance of rehabilitation protocols that not only target muscle strength but also emphasize the restoration of proprioception, neuromuscular timing, and coordination to achieve more normalized dynamic movement patterns.

Participants in the case group showed significantly greater body mass compared to the control group, which may be attributed to lower levels of physical activity. Although the TAS results indicated reduced physical activity levels after surgery with the PT autograft, this finding was not statistically significant. Nonetheless, this decrease could suggest a meaningful decline in athletic or daily performance, potentially due to differences in postoperative recovery trajectories, graft-specific rehabilitation challenges, or subjective factors such as anterior knee pain, which is more frequently reported in patients with PT autografts. It is important to note that such questionnaires are not exempt from limitations. For example, while participants may engage in the same types of activities post-surgery, their performance may be adversely affected, a nuance that this scale does not assess.

Psychological factors play a crucial role in sports participation, and fear of pain and re-injury can deter patients from engaging in certain movements [44,45]. Several participants expressed uncertainty about their ability to perform the side-hop test during data collection; however, they successfully completed it after attempting it. Compensatory strategies, such as increased joint stiffness, altered limb loading, and reduced knee flexion angles during dynamic tasks, may persist even in the absence of physical impairments, contributing to asymmetrical loading and an increased risk of re-injury [46,47]. As such, integrating psychological screening tools and addressing fear-avoidance behaviors through cognitive–behavioral interventions, graded exposure, and patient education may be essential components of a comprehensive rehabilitation program. Recognizing and treating psychological readiness as a modifiable factor could enhance both functional recovery and movement quality, ultimately supporting safer and more successful return-to-sport outcomes [48]. While these psychological factors are often overlooked in most studies of this nature, they can provide valuable context for the results, such as the extended duration contact with the ground between jumps.

Overall, neither autograft technique demonstrated significant repercussions in terms of knee kinematics, although some slight tendencies were observed. Unlike Welling et al. [36], we did not find differences in flexion between the case and control groups; however, the time since surgery was considerably shorter (8–10 months) compared to the post-surgery period in this study (12–48 months). This may lead us to conclude that flexion could be restored with a longer recovery period. A similar situation occurred with the knee valgus analysis, where Asaeda et al. [12] reported increased degrees of valgus alignment during initial contact, which does not align with our findings of non-significant differences; again, their evaluation was conducted closer to the time of surgery than ours. Furthermore, both studies utilized marker-based systems, which can also account for the discrepancies observed between their findings and ours. The contrast with our results may arise from various factors beyond the time since surgery and motion capture technology, which are challenging to control. First, the choice between using HT or PT is a clinical decision that may be influenced by the patient’s pre-lesion varus/valgus angle, which could explain the slight tendencies observed. This decision may also be significantly affected by the surgeon’s personal preferences and the team’s expertise with the chosen technique. Although patients with concomitant extra-lateral tenodesis techniques and meniscus sutures were excluded from this study, those with meniscus tears were included, and this anatomical variation may impact the results obtained. Furthermore, the medical records do not provide a comprehensive description of the surgical technique, fixation method, and graft tension—details that can lead to differing outcomes [1].

The motion capture technology that we employed may have had an impact on results. Marker-based motion capture systems are highly reliant on the analyst’s expertise and can be hindered by anatomical features (e.g., increased mass around the pelvis) or clothing (e.g., loose clothing). In contrast, markerless systems mitigate these issues, leading to more accurate results and greater inter-participant reliability. These advancements may have highlighted the lack of differences between HT and PT, especially given that more than 12 months have elapsed since surgery.

While the differences in frontal plane knee kinematics between graft types were not statistically significant, the observed trends—greater valgus in the PT group and greater varus in the HT group—may still warrant clinical consideration, particularly concerning dynamic joint loading and long-term joint health. However, we acknowledge that these findings must be interpreted with caution. Although markerless systems have demonstrated good agreement with marker-based systems in the sagittal plane, reduced accuracy has been shown in the frontal and transverse planes, particularly for joint rotations involving small angular displacements [49,50]. Therefore, while these kinematic trends may hold clinical relevance, especially regarding injury risk or the development of osteoarthritis, readers should interpret these results within the context of the technical limitations of markerless systems and the exploratory nature of our analysis.

Indeed, a previous study using a markerless approach [11] also found no significant differences between the HT and PT groups during drop jumps. In that study, participants dropped from an elevated platform and, upon contacting the ground, were required to control their landing and perform a small jump. In this study, the kinematics of landing were also analyzed, but participants had greater control over the movement performed. The consistent results across both studies further reinforce the likely absence of differences between HT and PT 12 months post-surgery.

Our findings, while preliminary, underscore the importance of considering not only structural integrity but also the potential implications for neuromuscular performance during functional tasks. Clinically, rehabilitation protocols for PT recipients may benefit from an increased emphasis on frontal plane control and plyometric retraining, whereas HT recipients may require targeted strategies to address deficits in knee extension strength. Although the current sample size limits definitive conclusions, the identified trends provide a rationale for individualized rehabilitation approaches and highlight the need for longitudinal studies with larger cohorts to validate these biomechanical distinctions.

The post hoc power analysis conducted revealed low power across most comparisons (ranging from 0.06 to 0.38); moderate effect sizes were observed for key parameters, such as maximum valgus (g = 0.60) and flexion at maximum valgus (g = 0.65). These results suggest that, although our study may have been underpowered to achieve statistical significance with the current sample size (n = 28), meaningful biomechanical differences between graft types may still exist. Notably, the moderate effect sizes in valgus-related measures align with the observed trends, supporting the clinical relevance of subtle alterations in knee stability following ACLR. Consequently, these findings emphasize the necessity for future research involving larger sample sizes to further investigate these biomechanical distinctions.

Limitations

Among the limitations of this study, the small sample size and the limited representation of male and female individuals hindered a comprehensive analysis of the effects of sex on the results. Additionally, the sample size also restricted the investigation of lower limb dominance’s impact on the outcomes [51]. Concurrently, the sample was only recruited from a single hospital, which means that all case group participants underwent surgery with similar teams and techniques. A multi-center approach could uncover interactions that are currently unclear. Moreover, since participants from both groups were volunteers, this may have resulted in a less representative sample, thereby limiting the generalizability of the findings. Psychological factors, such as kinesiophobia, should be considered in future studies. Similarly, the inclusion of additional scales, questionnaires, and a deeper understanding of the surgical techniques implemented is required to enhance the context of patient outcomes.

5. Conclusions

The outcomes of using HT or PT autographs during ACLR have been extensively discussed in the literature. In this study, we demonstrated that no significant differences in side-hop analysis were observed between these techniques 12 months post-surgery. The absence of differences between the intervention groups, as well as between these groups and the control group, may suggest that both autografts facilitate the restoration of knee kinematics. This restoration likely relies on neuromuscular and proprioceptive mechanisms that overcome the anatomical alterations, indicating that both techniques promote similar levels of functionality. However, it is crucial to note that these results may not reflect true biomechanical equivalence, which could impact return-to-sport readiness.

The use of a markerless motion capture system endowed a high degree of ecological validity to the data collection process, which may be a contributing factor to the lack of observed differences. This conclusion is supported by the minimal differences between groups, which may exceed the capacity of an analyst to accurately identify using conventional marker-based methods.

Future Research

Subsequent studies could expand upon our findings by investigating the long-term development of biomechanical effects beyond the 12-month post-surgery timeframe. Examining whether these differences persist, diminish, or evolve over time would yield an important understanding of long-term graft effectiveness. Moreover, incorporating neuromuscular evaluations—such as electromyography—may offer a more comprehensive understanding of the control strategies associated with each graft type during dynamic activities. Additional validation of markerless motion capture systems through direct comparison with traditional marker-based technologies could enhance their clinical applicability. Ultimately, correlating biomechanical factors with return-to-sport status could help establish clinically relevant thresholds and guide personalized rehabilitation approaches.