Differential Quadriceps Muscle Activation Preoperatively and at the Time of Return to Sport After Medial Patellofemoral Ligament Reconstruction

Abstract

Background:

Patients with patellofemoral instability (PFI) often demonstrate altered movement patterns and strength deficits. The vastus medialis plays an important role in patellar stabilization as a dynamic medial force vector on the patella.

Hypothesis:

It was hypothesized that limb-to-limb asymmetries in muscle activity would be present preoperatively and at the time of return to sport (RTS).

Study Design:

Cross-sectional study; Level of evidence, 3.

Methods:

A total of 25 adolescents with unilaterally treated PFI were evaluated before arthroscopic assisted medial patellofemoral ligament reconstruction with an allograft. A separate group of 16 patients with PFI were examined for RTS clearance. All participants completed an unweighted squat task in a motion capture laboratory. Surface electromyography sensors were placed on the rectus femoris, vastus medialis, semitendinosus, tibialis anterior, and gastrocnemius. Electromyography data were filtered and rectified, and a linear envelope was created. Peak activity of each muscle was analyzed. The Wilcoxon signed-rank test was used to compare muscle activity between limbs (α = 0.05).

Results:

Preoperatively, peak rectus femoris activity of the involved limb was 40% less than that of the contralateral limb (P = .006). At the time of RTS, peak vastus medialis activity was 73% less in the involved limb than in the contralateral limb (P = .001).

Conclusion:

A preoperative reduction in rectus femoris activity of the involved limb indicated that there might be compensations in movement patterns among those with PFI. At the time of RTS after medial patellofemoral ligament reconstruction, vastus medialis discrepancies suggested that there might be a muscle imbalance after surgery and continue through physical therapy. A greater emphasis on vastus medialis training may prove beneficial during the rehabilitation process and help to prevent the reduction in activity observed at the time of RTS.

Patellofemoral instability (PFI), a condition in which the patella dislocates from the trochlear groove, affects 1 in 1000 children aged between 9 and 15 years, with upwards of 80% of patients with PFI experiencing recurrent dislocations.^17,18 Patients with PFI often demonstrate altered knee morphology such as trochlear dysplasia, patella alta, patellar tilt, and genu valgum. ³⁹ These anatomic variations have been specifically associated with recurrent instability. ²⁹ Further, patients with PFI often exhibit movement abnormalities including decreased hip flexion, reduced knee flexion, and decreased quadriceps recruitment during gait.^1,7,18

Both operative and nonoperative approaches may be recommended for PFI treatment. A redislocation is dependent on a variety of factors; however, nonoperative treatment has been proven to be largely ineffective in preventing recurrence, while redislocation rates after operative treatment have been reported to be as low as 2.4%.^{3,8,15,25,37,41} The medial patellofemoral ligament (MPFL), which inserts on the medial border of the patella, is recognized as the primary soft tissue medial patellar stabilizer and is injured in most cases of lateral dislocations.^2,38 Therefore, surgical reconstruction of the MPFL is a popular patellar stabilization technique, resulting in lower rates of recurrence than nonoperative treatment.^3,37,41

Of those returning to sport participation after MPFL reconstruction, approximately half will return at a lower activity level compared with their preinjury level. ²⁴ Reduced activity levels may arise from functional deficits, such as quadriceps dysfunction and limited knee range of motion, potentially due to insufficient rehabilitation guidelines after MPFL reconstruction.^14,34 Additionally, muscle activation and strength deficits are often reported after surgical reconstruction of the structures surrounding the knee.^22,33,40 Specific associations have been identified between the role of the vastus medialis in the knee extensor mechanism and patellar instability.^5,12,23 Thus, an investigation of limb-to-limb asymmetries in muscle activity before surgery and at the time of return-to-sport (RTS) clearance is necessary to identify altered muscle function and resultant mobility deficits, allowing for improved rehabilitation strategies.

Limited research exists exploring the biomechanics of young patients with PFI after MPFL reconstruction, and even less investigation has focused on muscle activity patterns. The purpose of this study was to identify limb-to-limb asymmetries in muscle activity in adolescent patients with PFI both preoperatively and at the time of RTS clearance to inform rehabilitation strategies after MPFL reconstruction. It was hypothesized that limb-to-limb asymmetries in peak muscle activity would be present both preoperatively and at RTS clearance.

Methods

Participants

There were 2 separate groups of patients investigated. One group consisted of adolescents, aged 10 to 18 years, with unilaterally treated PFI who were recruited before surgery (preoperative group). A separate group of adolescents that underwent arthroscopic assisted MPFL reconstruction with an allograft for PFI were recruited for study participation when physician clearance was provided to initiate the progression to RTS, that is, physician discretion to return to any level of activity (RTS group). All MPFL reconstruction procedures were performed at a single sports medicine center by 2 physicians. Participants with musculoskeletal or neuromuscular diagnoses other than unilaterally treated PFI, those with moderate to severe effusion, or those with a previous condition/injury causing an inability to perform a bodyweight squat were excluded from this study. Only surgically treated patients were included in analyses, with all nonoperatively managed participants being excluded. Further, those who received tibial tubercle osteotomy were excluded because of anticipated longer rehabilitation times and altered neuromuscular changes.^21,28 Lastly, participants in the RTS group who were more than 2 months past clearance were excluded.

Approval from a regional institutional review board was granted before the initiation of study procedures. Informed consent/assent was obtained from all participants and parents, if applicable, before participation. Testing was performed in a motion capture laboratory from July 2020 through May 2025, and all participants wore their personal athletic footwear and attire. The sample size was determined by the number of eligible participants with complete data available for analysis during the study period.

Data Acquisition and Task Procedure

Patients’ descriptive data were recorded alongside injury details, such as the total number of instability events and the date of the most recent instability event. Additionally, the date of surgery, as well as clearance date, was collected for the RTS group. Anthropometric data, including body weight (kg) and height (cm), were collected upon arrival for testing. Three-dimensional biomechanical data were collected using motion capture in the preoperative group before MPFL reconstruction and in the RTS group once physician clearance was provided. Surface electromyography (EMG) sensors (Trigno Wireless; Delsys) were placed bilaterally on the rectus femoris, vastus medialis, semitendinosus, gastrocnemius, and tibialis anterior (Figure 1).

Figure 1.

Green rectangles indicate the placement of surface electromyography sensors. Locations comprised the rectus femoris (RF), vastus medialis (VM), semitendinosus (ST), gastrocnemius (GAS), and tibialis anterior (TA).

At the time of testing, participants completed 3 trials of a bodyweight squat task (Figure 2). To perform the squat, participants began standing with the feet shoulder-width apart, raised their arms above their head while holding an unweighted bar, and squatted as low as they felt comfortable before returning to a standing position. Squat trials were deemed unsuccessful if the participant lost balance and/or failed to return to the standing position.

Figure 2.

Demonstration of the bodyweight squat task at the lowest squat depth in sagittal (left) and coronal (right) views.

Data Processing and Analysis

A custom MATLAB program (Version 2022a; MathWorks) was employed to extract muscle activity data of the lower extremities throughout the entirety of the squat task. EMG data were rectified and filtered using a fourth-order 20- to 450-Hz bandpass Butterworth filter to compute a linear envelope before identifying peak muscle activity. A representative trial was chosen for each participant based on the squat with the lowest squat depth. Squat depth was calculated as the change in the vertical height of the hip joint center from standing to the minimum sacrum position and then normalized by standing sacrum height.

Statistical Analysis

Descriptive statistics, including the mean and standard deviation, were computed for each group and limb. Within each group, the Wilcoxon signed-rank test was performed to compare involved and contralateral limbs. Percentage differences between the involved and contralateral limbs were calculated to describe limb asymmetry. To evaluate between-group differences in limb asymmetry, the Mann-Whitney U test was used to compare percentage difference values between the preoperative and RTS groups. Between-group comparisons of kinematic measurements were also performed using the Mann-Whitney U test, with involved and contralateral limb values analyzed separately. Finally, between-group comparisons of descriptive and anthropometric variables were performed using the Mann-Whitney U and chi-square tests. Effect sizes (r) were calculated as the absolute value of the standardized test statistic divided by the square root of the number of paired observations. These effect sizes were then categorized as small, medium, and large, determined as 0.1-0.3, 0.3-0.5, and ≥0.5, respectively.^9,13,16 Effect sizes for chi-square comparisons were computed as the Cramer V. All statistical analyses were conducted using SPSS Statistics (Version 24.0; IBM) with a significance level (α) of 0.05.

Results

Participants’ Characteristics

The preoperative group consisted of 25 adolescents with unilaterally treated PFI (13 female; mean age, 14.1 ± 2.1 years; mean height, 164.6 ± 12.7 cm; mean weight, 58.5 ± 14.0 kg) who were evaluated before MPFL reconstruction. Preoperative testing was performed at a mean 58.4 ± 67.9 days after the most recent instability event. A separate group of 16 patients with PFI (9 male; mean age, 14.4 ± 0.9 years; mean height, 167.6 ± 9.5 cm; mean weight, 63.6 ± 14.1 kg) were included in the RTS group and were tested once physician clearance was provided (25.4 ± 4.5 weeks from surgery to RTS clearance and 1.3 ± 3.2 weeks from RTS clearance to testing). Motion capture testing ranged from 0 to 43 days after RTS clearance. The characteristics of the preoperative and RTS groups, as well as instability status and baseline anatomic measurements, can be found in Table 1, with no significant differences observed between groups. Squat depth did not significantly differ between the preoperative and RTS groups (P = .922). Additionally, no significant kinematic differences were observed between the preoperative and RTS groups, in either limb, at the hip, knee, or ankle (Table 2).

Table 1.

Descriptive and Anthropometric Data ^a

	Preoperative (n = 25)	RTS (n = 16)	P Value	Effect Size (r)
Sex, n			.751	0.081
Female	13	7
Male	12	9
Age, y	14.1 ± 2.1	14.4 ± 0.9	.682	0.067
Height, cm	164.6 ± 12.7	167.6 ± 9.5	.333	0.152
Weight, kg	58.5 ± 14.0	63.6 ± 14.1	.333	0.152
Instability status, %			.521	0.178
First-time dislocation (1)	24	38
Minimal recurrence (2-3)	48	31
Excessive recurrence (>3)	28	31
TT-TG distance, mm	12.11 ± 3.23	13.93 ± 4.06	.146	0.236
Caton-Deschamps index	1.24 ± 0.24	1.29 ± 0.22	.467	0.117
Patellar tilt, deg	16.55 ± 5.61	20.18 ± 8.18	.323	0.162
Trochlear depth, mm	2.76 ± 1.04	2.96 ± 0.94	.618	0.081

^aData are reported as mean ± SD unless otherwise specified. RTS, return to sport; TT-TG, tibial tubercle–trochlear groove.

Table 2.

Kinematic Measurements ^a

	Involved Limb		Contralateral Limb
	Preoperative	RTS	Preoperative	RTS
Hip flexion	94.68 ± 14.33	89.47 ± 26.53	94.45 ± 14.57	86.30 ± 26.35
Hip abduction	–19.77 ± 7.52	–17.78 ± 7.47	–16.85 ± 9.51	–14.36 ± 8.90
Hip internal rotation	15.00 ± 10.74	15.11 ± 9.36	13.08 ± 11.47	9.61 ± 9.24
Knee flexion	107.47 ± 16.77	103.55 ± 31.31	108.85 ± 17.69	102.99 ± 31.64
Knee abduction	–3.98 ± 4.06	–4.97 ± 2.97	–3.67 ± 3.58	–4.17 ± 2.86
Knee external rotation	–21.97 ± 7.99	–21.29 ± 5.53	–22.67 ± 6.23	–19.79 ± 7.16
Ankle dorsiflexion	27.83 ± 5.44	28.33 ± 11.65	31.23 ± 5.03	31.02 ± 11.42
Ankle eversion	–15.41 ± 10.75	–15.28 ± 6.95	–12.80 ± 9.19	–11.13 ± 6.28
Ankle external rotation	16.67 ± 8.24	16.41 ± 6.27	13.90 ± 6.54	12.93 ± 6.48

^aData are reported as mean ± SD (in degrees). All P values are >.05. RTS, return to sport.

Muscle Activity

The percentage difference in semitendinosus muscle activity was significant between the preoperative and RTS groups (P = .028). No other differences were observed between the groups (P≥ .248). Within the preoperative group, peak rectus femoris activity of the involved limb was 40% less than that of the contralateral limb (P = .006); however, no differences in muscle activity were observed between limbs for any other muscles (Table 3). Within the RTS group, peak vastus medialis activity was 73% less in the involved limb compared to the contralateral limb (P = .001), with no difference in peak rectus femoris activity. No significant limb-to-limb differences in muscle activity were observed in the preoperative or RTS group for the semitendinosus, tibialis anterior, or gastrocnemius (P≥ .063).

Table 3.

Peak Muscle Activity ^a

	Involved Limb	Contralateral Limb	P Value	Effect Size (r)	Percentage Difference
Preoperative
Rectus femoris	36.1 ± 23.3	54.2 ± 36.9	.006	0.55	–40.1
Vastus medialis	69.9 ± 72.4	88.9 ± 61.5	.098	0.33	–24.0
Semitendinosus	14.0 ± 11.9	18.9 ± 29.6	.657	0.09	–29.6
Tibialis anterior	66.8 ± 49.5	75.6 ± 53.0	.253	0.23	–12.4
Gastrocnemius	19.9 ± 11.3	16.6 ± 8.2	.581	0.11	18.2
RTS
Rectus femoris	49.7 ± 23.2	67.5 ± 38.4	.063	0.47	–30.4
Vastus medialis	40.8 ± 21.5	88.4 ± 37.5	.001	0.80	–73.7
Semitendinosus	10.8 ± 5.9	13.6 ± 7.9	.079	0.44	–23.0
Tibialis anterior	111.1 ± 101.3	109.4 ± 51.4	.352	0.23	–1.5
Gastrocnemius	24.4 ± 19.8	19.9 ± 22.5	.326	0.25	20.3

^aData are reported as mean ± SD (in microvolts). Significant differences are denoted in bold. RTS, return to sport.

Discussion

The purpose of the current study was to investigate asymmetries in lower extremity muscle activity during a squat task in adolescents with PFI before and after MPFL reconstruction. Significant asymmetries were observed in quadriceps activity both preoperatively and at RTS clearance, supporting the original hypothesis. In the preoperative group, the rectus femoris demonstrated significantly reduced peak muscle activity in the involved limb compared to the contralateral limb. No other significant findings were observed at the preoperative time point. Alternatively, in the RTS group, peak muscle activity of the vastus medialis was significantly less in the involved limb compared to the contralateral limb, with no other significant differences observed. Overall, the involved limb, compared to the contralateral limb, exhibited reduced peak muscle activity across all muscles, except for the gastrocnemius, at both time points. However, the statistically significant disparity in peak activity between limbs shifted from the rectus femoris in the preoperative group to the vastus medialis in the RTS group.

There is limited research exploring lower extremity muscle activity during dynamic movements in patients with PFI, especially among pediatric populations. Much of the existing literature focuses on gait, which does not elicit the same strength or mobility requirements as more dynamic sport-like tasks, such as the squat. One preoperative investigation of participants aged between 9 and 30 years found that the vastus medialis presented weakness in the involved limb compared to the uninvolved limb during a nonweightbearing movement, with no significant differences in the rectus femoris or vastus lateralis. ⁴ These findings differ from those of the present study, which revealed rectus femoris weakness at the preoperative time point rather than vastus medialis weakness. One explanation for the reduction in peak rectus femoris activity in the involved limb is injury-mediated quadriceps muscle inhibition. Research has shown that pain or injuries to the knee can result in an inhibition of the efferent signals being sent to the involved leg muscles, in this case, the rectus femoris.^10,30 Additionally, this muscle discrepancy highlights potential compensation techniques of a PFI population as the lower extremities are loaded and put through a greater range of motion.

In another study, researchers observed increased activity of the vastus medialis during midstance, followed by a decrease during the swing phase of gait, in patients with a history of lateral patellar subluxations, or partial dislocations, compared to healthy controls. ²⁶ However, the age range of these participants was notably wide, ranging from 17 to 60 years. Other studies have concluded insufficient or inconsistent results of preoperative muscle activity in a PFI population. ³⁶ Observed differences in quadriceps activity may be related to anatomic abnormalities and/or movement pattern differences observed in patients with PFI before reconstruction, such as reduced hip and knee flexion or quadriceps avoidance gait.^7,18,32 Unlike gait, a squat requires a greater range of motion of the knee, which may increase the demand on the quadriceps muscles. The limb-to-limb differences in preoperative rectus femoris activity observed in the present study may suggest the existence of movement pattern abnormalities. However, significant differences were not observed in kinematics between groups, suggesting that kinetic differences, such as offloading of a limb, may provide further insight.

After MPFL reconstruction in adolescents, altered muscle activation has been observed in the lower extremities. Specifically, the onset of vastus medialis activation was significantly delayed, compared to a healthy control group, when performing a maximal-effort isometric knee extension task in 30° of flexion. ²⁸ These findings, along with altered peak activity of the vastus medialis observed in the present study, provide further evidence of the impact that surgery for patellar instability, and possibly specifically MPFL reconstruction, may have on the medial quadriceps. Several studies have shown that quadriceps strength deficits and activation delays persist for several years after surgery.^20,22,31 While not statistically different, rectus femoris activity in the RTS group did approach significance, further reiterating knee extensor abnormalities. Despite detailed rehabilitation programs for general quadriceps strength after surgery, these long-term activation deficits may be, in part, caused by the lasting effects of MPFL reconstruction, specifically on the vastus medialis. ¹¹ The current investigation adds to the growing body of evidence that patients with PFI may face muscle weakness and disruptions in medial quadriceps muscle activity after surgery, whether it be altered activity or delayed activation, and may benefit from rehabilitation that focuses on the vastus medialis. ³⁵

The findings of the present study highlight the importance of the vastus medialis and the immediate muscle effects of MPFL reconstruction in patients with PFI. Despite limited dissection of the soft tissue attachments along the medial border of the patella, inherent to the technique employed within the study cohort, as with many contemporary techniques, MPFL reconstruction may affect vastus medialis function by temporary alteration of this insertional tissue, possibly leading to lasting effects on the muscle. ¹¹ While structures of the knee can be altered both during dislocations and surgery, there may be rehabilitation strategies available to mitigate functional detriments, such as preferential training. ²⁷ Researchers have found that patients were able to favor vastus medialis or vastus lateralis activity with EMG feedback and focused visual activation. ¹⁹ While quadriceps muscle imbalance in those with PFI is not uncommon, the shift in activation deficiencies between the limbs is notable. ⁶ More specifically, the transition of reduced activation from the rectus femoris in the preoperative group to the vastus medialis in the RTS group is not fully understood. Preoperative mechanics, whether anatomic or inherently avoidant, may explain some differences seen before surgery. Conversely, the operative procedures associated with MPFL reconstruction may involve the surrounding structures, including the vastus medialis, more than previously understood and lead to lasting impacts on the medial quadriceps musculature.

Limitations

Several limitations to this study should be noted. First, kinetic and strength data were not included in the present analysis. Therefore, it cannot be determined whether differences in muscle activity reflect changes in force mechanics. This distinction is important because previous research has correlated biomechanical movement patterns of patients with PFI to anatomic abnormalities of the knee, such as trochlear dysplasia and patellar tilt. ³² Muscle activity alone leaves room for interpretation, and kinetics may help to explain why these differences exist. Second, the timing of muscle activity during the task was not analyzed. The present results provide valuable insight into the peak activity achieved in the lower extremity musculature; however, it is impossible to conclude when those peaks occur throughout the task. Previous investigations have found the delayed onset of vastus medialis activity in instability populations. ²⁸ Further research should investigate whether one limb peaks earlier in the descent or ascent phase than the other limb, providing insight into when the quadriceps muscle is activated.

Because of the sample size and patient visits, 2 separate groups were used in the present analysis, which limited comparisons between the preoperative and RTS groups. This study should be interpreted as exploratory in nature, and the findings provide preliminary estimates of the effect magnitude to inform future adequately powered studies. Regarding the participants examined for this study, the cohort included patients ranging from first-time dislocators (24%-38%) to those with several instability events (28%-31%). A comparison between groups discerning the number of instability events may reveal further insights into lower extremity muscle activity patterns. Because of the limited sample size of the present study, subgroup analyses investigating the number of events were not pursued. Future studies should further evaluate differences between recurrent and first-time dislocators.

Conclusion

In patients with patellar instability, limb differences in peak rectus femoris activity existed preoperatively. Specifically, a 40% difference in limbs was observed during a bodyweight squat, with the involved limb demonstrating significantly reduced activity. Compensation in movement patterns specific to PFI may provide some explanation for differences before surgery. At RTS clearance, a 73% intralimb difference in peak vastus medialis activity was present, with significantly reduced activity in the involved limb. MPFL reconstruction appears to have an impact on the medial quadriceps, and therefore, an increased focus on vastus medialis rehabilitation could prove beneficial for long-term success of the knee. Future work should further investigate the lower limb muscle activity of adolescents with PFI during dynamic movements and consider comparisons to healthy controls.