Author ORCID Identifier

https://orcid.org/0000-0002-8540-7461

Date Available

11-19-2027

Year of Publication

2025

Document Type

Doctoral Dissertation

Degree Name

Doctor of Philosophy (PhD)

College

Education

Department/School/Program

Exercise Science

Faculty

Dr. Brian Noehren

Faculty

Dr. Haley Bergstrom

Abstract

Background: Restoring quadriceps strength is a primary goal after anterior cruciate ligament reconstruction (ACLR) due to the large body of evidence linking knee extensor weakness to negative long-term outcomes (i.e., ACL reinjury, knee osteoarthritis). Despite best efforts, current ACL rehabilitation protocols have proven ineffective for restoring symmetrical knee extensor strength prior to return to sport, particularly in individuals with a bone-patellar tendon-bone (BPTB) graft. Persistent knee extensor weakness after ACLR is often attributed to neuromuscular alterations in the quadriceps muscle (e.g., activation, atrophy), but it remains unclear how surgical disruption of the donor-site patellar tendon impacts force production. Additionally, clinicians often prescribe closed-chain tasks (e.g., squats, lunges) to strength the quadriceps after ACLR, yet these are frequently performed with knee-avoidant mechanics that shift loading demands from the knee to the hip. Such compensations limit the effectiveness of exercise for restoring quadriceps function and may reinforce maladaptive movement strategies. This critical knowledge gap represents a significant opportunity to define the modifiable factors underlying knee extensor weakness after a BPTB graft and provide the insights needed to develop more targeted and specific interventions throughout the rehabilitation process.

Objective: Aim 1: To identify which quadriceps muscle and patellar tendon properties are associated with knee extensor peak torque and rate of torque development in individuals 3-4.5 months following ACLR with BPTB graft. Aim 2: To evaluate the efficacy of various intervention strategies to improve knee extensor and patellar tendon loading during bilateral squats and rearfoot elevated split squats in individuals 3-4.5 months following ACLR with BPTB graft.

Participants: Twenty-five participants (age: 18.4 ± 2.8 years; 56% female; time since surgery: 3.9 ± 0.4 months) with primary ACLR using a BPTB graft

Methods: Aim 1: Vastus lateralis (VL) muscle cross-sectional area (CSA), stiffness, and quality (echo intensity) were assessed using ultrasound B-mode and shear wave elastography. Patellar tendon thickness, CSA, and regional stiffness were assessed in the same manner. Knee extensor peak torque and rate of torque development (RTD) were evaluated using isometric testing on an isokinetic dynamometer. Quadriceps activation percentage was assessed using interpolated twitch technique. Aim 2: Three-dimensional motion analysis used to analyze knee joint biomechanics during bilateral squats and rearfoot elevated split squats (RFESS). Participants were analyzed across four squat conditions: (1) control, (2) heel wedge, (3) superimposed neuromuscular electrical stimulation (NMES), and (4) combination (combo) (i.e., NMES + heel wedge). Biomechanical variables of interest included peak knee flexion angle, peak internal knee extensor moment, knee extensor angular impulse, and patellar tendon load.

Statistical Analysis: Aim 1: Paired t-tests examined between-limb differences in muscle and tendon properties. Partial correlations controlling for body mass were used to identify candidate predictors, and hierarchical linear regression models were used to determine predictors of peak torque and RTD in the ACL-involved limb. Aim 2: Paired t-tests were used to compare knee joint mechanics and patellar tendon load between limbs. A repeated measures ANOVA was used to determine differences across conditions within the ACL-involved limb, using the control condition as the reference.

Results: Aim 1: Compared to the non-involved limb, the ACL-involved limb VL was significantly smaller (- 4.3cm2), had worse muscle quality (+8.0 au), lower stiffness (-0.32 m/s), and lower quadriceps activation percentage (-19.9%). The donor-site patellar tendon was larger (+ 1.03 cm2), thicker (+ 3.06 mm), and had a stiffer lateral region (+1.66 m/s) compared to the non-involved tendon. In regression models that controlled for body mass, VL quality and VL CSA emerged as the strongest predictors of both peak torque and RTD. Together, VL quality and VL CSA accounted for a substantial portion of the variance in peak torque (R² = 0.477, adjusted R² = 0.402; p < .001) and RTD (R² = 0.519, adjusted R² = 0.450; p < .001), with VL quality emerging as a consistent independent predictor. Patellar tendon CSA, thickness, and medial stiffness had moderate negative partial correlations with RTD. In the regression model controlling for body mass, medial stiffness and CSA resulted in the best model fit (R² = 0.297, adjusted R2 = 0.196, p = .056) but the overall model did not meet statistical significance. Aim 2: The ACL-involved limb under-loaded the knee extensors and patellar tendon relative to the non-involved limb, despite demonstrating relatively symmetrical knee flexion angles. These between-limb differences were significant across all conditions, but the addition of a heel wedge exacerbated the asymmetries, particularly during bilateral squats. The heel wedge and combo conditions augmented knee extensor and patellar tendon load the most in the ACL-involved limb during the bilateral squat, with more modest effects from the superimposed NMES. There was little to no effect of condition on knee extensor or patellar tendon load during the RFESS, with only angular impulse being significantly higher during the wedge condition.

Conclusions: VL muscle size and quality are the primary predictors of knee extensor peak torque and RTD. Patellar tendon size and stiffness had a modest effect on RTD, with future work needed to further establish this relationship. During ACL rehabilitation, adding a heel wedge and/or superimposed NMES to a bilateral squat can make the exercise more demanding on quadriceps muscle and patellar tendon, but may exacerbate asymmetrical loading strategies. The RFESS provided little stimulus to the knee extensors and patellar tendon across all conditions and may not be a suitable quadriceps strengthening exercise for individuals in the early phases of ACL rehabilitation. These findings may help guide clinicians by providing actionable strategies to increase knee extensor demand and patellar tendon loading in individuals with a BPTB graft.

Digital Object Identifier (DOI)

https://doi.org/10.13023/etd.2025.522

Download

Available for download on Friday, November 19, 2027

Contact Author

Share

COinS