Abstract
Background:
Recovery of quadriceps muscle volume and function is a primary rehabilitation focus after anterior cruciate ligament reconstruction (ACLR). However, limited evidence exists on the recovery of other lower extremity muscles. Early activity limitations and persistent postoperative biomechanical alterations may contribute to volume changes in lower extremity muscles.
Purpose/Hypothesis:
The purpose of this study was to characterize lower extremity muscle volume changes from preoperative baseline to 12 months post-ACLR. It was hypothesized that the quadriceps would show the largest, most persistent atrophy. It was further hypothesized that other muscles would show atrophy early post-ACLR but that most would recover to preoperative levels by 12 months except for the gastrocnemii, which have previously demonstrated long-term atrophy.
Study Design:
Case series; Level of evidence, 4
Methods:
A total of 25 patients (mean age, 26.8 ± 9.0; 52% female) undergoing ACLR with quadriceps tendon autograft had bilateral lower extremity magnetic resonance imaging scans within 9 days prior to ACLR as well as 3 and 12 months post-ACLR. Muscle volumes for 34 bilateral muscles were determined using an automated segmentation algorithm. Volume was normalized to height-mass product. Standardized mean differences (SMDs) and paired t tests assessed between-limb asymmetries. Linear mixed-effects models with a fixed effect of time and random effect for patient assessed longitudinal changes.
Results:
Significant operative limb deficits were present in 6 of 34 (17.6%) muscles preoperatively, 16 of 34 (47.1%) at 3 months, and 10 of 34 (29.4%) at 12 months. Vasti deficits were the most pronounced and persistent (SMD: preoperative, 0.4-0.6; 3 months, 1.1-1.2; 12 months, 0.9; all P < .03). Significant deficits also persisted in the gastrocnemii (SMD: 0.4-0.5 at 12 months; P < .03). The gluteus maximus of the operative limb was significantly larger at 12 months (SMD: 0.18; P = .045).
Conclusion:
Operative limb muscle volume deficits were greatest at 3 months post-ACLR. Significant deficits of the quadriceps and gastrocnemii persisted at 12 months post-ACLR, while the operative limb gluteus maximus was larger than the nonoperative limb at 12 months.
Recovery of lower extremity muscle function following anterior cruciate ligament (ACL) reconstruction (ACLR) is a critical focus of rehabilitation. Specifically, the quadriceps muscles have received the most attention owing to the abundance of literature documenting muscle atrophy,3,21 inhibition, 22 and reduced torque-generating capacity postoperatively.5,22,38 Better quadriceps function has been linked to more symmetrical movement patterns,10,17,34 better self-reported function,2,25,30,43 reduced reinjury, 11 and better long-term cartilage health. 31 Despite the significant focus on the quadriceps muscles, there is limited evidence highlighting how other lower extremity muscles are affected post-ACLR.
It is plausible that early, global inactivity postoperatively and persistent alteration in movement mechanics across tasks post-ACLR may contribute to muscle adaptations throughout the lower extremity. For instance, disuse muscle atrophy occurs rapidly following a brief period of immobilization. 14 Although patients post-ACLR are not completely immobilized, their physical activity levels decline substantially in the early postoperative phases, which may contribute to more global atrophy of the lower extremity muscles.6,41 Additionally, individuals following ACLR often adapt movement patterns that reduce the forces needed from the quadriceps and compensate with the ankle and/or hip musculature.4,16,18,28,35,37 These altered movement patterns may lead to changes in muscle mass distribution throughout the lower extremity. Supporting this theory, a recent cross-sectional study of NCAA Division I collegiate athletes 2 years post-ACLR observed gastrocnemii atrophy and fibularis hypertrophy in addition to expected quadriceps atrophy. 15 However, it remains unknown how individual lower extremity muscle volumes change throughout the initial year post-ACLR. Understanding the trajectory of recovery of all lower extremity muscle volumes may better inform the timing in which hypertrophy-based interventions should be employed or if certain muscle groups beyond the quadriceps should be directly targeted.
Historically, it has been challenging to characterize the muscle volume of more than a few muscles given the reliance on magnetic resonance imaging (MRI) manual segmentation, which is an arduous, time-consuming process. 32 As such, much of the literature to date has focused solely on the quadriceps and/or the hamstring musculature in relatively small samples.3,7 Further, there is considerable variability in methodology for determining muscle size and when assessments occurred postoperatively.3,7 Recent advances in automated machine learning segmentation methods have removed this barrier and enabled researchers and clinicians to assess individual muscle volumes of the entire lower extremity musculature at scale.13,15,27
The purpose of this study was to characterize lower extremity muscle volume changes from preoperative baseline through 12 months post-ACLR. We hypothesized that the quadriceps muscles (ie, vastus lateralis, vastus medialis, vastus intermedius, and rectus femoris) would demonstrate the largest amount of atrophy early post-ACLR and partially recover by 12 months. We further hypothesized that atrophy would be apparent in muscles other than the quadriceps early, but most would recover to preoperative levels by 12 months except for the gastrocnemii muscles, which demonstrated long-term atrophy in a previous investigation. 15
Methods
Participants
This study is a secondary analysis of an institutional review board–approved, registered clinical trial (NCT04519801). Patients between the ages of 15 and 45 with an ACL rupture confirmed via physical examination and knee MRI, planning to undergo isolated ACLR using a quadriceps tendon autograft, were recruited as part of a randomized controlled trial comparing 2 approaches to early post-ACLR rehabilitation. Patients were excluded if they had revision or previous contralateral ACLR, bilateral knee injuries, multiligament knee injuries, and/or concomitant procedures at the time of ACLR that necessitated postoperative weightbearing restrictions. All patients provided written informed consent.
Data Collection and Analysis
Coronal scout and axial spin-echo T1-weighted MRI scans of both lower extremities from the 12th thoracic vertebrae to below the ankle joint were performed within 9 days prior to ACLR (preoperative baseline) and at 3 and 12 months post-ACLR using a 3T Discovery 750 MRI (GE Healthcare) for most scans. Scan parameters were selected based on the recommendation from the company responsible for the segmentation analyses (Springbok Analytics). 33 Specific scanning parameters for the 3T Discovery 750 were as follows: echo time (TE)/repetition time (TR)/α, 3.38 ms/550 ms/70; field of view, 480 mm × 480 mm; slice thickness, 5 mm; and in-plane spatial resolution, 1.875 mm × 1.875 mm. Because of institutional equipment changes, the last 7 patients imaged at the 12-month time point were collected on a 3T Siemens Magnetom Vida (Siemens Healthcare). Specific scanning parameters for the 3T Siemens Magnetom Vida were the following: TE/TR/α, 1.23 ms/3.97 ms/12; field of view, 500 mm × 500 mm; slice thickness, 5 mm; in-plane spatial resolution, 1.95 mm × 1.95 mm.
For data analysis, MRI scans were first preprocessed to produce continuous, axial 3-dimensional images. Next, 88 individual lower body muscles and bones were segmented by labeling of each pixel using a combination of an artificial intelligence–based algorithm and manual quality assurance by an individual blinded to the surgical limb (Springbok Analytics).13,27,33 All muscle volumes were normalized to the height mass product. 13 A total of 34 bilateral muscles were included in the statistical analysis.
Statistical Analysis
Descriptive statistics, such as means and standard deviations for continuous variables, and frequencies with percentages for categorical variables, were used to describe the patient population. At each time point, bilateral volumes for each of the 34 lower limb muscles were standardized to Z scores across the cohort to facilitate comparison across muscles. Standardized mean differences (SMDs) between limbs in muscle volume Z scores were calculated (surgical-nonsurgical) and associated 95% CIs were visualized. Given that the data were normally distributed, parametric statistics were used. Significant between-limb differences were assessed using paired t tests. Separately, between-limb asymmetries were calculated as the difference between the surgical and nonsurgical limb volumes, divided by the mean volume between the limbs. Linear mixed-effects models—with a fixed effect of time and a random effect for participant—were used to assess changes in between-limb differences over time with Tukey adjusted P values reported. Statistical significance was set at α = .05. All statistical analyses were performed using R and MATLAB programming languages (Version 4.5.0; R Core Team; MATLAB 2024a; Mathworks).
Results
In total, 25 patients completed lower extremity MRIs preoperatively (4.2 ± 2.6 days prior to ACLR), 3 months (115 ± 9.2 days), and 12 (394.1 ± 7.8 days) months post-ACLR. Patient demographics can be found in Table 1. Figure 1 shows a representative 2-dimensional visualization of the 3-dimensional muscle volume segmentation over the 3 time points.
Preoperative Patient Demographics a
Data are presented as mean ± SD or n (%) unless otherwise indicated. IKDC, International Knee Documentation Committee score; MRI, magnetic resonance imaging.

Representative patient example of the lower extremity muscle volume segmentation for between-limb asymmetries preoperatively (left), 3 months post-ACLR (middle), and 12 months post-ACLR (right). INV, involved limb; UNI, uninvolved limb.
Across the entire cohort, significant surgical limb deficits were present in 6 of 34 (17.6%) muscles preoperatively, 16 of 34 (47.1%) muscles at 3 months postoperative, and 10 of 34 (29.4%) muscles at 12 months post-ACLR (Figure 2). Deficits in surgical limb vasti (ie, vastus lateralis, vastus medialis, and vastus intermedius) muscles were most pronounced across all time periods (SMD: preoperatively, 0.4-0.6; 3 months, 1.1-1.2; 12 months, 0.9; all P values < .03). Similarly, significant deficits in the surgical limb rectus femoris were observed at both post-ACLR time periods (SMD: 3 months, 0.9; 12 months, 0.6; both P < .001), but to a lesser extent than the other vasti muscles. Significant surgical limb deficits in medial and lateral gastrocnemius muscles were also identified at all time periods (SMD: preoperatively, 0.3-0.4; 3 months, 0.6-0.7; 12 months, 0.4-0.5; all P < .03). The surgical limb flexor digitorum longus was significantly larger than the nonsurgical limb preoperatively (SMD: 0.26; P = .007), and the surgical limb gluteus maximus was significantly larger than the nonsurgical limb at 12 months post-ACLR (SMD: 0.18; P = .045).

Between-limb muscle volume differences across 34 lower extremity muscles (A) preoperatively, (B) 3 months post-ACLR, and (C) 12 months post-ACLR and associated 95% CIs.
Longitudinal changes for all muscles that had a significant between-limb difference are visualized in Figure 3, which highlights persistent atrophy of the quadriceps and gastrocnemii over time.

Longitudinal changes in all lower extremity muscles that were found to have significant between-limb differences at any time point. The points represent the least square means, and the error bars are the 95% CIs about the mean. Asterisk indicates statistically significant between-limb difference compared with pre-ACLR (P < .05). Dagger indicates statistically significant between-limb difference compared with 3 months post-ACLR.
Discussion
This study provides a comprehensive characterization of lower extremity muscle volume changes following ACLR with quadriceps autograft, with assessments spanning from preoperative baseline to 12 months postoperative. The primary findings confirm our hypothesis of significant and persistent quadriceps muscle atrophy, consistent with previous literature.3,7 In addition, we observed notable muscle volume changes in several other lower extremity muscles, particularly the gastrocnemii and gluteus maximus. To our knowledge, this is the first study to comprehensively assess complete lower extremity muscle volume changes from preoperative baseline to 12 months post-ACLR. These findings have implications for exercise selection and dosage to mitigate or restore symmetrical lower extremity muscle volume post-ACLR.
As expected, the quadriceps muscles exhibited the most pronounced atrophy throughout the continuum of care, with deficits evident preoperatively and persisting through 12 months post-ACLR. These findings align with prior studies, highlighting the vulnerability of the quadriceps to disuse atrophy and neuromuscular inhibition following ACL injury and reconstruction, particularly in those receiving an extensor mechanism graft.3,7,21 Importantly, this cohort did not have postoperative weightbearing restrictions, suggesting that even in the absence of prescribed weightbearing restrictions, significant atrophy occurs. These findings have implications for torque production capacity. Restoration of quadriceps strength is a key focus of ACLR rehabilitation, and a recommended criterion used as part of the return to sport decision-making. 1 As muscle volume is a fundamental contributor to a muscle's ability to produce torque,8,19,39 significant and persistent atrophy post-ACLR can delay an athlete's recovery. Moreover, increasing muscle volume takes considerable time and requires consistent properly dosed strength training.12,24 The difficulty of restoring quadriceps volume and strength can be compounded by underlying neuromuscular inhibition20,42 and/or central mechanisms of inhibited quadriceps efferent muscle drive, 38 which restricts the muscle's capacity to generate maximal voluntary torque and thus reduces the mechanical tension required for effective hypertrophy. This is likely why it is not uncommon to see persistent quadriceps impairments beyond 2 years post-ACLR, particularly in those with an extensor mechanism graft. 9 These findings underscore the importance of addressing muscle inhibition and prescribing targeted hypertrophy-based training during rehabilitation. There is a clear need to investigate strategies to mitigate the extent of muscle volume loss following ACL injury and surgical reconstruction.
Beyond the quadriceps, the gastrocnemii muscles also demonstrated consistent surgical limb deficits across all time points, which has also been observed in Division I collegiate football players 27.9 ± 19.0 months post-ACLR. 15 The fact that the gastrocnemii muscles are biarticular and cross the knee joint may play a role in why muscle volume deficits of the surgical limb were observed and why these muscles show greater deficits than the uniarticular soleus muscles. Persistent atrophy of the gastrocnemii may be a result of disuse atrophy, altered movement mechanics, or persistent muscle inhibition in an attempt to reduce the amount of strain on the ACL, given the posterior attachment of the gastrocnemii on the femur. 23 These findings suggest that gastrocnemii-specific hypertrophy training may be warranted to restore muscle volume symmetry and optimize lower limb function in patients following ACLR.
Interestingly, the gluteus maximus was significantly larger on the surgical limb at 12 months post-ACLR. This may represent compensatory hypertrophy that develops due to increased reliance on hip extensors in response to impaired quadriceps function in the later stages of rehabilitation when patients are returning to more high-demand activities. Such adaptations could reflect a shift toward hip-dominant movement strategies, which have been observed in post-ACLR populations.4,16,29,35,40 These compensations may also result from or contribute to quadriceps underutilization, inhibiting full recovery of the quadriceps. As such, in addition to hypertrophy-based quadriceps strengthening, it may be necessary to alter an individual's movement pattern to be less reliant on the hip extensors to fully restore quadriceps function.
The observed muscle volume asymmetries, particularly at 3 months post-ACLR, suggest a period of global disuse atrophy early in recovery. Although recovery was evident by 12 months in some of the muscles, significant asymmetries remained, highlighting the need for ongoing rehabilitation efforts beyond the typical return-to-sport timeline. Pairing volumetric assessments with strength testing could further elucidate the relationship between muscle atrophy and torque production, guiding more effective rehabilitation strategies. Further, future studies should explore limb-specific muscle volume changes over time, compare ACLR cohorts with matched healthy controls, and determine whether muscle volume changes are related to long-term joint health concerns or reinjury risk.
Another interesting finding to highlight is the fact that muscle atrophy was evident in 6 of 34 (17.6%) muscles preoperatively. Specifically, the vasti muscles, gastrocnemii, and biceps femoris long head were all smaller on the injured limb, suggesting that atrophy can occur rapidly after injury. Although these findings may have been confounded by the variable time from injury to preoperative MRI, prehabilitation may be important to consider and should incorporate strategies to mitigate atrophy of these muscles. Additionally, hypertrophy of the injured limb flexor digitorum longus was observed preoperatively. Although the magnitude of this asymmetry is small, with questionable clinical value, this asymmetry may be a result of altered gait strategies following injury.
Limitations
It is important to highlight limitations of the current study. Specifically, this is a secondary analysis of a randomized controlled trial and included only patients who underwent ACLR with a quadriceps tendon autograft. Therefore, findings may not be generalizable to individuals undergoing ACLR with hamstring or patellar tendon autografts. It would be expected to see different patterns of muscle atrophy, particularly at the graft harvest site for hamstring tendon autografts.26,36 Given that this was a secondary analysis, this study is exploratory in nature and therefore no a priori power analysis was performed and no correction for multiple comparisons was utilized when assessing the SMD between limbs (see Figure 2). Future confirmatory studies are needed to corroborate these findings. While all patients completed a standardized rehabilitation protocol for the first 3 months postoperatively under the guidance of a single lead physical therapist (J.S.), rehabilitation was not controlled between 3 and 12 months post-ACLR as patients went to other physical therapy clinics throughout the community in that window of time. Additionally, muscle volume was assessed at 3 distinct time points. Imaging more frequently may provide a better estimate of the typical trajectory of muscle volume changes post-ACLR. Further, 7 of the 12-month scans were acquired on a different MRI because of institutional equipment changes. Last, while advanced segmentation techniques enabled comprehensive muscle volume analysis, the cost and issues pertaining to accessibility may limit the generalizability of these findings to routine clinical practice.
Conclusion
Surgical limb muscle volume deficits were most pronounced at 3 months post-ACLR, indicating potential disuse atrophy throughout the lower extremity in the early postoperative period. While the magnitude of asymmetry reduced from 3 to 12 months post-ACLR, significant asymmetries persisted, most notably in the quadriceps and the gastrocnemii muscles. An exception to this was that greater surgical limb gluteus maximus muscle volume was observed at 12 months post-ACLR.
Footnotes
Acknowledgements
The authors would like to acknowledge the contributions of Josh Mckinley, PTA, to the postoperative care of the patients included in this study.
Final revision submitted February 20, 2026; accepted March 8, 2026.
One or more of the authors has declared the following potential conflict of interest or source of funding: A.J.S. is a consultant for Stryker, an advisory board member for Springbok Analytics, and has received hospitality payments from Stryker and MedInc.
Ethical approval for this study was obtained from the institutional review board of Regional Health Command-Central (No. C.2020.053).
