Muscle fiber cross-sectional area is associated with quadriceps strength and rate of torque development after ACL injury

Abstract

The purpose of this study was to investigate the relationship between muscle fiber-type specific properties of the vastus lateralis (VL) and quadriceps muscle performance in individuals following an anterior cruciate ligament (ACL) tear. 26 participants (22.0 ± 5.4 years) were included in this cross-sectional study and all data was collected prior to ACL reconstruction. Quadriceps peak torque (QPT) and early (0–100ms) and late (100–200ms) rate of torque development (RTD) were obtained from maximal voluntary isometric quadriceps strength testing. Muscle fiber cross-sectional area (fCSA) and percent fiber-type distribution (FT%) were evaluated through immunohistochemical analysis of a muscle biopsy. Between-limb differences in fiber characteristics were assessed using paired t-tests (with α-level 0.05). Relationships between fiber-specific properties and quadriceps muscle performance were determined using separate multiple linear regression analyses for ACL-injured and non-injured limbs. There were significant differences in fCSA between ACL-injured and non-injured limbs across all fiber types, but no differences in FT%. Type 1 fCSA, type 2a fCSA and their interaction effect were the explanatory variables with the strongest relationship to all performance outcomes for the ACL-injured limb. The explanatory variables in the ACL-injured limb had a significant relationship to QPT and late RTD, but not early RTD. These findings suggest that QPT and late RTD are more heavily influenced by fCSA than FT% in ACL-injured limbs. This work serves as a foundation for the development of more specific rehabilitation strategies aimed at improving quadriceps muscle function prior to ACL reconstruction or for individuals electing non-surgical management.

INTRODUCTION

Quadriceps weakness develops rapidly following an anterior cruciate ligament (ACL) tear and can have pervasive effects on long-term outcomes if left unresolved (10, 19, 24). Failure to properly restore quadriceps strength prior to elective ACL reconstruction (ACLR) is associated with persistent post-operative quadriceps weakness and poor self-reported knee function (10, 19, 24). Although the literature recommends a period of rehabilitation prior to ACLR focused on restoring quadriceps function, effective strategies to combat quadriceps strength deficits remain elusive (1). Previous work has shown that injury to the ACL potentiates maladaptive morphological alterations of the quadriceps muscle, but it remains unclear how these morphological changes influence muscular performance (i.e., peak strength and rate of torque development) (18). Understanding how the fiber type-specific characteristics of the quadriceps muscle affect muscular performance after an ACL tear will have important implications for prescribing appropriate rehabilitation exercises aimed at restoring muscular function prior to surgery.

Quadriceps muscle atrophy is commonly observed after an ACL injury and is thought to be a significant contributor to quadriceps weakness (9, 21, 22). While total quadriceps muscle cross-sectional area (CSA) may provide insight into maximal strength capacity, CSA alone does not provide information regarding the fiber-type composition of the muscle (9, 14). The mechanical properties of muscle are highly dependent on its fiber type-specific properties (i.e., type, size, and percent distribution) and previous work has shown that these properties are altered in the quadriceps muscle of the involved limb after an ACL tear (29, 32, 45). After an ACL injury, there is selective atrophy and loss of type 2a fibers in the vastus lateralis (VL) muscle (23, 32). Considering type 2a fibers have high peak force and power production capabilities, selective atrophy and loss of these fasttwitch fibers is likely contributing to the deficits in maximal quadriceps strength and rate of torque development (RTD) commonly observed after ACL injury (7, 26, 28, 29). However, no work to date has explored the relationship between VL fiber type-specific properties and muscular function in an ACL-injured population and these claims remain hypothetical.

Therefore, the purpose of this study was to identify which fiber type-specific properties (fiber CSA [fCSA] and percent fiber-type distribution [FT%]) were the strongest predictors of quadriceps muscle performance (peak torque, early RTD, late RTD) in individuals who had sustained a unilateral ACL tear. We also evaluated between-limb differences in the fiber-specific properties of the VL. We hypothesized that the ACL-injured limb would have significantly smaller type 2a fCSA and lower FT% compared to the non-injured limb. Additionally, we hypothesized that type 2a fCSA would have the strongest relationship to quadriceps peak torque and type 2a/x fCSA and FT% would have the strongest relationship to both early and late RTD. Gaining a deeper insight into the relationship between VL fiber composition and quadriceps muscle performance will help to guide exercise selection aimed at improving specific parameters of quadriceps muscle performance in individuals who have sustained an ACL tear.

METHODS

Experimental Approach to the Problem

This cross-sectional study is a secondary analysis of previously collected baseline data from an ongoing clinical trial. All strength testing and muscle biopsies took place at the University of Kentucky as part of the baseline testing. At the time of the baseline testing, the participants had not received any formal physical therapy for their ACL injury. The time period of testing and biopsy collection for this study was prior to elective surgical ACLR. Peak quadriceps strength was selected as a dependent variable because of its strong association to knee function and patient-reported outcomes in individuals who have had an ACL injury (44). Quadriceps RTD was selected as a dependent variable because of its association to both athletic performance and knee function following ACL injury (43). Although RTD is commonly measured from onset of contraction, assessing the late phase of RTD during a consecutive time period (i.e., 100–200 ms) may better highlight the specific contributions of the investigated muscular factors (43). The fCSA and FT% of the VL were selected as independent variables due to their potential to influence muscular performance (28, 29). The VL was specifically chosen for the site of muscle biopsy due to its mixed fiber type composition, accessibility, and for a more direct comparison to previous studies (16, 23, 28, 42).

Subjects

A total of 26 individuals participated in this study following the approved protocol (Table 1). No a priori sample size calculation was performed for this analysis, since the dataset was taken from an ongoing clinical trial which was powered for a different primary outcome specific to the trial (11). The participants included in this study were all of the individuals to date who had been informed of the risks and benefits of the study and provided written informed consent or parental consent and participant assent for participation in the ongoing clinical trial. Inclusions criterion for the clinical trial were 15–35 years old with a unilateral ACL tear. All ACL tears were confirmed by clinical evaluation and diagnostic testing performed by orthopedic surgeons from the same practice. Participants were excluded if they had any previous history of ACL injury or reconstruction, any previous surgeries or conditions that may affect their gait, if they were skeletally immature, or had a body mass index > 35 kg/m². All participants were full weight-bearing at the time of data collection and were ambulating without an assistive device. All participants were recruited between 2015–2020. This study was approved by the University of Kentucky’s Institutional Review Board for Human Subjects (IRB #42791).

TABLE 1:

Participant Characteristics^a

Characteristic	Ages 15–18	Ages 19–24	Ages 25–32
Sample size, n	10	8	8
Age, y, mean (SD)	16.4 (0.9)	22.1 (1.8)	28.8 (2.2)
Sex, male/female, (% female)	4/6 (60.0)	3/5 (62.5)	6/2 (25.0)
Mass, kg, mean (SD)	73.2 (24.6)	74.4 (16.0)	81.3 (11.0)
Height, m, mean (SD)	1.7 (0.1)	1.7 (0.1)	1.8 (0.1)
BMI, kg/m2, mean (SD)	24.3 (5.5)	25.5 (3.0)	26.7 (2.1)
Pre-injury Tegner, median (min-max)	9 (5–10)	7 (5–10)	8 (5–10)
Time since ACL injury, days, mean (SD)	17.9 (13.8)	28.8 (16.6)	27.3 (11.3)

^aACL = anterior cruciate ligament; BMI = body mass index

Procedures

Maximal Voluntary Isometric Quadriceps Strength

Participants completed maximal voluntary isometric knee extension strength testing of the ACL-injured and non-injured limbs using a Biodex System 4 isokinetic dynamometer (Biodex Medical Systems Inc., Shirley, NY). Each participant was tested on the non-injured limb first, followed by the ACL-injured limb. One submaximal effort practice trial was allowed to familiarize the patient with the testing procedure (26). The participants then performed four 5 second trials separated by 30 seconds of rest. Participants were instructed to kick out as hard and fast as possible and maximal verbal encouragement was given to ensure full effort. Participants were secured in the dynamometer by a shoulder, lap, and thigh strap to ensure the isometric contraction was isolated to the knee joint (12). All trials were performed with the knee at 90° and hip flexed to 85°. The torque signal was sampled at 100 Hz and processed using a custom MATLAB code (MathWorks Inc, Natick, MA). Quadriceps peak torque was quantified as the single highest torque value recorded for each individual trial (Figure 1). RTD was calculated as the slope of the torque-time curve from 0–100 ms and 100–200 ms (Figure 1) (12, 20). Recent work has shown that there is strong correlation between RTD sampled at 100 Hz and 2000 Hz for the investigated time phases of 0–100 ms and 100–200 ms (6). Sampling RTD at 100 Hz may also be more clinically feasible as this is the standard sampling frequency of commercial isokinetic dynamometers. The torque signal was filtered using a fourth-order, low-pass Butterworth zero-lag digital filter with a 24 Hz cut-off frequency. Contraction onset was defined by an absolute threshold of 5 Nm following previously published methods (17). Trials were excluded if the participant had uncontrolled pre-tension or if the participant started the contraction early. The average peak torque and RTD of the four trials was used for subsequent analysis.

FIGURE 1. Torque-time graph from a single participant during isometric quadriceps strength testing.

Peak torque is the highest single torque reading and rate of torque development is the slope of the torque-time curve during the respective time period. RTD = Rate of torque development.

Muscle Biopsy Processing and Immunohistochemical Analysis

The muscle biopsy from the VL was collected on the same day following strength testing. Percutaneous muscle biopsies from the VL were performed using a Bergström 5 mm muscle biopsy needle with suction (40). Approximately 50 mg was mounted in tragacanth gum on cork and flash frozen in liquid nitrogen-cooled 2-methylbutane. Samples were stored at −80°C until processing.

Seven-micron cross-sections were cut in a cryostat and sections were allowed to air dry for 1 hour. For fiber typing, methods were adapted from prior publications (2, 13, 31). Unfixed slides were incubated overnight at room temperature with antibodies against myosin heavy chain isoforms type 1 (#BA.D5, IgG2B), type 2a (#SC.71; IgG1) and type 2x (#6H1, IgM) from Developmental Studies Hybridoma Bank (Iowa City, IA) and laminin (#L9393, Sigma, St. Louis, MO). The next day slides were incubated with immunoglobulin-specific secondary antibodies: goat anti-mouse IgG2b AF647 (#A21242), goat anti-mouse IgG1 AF488 (#A21121), goat anti-mouse IgM AF555 (#A21426) and goat anti-rabbit AF350 (#A11046) all from ThermoFisher (Waltham, MA). Slides were post-fixed in methanol prior to mounting with fluorescent mounting media (Vectashield, Vector Labs, #H-1000, Burlingame, CA). Whole cross-sectional images of the muscle biopsies were captured at ×100 total magnification using the tiles and stitching functions on an Axioimager M2 upright microscope (Zeiss). We quantified fiber type distribution as type 1, type 2a and type 2a/2x (all 2x-expressing fibers concurrently expressed 2a, and are denoted as type 2a/2x) as relative percent distribution by normalizing the specific fiber type relative to the total number of fibers. We also directly assessed CSA (μm²) of each fiber in a fiber-type specific and pooled manner similar to how we have published previously (13). Figure 2 shows representative images from the immunohistochemical analysis.

FIGURE 2. Representative images of fiber type immunohistochemistry.

A) type 1 fibers; B) type 2a fibers; C) type 2x fibers; D) laminin staining; E) merged image denoting all fiber types; F) representative image of entire muscle biopsy cross-section denoting all fiber types. Scale bar = 100 μm (A-E); 1,000 μm (F)

Statistical Analyses

All analyses were conducted using R version 4.1.2 (R Foundation for Statistical Computing, Vienna, Austria), with statistical significance defined using an α-level of 0.05 throughout. Descriptive statistics were calculated for all demographic variables and reported using means and standard deviations for quantitative variables and counts and percentages for categorical variables. Paired t-tests were used to assess differences in muscle fiber characteristics between ACL-injured and non-injured limbs, and were reported along with test statistics and degrees of freedom (measures of effect size) and 95% confidence intervals for the mean paired differences.

To assess the relationship between each of the muscular performance outcomes (QPT, early RTD, and late RTD) and the fiber-specific properties of the VL for the ACL-injured limbs, separate multiple linear regression models were built using an subset selection algorithm (30). The first set of candidate explanatory variables included type 1, type 2a, and type 2a/2x fCSA, along with all possible two-way statistical interaction effects. For each of the three outcomes, subset selection was performed using the “leaps” package in R to identify the candidate variable subset whose corresponding linear model yielded the largest value of adjusted R² (25). The process was repeated for the second set of candidate explanatory variables: type 1, type 2a, and type 2a/2x FT%, along with all possible two-way statistical interactions. The model with the highest overall value of adjusted R² was chosen as the final model for each outcome. After final model selection, pooled fCSA was considered as a possible covariate for each of the chosen models. Partial F-tests were used to evaluate significance of the selected explanatory variable set after accounting for pooled fCSA. To investigate whether or not the same relationships held in the data from the non-injured limb, separate models were fit using the same subset of explanatory variables from the final ACL-injured models. For all models, the validity of assumptions was assessed using univariate analysis and data visualization techniques for the response and candidate explanatory variables, along with overall residual analysis for the selected models. 95% confidence intervals for the value of R² for each regression model were computed using a bias-corrected Smithson estimator (15).

RESULTS

Descriptive Statistics

Table 1 shows the descriptive analysis for the demographic variables and participant characteristics. Quadriceps peak torque and early and late RTD values for the ACL-injured and non-injured limbs can be found in Table 2.

TABLE 2:

Muscular Performance Outcomes^a

Outcome	ACL-injured Limb	Non-injured Limb
Quadriceps Peak Torque (Nm)	156.7 ± 64.4	216.3 ± 74.7
Early RTD 0–100ms (Nm/s)	430.4 ± 250.2	701.3 ± 378.8
Late RTD 100–200ms (Nm/s)	325.2 ± 170.5	468.2 ± 175.6

^aThe values were dervied from the average of four maximal voluntary isometric knee extension test trials. Data is presented as the mean and standard deviaion. RTD = Rate of torque development; ACL = anterior cruciate ligament

Vastus Lateralis Muscle Characteristics

The results of the paired t-tests for the between-limb differences in fiber-type characteristics of the VL can be found in Table 3 and can be visualized in Figure 3 for both the non-injured and ACL-injured limbs, along with test statistics and degrees of freedom (measures of effect size) and 95% confidence intervals.

TABLE 3:

Vastus Lateralis Fiber Type-Specific Characteristics^a

Characteristic	Fiber Type	ACL-injured Limb	Non-injured Limb	p b	Test Statisticc	95% Mean Difference CI
Frequency (%)	Type 1	43.4 ± 11.1	40.8 ± 10.2	.231	1.23	(–6.9, 1.7)
	Type 2a	41.0 ± 11.2	43.0 ± 10.7	.375	0.90	(–2.6, 6.7)
	Type 2a/x	15.6 ± 8.2	16.2 ± 9.0	.765	0.30	(–3.1, 4.1)
Cross-Sectional Area (μm 2 )	Type 1	4205 ± 812	4705 ± 832	.005	3.05	(162, 837)
	Type 2a	4137 ± 987	5165 ± 1118	< .001	5.79	(662, 1393)
	Type 2a/x	3773 ± 1290	4535 ± 222	.032	2.28	(72, 1452)
	Pooled	3994 ± 780	4790 ± 883	< .001	4.68	(445, 1145)

^aData presented as the mean and standard deviation. ACL = anterior cruciate ligament; CI = confidence interval.

^bPaired t-tests with 25 degrees of freedom (df), significance level α = .05.

^cTest statistic (t-statistic with 25 df) represents the effect size.

FIGURE 3. Vastus lateralis fiber type cross-sectional area.

Fiber type-specific cross-sectional area (CSA) for the ACL-injured and non-injured limbs with their individual data points overlaid in box and whisker plots. Median values are represented by the bold center line. ACL = anterior cruciate ligament; significance level α = .05.

Multiple Linear Regression Analysis

For QPT, the candidate model selected included the following explanatory variables: type 1 fCSA, type 2a fCSA, and their interaction effect. These terms explained 63.56% (F(3,22) = 12.791, p < 0.001) of the variability in QPT in the non-injured limb and 54.53% in the ACL-injured limb (F(3,22) = 8.796, p < 0.001) (Table 4). The relationship between type 1 fCSA, type 2a fCSA, and QPT in the ACL-injured limb can be visualized in a three-dimensional (3D) scatter plot available as Supplemental Digital Content 1.

TABLE 4:

Candidate Multiple Linear Regression Model^a

	ACL-injured Limb			Non-injured Limb
Outcome Variable	R2	95% CI	p b	R2	95% CI	p b
Quadriceps Peak Torque	0.545	(0.293, 0.744)	< .001	0.636	(0.402, 0.798)	< .001
Early RTD (0–100 ms)	0.235	(0.130, 0.527)	.11	0.382	(0.154, 0.639)	.012
Late RTD (100–200 ms)	0.363	(0.142, 0.625)	.017	0.559	(0.309, 0.752)	< .001

^aModel using the explanatory variables: type 1 cross-sectional area (CSA), type 2a CSA, and their interaction effect; ACL = anterior cruciate ligament; RTD = rate of torque development; CI = confidence interval.

^bSignificance level α = .05.

For both early (0–100ms) and late (100–200ms) RTD, the candidate model selected included the following explanatory variables: type 1 fCSA, type 2a fCSA, and their interaction effect. These terms explained 38.23% (F(3,22) = 4.539, p = 0.012) of the variability in early RTD in the non-injured limb and 23.51% (F(3,22) = 2.25, p= 0.11) in the ACL-injured limb (Table 4). The candidate model failed to reach significance in the ACL-injured limb. For late RTD, these terms explained 55.94% (F(3,22) = 4.931, p < 0.001) of the variability in late RTD in the non-injured and 36.26% (F(3,22) = 4.172, p = 0.017) in the ACL-injured limb (Table 4) (SDC Figures 2 and 3). The relationship between type 1 fCSA, type 2a fCSA, and early and late RTD in the ACL-injured limb can be visualized in a 3D scatter plot available as Supplemental Digital Content 2 and 3 respectively.

The model which contained pooled fCSA alone failed to provide the level of granularity achieved in the candidate model using type 1 fCSA, type 2a fCSA, and their interaction effect. This indicated that fiber type-specific fCSA is better able to distinguish strength predictions across all participants when compared the models using pooled fCSA alone.

DISCUSSION

The primary objective of this study was to investigate the relationship between muscle fiber-type specific properties (fCSA and FT%) of the VL and quadriceps strength-related outcomes after ACL injury (QPT, early RTD, and late RTD). We also sought to determine the differences in fiber type-specific CSA and FT% between ACL-injured and non-injured limbs. The main findings of this study were that type 1 fCSA, type 2a fCSA and their interaction effect had the strongest relationship to quadriceps muscle performance (QPT, early RTD, and late RTD) in both ACL-injured and non-injured limbs (Table 4). Additionally, we found that participants’ ACL-injured limbs had significantly smaller fCSA across all fiber types compared to the non-injured limb (Table 3, Figure 3). Fiber type percent distribution had no significant effects on muscular performance and we found no between-limb differences in FT%. Overall, our findings suggest that type 1 and type 2a fCSA have a strong relationship to quadriceps muscle performance after ACL injury.

We found that type 1 fCSA, type 2a fCSA and their interaction effect predicted over half of the variance in QPT in both the ACL-injured and non-injured limbs (Table 4). These findings support prior work suggesting that peripheral muscular factors are strong contributors to maximal strength in both uninjured individuals and those after ACLR. Although the relationship between type 2a fCSA and peak strength was included in our initial hypothesis, the addition of type 1 fCSA to the candidate model was surprising based on their slow twitch characteristics (14, 21). There are several potential explanations for these results. In contrast to type 2a/x fiber expression which is highly variable based on athletic background and activity level, type 1 fiber expression remains relatively constant and exhibits less variance in the VL than type 2 fibers (41). Type 1 fibers also comprise almost half of the fiber pool of the VL, making them large contributors to the total muscular CSA (42). Additionally, type 1 fibers have been found to produce equivocal amounts of force per CSA when compared to the larger type 2a fibers (45).

Early RTD (0–100ms) had the weakest relationship to the fiber-specific variables (type 1 fCSA, type 2a fCSA, and their interaction effect) compared to the other muscular performance outcomes (Table 4). Early RTD is thought to be governed primarily by neural factors such as motor unit recruitment and discharge rate, which may help to explain these findings (26). After ACLR, deficits in early RTD in the injured limb have shown correlations to reductions in corticospinal excitability, which may help to explain why the ACL-injured limb failed to reach significance (37).

While it has been hypothesized that type 2a/x fiber size and FT% play a role in the development of early RTD, research in this area is limited and has been conducted primarily on small homogenous groups of healthy males (26, 28). For example, Methenitis et al. found a moderate relationship between early rate of force development (RFD) and type 2x fiber size and percent distribution, but the results are only generalizable to power-trained males (n = 10) (28). Other work by Hvid et al. found a relationship between type 2 fiber size and early RTD in young males, but the sample size was small (n = 11), and the relationship was only observed prior to the study intervention (16). Although the current study did not find any effect of type 2a/x fiber size or FT% on early RTD in non-injured limbs, the heterogeneity of our participants in regard to sport activity and biological sex make our findings more generalizable.

In contrast to early RTD, late RTD is thought to be more dependent on peripheral muscular properties that influence maximal strength capacity (i.e., type 2 fCSA and FT%) (7). Our results found that type 2a fCSA and type 1 fCSA play a significant role in the development of late RTD (100–200ms), particularly in non-injured limbs. These findings support previous work suggesting that peripheral muscular factors are the main determinants of later phases of RTD; however, the contribution of type 1 fibers to late RTD has not yet been established (28, 29). In contrast to previous work suggesting that increased type 1 fCSA is negatively correlated to RTD, we found that type 1 fCSA had a positive relationship with late RTD (5, 28, 38). The lower activation threshold of type 1 fibers and the ability of slow-twitch fibers to resist fatigue during repeated maximal effort contractions may help to explain these novel findings (5, 26, 38).

The predictor variables explained 23–36% of the variance in early and late RTD respectively, suggesting that other factors may also influence quadriceps RTD after an ACL tear, particularly in the early phase. Arthrogenic muscle inhibition (AMI) is common after an intra-articular knee injury and can cause alterations in the neural input to the quadriceps muscle (36). Acute manifestations of AMI decrease the number of available motor units within the quadriceps and impair voluntary motor unit recruitment (33, 36). Considering RTD is influenced by the ability to actively recruit motor neurons and the speed at which they discharge, AMI would likely impair RTD in the ACL-injured limb (8, 26). We also cannot rule out that psychological factors such fear, depression, and anxiety may have negatively impacted effort and confidence during testing of the ACL-injured limb (27).

Despite our initial hypothesis that only type 2a fibers would be smaller in the ACL-injured limb, all fiber types were significantly smaller compared to the non-injured (Figure 3). These findings are in contrast to previous studies investigating morphological changes in the VL after ACL injury and reconstruction showing selective loss of type 2 fCSA (23, 32). However, these studies either had small sample sizes or only investigated post-operative changes in fiber size, which limits the comparison of the findings to the results of the current study (23, 32). A more robust body of research exists regarding fiber-type specific changes after periods of immobilization, which finds similar patterns of atrophy across all fiber types (16). Despite these similarities, the mechanisms driving muscular atrophy following immobilization of healthy limbs differ from those observed following a traumatic joint injury (22).

Muscular atrophy as a result of immobilization is in response to muscular inactivity and is primarily driven by an increase in protein catabolism, leading to an imbalance between protein breakdown and synthesis (3, 4). In contrast, loss of quadriceps muscle mass after an ACL tear is mediated by deficits in protein synthesis (18). Additionally, myostatin expression within the quadriceps is upregulated after an ACL tear, further contributing to suppressed muscular growth (18, 34). These alterations in protein anabolic signaling and myostatin expression have been observed as soon as one day after injury, suggesting that quadriceps atrophy begins almost immediately after an ACL tear (18). Although we cannot rule out that participants in this study had a brief period of disuse immediately following the initial injury, it is likely that the widespread fiber atrophy we observed was a direct result of the ACL tear.

We observed no differences in fiber distribution between limbs; however, it is possible that the time between initial ACL injury and muscle biopsy (24.1 ± 14.8 days) was not long enough for these fiber-type transitions to occur (Table 3) (22). Previous work by Noehren et al. observed a reduction in type 2a fiber frequency and increase in the abundance of type 2a/x in the VL after ACLR (32). The findings by Noehren et al. after ACLR are in contrast to the typical pattern of fiber-type transition observed after periods of disuse or inactivity, which involves a decrease in the percentage of type 1 fibers and increase in type 2 fiber frequency (4, 35). These conflicting results suggest that the pattern and degree of fiber-type transition likely differs based on the context of the situation (i.e., traumatic injury versus disuse) and the length of time since injury (16).

This work is not without limitations. One limitation was the variability amongst literature for defining periods of “early” and “late” RTD (26). Since the timeframes investigated vary from study to study, it made comparison of our findings to previous work more challenging. Another limitation of this study is the reference of the non-injured limb as a baseline for normal or healthy. However, considering our data was collected within days to weeks of the initial injury, it is reasonable to suggest that the non-injured limb had not yet undergone alterations in strength and RTD (43). We also acknowledge that we do not have pre-injury data and therefore cannot reasonably speculate that ACL-injured limbs had fiber-specific atrophy of the quadriceps, only that we observed smaller fCSA compared to non-injured limbs at the time of the muscle biopsy. Additionally, it was beyond the scope of this study to consider the effects of psychological factors, neural factors, and other muscular properties that could influence performance of the ACL-injured limb. Lastly, RTD measures were collected at 100 Hz, which is lower than the recommended 1000 Hz; however, recent work found high agreement between RTD measures at high and low sampling rates (6).

In conclusion, the strong relationship between type 1 and type 2a fiber size and muscular performance highlights the importance of both mitigating atrophy and restoring muscle size after ACL injury. With the increasing number of individuals electing non-operative management following ACL injury, understanding the nature and magnitude of muscular changes becomes even more important for improving patient outcomes (10, 22). The findings of this study provide a critical framework for future studies to investigate the efficacy of specific rehabilitation strategies aimed at restoring quadriceps muscle function after ACL injury (10, 22).

PRACTICAL APPLICATIONS

Practitioners working with individuals prior to ACLR or with those choosing non-operative management should consider the underlying muscular factors affecting quadriceps muscle performance. The findings of this study suggest that exercises focused on hypertrophy may be warranted due to the significantly smaller fCSA in the ACL-injured limb and the strong relationship between fiber size and muscular performance. It should be taken into consideration that the intensity of an exercise must be relatively high (> 65% of 1RM) and repetitions performed within close proximity of muscular failure in order to effectively train for muscular hypertrophy (39). Resistance training at high levels of volume and intensity should only be initiated once an individual has obtained a quiet knee (i.e., little to no swelling or pain). Practitioners should work to alleviate acute knee symptoms with evidence-based interventions such as cryotherapy and transcutaneous electrical nerve stimulation in order to mitigate the negative effects of AMI on muscular performance (33).

Supplementary Material

Supplemental Digital Content 2

3D scatter plot of the relationship between type 1 and type 2a fiber cross-sectional area (CSA) and early rate of torque development (RTD) for anterior cruciate ligament (ACL) injured limbs. Type 1 CSA is represented on the x-axis, type 2a CSA on the y-axis, and early RTD on the z axis.

Supplemental Digital Content 3

3D scatter plot of the relationship between type 1 and type 2a fiber cross-sectional area (CSA) and late rate of torque development (RTD) for anterior cruciate ligament (ACL) injured limbs. Type 1 CSA is represented on the x-axis, type 2a CSA on the y-axis, and late RTD on the z axis.

Supplemental Digital Content 1

3D scatter plot of the relationship between type 1 and type 2a fiber cross-sectional area (CSA) and quadriceps peak torque for anterior cruciate ligament (ACL)-injured limbs. Type 1 CSA is represented on the x-axis, type 2a CSA on the y-axis, and peak torque on the z axis.