Substantial atrophy of the quadriceps femoris muscle is observed in patients after anterior cruciate ligament reconstruction (ACL-R) persisting even 1 yr after ACL surgery and completion of a 6-month guided rehabilitation (1,2). The importance of quadriceps strength, which is strongly correlated with quadriceps muscle mass, is overall accepted for a successful return to sports after ACL-R (3,4) and for coping with ACL injury without surgery (5,6). After ACL injury, the loss of muscle mass is attributed to impaired neuromuscular function (5) and postsurgery inflammation (7). However, the loss of muscle mass is certainly aggravated by postsurgery immobilization (8) and probably also by injuries due to the graft surgery (9), especially with the use of quadriceps tendon or patella tendon autografts (10). Physical rehabilitation programs starting immediately after surgery and lasting from 12 to 24 wk are designed to support the regeneration of tendinous and muscle tissue at the harvest sites; however, it is only after 3–4 months postoperatively that the focus of such rehabilitation programs is on muscular volume and strengthening (1).
To date, little is known about morphological and cellular alterations in the quadriceps femoris muscle after ACL injury or ACL-R. Muscle fiber atrophy has repeatedly been described with inconsistent findings whether type I or type II fibers are most affected (11–13). In a recently published study, biopsies of the vastus lateralis muscle were simultaneously obtained from the injured and the uninjured leg some days before ACL-R, which was conducted within 2 months after ACL injury. A significantly reduced satellite cell (SC) number was observed in the injured compared with the uninjured leg. SC number was also diminished after surgery and rehabilitation at the time point when athletes resumed sport-specific drills. At this time point, the proportion of type IIA fibers was decreased, too (13). The same research group found increased fibroblast and fibrogenic/adipogenic progenitor cell content in the biopsies obtained before surgery and concluded that these findings might explain skeletal muscle maladaptation after ACL injury (14). The need to efficiently counteract quadriceps atrophy and weakness after ACL-R initiated a study on perioperative testosterone supplementation as a useful adjunct therapy in healthy men (15). Furthermore, the effects of pharmacological inhibition of myostatin after ACL injury were investigated in an animal study (16). However, in athletes, these pharmacological interventions are no options to promote muscle regeneration after ACL-R or ACL injury as testosterone supplementation and the pharmacological inhibition of myostatin are found on the World Anti-Doping Agency prohibited list (17). Instead, the focus should be on optimizing quadriceps strength training during rehabilitation after ACL-R.
To our best knowledge, there are no studies on the effects of supervised progressive quadriceps strength training after early rehabilitation after ACL-R on muscle regeneration and the morphological and cellular changes induced by such training are unknown. We and others have shown that combined concentric/eccentric strength training with eccentric overload (CON/ECC+) is superior to conventional concentric/eccentric strength training (CON/ECC) with regard to the development of muscle hypertrophy and a fast muscle phenotype (18–21). Furthermore, differently affected SC activity by the contraction mode of skeletal muscle is reported with a greater increase in SC content in type II muscle fibers after eccentric exercise (22,23). Microinjuries, induced by functional overload or electrical stimulation, are thought to stimulate SC activation, proliferation and, eventually, differentiation with myofiber formation (22,24–26). We hypothesized that CON/ECC+ training enhances muscle regeneration during rehabilitation after ACL-R compared with CON/ECC training due to increased SC activity with a greater increase in muscle mass and development of a faster muscle phenotype.
Sixty-eight recreational athletes who underwent standardized ACL-R (55 men, 13 women) with either quadriceps tendon (QUAD; n = 32) or semitendinosus tendon autograft (SEMI; n = 26) volunteered to participate in a 12-wk strength training program after finishing 12 wk of early rehabilitation and after being cleared for strength training by an orthopedic surgeon. They were randomly assigned to 12-wk supervised quadriceps training performed either as conventional concentric/eccentric (CON/ECC) leg-press training (Compass; Proxomed, Wolfratshausen, Germany) or as computer-guided leg-press training (IsoMed 2000; D&R Ferstl, Hemnau, Germany) with eccentric overload (CON/ECC+). The sampling of muscle biopsies from the vastus lateralis muscle of the injured leg was optional. Fifty-eight of the 68 athletes participated regularly in the strength training, and from a subgroup of 37 athletes, muscle biopsies from the vastus lateralis muscle could be obtained before and after the 12-wk training period (CON/ECC: n = 16, age, 26 ± 5 yr; height, 178.7 ± 5.9 cm; weight, 79.0 ± 12.7 kg; CON/ECC+: n = 21; age, 24 ± 4 yr; height, 182.3 ± 6.8 cm; weight, 87.0 ± 15.7 kg). The study was approved by the local ethics committee and conducted in accordance with the Declaration of Helsinki. Each subject provided written informed consent to participate.
At the time point of ACL surgery, otherwise healthy recreational athletes who met the inclusion criteria (age, 18–35 yr; first traumatic ACL rupture within the last year without further serious knee injury or signs of arthrosis) were asked to participate in the study. After finishing the regularly scheduled 12-wk early rehabilitation with physiotherapy treatment and after clearance for strength training and coordinative exercises by an orthopedic surgeon, magnetic resonance imaging (MRI) of both thighs for determination of muscle cross-sectional area (MCSA), and isokinetic strength tests were performed. The same tests were conducted after completing the 12-wk quadriceps strength training. Biopsies from the vastus lateralis muscle of the injured leg were obtained before and after the 12-wk training period at least 2 d after the last training session or 2 d after the strength tests, respectively.
Progressive resistance training was conducted twice a week (Monday and Thursday or Tuesday and Friday, respectively) for 12 wk under the supervision of a certified rehabilitation coach. Before each training session, the subjects completed a standardized warm-up program and exercises for improvement of coordination. The subjects then performed one-legged concentric/eccentric leg-press training on either a conventional device (Compass; Proxomed; CON/ECC) or on a computer-driven isokinetic device that allows for concentric/eccentric training with eccentric overload (IsoMed 2000; D&R Ferstl; CON/ECC+). The sitting position was similar in both leg-press situations with a hip angle of approximately 75°–80°. Shoulders were fixed with pads. The motion ranged from approximately 90° to 0° of knee flexion (0°, full knee extension). All subjects performed six sets of eight repetitions at the eight-repetition maximum; that is, the load was chosen to cause exhaustion after eight repetitions on both training devices. The series of eight repetitions was separated by 90-s rest.
In the case of CON/ECC, load was controlled by weight. The same absolute load was applied in the concentric and eccentric phases. After 6, 12, and 18 training sessions, the applied weight was enhanced to adjust for eight-repetition maximum. That led to an average of 2.2-fold increase in the load throughout the study (Table 1).
In CON/ECC+, load was controlled by velocity. At the beginning of the training period, the leg-press plate moved with the same velocity (200 mm·s−1) during the concentric and eccentric phases. After 6, 12, and 18 training sessions, velocity was increased during the eccentric and decreased during the concentric phase to 400 and 100 mm·s−1, respectively (Table 1). Subjects pressed against the leg-press plate moving away from the body during the concentric phase. During the eccentric phase, the leg-press plate moved toward the subjects’ body against the resisting extensor muscle group. Subjects were instructed to develop maximal force onto the leg-press plate during the eight repetitions in the concentric and eccentric phases, which resulted in an average of 1.14-fold higher eccentric than concentric load at the beginning and an average of 1.24-fold higher eccentric load at the end of the 12-wk training period. Throughout the study, the subjects were not allowed to perform any other training. Furthermore, they had to refrain from the consumption of any nutritional supplements, especially from protein or creatine supplements.
Isokinetic muscle strength testing
After a standardized warm-up procedure on a bicycle ergometer (10 min at 1.5 W·kg−1 bodyweight, 60–70 U·min−1), maximum strength was assessed using the leg-extension module of the IsoMed 2000 (D&R Ferstl). For familiarization, patients completed a five-repetition warm-up of leg extension in a concentric mode with increasing but submaximal intensity. After 1-min rest, peak torque was determined out of three repetitions at an angular velocity of 60°·s−1 and out of five repetitions at 180°·s−1 between 10° and 90° of knee flexion for both legs separately. Subjects were seated upright with 90° of hip flexion and with their arms folded across the chest. They were secured with shoulder pads and with a belt across the hip; the thigh was fixed with a belt as well. The axis of rotation was set in the middle of the lateral femoral condyle. The height of the force application pad was set in the middle of the tibia for secure leg-extension testing in ACL patients.
MRI of both thighs was performed in the supine position on a 3.0-T clinical whole-body magnetic resonance (MR) system (MAGNETOM Verio; Siemens Healthineers, Erlangen, Germany) using the manufacturer’s standard body-phased array coil for signal reception. The imaging protocol comprised an axial T1-weighted spin-echo sequence (repetition time (TR)/echo time (TE) in milliseconds, 700/15) and a coronal T1-weighted turbo-spin-echo sequence (TR/TE, 852/12) with a section thickness of 8 mm. Moreover, an axial short tau inversion recovery sequence (TR/TE, 2690/38; inversion time TI, 210 ms) and a coronal short tau inversion recovery sequence (TR/TE, 3940/38; inversion time TI, 210 ms), both with a section thickness of 8 mm, were also applied.
All MR images were transferred and displayed as softcopies in fully electronic monitored fashion using a picture archiving and communication system (PACS; Centricity PACS 4.0; GE Healthcare, Barrington, IL) with large-screen high-resolution displays, which easily enabled the review of eight images simultaneously. The MR image analysis was performed by two experienced readers in consensus. Each reader also had the opportunity to individually select the different MR sequences using the PACS to, for example, clarify the facial boundaries of the quadriceps muscle. All MR examinations were jointly randomized and presented to the readers. Identifying parameters such as patient’s name were omitted. Both readers were blinded to the clinical and training data of the subjects.
Cross-sectional area of both quadriceps femoris muscles (MCSA) was determined at 20-cm distance from the trochanter major on axial T1-weighted images before and after the training period according to previous recommendations (20). MCSA was recorded separately for the right and left thighs. MCSA of the whole quadriceps femoris muscles was determined, because the facial boundaries between the lateral and deep vastus muscles could not always be clearly identified, as described previously (19).
Muscle biopsy sampling
Before and after the 12-wk training period, muscle biopsy samples were taken from the same region at mid-thigh level of the vastus lateralis muscle of the injured leg under local anesthesia, using the Bergström technique (27). The muscle tissue was immediately freed from blood and visible connective tissue, rapidly frozen in isopentane cooled by liquid nitrogen and subsequently stored at −80°C. To avoid residual effects from preceding biopsies, biopsy sites were spaced apart by approximately 1 cm from proximal to distal.
All biopsies were assigned a random unique identification number, thereby blinding the investigator to subject identity and time point. Immunofluorescent analyses were performed on serial sections for SC number, activated SC number, fiber-type distribution, and myofibers expressing neonatal myosin (MHCneo). For all these analyses, transverse sections (10 μm) were cut in a cryostat (−20°C), mounted on Super Frost Plus slides (ThermoFisher Scientific, Waltham, MA) and allowed to dry for 20 min. Sections were then fixed in a 4% paraformaldehyde (Roti-Histofix 4%, catalog no. P087; Carl Roth, Karlsruhe, Germany) for 10 min, washed 3 × 5 min in phosphate-buffered saline (PBS) and blocked for 10 min at room temperature (RT) in PBS containing 0.5% Triton (Triton X-100, catalog no. 108643; Merck Millipore, Billerica, MA) and 1% bovine serum albumin (Albumin Bovine Fraction V, catalog no. 11922; Serva Electrophoresis, Heidelberg, Germany).
For determination of SC number, staining of Pax7, laminin, type II myosin, and nuclei was done on the same section. Sections were incubated overnight at RT with the primary antibodies anti-Pax7 (1/500, catalog no. MAB1675; Bio-Techne (R&D Systems), Minneapolis, MN), anti-Laminin (1/1000, catalog no. ab11575; Abcam, Cambridge, UK), and anti–Fast Myosin Skeletal Heavy chain (catalog no. ab91506; Abcam). The next day, the sections were washed in PBS followed by incubation in appropriate secondary antibodies: Alexa Fluor 488 goat–antimouse (Pax7; 1/1000, catalog no. A.11029; ThermoFisher (Invitrogen)) and Alexa Fluor 568 goat–antirabbit (Laminin and myosin heavy-chain (MHC) II; 1/1000, catalog no. A.11036; ThermoFisher (Invitrogen)) for 2 h at RT. Then, the sections were washed in PBS, and nuclei were visualized with bisBenzimide H33258 (catalog no. B2883; Sigma-Aldrich, St. Louis, MO). Finally, sections were washed in PBS, were allowed to dry completely in the dark, and were then fixed with FluorSave Reagent (catalog no. 345789; Merck Millipore) at 4°C for 24 h. Staining specificity was confirmed using appropriate negative controls.
For the analysis of myogenic activation, sections were incubated overnight at RT with primary antibodies against Pax7 (1/500, catalog no. MAB1675; Bio-Techne (R&D Systems)) and against MyoD (1/500, catalog no. ab133627; Abcam). The secondary antibodies Alexa Fluor 488 goat–antimouse and Alexa Fluor 568 goat–antirabbit were used for visualization of Pax7 and for MyoD, respectively, and the staining procedure was finished as described prevously.
To enable categorization into type I, type II, or hybrid fibers, sections were incubated overnight at RT with the primary antibodies anti-Myosin (Skeletal, Slow; MHC I, 1/6000, catalog no. M8421; Sigma-Aldrich) and anti–Fast Myosin Skeletal Heavy chain (MHC II, 1/1000, catalog no. ab91506; Abcam). The secondary antibodies Alexa Fluor 488 goat–antimouse and Alexa Fluor 568 goat–antirabbit were used for visualization of MHC I and MHC II, respectively, and the staining procedure was finished as described previously.
Myofibers expressing MHCneo
Regeneration/remodeling was assessed by staining of sections for neonatal MHC. Sections were incubated overnight with the primary antibodies anti-MHCneo (1/75, NCL-MHCn; Leica Biosystems, Wetzlar, Germany), anti-Laminin (1/1000, catalog no. ab11575; Abcam), and anti–Fast Myosin Skeletal Heavy chain (catalog no. ab91506; Abcam). The secondary antibody Alexa Fluor 488 goat–antimouse was used for visualization of MHCneo and Alexa Fluor 568 goat–antirabbit for laminin and for MHC II. The staining procedure was finished as described previously.
Images were obtained at ×20 and at ×40 magnification using the fluorescence microscope Axio Observer and the appropriate software program Zen (Carl Zeiss, Oberkochen, Germany). On average, 484 ± 286 myofibers per section were analyzed before and 628 ± 316 myofibers per section after the 12-wk training period. The criterion for an SC was simultaneous Pax7 and myonucleus (bisBenzimide H33258) staining within the myofiber (sublaminar). An activated SC was characterized by simultaneous MyoD, Pax7, and bisBenzimide staining. The number of Pax7-positive (Pax7+) cells (SC) associated with MHC I+ (type I) or MHC II+ (type II) fibers was quantified, and the number of SC divided by the total number of myofibers was categorized as SC per fiber. MyoD-positive SC was classified as activated SC and divided by the total number of myofibers. The number of MHCneo-positive fibers was expressed relative to total fiber number from the section. Fiber cross-sectional areas (FCSA) were measured in images obtained at ×20 magnification using the software program Image J for Windows (National Institute of Health, Bethesda, MD). Only fibers cut perpendicular to their longitudinal axis were used for determination of fiber size. The goal was to measure 50 fibers of each type, but this could not be realized in some cases. On average, FCSA of 44 type I, 42 type II, and 26 hybrid fibers could be determined in 14 subjects of the CON/ECC and in 12 subjects of the CON/ECC+ training group, respectively.
All statistical tests were processed using SPSS 24.0 software for Windows (SPSS Inc., Chicago, IL). The effects of strength training were assessed using mixed ANCOVA repeated-measures analysis with time (before and after the 12-wk training period) as a within-subject variable and training group (CON/ECC and CON/ECC+) as a between-subject variable. Type of ACL autograft (QUAD and SEMI) was entered as a covariate. First, the normal distribution of all data was checked using the Kolmogorov–Smirnov test. This test failed for activated SC and myofibers expressing MHCneo, and these data could not be transformed to normal distribution. Therefore, the Wilcoxon signed-rank test was applied to test these data for within-subject effects (before and after training) and the Mann–Whitney U test was used to test for between-subject effects (CON/ECC vs CON/ECC+). In addition, effect sizes were determined by partial eta-squared (ηp2). Post hoc analyses were conducted where appropriate, and a Bonferroni correction was used for multiple comparisons. To investigate the relationship between selected parameters, Pearson product–moment correlations were calculated. For all analyses, a value of P < 0.05, set a priori, was considered to represent statistical significance. All data are presented as mean ± SD.
There was a significant (P = 0.002) time–group interaction for the MCSA measured by MRI (F(1.32) = 11.266, np2 = 0.260) and a significant (P < 0.001) main time effect (F(1.32) = 18.221, np2 = 0.363). Post hoc testing revealed a significant (P < 0.001) increase in MCSA after the 12-wk training period in both training groups, by an average 4% in CON/ECC and 11% in CON/ECC+. No significant group effect occurred; however, after the 12-wk training period, there was a trend (P = 0.052) toward greater MCSA in the CON/ECC+ training group (Fig. 1). Type of autograft did not significantly affect the test results.
When MCSA of the injured leg was related to MCSA of the uninjured leg (limb symmetry index, or LSI), a significant (P = 0.008) time–group interaction (F(1.32) = 8.011, np2 = 0.200) and a significant (P = 0.001) main time effect (F(1.32) = 13.554, np2 = 0.298) were observed. No significant group effect occurred. Type of autograft did not significantly affect MCSA.
No significant time–group interactions were observed for the FCSA of type I, type II, and hybrid fibers. There were significant main time effects for the FCSA of all fiber types (F(1.23) = 4.531, np2 = 0.165, P < 0.05 (type I); F(1.23) = 6.311, np2 = 0.215, P < 0.05 (type II); and F(1.18) = 10.408, np2 = 0.366, P < 0.01 (hybrid), respectively). Post hoc testing revealed a significant (P ≤ 0.001) increase in the FCSA of all fiber types after the 12-wk training period in both training groups. No significant group effect occurred (Fig. 1). Type of autograft significantly (P < 0.05) affected FCSA of type I fibers (F(1.23) = 5.545, np2 = 0.194), with significantly greater FCSA in the SEMI group (4637 ± 736 vs 5213 ± 1037 mm2) than in the QUAD group (3676 ± 836 vs 4265 ± 836 mm2), but did not have an effect on FCSA of type II and hybrid fibers.
The FCSA of all fiber types were significantly correlated with MCSA in the CON/ECC group (type I: r = 0.636, P < 0.001; type II: r = 0.627, P < 0.001; hybrid: r = 0.448, P < 0.05). In the CON/ECC+ group, significant correlations between FCSA of type II (r = 0.623, P < 0.01) and of hybrid fibers (r = 0.661, P < 0.01) were found; for type I fibers, the level of significance was not reached (r = 0.434, P = 0.056). No significant correlations were observed between the changes in MCSA (ΔMCSA) and the changes in FCSA (ΔFCSA) of all fiber types in both training groups.
No significant time–group interaction was observed for the peak torque measurement during leg extension at angular velocities of 60°·s−1 and 180°·s−1. There were significant (P < 0.001) main time effects for the test results at both angular velocities (F(1.34) = 17.904, np2 = 0.345 (60°·s−1) and F(1.34) = 19.454, np2 = 0.364 (180°·s−1), respectively). Post hoc testing showed significant (P < 0.001) increases in the peak torque at both angular velocities in both training groups. No significant group effect occurred (Fig. 1). Type of autograft significantly (P < 0.05) affected the test results at both angular velocities (F(1.34) = 4.729, np2 = 0.122 (60°·s−1) and F(1.34) = 5.189, np2 = 0.132 (180°·s−1), respectively), with a higher peak torque at both angular velocities in the SEMI group (143.7 ± 48.3 vs 186.1 ± 44.6 N·m at 60°·s−1 and 116.1 ± 30.7 vs 142.2 vs 30.5 N·m at 180°·s−1) than in the QUAD group (107.4 ± 32.4 vs 167.2 ± 44.2 N·m at 60°·s−1 and 72.3 ± 16.5 vs 88.7 vs ± 14.3 N·m at 180°·s−1).
When peak torque measurements of the injured leg were related to the respective measurements of the uninjured leg (LSI), no significant time–group interactions were found for the test results obtained at both angular velocities. There were significant (P < 0.001) main time effects for the test results at both angular velocities (F(1.34) = 19.454, np2 = 0.364 (60°·s−1) and F(1.34) = 16.926, np2 = 0.332 (180°·s−1), respectively). Post hoc testing showed significant (P < 0.001) increases in the peak torque related to the respective measurements of the uninjured leg at both angular velocities in both training groups. No significant group effect occurred (Fig. 1). Type of autograft significantly (P < 0.05) affected the test results at an angular velocity of 180°·s−1 (F(1.34) = 4.585, np2 = 0.119), with higher values in the SEMI (72.3% ± 16.5% vs 88.7% ± 14.3%) than in the QUAD group (58.1% ± 15.2% vs 80.6% ± 20.7%).
Peak torque at both angular velocities was significantly (P < 0.001) correlated with MCSA in both training groups (60°·s−1: r = 0.682 (CON/ECC), r = 0.619 (CON/ECC+); 180°·s−1: r = 0.799 (CON/ECC), r = 0.772 (CON/ECC+). However, the increase in peak torque at an angular velocity of 180°·s−1 was significantly (P < 0.01) correlated with the increase in MCSA only in the CON/ECC+ group (r = 0.664, CON/ECC: r = 0.122). The correlation between the increase in peak torque at an angular velocity of 60°·s−1) did not reach the level of significance (r = 0.391, P = 0.098 (CON/ECC+); r = 0.031 (CON/ECC)).
No significant time–group interactions were observed for the SC number in relation to total myofiber number or related to type I or type II fibers, respectively. There were no significant time effects and no significant group effects for these parameters. Type of autograft did not significantly affect SC number (Fig. 2, Table 2).
Myogenic cell activation
Only after 12 wk of CON/ECC+ training, there was a tendency (P = 0.071) for activated SC number (Pax7+/MyoD+) to be increased. However, before training, the number of activated SC was significantly (P < 0.05) lower in the CON/ECC+ group compared with the CON/ECC group (Fig. 2, Table 2). Although no significant correlation was observed between the total SC number per fiber and the number of activated SC number, the changes in the number of activated SC number (ΔPax7+/MyoD+) were significantly correlated with the changes in total SC number (ΔPax7+) per fiber in the CON/ECC training group only (r = 0.564, P < 0.05; CON/ECC+: r = 0.278).
Indices of Remodeling
No significant time–group interactions were observed for the percentage of type I, type II, and hybrid fibers. A significant (P < 0.05) main time effect was found for type I fibers (F(1.41) = 1.630, np2 = 0.038). Post hoc testing revealed a significant (P < 0.05) increase in the number of type I fibers in the CON/ECC+ training group only. No main time effects for type II and hybrid fibers were observed. However, there was a tendency for a decrease in the number of hybrid fibers in both training groups (CON/ECC: P = 0.055; CON/ECC+: P = 0.05) No significant group effect occurred (Fig. 3). Type of autograft did not significantly affect fiber-type distribution.
Myofibers expressing MHCneo
After the 12-wk training period, the number of MHCneo and MHC I expressing myofibers (MHCneo+/MHC I+) was significantly (P < 0.05) higher in the CON/ECC+ compared with the CON/ECC training group. MHCneo+/MHC I+ fibers showed a nonsignificant decrease after 12-wk CON/ECC training, whereas a nonsignificant increase was observed after CON/ECC+ training. No significant differences (after vs before training or between training groups) were observed for MHCneo+/MHC II+ fibers (Fig. 3).
To our knowledge, this is the first study to investigate the effects of supervised quadriceps strength training on muscle regeneration after ACL-R. Twelve weeks of one-legged quadriceps strength training performed either as conventional leg-press training or as leg-press training with eccentric overload induced significant increases in quadriceps muscle mass and in quadriceps strength. The regain of muscle mass was significantly enhanced in the CON/ECC+ compared with the CON/ECC training group. However, although the increases in MCSA were significantly correlated with the increases in quadriceps strength in the CON/ECC+ training group only, the increase in quadriceps strength was not enhanced after CON/ECC+ training compared with conventional leg-press training. The lack of difference in the development of quadriceps strength between both training groups might, at least partly, be explained by the surprising significant increase in the proportion of type I fibers in the CON/ECC+ training group. Although we found signs of myogenic cell activation and remodeling, our hypothesis that eccentric-overload training would induce an increased SC activation thereby promoting muscle hypertrophy and a shift toward a faster muscle phenotype is not supported for rehabilitation strength training after ACL-R.
Asymmetry in quadriceps muscle volume and in quadriceps strength due to the substantial atrophy of the quadriceps femoris muscle after ACL-R is regarded as a main risk factor for secondary knee injury (1,3,4,28) and probably provides an explanation for the relatively low proportion of athletes successfully returning to sports after ACL-R. It was reported that less than 50% of patients who practice sports recreationally or at an amateur level and 83% of elite athletes reach their preinjury performance level (29,30). A quick and efficient regain of the lost quadriceps muscle mass and of quadriceps strength after ACL-R seems to be essential, especially for elite athletes. This regain must be achieved by optimized strength training because the antidoping regulations do not allow for the enhancement of muscle regeneration by pharmacological intervention as, for example, testosterone supplementation (15) or inhibition of myostatin (16,17).
In the present study, an average LSI for MCSA of 84% (CON/ECC) and 82% (CON/ECC+) was found before training. After training, the LSI for MCSA approached 100% with an enhanced improvement in the CON/ECC+ training group. However, a distinctly lower LSI was observed for leg extension strength with 57% and 52% (peak torque at angular velocity of 60°·s−1 in the CON/ECC the CON/ECC+ groups, respectively) and 67% and 62% (peak torque at an angular velocity of 180°·s−1). Also, after 12 wk of supervised strength training, the LSI values for the peak torque at both angular velocities (81% and 77% at angular velocity of 60°·s−1 and 84% at an angular velocity of 180°·s−1 in both groups) were still below the recommended value of 90% for return to sports (28). The negative effects on quadriceps strength were aggravated in the patients with QUAD autograft, suggesting that they might even need a longer rehabilitation for return to sports.
Despite the strong correlation between MCSA and quadriceps strength in both training groups and the significant correlation between the increases in MCSA and in quadriceps strength in the CON/ECC+ training group, there was quite a discrepancy between the LSI for quadriceps mass and quadriceps strength. This discrepancy might be due, on one hand, to a still impaired neuromuscular function, which is observed after ACL injury (5), and, on the other hand, to muscle quality or muscle phenotype, respectively. In the only study that, to our knowledge, so far investigated muscle regeneration after ACL-R, a significantly reduced FCSA of type IIA fibers compared with the uninjured leg at the time of surgery was reported as well as a lack of increase in the FCSA of all fiber types after rehabilitation. There was even a significant decrease in type IIA fibers after rehabilitation at the time of resumption of sport-specific drills (13). In our study, supervised leg-press training started after 12 wk of early rehabilitation when subjects were first cleared for coordination and strength training, probably round about the time point of the second biopsy in the study of Noehren et al. (13). In both our training groups, considerable muscle hypertrophy could be induced with a significant increase in the FCSA of all fiber types and in MCSA. However, eccentric-overload training, which led to a significantly greater increase in MCSA than conventional leg-press training, at the same time induced a less favorable slower muscle phenotype for strong and fast movements. This finding is in sharp contrast to the results of our previous studies with healthy untrained subjects (19) and athletes accustomed to regular strength training (20), where we found a shift toward a stronger and faster muscle phenotype after eccentric overload compared with conventional quadriceps strength training. The discrepancy might be explained by the distinctly lower eccentric overload with an average 1.2-fold higher eccentric than concentric load compared with a 1.9-fold higher eccentric load in healthy untrained subjects (19) and 2.3-fold higher load in strength training–experienced athletes (20). The lower eccentric overload in the present study is probably due to the still impaired neuromuscular function causing a reduced recruitment of fast myofibers.
The relatively low eccentric overload might also provide an explanation for the lack of increase in SC number, which we expected to occur especially in CON/ECC+ because of the increase in gene expression of myogenic regulatory factors in one of our previous studies (20). Increases in SC number have repeatedly been reported after 11–12 wk of quadriceps strength training (31–33), and there is no doubt that SC number plays an important role regarding skeletal muscle regeneration (34). SC activation occurs after training and postexercise microinjuries (24). It was reported that a single bout of exhaustive eccentric, but not a single bout of exhaustive concentric knee contractions, caused significant increases in total SC number (Pax7+ cells) and in MyoD+ nuclei in the vastus lateralis muscle of untrained young men (22). In another study, high-force eccentric exercises induced a significant increase in SC number and SC activation status related to type II, but not to type I myofibers in recreational active young men (23). On the contrary, 12 wk of concentric knee extensor strength training led to greater SC proliferation compared with eccentric knee extensor strength training (31). Taken these results together, it could be assumed that CON/ECC+ strength training might be a strong stimulus for SC activation and proliferation. However, during rehabilitation strength training after ACL-R, SC number did not change significantly. Besides the already mentioned neuromuscular impairment, degenerative changes in knee extensor muscles after ACL-R with collagen accumulation, increased fibroblast and fibrogenic/adipogenic progenitor cell content might explain this surprising result (14). Furthermore, increased myonuclear turnover as indicated by myonuclear apoptosis in 40% of the patients in the study of Fry et al. (14) and perhaps also an increased SC turnover as observed in a study with severely burned children (35) could be the reasons for the unchanged SC number after the 12-wk strength training period.
We indeed, however, observed moderate signs of myogenic cell activation and of remodeling. The activated SC numbers tended to be increased after CON/ECC+ training only. Furthermore, the significantly greater number in myofibers expressing MHCneo and MHC I in the CON/ECC+ group after the 12-wk training indicates the transfer of regenerative myofibers to type I fibers. In the CON/ECC group, the change in the proportion of activated SC number (ΔPax7+/MyoD+) was correlated with the change in SC number (ΔPax7+), showing that in individuals with an increasing SC number, the activated SC number also increased. However, due to a considerable interindividual variation, no further significant results concerning myogenic cell activation and remodeling emerged. Especially with regard to the interindividual variation, it is a drawback of our study that we did not obtain biopsies of the uninjured leg. A further limitation is the investigation of only steady-state levels of SC and not of acute responses to strength training. An enhanced acute increase of activated SC number to a single bout of resistance exercise was recently shown in untrained healthy individuals after they had performed 16 wk of supervised strength training (36). The acute response to resistance exercise probably gives further insight into the ability of the skeletal muscle to adequately response to a training stimulus.
In conclusion, even 24 wk after ACL-R, after completion of a 12-wk early rehabilitation and after 12 wk of supervised strength training, quadriceps strength was still impaired and according to recently published recommendations, athletes were not yet ready for return to sports. It seems that the previously described negative alterations within knee extensor muscles compromise muscle regeneration, apparently also influenced by the type auf autograft. CON/ECC+ quadriceps strength training induced an enhanced muscle hypertrophy compared with CON/ECC, however, with a concomitant increase in slow type I fibers and without the hypothesized enhancing effect on SC activation, proliferation, and differentiation. Therefore, some more work is needed to understand skeletal muscle maladaptation and to optimize quadriceps strength training in the rehabilitation after ACL-R.
We thank Dr. Jan-Paul Flacke and Steffi Kraushaar for their assistance in the study.
The study was funded by the Dietmar Hopp Foundation (Project 23011193).
The authors have no conflict of interests to report. The results of the present study do not constitute endorsement by the American College of Sports Medicine and are presented clearly, honestly, and without fabrication, falsification, or inappropriate data manipulation.
1. Lindstrom M, Strandberg S, Wredmark T, Fellander-Tsai L, Henriksson M. Functional and muscle morphometric effects of ACL reconstruction. A prospective CT study with 1 year follow-up. Scand J Med Sci Sports
2. Marcon M, Ciritsis B, Laux C, et al. Quantitative and qualitative MR-imaging assessment of vastus medialis muscle volume loss in asymptomatic patients after anterior cruciate ligament reconstruction. J Magn Reson Imaging
3. Grindem H, Snyder-Mackler L, Moksnes H, Engebretsen L, Risberg MA. Simple decision rules can reduce reinjury risk by 84% after ACL reconstruction: the Delaware-Oslo ACL cohort study. Br J Sports Med
4. Zwolski C, Schmitt LC, Quatman-Yates C, Thomas S, Hewett TE, Paterno MV. The influence of quadriceps strength asymmetry on patient-reported function at time of return to sport after anterior cruciate ligament reconstruction. Am J Sports Med
5. Williams GN, Buchanan TS, Barrance PJ, Axe MJ, Snyder-Mackler L. Quadriceps weakness, atrophy, and activation failure in predicted noncopers after anterior cruciate ligament injury. Am J Sports Med
6. Williams GN, Snyder-Mackler L, Barrance PJ, Buchanan TS. Quadriceps femoris muscle morphology and function after ACL injury: a differential response in copers versus non-copers. J Biomech
7. Mendias CL, Lynch EB, Davis ME, et al. Changes in circulating biomarkers of muscle atrophy, inflammation, and cartilage turnover in patients undergoing anterior cruciate ligament reconstruction and rehabilitation. Am J Sports Med
8. Brooks NE, Myburgh KH. Skeletal muscle wasting with disuse atrophy is multi-dimensional: the response and interaction of myonuclei, satellite cells
and signaling pathways. Front Physiol
9. Irie K, Tomatsu T. Atrophy of semitzendinosus and gracilis and flexor mechanism function after hamstring tendon harvest for anterior cruciate ligament reconstruction. Orthopedics
10. Han HS, Seong SC, Lee S, Lee MC. Anterior cruciate ligament reconstruction. Quadriceps versus patellar autograft. Clin Orthop Relat Res
11. Baugher WH, Warren RF, Marshall JL, Joseph A. Quadriceps atrophy in the anterior cruciate insufficient knee. Am J Sports Med
12. Nakamura T, Kurosawa H, Kawahara H, Watarai K, Miyashita H. Muscle fiber atrophy in the quadriceps in knee-joint disorders. Histochemical studies on 112 cases. Arch Orthop Trauma Surg
13. Noehren B, Andersen A, Hardy P, et al. Cellular and morphological alterations in the vastus lateralis muscle as the result of ACL injury and reconstruction. J Bone Joint Surg Am
14. Fry CS, Johnson DL, Ireland ML, Noehren B. ACL injury reduces satellite cell abundance and promotes fibrogenic cell expansion within skeletal muscle. J Orthop Res
15. Wu B, Lorezanza D, Badash I, et al. Perioperative testosterone supplementation increases lean mass in healthy men undergoing anterior cruciate ligament reconstruction: a randomized controlled trial. Orthop J Sports Med
16. Wurtzel CN, Gumucio JP, Grekin JA, et al. Pharmacological inhibition of myostatin protects against skeletal muscle atrophy and weakness after anterior cruciate ligament tear. J Orthop Res
17. World Anti-Doping Agency (WADA) Web site [Internet]. Montreal (Canada): WADA prohibited list 2017 [cited 2017 Jan 1]. Available from: https://www.wada-ama.org/en/prohibited list
18. English KL, Loehr JA, Lee SM, Smith SM. Early-phase musculoskeletal adaptations to different levels of eccentric resistance after 8 weeks of lower body training. Eur J Appl Physiol
19. Friedmann B, Kinscherf R, Vorwald S, et al. Muscular adaptations to computer-guided strength training with eccentric overload. Acta Physiol Scand
20. Friedmann-Bette B, Bauer T, Kinscherf R, et al. Effects of strength training with eccentric overload on muscle adaptation in male athletes. Eur J Appl Physiol
21. Norrbrand L, Fluckey JD, Pozzo M, Tesch PA. Resistance training using eccentric overload induces early adaptations in skeletal muscle size. Eur J Appl Physiol
22. Hyldahl RD, Olson T, Welling T, Groscost L, Parcell AC. Satellite cell activity is differentially affected by contraction mode in human muscle following a work-matched bout of exercise. Front Physiol
23. Cermak NM, Snijders T, McKay BR, et al. Eccentric exercise increases satellite cell content in type II muscle fibers. Med Sci Sports Exerc
24. Bazgir B, Fathi R, Rezazadeh VM, Mozdziak P, Asgari A. Satellite cells
contribution to exercise mediated muscle hypertrophy
and repair. Cell J
25. Crameri RM, Aagaard P, Qvortrup K, Langberg H, Olesen J, Kjaer M. Myofibre damage in human skeletal muscle: effects of electrical stimulation versus voluntary contraction. J Physiol
. 2007;583(Pt 1):365–80.
26. Fujimaki S, Machida M, Wakabayashi T, Asashima M, Takemasa T, Kuwabara T. Functional overload enhances satellite cell properties in skeletal muscle. Stem Cells Int
27. Bergstrom J. Percutaneous needle biopsy of skeletal muscle in physiological and clinical research. Scand J Clin Lab Invest
28. Dingenen B, Gokeler A. Optimization of the return-to-sport paradigm after anterior cruciate ligament reconstruction: a critical step back to move forward. Sports Med
29. Notarnicola A, Maccagnano G, Barletta F, et al. Returning to sport after anterior cruciate ligament reconstruction in amateur sports men: a retrospective study. Muscles Ligaments Tendons J
30. Lai CCH, Ardern CL, Feller JA, Webster KE. Eighty-three per cent of elite athletes return to preinjury sport after anterior cruciate ligament reconstruction: a systematic review with meta-analysis of return to sport rates, graft rupture rates and performance outcomes. Br J Sports Med
31. Farup J, Rahbek SK, Riis S, Vendelbo MH, Paoli Fd, Vissing K. Influence of exercise contraction mode and protein supplementation on human skeletal muscle satellite cell content and muscle fiber growth. J Appl Physiol (1985)
32. Hanssen KE, Kvamme NH, Nilsen TS, et al. The effect of strength training volume on satellite cells
, myogenic regulatory factors, and growth factors. Scand J Med Sci Sports
33. Mackey AL, Holm L, Reitelseder S, et al. Myogenic response of human skeletal muscle to 12 weeks of resistance training at light loading intensity. Scand J Med Sci Sports
34. Dumont NA, Bentzinger CF, Sincennes MC, Rudnicki MA. Satellite cells
and skeletal muscle regeneration. Compr Physiol
35. Fry CS, Porter C, Sidossis LS, et al. Satellite cell activation and apoptosis in skeletal muscle from severely burned children. J Physiol
36. Nederveen JP, Snijders T, Joanisse S, et al. Altered muscle satellite cell activation following 16 wk of resistance training in young men. Am J Physiol Regul Integr Comp Physiol