Search

Lumbar disc herniation (LDH) is a frequently encountered pathologic condition in orthopedic daily practice. Discectomy is considered when patients with LDH experience persistent limb or lumbar pain or neurologic deficits. Various minimally invasive techniques are available for discectomy. Among these techniques, full-endoscopic lumbar discectomy (FED) is one of the least invasive options. During FED, removal of LDH is accomplished using 2 major approaches: transforaminal (TF) or interlaminar (IL). The prototype FED was percutaneous nucelotomy. An endoscope was integrated to percutaneous nucelotomy and TF endoscopic lumbar discectomy (TELD) was first derived. IL endoscopic lumbar discectomy (IELD) was introduced years after TELD. TELD and IELD can compensate for the shortcomings of each other and endoscopic spine surgeons need to acquire both techniques to deal with all type of LDHs. Because of its long history, the TF approach seems to represent the major approach for FED, but the IL approach has numerous benefits in particular types of LDH. The present article focuses on IELD and reviews the history, surgical techniques, indications and contraindications, clinical outcomes, and complications. This review will contribute to improved understanding of IELD as an important technique in full-endoscopic spine surgery.

Background Femoral rotational alignment is considered an essential factor for total knee arthroplasty because malrotation of femoral components results in poor outcomes. To obtain proper alignment, we developed a superimposable computed tomography (CT) scan-based template to intraoperatively determine the accurate surgical epicondylar axis (SEA), and evaluated the effectiveness of this CT template. Methods In the experimental group (n = 55), three serial slices of CT images, including medial and lateral epicondyles, were merged into a single image, and SEA was overlaid. SEA was traced onto an image of an assumed distal femoral resection level; this combined image was then printed out onto a transparent film as a CT template. Following a distal femoral resection in TKA, SEA was duplicated onto the femoral surface. Thereafter, the posterior condyle was resected parallel to this SEA. In the control group (n = 53), posterior condyles were resected at three degrees of the external rotation from the posterior condylar line (PCL). A posterior condylar angle (PCA) between PCL of the femoral component and SEA was postoperatively evaluated. Positive values indicated external rotation of the femoral component from the SEA. Results In the experimental group, PCA was 0.01° ± 1.61°, and three cases were considered as outliers (greater than three degrees or less than −3 degrees). Conversely, in the control group, PCA was 0.10° ± 2.4°, and 12 cases were considered as outliers. Consequently, dispersion of PCA data was significantly smaller in the experimental group (P = 0.004). Conclusions The CT template accurately determined intraoperative SEA.

We describe a rare case of a fracture of the medial tubercle of the posterior process of the talus in a 16-year-old male athlete who fell during basketball practice. The patient presented to our orthopedic clinic when pain and swelling had persisted despite 2 weeks of anti-inflammatory medication and rest. Computed tomography and magnetic resonance imaging scans revealed a fracture of the posteromedial tubercle of the talus and a small amount of retained fluid in the joint. Immobilization with a below-the-knee cast and non-weightbearing for 4 weeks, with a gradual return to full activity, was successful. At the 1-year follow-up evaluation, the patient expressed no complaints. Fracture of the posteromedial tubercle of the talus will usually result in a misdiagnosis or delayed diagnosis owing to the insidious onset of symptoms. We believe the present fracture configuration resulted from the vertical compression force that occurred on landing by posterior medial ankle impingement in plantarflexion-supination, modifying the conventional concept of the posteromedial tubercle fracture. We also present a suggested classification with a flowchart diagram.

We present a case of reverse Z-effect phenomenon in a basicervical femoral fracture using a cephalomedullary nail together with two superior antirotation screws and evaluate the procedure. An 86-year-old woman fell in her home and could not stand due to right hip joint pain. X-ray and CT imaging showed a right basicervical femoral fracture (AO/OTA classification; 31B3), and open reduction and internal fixation (OR/IF) was performed with a cephalomedullary nail and two superior antirotation screws (TES Nail, HOMS, Tokyo, Japan). Two months later, X-ray showed penetration of the femoral head by the inferior lag screw with lateral migration of the two superior antirotation screws; the so-called “reverse Z-effect”; without any trauma. We performed the exchange of a shorter inferior lag screw for the longer one, and replaced the sliding type end cap with one of rocking type. The reverse Z-effect has been reported in cases with two lag screws, i.e. one inferior and one superior, in the past; however, to our knowledge, there has been no case reported in the literature using two superior antirotation screws together with one inferior lag screw.

Background Precise biomechanical knowledge of individual components of the MCL is critical for proper MCL release during TKA. This study was to define the influences of the deep MCL and the POL on valgus and rotatory stability in TKA using cadaveric knees. Methods This study used six fresh-frozen cadaveric knees. All TKA procedures were performed using a cruciate-retaining TKA with a CT-free navigation system. We did a sequential sectioning on each knee, S1; femoral arthroplasty only, S2; medial half tibial resection with spacer, S3; anterior cruciate ligament cut, S4; tibial arthroplasty, S5; release of the dMCL, S6; release of the POL. The navigation system monitored motion after application of 10 N-m valgus loads and 5 N-m internal and external rotation torques to the tibia at 0°, 20°, 30°, 60°, and 90° of knee flexion for each sequence. Results There were no significant differences in medial gaps. Internal rotation angles significantly increased after S2 at 0°, 20°, and 30°, and after S6 at 90°compared with those after S1. External rotation angles significantly increased after S3 at 0°, S4 at 60°, S5 at 0°, 30°and 90°, and after S6 at 30°, 60° compared with those after S1. Conclusion Significant increases of rotatory instability were seen on release of the dMCL, and then further increased after release of the POL. Surgical approach of retaining the dMCL and POL has a possibility to improve the outcome after primary TKA.

The aim of this study is to evaluate the relationship between patellofemoral contact stress and intraoperative knee kinematic patterns after mobile bearing total knee arthroplasty (TKA). Medial osteoarthritic knees of forty-six posterior-stabilized total knee prostheses were evaluated using a computed tomography-guided navigation system. Subjects were divided into two groups based on intraoperative knee kinematic patterns: the medial pivot group (n=19) and the non-medial pivot group (n=27). Mean intraoperative patello-femoral contact stress was significantly lower in the medial pivot group than in the non-medial pivot group (1.7MPa vs. 3.2MPa, P

The aim of this study was to evaluate the relationship between clinical results including patient-reported outcomes and intraoperative knee kinematic patterns after total knee arthroplasty (TKA). A cross-sectional survey of forty consecutive medial osteoarthritis patients who had a primary TKA using a CT-based navigation system was conducted. Subjects were divided into two groups based on intraoperative kinematic patterns: a medial pivot group (n = 20) and a non-medial pivot group (n = 20). Subjective outcomes with the new Knee Society Score and clinical outcomes were evaluated. The functional activities, patient satisfaction and the knee flexion angle of the medial pivot group were significantly better than those of the non-medial pivot group. An intraoperative medial pivot pattern positively influences deep knee flexion and patient-reported outcomes.

The aim of this study was to clarify the risk of knee flexion contracture associated with a posterior femoral condylar offset after total knee replacement (TKR). Radiographs from 100 healthy Japanese volunteers were included in the study. We evaluated femoral component posterior offset in various implants and compared them with the normal Japanese knee. Posterior offset of the femoral condyle is up to a maximum of 4.7 times greater than that of the healthy Japanese knee in all knee implants. Excess posterior offset of the femoral condyle in TKR prostheses may cause knee joint flexion contracture due to the relative shortening of the posterior soft tissue.

The present study assessed the midterm results of reconstruction for rheumatoid forefoot deformity with arthrodesis of the first metatarsophalangeal (MTP) joint, scarf osteotomy, resection arthroplasty of the metatarsal head of the lesser toes, and surgical repair of hammertoe deformity (arthrodesis of the proximal interphalangeal joint). Special focus was placed on the sagittal alignment of the first metatarsophalangeal joint after arthrodesis. We retrospectively evaluated the postoperative clinical outcomes and radiographic findings for 16 consecutive female patients (20 feet) with symptomatic rheumatoid forefoot deformities. The mean duration of follow-up was 7.9 (range 4 to 13) years. All first MTP joints and first metatarsal bones were fused successfully. The mean value of the American Orthopaedic Foot and Ankle Society and Japanese Society for Foot Surgery clinical scores significantly improved overall, except for 2 patients (10%), who complained of first toe pain at the final follow-up visit owing to sagittal misalignment of the fused first MTP joint. Sagittal alignment of the first metatarsal varies greatly because of the rheumatoid midfoot and hindfoot deformities. Therefore, inclination of the first metatarsal should be considered when determining the first MTP joint sagittal fusion angle.

The aim of this randomized prospective study was to clarify risks associated with a drain-clamping method using tranexamic acid and carbazochrome sodium sulfonate hydrate after total knee arthroplasty (TKA). Subjects comprised 100 patients scheduled to undergo TKA, randomized into 2 groups: 50 patients received the drain-clamping method using tranexamic acid and carbazochrome sodium sulfonate hydrate and 50 patients received drain-clamping with saline. Although bleeding volume was significantly lower in the group with tranexamic acid and carbazochrome sodium sulfonate hydrate, risk of asymptomatic deep venous thrombosis as detected by ultrasonography was comparable between groups. Tranexamic acid and carbazochrome sodium sulfonate hydrate in the drain-clamping method help reduce bleeding after TKA without increasing the risk of deep venous thrombosis.

Controversies exist in clinical study concerning the effect of rotating platform on patellar tracking. The aim of this in vivo study was to compare tibial rotation, patellar tracking, and patellofemoral contact stress in mobile and fixed-bearing platform intraoperatively in the same knee. Sixty-six knees of posterior-stabilized total knee prostheses were evaluated using a computed tomography-guided navigation system. Medial shift and lateral tilt of patella were significantly smaller in mobile knee. Averaged maximum contact stress was significantly smaller in mobile knee than fixed knee. However, tibial rotation during flexion has no significant difference. This study showed that mobile platform total knee arthroplasty significantly improved patellar tracking and decreased patellofemoral contact stress.

New tissue-engineering technology was developed to create a cartilage-like tissue in a three-dimensional culture using atelocollagen gel. The minimum 2-year followup outcome of transplanting autologous chondrocytes cultured in atelocollagen gel for the treatment of full-thickness defects of cartilage in knees was reported from the single institution. The present multicenter study was conducted to determine clinical and arthroscopic outcomes in patients who underwent atelocollagen-associated autologous chondrocyte implantation for the repair of chondral defects of the knees.At six medical institutes in Japan, we prospectively evaluated the clinical and arthroscopic outcomes of transplanting autologous chondrocytes cultured in atelocollagen gel for the treatment of full-thickness defects of cartilage in 27 patients (27 knees) with cartilage lesions on a femoral condyle or on a patellar facet over 24 months.The Lysholm score significantly increased from 60.0 +/- 13.7 points to 89.8 +/- 9.5 points (P = 0.001). Concerning the ICRS grade for arthroscopic appearance, 6 knees (24%) were assessed as grade I (normal) and 17 knees (68%) as grade II (nearly normal). There were few adverse features, except for detachment of the graft in two cases.We concluded that transplanting chondrocytes in a newly formed matrix of atelocollagen gel can promote restoration of the articular cartilage of the knee.

Sialic acid, which is located at the end of the carbohydrate moiety of cell surface glycoconjugates, is involved in many biologic responses, such as intercellular reactions and virus-cell fusion, especially in hematopoietic cells. Here we provide experimental evidence that the sialic acid of cell surface glycoconjugates has a role in osteoclast differentiation. Lectin histochemical study demonstrated the existence of both alpha (2,3)-linked-sialic acid and alpha (2,6)-linked-sialic acid in mouse bone marrow-derived macrophages and in the RAW264.7 macrophage cell line, which are osteoclast precursors. Flow cytometric analysis of surface lectin staining revealed the kinetics of these sialic acids during osteoclastogenesis: alpha (2,3)-linked-sialic acid was abundantly expressed throughout osteoclastogenesis, whereas alpha (2,6)-linked-sialic acid levels declined at the terminal stage of osteoclast differentiation. To investigate the role of sialic acid in osteoclast differentiation, we performed an osteoclastogenesis assay with or without exogenous sialidase treatment. Desialylated cells formed TRAP-positive mononuclear cells, but did not become multinuclear cells despite the normal expression of osteoclast markers such as cathepsin K, integrin beta3, and nuclear factor-ATc1. Flow cytometric analysis also demonstrated that exogenous sialidase effectively removed alpha (2,6)-linked-sialic acid, but only slightly changed the alpha (2,3)-linked-sialic acid content, suggesting that alpha (2,6)-linked-sialic acid might be involved in osteoclast differentiation. Findings from knockdown analysis using small interfering RNA oligonucleotides against alpha 2,6-sialyltransferase support this idea: alpha (2,6)-linked-sialic acid-deficient cells markedly inhibit the formation of multinuclear osteoclasts. Our findings suggest that alpha (2,6)-linked-sialic acid of cell surface glycoconjugates has a role in osteoclast differentiation, possibly via its role in the cell-cell fusion process.

It has been shown that musculoskeletal tissues undergo dynamic tissue remodeling by a process that is quite sensitive to the mechanical environment. However, the detailed molecular mechanism underlying this process remains unclear. We demonstrate here that after denervation-induced mechanical stress deprivation, tendons undergo dynamic tissue remodeling as evidenced by a significant reduction of the collagen fibril diameter. Importantly, the transient up-regulation of osteopontin (OPN) expression was characteristic during the early phase of tendon tissue remodeling. Following this dynamic change of OPN expression, matrix metalloproteinase (MMP)-13 expression was induced, which presumably accounts for the morphological changes of tendon by degrading tendon collagen fibrils. The modulation of MMP-13 expression by OPN was specific, since the expression of MMP-2, which is also known to be involved in tissue remodeling, did not alter in the tendons under the absence or presence of OPN. We also demonstrate that the modulation of MMP-13 expression by OPN is due to the signaling through cell surface receptors for OPN. Thus, we conclude that OPN plays a crucial role in conveying the effect of denervation-induced mechanical stress deprivation to the tendon fibroblasts to degrade the extracellular matrices by regulating MMP-13 expression in tendon fibroblasts.

In the present study, we have developed a novel and versatile method for the preparation of chitosan-peptide complex based on the selective reaction of chitosan with 2-iminothiolane. The new type of SH-chitosan derivative showed an excellent solubility to aqueous solution even in the alkaline conditions. This characteristic greatly facilitated further modification study of chitosan with a variety of bioactive substances. A synthetic peptide, RGDSGGC containing RGDS moiety that is known as one of the most important cell adhesive peptides, was readily coupled by disulfide bonds formation with sulfhydryl groups of SH-chitosan in the presence of dimethyl sulfoxide. Next, the effect of the introduction of RGDSGGC moiety to chitosan on cell adhesion and proliferation activity of chondrocytes and fibroblasts were evaluated. As a result, it was suggested that this polysaccharide-peptide conjugate exhibited excellent capacities for both cell adhesion and cell proliferation of chondrocytes and fibroblasts. Considering the growing importance of the biocompatible scaffolds in the recent tailored tissue engineering technique, these results indicate that the present strategy of 2-iminothiolane-based conjugation of polysaccharides with biologically active peptides will become a key and potential technology to develop desirable scaffold materials for the tissue regenerations.

Secondary collapse of the femoral head after transtrochanteric rotational osteotomy (TRO) for osteonecrosis of the femoral head is a major complication leading to failure. We investigated the relation between the extent and shape of a newly established healthy weight-bearing area and the incidence of secondary collapse.Altogether, 38 hips in 32 patients with a mean age of 33 years at the time of surgery were followed for a mean of 4 years.The outcome was satisfactory in 22 hips (58%) and unsatisfactory in 16 hips (42%), with 9 requiring total hip arthroplasty. The extent of the newly established healthy weight-bearing area as depicted on the immediate postoperative radiograph was significantly greater in the satisfactory group than in the unsatisfactory group. The rate of secondary collapse was significantly higher when the newly established weight-bearing area was thin, like a beak ("beak-shaped" healthy area). The incidence of a beak-shaped healthy area was significantly higher in the steroid group than in nonsteroid group.We concluded that we should be cautious when the newly established healthy weight-bearing area exhibits this beak shape, even if its extent in the weight-bearing area is large, because of the higher risk of postoperative secondary collapse. In such cases, alteration of the after-treatment program (e.g., prolonging the non-weight-bearing period) should be considered.

Among many colorimetric methods for carbohydrate analysis, the phenol-sulfuric acid method is the easiest and most reliable method. It has been used for measuring neutral sugars in oligosaccharides, proteoglycans, glycoproteins, and glycolipids. This method is used widely because of its sensitivity and simplicity. In its original form, it required 50-450 nmol of monosaccharides or equivalent for analysis and thus is inadequate for precious samples. A scaled-down version requiring only 10-80 nmol of sugars was reported previously. We have now modified and optimized this method to use 96-well microplates for high throughput, to gain greater sensitivity, and to economize the reagents. This modified and optimized method allows longer linear range (1-150 nmol for Man) and excellent sensitivity. Moreover, our method is more convenient, requiring neither shaking nor covering, and takes less than 15 min to complete. The speed and simplicity of this method would make it most suitable for analyses of large numbers of samples such as chromatographic fractions.

Following knee ligament injury, clinical and experimental investigations usually focus on the injured ligament, and uninjured ligaments of the same joint are largely ignored and presumed to remain unchanged. The purpose of this study was to characterize changes in mRNA levels for a relevant subset of molecules in the uninjured knee ligaments following combined unilateral medial collateral ligament (MCL) and partial anterior cruciate ligament (ACL) transection in sheep. Semiquantitative reverse transcription-polymerase chain reaction was performed for collagen types I, III, and V; matrix metalloproteinase-13 (MMP-13); and tissue inhibitor of metalloproteinase-1 for both injured and uninjured knee ligaments at 6 and 12 weeks after injury. Collagen type I, III, and V mRNA levels were significantly increased in MCL scars at 6 weeks as well as in the uninjured lateral collateral ligament and the anteromedial band of the ACL (AM-ACL). MMP-13 mRNA levels were also elevated in the MCL at 6 and 12 weeks and in the AM-ACL 6 weeks after injury. In contrast, significant changes in the posterior cruciate ligament were not detected at either time point, indicating specificity in the transient alterations. These results suggest that following injury, responses occur in uninjured ligaments that are specific although transient in nature. These responses may be an adaptive attempt to preserve function until the scar tissue can stabilize the mechanical environment following injury.

s / Osteoarthritis and Cartilage 21 (2013) S63–S312 S104 178 THE POSITION OF THE CENTER OF THE MASS FROM THE KNEE CENTER CORRELATES CLOSELY WITH THE KNEE ADDUCTION MOMENT IN PATIENTS WITH KNEE OSTEOARTHRITIS N. Takeda y,z, R. Niki x, T. Kobayashi y, T. Ino k, M. Yamanaka y, T. Majima y. yHokkaido Univ., Sapporo, Japan; zHokuto Hosp., Obihiro, Japan; xObihiro Kousei Hosp., Obihiro, Japan; kHakodate Orthopedic Clinic, Hakodate, Japan Purpose: The external knee adduction moment (KAM) has been demonstrated to be a risk factor for the progression of knee osteoarthritis. The KAM can be changed by the relative position of the knee center, the center of pressure (COP), and the center of the mass (COM) during walking. Few previous gait analysis studies have examined whether the distance from the COM to the knee center are variables that influence the KAM or its relationship with the kinematics of the trunk and lower limbs. Therefore, the objective of this study was to clarify the relationships among the KAM, COM and other kinematic variables. Methods: Nineteen subjects with knee osteoarthritis, eighteen females and one male, with an average age of 70.9 years (range: 5079 years) were recruited from the university hospital. The study was approved by the institutional board, and all subjects gave their written informed consent for their participation. All subjects had a radiograph taken while standing on one foot. The femorotibial joint in all radiographs was evaluated using the Kellgren and Lawrence scale (K/L). Two subjects had a K/L score of 2, nine had a K/L of 3, five had a K/L of 4. The knees with more symptoms were investigated. The average femorotibial angle (FTA) was 178.9 (range: 174 -187 ). Gait data were collected with a six camera motion analysis system and two force plates that were time synchronized and sampled at 120 Hz and 1200 Hz, respectively. Modified Helen Hays marker sets with 31 retroreflective markers were attached. The subjects were instructed to walk on a 10 m walkway at a selfselected speed. A total of three trials with clean, single force plate strikes from the study limb were collected. The external first peak knee adduction moment in the stance phase (1st KAM), COM, the distance from the COM to the center of the knee in the frontal plane (COM-K), lateral trunk tilt, lateral pelvic tilt, adduction angle at the hip joints, toeout angle of the foot, gait width and the gait speed were analyzed. The external moment was calculated with OrthoTrak. The kinetic data were calculated using the Matlab R2009b software program. All angles were reported during the stance phase of the gait for the affected limb at the first peak knee adduction moment. Values for each gait variable were obtained by averaging them across the trials. We used Pearson correlation coefficients to examine the relationships among the 1st KAM, FTA, COM-K, trunk tilt angle, pelvic tilt angle, hip adduction angle, leg heel angle, heel floor angle and toe-out angle. We used a multiple regression analysis to evaluate the amount of variance in the 1st KAM explained by FTA, COM-K, pelvic tilt angle and toe-out angle. Results: The 1st KAM was significantly correlated with the COM-K (r1⁄40.64, p1⁄40.014) and the pelvic tilt angle (r1⁄4-0.458, p1⁄40.043). The COM-K was also significantly correlated with the trunk tilt angle (r1⁄4-0.46, p1⁄40.040) and hip adductor angle (r1⁄4-0.60, p1⁄40.005). In a multiple regression analysis, the COM-K and pelvic tilt angle were independently correlated with the 1st KAM (R2 1⁄40.445). Conclusions: The 1st KAMwas more closely correlated with the COM-K thanwith the foot-related variables. Modifications to the position of the COM, including the pelvic tilt, trunk tilt, and hip adduction are important to reduce the 1st KAM for patients with knee osteoarthritis. The factors associated with the 1st knee adduction moment as determined by a multivariate analysis. Predictor variables b-coefficient Standard error p value COM-K 0.0050 0.002 0.006 Pelvic tilt 0.042 0.015 0.014 Femorotibial angle -0.011 0.009 0.232 Toe-out angle 0.009 0.006 0.133 179 CHANGES TO ROTATIONAL LOADING AFTER ACL INJURY AND RECONSTRUCTION DURING STANDING TARGET MATCHING A.S. Lanier y,z, K. Manal y,z, T.S. Buchanan y,z. yUniv. of Delaware, Newark, DE, USA; zDelaware Rehabilitation Inst., Newark, DE, USA Purpose: Nearly half of patients with anterior cruciate ligament injury will eventually get osteoarthritis (OA). Frontal plane knee measures are commonly used to explain the high incidence of OA but rotational loading may also be a contributing factor. Transverse knee moment (TKM) can help understand rotational instability common after ACL injury and correlates to cartilage loss. Measuring transverse knee moment during tasks that require dynamic rotational loading, such as a run cut maneuver, post ACL injury and reconstruction risks re-injury. Our lab uses a standing target matching protocol that safely evaluates rotational loading as it requires sub-maximal forces. In standing target matching subjects generate controlled shear forces to control a cursor. Our goal was to use standing target matching to understand rotational loading in ACL deficient (ACL-d) and reconstructed (ACL-r) subjects. We hypothesized that TKM measured during standing target matching would be greater in ACL-d subjects than healthy subjects and similar between healthy and ACL-r subjects. Methods: This study included 8 healthy, 8 ACL-d and 8 ACL-r subjects, all of similar age, mass and BMI. Reconstruction types included cadaveric allograft or hamstrings autograft. ACL-d subjects were testedwithin 6 months of injury. ACL-r subjects were tested 6 months to 1 year of surgery. For the standing target matching subjects stood barefoot on 2 force plates (OR-6 AMTI, Watertown, MA, USA), each limb on a force plate. Subjects controlled a cursor presented on a screen in front of them. The cursor was controlled by shear forces generated by a single limb. The goal of standing target matching was to create the forces to place the cursor on a location on the screen designated with a target. Shear direction controlled the cursor's movement in that direction if subjects created shear in the anterior/posterior direction the cursor moved towards the top/bottom of the screen, respectively, and AP shear forces moved the cursor left and right. 18 targets were presented at 20 increments around a circle; target position order was randomized. Target position from the center of the screen was 50% of the minimum shear MVC of the anterior, posterior, medial, and lateral directions. Cursor size was based on weight distribution; requiring even distribution of body weight. Standard motion capture techniques were used to collect kinematic and GRF data during target matching. Results: ACL-d and ACL-r limbs have significantly larger TKM when compared to limbs of healthy subjects when matching targets (Figure 1). The ACL-d limbs have greater internal rotation moment (designated as a positive TKM) at 5 target positions which require medial shear forces (140 -220 ). Figure 1. Avg. TKM SE normalized to mass*height: ACL-d (left), ACL-r (right) & healthy subjects. * p