Development of a machine learning model to predict lateral hinge fractures by analyzing patient factors before open wedge high tibial osteotomy.

PURPOSE: Several methods have been developed to prevent lateral hinge fractures (LHFs), using only classic statistical models. Machine learning is under the spotlight because of its ability to analyze various weights and model nonlinear relationships. The purpose of this study was to create a machine learning model that predicts LHF with high predictive performance. METHODS: Data were collected from a total of 439 knees with medial osteoarthritis (OA) treated with Medial open wedge high tibial osteotomy (MOW-HTO) from March 2014 to February 2020. The patient data included age, sex, height, and weight. Preoperative, determined, and modifiable factors were categorized using X-ray and CT data to create ensemble models with better predictive performance. Among the 57 ensemble models, which is the total number of possible combinations with six models, the model with the highest area under curve (AUC) or F1-score was selected as the final ensemble model. Gain feature importance analysis and the Shapley additive explanations (SHAP) feature explanation were performed on the best models. RESULTS: The ensemble model with the highest AUC was a combination of a light gradient boosting machine (LGBM) and multilayer perceptron (MLP) (AUC = 0.992). The ensemble model with the highest F1-score was the model that combined logistic regression (LR) and MLP (F1-score = 0.765). Distance X was the most predictive feature in the results of both model interpretation analyses. CONCLUSION: Two types of ensemble models, LGBM with MLP and LR with MLP, were developed as machine learning models to predict LHF with high predictive performance. Using these models, surgeons can identify important features to prevent LHF and establish strategies by adjusting modifiable factors. STUDY DESIGN: Retrospective cohort study.

Is there any benefit in the combined ligament reconstruction with osteotomy compared to ligament reconstruction or osteotomy alone?: Comparative outcome analysis according to the degree of medial compartment osteoarthritis with anterior or posterior cruciate ligament insufficiency.

INTRODUCTION: The purpose of this study was to compare the outcomes of middle-aged patients with anterior cruciate ligament (ACL) or posterior cruciate ligament (PCL) insufficiency by assessing different groups: high tibial osteotomy (HTO), HTO with combined ligament reconstruction, and isolated ligament reconstruction according to the alignment change and medial compartment osteoarthritis (OA). MATERIALS AND METHODS: From 2014 to 2019, middle-aged (40-65 years) patients with knee instability were enrolled in this retrospective study. They were categorized into three groups: group I, HTO; group II, HTO with combined ACL or PCL reconstruction; and group III, isolated ligament reconstruction. Radiological outcomes, including Kellgren-Lawrence grade, mechanical femorotibial angle (mFTA), weight-bearing line (WBL) ratio, and posterior tibial slope were compared. Knee stability and clinical outcomes were also compared. RESULTS: Seventy-nine patients completed the final assessment. Group I was older than other two groups (p = 0.006). Groups I and II had a higher body mass index (p = 0.043) and more preoperative varus alignment than group III (p < 0.001). OA severity was ranked in the order of group I, II, and III (p < 0.001). Group I showed more valgus alignment than group II after HTO (p = 0.024 for mFTA and 0.044 for WBL ratio, respectively). Compared to their preoperative status, all three groups showed significant improvement in knee stability (p < 0.001); however, group I showed inferior knee stability regardless of ACL or PCL reconstruction (p < 0.001 and 0.043, respectively). All clinical scores significantly improved in the three groups (p < 0.001), and they showed comparable clinical outcomes in the final assessment. CONCLUSIONS: Our strategy in managing middle-aged patients with knee instability according to the varus alignment and medial degeneration showed favorable stability and clinical outcomes. Middle-aged patients with knee instability should be managed with different strategies depending on their status. LEVEL OF EVIDENCE: Case-control study; Level-III.

Excessively Increased Joint-Line Obliquity After Medial Opening-Wedge High Tibial Osteotomy Is Associated With Inferior Radiologic and Clinical Outcomes: What Is Permissible Joint-Line Obliquity.

PURPOSE: This study aimed to evaluate the permissible joint-line obliquity (JLO) based on radiologic and clinical outcomes with midterm follow-up after medial open-wedge high tibial osteotomy (MOWHTO). METHODS: Patients who had undergone MOWHTO from March 2014 to May 2016 were retrospectively evaluated. They were divided into 4 groups based on JLO as represented by postoperative medial proximal tibial angle (MPTA). Radiologic parameters including MPTA, joint-line orientation angle (JLOA), joint-line convergence angle (JLCA), posterior tibial slope, weightbearing line ratio (WBLR), and coronal translation were analyzed. Clinical outcomes were evaluated with American Knee Society Score (AKSS), Western Ontario and McMaster University Index, and short-form 36 health survey (SF-36). The changes between preoperation and final follow-up in JLOA and MPTA were defined as ΔJLOA and ΔMPTA. RESULTS: A total of 135 knees were finally included (MPTA ≤90.32° as group I; 90.33° to 92.62° as group II; 92.74° to 95.22° as group III; and ≥95.23° as group IV). The last follow-up MPTA, JLOA, and JLCA values were different between the groups (P < .001, P < .001, and P = .015, respectively). WBLR and JLOA positively correlated with MPTA; however, WBLR showed an abrupt increase at MPTA >96.5°, and the JLOA distribution tended to be greater than the regression line at MPTA >96°. Moreover, ΔJLOA was not as large as ΔMPTA. The percentage of patients attaining a minimal clinically important difference was significantly lower in the AKSS-functional score and SF-36 physical component summary in group IV (P = .008 and 0.021, respectively). CONCLUSION: The JLOA did not change as much as the MPTA, but an MPTA >95.2° abruptly increased the JLOA and valgus overcorrection after MOWHTO. Poor clinical outcomes were more evident in excessive MPTA (>95.2°) than in mildly undercorrected or properly corrected MPTA (<95.2°). LEVEL OF EVIDENCE: III, retrospective cohort study.

Additional tying on the adjustable-loop device improves the outcomes of anterior cruciate ligament reconstruction using hamstring autograft.

PURPOSE: The purpose of this study was to verify the efficacy of a novel technique for additional tying on the adjustable-loop device to prevent stress concentration on the graft loop end and gradual loop lengthening. METHODS: A total of 124 patients who underwent anterior cruciate ligament reconstruction using hamstring autografts from 2014 to 2017 were included in this retrospective study. After 1:1 propensity score matching, two groups were formed (group I: 50 patients without tying vs. group II: 50 patients with tying). Anterior laxity was evaluated using side-to-side differences. Tunnel length, loop length, and graft-tunnel gap were measured using follow-up magnetic resonance imaging. The signal-to-noise ratio was calculated at the loop end, loop inner side, tunnel entrance, and graft mid-substance. The clinical outcomes were assessed using the International Knee Documentation Committee score, Lysholm score, pivot shift test, and Lachman test. RESULTS: The average follow-up period was 63.2 ± 4.8 and 53.8 ± 11.9 months in groups I and II, respectively. Anterior laxity showed that side-to-side differences improved significantly 6 months postoperatively in both the groups. Although the anterior laxity improved in group II (2.9 ± 1.0 to 1.6 ± 0.8, p < 0.001), it deteriorated in group I (2.5 ± 1.5 to 3.3 ± 1.3 mm, p < 0.001) at the final follow-up. The graft-tunnel gap was significantly larger in group I (p < 0.001). The signal-to-noise ratios of the loop end and loop inner side were significantly higher in group I (p < 0.001 and p = 0.020, respectively). The clinical outcomes at the final follow-up were not significantly different between the groups. CONCLUSION: The additional tying on the adjustable-loop device was not superior to the control group in clinical stability examination or outcome. However, it was effective in anterior laxity measured by stress radiographs, preventing stress on the adjustable-loop device, and gradual graft loop lengthening. LEVEL OF EVIDENCE: Level III.

Retro-Tubercle Biplanar Opening Wedge High Tibial Osteotomy Is Favorable for the Patellofemoral Joint But Not for the Osteotomized Tubercle Itself Compared With Supra-Tubercle Osteotomy.

PURPOSE: To identify whether retro-tubercle opening-wedge high tibial osteotomy (RT-OWHTO) produces more favorable radiographic outcomes on patellofemoral joint alignment and clinical outcomes than supra-tubercle opening-wedge high tibial osteotomy (ST-OWHTO). METHODS: From January 2017 to July 2018, patients who underwent biplanar OWHTO were allocated to 1 of 2 groups (ST-OWHTO and RT-OWHTO). Plain radiographs and computed tomography were used to analyze patellofemoral alignment and other radiologic parameters representing osteotomy configurations. Clinical outcomes were assessed using American Knee Society Score and Western Ontario and McMaster Universities Osteoarthritis Index. RESULTS: In total, 50 knees that underwent ST-OWHTO and 44 knees that underwent RT-OWHTO were enrolled. Patellar height was significantly decreased only after ST-OWHTO (Caton-Deschamps ratio: P = .007; Blackburne-Peel ratio: P = .012). Patellar tilt angle was decreased in both groups (P = .009 and .004, respectively). Postoperative posterior tibial slope (PTS) (P = .013), PTS (Δ) (P < .001), retro-tuberosity gap distance (P = .001), and retro-tuberosity tip distance (P = .001) were significantly larger in RT-OWHTO. Retro-tuberosity tip distance was significantly correlated with retro-tuberosity gap distance (P = .002), thickness of second plane osteotomy fragment (P = .027), and anterior osteotomy ratio (P = .031) in ST-OWHTO. In RT-OWHTO, it was significantly correlated with PTS (△) (P < .001), retro-tuberosity gap distance (P < .001), and sagittal angle of bi-planar osteotomy (P = .005). There were 2 cases of tibial tuberosity fracture, 9 cases of delayed union on second plane osteotomy and 5 cases of tuberosity protrusion in RT-OWHTO. CONCLUSIONS: Although the RT-OWHTO technique maintains patellofemoral joint alignment, no difference in clinical outcome was detected. The RT-OWHTO has increased risk of tuberosity fracture, delayed union, and prominent tibial tuberosity. The surgeon should consider these negative aspects of the technique and consider adjusting additional stabilization. LEVEL OF EVIDENCE: Level III, retrospective cohort study.

Difference of preoperative varus-valgus stress radiograph is effective for the correction accuracy in the preoperative planning during open-wedge high tibial osteotomy.

PURPOSE: This study aimed to evaluate (1) the efficacy of varus-valgus stress radiographs to adjust the preoperative soft-tissue imbalance and (2) ascertain whether varus-valgus stress radiographs are effective for the correction accuracy in the preoperative planning of the opening wedge high tibial osteotomy (OWHTO). METHODS: From February 2017 to December 2018, a total of 121 consecutive knees that underwent bi-planar OWHTO were enrolled in this retrospective analysis. Preoperative planning was performed using a weight-bearing line (WBL). Target WBL was determined according to the status of the medial compartments such as cartilage, meniscus, and preoperative arthritic grade. Preoperative varus-valgus stress radiographs were used to assess the preoperative mediolateral ligament imbalance. The final target correction length of the opening gap was determined by subtracting the difference between the varus-valgus stress radiographs (VVD). All patients were divided into two groups according to the preoperatively planned correction degree: (group A), smaller than average; (group B), larger than average. Patients were also divided into two other groups (VVD adjusted and neglected groups). RESULTS: Groups A and B were 56 and 54 knees, respectively. The preoperatively planned correction lengths of the opening gap were 9.33 ± 1.5 and 14.16 ± 3.96 mm, respectively (p < 0.01). Mean values of the VVD were 0.85 ± 0.72, and 1.27 ± 1.78 mm, respectively (p < 0.01). Correction errors were 2.17 ± 2.06 and 3.52 ± 2.16%, respectively (p < 0.01). Planned and final correction degrees were also significantly larger (p < 0.01, and p < 0.01, respectively), because the preoperative WBL ratio was significantly smaller in the VVD adjusted group (p < 0.01). CONCLUSION: The VVD values could reproduce the preoperative soft-tissue imbalance and it was more prominent as the correction degree increased. The strategy of subtracting the VVD as assumed soft-tissue imbalance in the preoperative planning worked well for the correction accuracy during OWHTO. LEVEL OF EVIDENCE: Level IV, retrospective study.

Effect of Isolated Medial Patellofemoral Ligament Reconstruction in Patellofemoral Instability Regardless of Predisposing Factors.

The purpose of this study was to evaluate the outcomes of isolated medial patellofemoral ligament (MPFL) reconstruction, regardless of the presence of predisposing factors. A total of 21 knees that underwent isolated MPFL reconstruction from March 2014 to August 2017 were included in this retrospective series. Radiographs of the series of the knee at flexion angles of 20, 40, and 60 degrees were acquired. The patellar position was evaluated using the patellar tilt angle, sulcus angle, congruence angle (CA), and Caton-Deschamps and Blackburne-Peel ratios. To evaluate the clinical outcome, the preoperative and postoperative International Knee Documentation Committee (IKDC) and Lysholm knee scoring scales were analyzed. To evaluate the postoperative outcomes based on the predisposing factors, the results were separately analyzed for each group. Regarding radiologic outcomes, 20-degree CA was significantly reduced from 10.37 ± 5.96° preoperatively to -0.94 ± 4.11° postoperatively (p = 0.001). In addition, regardless of the predisposing factors, delta values of pre- and postoperation of 20-degree CA were not significantly different in both groups. The IKDC score improved from 53.71 (range: 18-74) preoperatively to 94.71 (range: 86-100) at the last follow-up (p = 0.004), and the Lysholm score improved from 54.28 (range: 10-81) preoperatively to 94.14 (range: 86-100) at the last follow-up (p = 0.010). Isolated MPFL reconstruction provides a safe and effective treatment for patellofemoral instability, even in the presence of mild predisposing factors, such as trochlear dysplasia, increased patella height, increased TT-TG distance, or valgus alignment. This is a Level 4, case series study.

Comparison of the Radiologic, Morphometric, and Clinical Outcomes between Kinematically and Mechanically Aligned Total Knee Arthroplasty: A Propensity Matching Study.

The purpose of this study was to compare radiologic, morphometric, and clinical outcomes between kinematically aligned (KA) and mechanically aligned (MA) total knee arthroplasty (TKA) in Korean patients. Overall, 168 patients who underwent primary TKA were retrospectively reviewed, and propensity matching (age, sex, and body mass index) was performed as 1:3 ration (KA TKAs [n = 42]: MA TKAs [n = 126]). Joint-line orientation angle (JLOA), coronal and axial alignments of implants, hip-knee-ankle (HKA) angle, and patellar tilt angle were assessed using full-length standing radiograph, axial computed tomography (CT) scan, and plain radiographs. Morphometric assessment was performed by analyzing the intraoperative measurement of the femoral cut surface and femoral components fitting in five zones. Clinical outcomes more than 2 years of follow-up were evaluated with the Knee Society (KS) knee and functional scores, Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC) scores, and the Short-Form Health Survey (SF-36). In radiologic results, JLOA was more parallel to the floor in KA TKAs (KA: medial tilt 0.9 ± 1.5 degrees; MA: lateral tilt 1.7 ± 1.5 degrees, p < 0.05), and patellar tilt angle was closer to preoperative status after KA TKA (KA: 2.0 ± 1.6 degrees; MA;0.3 ± 1.2 degrees, p < 0.05). HKA angle and rotational mismatch were similar between two groups. In morphometric analysis, entire overhang of anterior femoral cutting surface was reduced in KA TKA compared with MA TKA (KA: 11.7 ± 6.2 mm; MA: 14.4 ± 5.9 mm, p < 0.05). However, both of MA and KA TKAs showed underhang in mediolateral dimension without difference. There were no significant differences in clinical scores between two groups. KA TKA showed more parallel JLOA to floor, closer patellar tilt to preoperative status, and better anterior flange fitting that can reproduce more natural knee kinematics compared with MA TKA. Although clinical outcomes assessed by conventional evaluating tools were similar between two groups, further evaluation focusing on the patellofemoral symptoms or unawareness of TKA is necessary to clarify the clinical benefit of KA TKA.

Analysis of Affecting Factors of the Fate of Medial Meniscus Posterior Root Tear Based on Treatment Strategies.

Meniscal tear is a common reason for patients to undergo knee operation, and the medial meniscus posterior root tear (MMPRT) is one of the most frequent kinds of meniscal tears. The purpose of this study was to analyze participants' factors (anthropometric and medical) to the fate of the MMPRT based on the treatment strategy. The hypothesis of this study was that treatment modalities from conservative treatment to final arthroplasty would be affected by participants' affecting factors. From July 2003 to May 2018, 640 participants were included. Groups were categorized according to the treatment strategies such as conservative treatment, arthroscopic surgery, high tibial osteotomy (HTO) and arthroplasty surgery. Participants' affecting factors were analyzed by one-way analysis of variance according to the four different treatment strategies and a correlation between affecting factors was also analyzed. Participants with K-L (Kellgren-Lawrence) Grade 4 and high BMI > 28.17 were appropriate candidates for arthroplasty, with K-L Grade 4 being a greater determining factor than high BMI. Participants with alignment factors such as low initial weight bearing line (WBL) (26.5%) and high delta WBL ratio (5.9%) were appropriate candidates for HTO, with the delta WBL ratio being a greater determining factor than initial WBL. Longer MRI-event times (1.44 year) and a lesser extent of meniscal extrusion (2.98 mm) were significantly associated with conservative treatment. Understanding the correlation of each affecting factor to the treatment strategy will help clinicians decide on the appropriate treatment for patients with MMPRT.

The Fate of Bone Marrow Lesions After Open Wedge High Tibial Osteotomy: A Comparison Between Knees With Primary Osteoarthritis and Subchondral Insufficiency Fractures.

BACKGROUND: Subchondral insufficiency fracture of the knee (SIFK) is characterized by a subchondral lesion that may lead to end-stage osteoarthritis (OA). In patients who have SIFK in a precollapse state with varus malalignment, a joint-preserving technique such as open wedge high tibial osteotomy (OWHTO) should be considered. PURPOSE: To evaluate the efficacy of OWHTO in primary OA and SIFK-dominant OA by clinical and radiological evaluations including magnetic resonance imaging (MRI). STUDY DESIGN: Cohort study; Level of evidence 3. METHODS: A total of 33 SIFK-dominant OA knees and 66 with primary OA that underwent biplanar OWHTO between March 2014 and February 2016 were included after 1:2 propensity score matching. The MRI Osteoarthritis Knee Score was used to assess bone marrow lesions (BMLs) preoperatively and at follow-up. The weightbearing line ratio, the hip-knee-ankle angle, and the joint line convergence angle were measured. The clinical outcomes assessed were range of motion, the American Knee Society Score, and the Western Ontario and McMaster University (WOMAC) score. RESULTS: The mean follow-up period was 41.2 ± 12.6 months. The distribution of preoperative BML grade in the SIFK-dominant OA group was significantly higher in both the femur and tibia (P < .001 and <.001, respectively) than that in the primary OA group. However, the difference was not significant postoperatively (femur, P = .425; tibia, P = .462). In both groups, postoperative BMLs showed significant improvement compared with preoperative BMLs (primary OA [femur, P < .001; tibia, P = .001] and SIFK-dominant OA [femur, P < .001; tibia, P < .001]). The WOMAC pain score was higher in the SIFK-dominant OA group preoperatively (primary OA, 7.0 ± 3.73; SIFK-dominant OA, 9.17 ± 2.6; P = .032) even though it was not different at the final follow-up (primary OA, 2.11 ± 1.7; SIFK-dominant OA, 1.79 ± 1.32; P = .179). CONCLUSION: OWHTO is an effective procedure not only for primary OA but also for SIFK-dominant OA. OWHTO can improve BMLs, which represent the main pathological feature of SIFK. Therefore, in patients who have SIFK with varus malalignment, OWHTO can be an attractive treatment option for preserving the joint and enhancing subchondral bone healing.

Meniscal and Cartilage Changes on Serial MRI After Medial Opening-Wedge High Tibial Osteotomy.

BACKGROUND: Assessments of the effects of realignment using opening-wedge high tibial osteotomy (OWHTO) on the medial, lateral, and patellofemoral compartments have been limited to cartilage evaluations. PURPOSE/HYPOTHESIS: The purpose was to evaluate the effects of OWHTO on the meniscus and cartilage of each compartment as a cooperative unit (meniscochondral unit) using serial magnetic resonance imaging (MRI). It was hypothesized that (1) favorable changes in the meniscochondral unit would occur in the medial compartment and (2) that changes in the patellofemoral and lateral compartments would be negligible. STUDY DESIGN: Case series; Level of evidence, 4. METHODS: Included were 36 knees that underwent OWHTO from March 2014 to February 2016 and had postoperative serial MRI. The MRI was performed at 19.9 ± 7.4 and 52.3 ± 8.3 months postoperatively, and the cartilage and meniscal changes were evaluated by highlighting the regions of interest. We evaluated the T2 relaxation times of each cartilage and meniscal area, the cross-sectional area of the menisci, and the extrusion of the medial meniscus (MM). The meniscochondral unit was assessed using subgroup analyses according to the status of the MM. RESULTS: Significant decreases were seen in T2 relaxation times in the medial femoral condyle (MFC) (P < .001) and medial tibial plateau (MTP) (P = .050), and significant increases were seen in the lateral femoral condyle (LFC) (P = .036). The change was more prominent in the MFC compared with the MTP and LFC (P = .003). No significant changes were observed in the lateral tibial plateau, patella, or trochlear groove. The area of the lateral meniscus (body and posterior horn) was decreased compared with preoperative MRI (P < .001 for both). The extent of MM extrusion decreased between the preoperative, first follow-up, and second follow-up MRIs (P < .001). CONCLUSION: OWHTO affected the medial compartment positively, the lateral compartment negatively, and the patellofemoral compartment negligibly. The effects were more prominent and consistent in the medial than in the lateral compartment.

Prediction and Development of Preventive Strategies for Lateral Hinge Fracture During Opening Wedge High Tibial Osteotomy Based on Osteotomy Configurations.

BACKGROUND: Lateral hinge fracture (LHF) is a major complication of opening wedge high tibial osteotomy (OWHTO) and may result in poor outcomes. PURPOSE/HYPOTHESIS: The purpose of this study was to develop preventive strategies by identifying factors that affect LHFs. We hypothesized that (1) each LHF type would have different affecting factors and that (2) specific operative strategies that can contribute to the prevention of each LHF type can be developed. STUDY DESIGN: Case-control study; Level of evidence, 3. METHODS: We retrospectively analyzed 261 consecutive knees treated with biplanar OWHTO between March 2014 and December 2017. Perioperative radiological variables that can affect LHFs were measured and divided into 2 categories: unmodifiable and modifiable. A regression model was developed, and subgroup analyses involving comparisons between the non-LHF group and each LHF group were performed. The weightbearing line (WBL) ratio was measured at 2 weeks and 1 year after surgery to determine the serial changes in each LHF type. RESULTS: A total of 66 knees (25.3%) were diagnosed with LHFs. From these, 26 (39.4%), 13 (19.7%), 15 (22.7%), and 12 (18.2%) showed type I, II, III, and I-variant LHFs, respectively. In the subgroup analysis, a larger posterior gap and distance X and a smaller fibular height (FH) were significant unmodifiable factors, while the retrotubercular thickness was a significant modifiable factor, for type I LHF. For type II LHF, a smaller lateral condylar slope and a larger distance X were significant unmodifiable factors, while the lateral distal fragment thickness and the osteotomy-condylar angle were significant modifiable factors. For type III LHF, a larger lateral condylar width and distance X and a smaller FH were significant unmodifiable factors, while the lateral proximal fragment thickness and the retrotubercular angle (RA) were significant modifiable factors. A smaller FH and a larger distance X were significant unmodifiable factors for type I-variant LHFs, while the lateral thickness ratio and the RA were significant modifiable factors. Between postoperative week 2 and 1 year, the WBL ratio decreased in cases with type I LHFs (P < .001) and increased in those with type II (P = .001) and type I-variant (P = .006) LHFs. CONCLUSION: Unmodifiable and modifiable factors for the development of LHFs after OWHTO differ among LHF types. To prevent LHFs, the causes of each LHF must be identified, the patient's specific geometry be considered in the preoperative planning, and the surgical technique be modified according to the modifiable factors. In addition, during the rehabilitation period after OWHTO, specific caution and close observation are necessary for alignment changes related to each LHF type.

Comparison of the Predicting Performance for Fate of Medial Meniscus Posterior Root Tear Based on Treatment Strategies: A Comparison between Logistic Regression, Gradient Boosting, and CNN Algorithms.

This study aimed to validate the accuracy and prediction performance of machine learning (ML), deep learning (DL), and logistic regression methods in the treatment of medial meniscus posterior root tears (MMPRT). From July 2003 to May 2018, 640 patients diagnosed with MMPRT were included. First, the affecting factors for the surgery were evaluated using statistical analysis. Second, AI technology was introduced using X-ray and MRI. Finally, the accuracy and prediction performance were compared between ML&DL and logistic regression methods. Affecting factors of the logistic regression method corresponded well with the feature importance of the six top-ranked factors in the ML&DL method. There was no significant difference when comparing the accuracy, F1-score, and error rate between ML&DL and logistic regression methods (accuracy = 0.89 and 0.91, F1 score = 0.89 and 0.90, error rate = 0.11 and 0.09; p = 0.114, 0.422, and 0.119, respectively). The area under the curve (AUC) values showed excellent test quality for both ML&DL and logistic regression methods (AUC = 0.97 and 0.94, respectively) in the evaluation of prediction performance (p = 0.289). The affecting factors of the logistic regression method and the influence of the ML&DL method were not significantly different. The accuracy and performance of the ML&DL method in predicting the fate of MMPRT were comparable to those of the logistic regression method. Therefore, this ML&DL algorithm could potentially predict the outcome of the MMRPT in various fields and situations. Furthermore, our method could be efficiently implemented in current clinical practice.

Effects of Total Knee Arthroplasty on Coronal and Sagittal Whole-Body Alignments: Serial Assessments Using Whole-Body EOS.

BACKGROUND: The aims of this study were to evaluate the effects of correcting lower limb alignment by total knee arthroplasty (TKA) on the spinopelvic alignment and to identify patients with difference in the knee joint between clinically measured passive motion and the actual standing posture. METHODS: In this retrospective study, 101 patients who underwent TKA and whose serial whole-body EOS X-ray were available were included. The relationship of the knee and spinopelvic alignment was analyzed by evaluating the parameters of standing anterior-posterior and lateral whole-body EOS X-ray. The differences between postoperative passive motion and weight-bearing posture in the knee joint were assessed in both coronal and sagittal planes. Furthermore, the causes of such differences were analyzed. RESULTS: Significant correlations between Δpelvic obliquity and coronal ΔHip-Knee-Ankle (HKA)Rt-Lt angle between the preoperative and 3-month and 1-year postoperative data (p < 0.001 and p < 0.005, respectively) and improved with coronal lower limb alignment close to neutral resulted in decreased pelvic obliquity (p < 0.001, ß = 0.085 and p = 0.005, ß = 0.065, respectively) were observed. The correlations between Δpelvic tilt (PT) and Δsacral slope (SS) and sagittal ΔHKARt-Lt angle were statistically significant (PT: p < 0.001 and p < 0.045; SS: p = 0.002 and p < 0.001, respectively). The improved sagittal alignment close to neutral resulted in decreased PT and increased SS. The difference between postoperative passive motion and the weight-bearing posture of the knee joint was correlated with lumbar lordosis and sagittal C7 plumb line-sacrum distance (p = 0.042 and p < 0.001, respectively). CONCLUSIONS: The correction of lower limb alignment with TKA affected pelvic parameters dominantly; however, there was little effect on the spinal alignment. Additionally, patients with anterior stooping or lumbar flat back demonstrated difference in extension between passive knee motion and standing. Therefore, rather than only focusing on changes in the knee alignment correction, knee surgeons should also evaluate the spinopelvic alignment before surgery to consider the prognosis of the standing and predict the possible changes in the whole-body alignment. This preoperative assessment may improve the prognosis of TKA.

Achilles Tendon Allograft for Superior Capsule Reconstruction in Irreparable Massive Rotator Cuff Tears.

BACKGROUD: Treatment remains a challenge in massive and irreparable rotator cuff tears (RCTs), and superior capsular reconstruction (SCR) has become an increasingly popular choice. The objective of this study was to evaluate clinical and radiological outcomes after SCR using an Achilles tendon allograft in irreparable massive RCTs. METHODS: From December 2015 to March 2018, 11 patients (mean age, 66.3 ± 5.8 years) with irreparable massive RCTs who underwent SCR using an Achilles tendon allograft were enrolled in this study. The range of motion (ROM), visual analog scale (VAS), clinical scores, muscle strength, and acromiohumeral distance (AHD) were measured preoperatively and at 3, 6, and 12 months, and final follow-up postoperatively. Magnetic resonance imaging (MRI) was performed preoperatively and at 6 months postoperatively to assess the global fatty degeneration index and graft failure. Ultrasonography was also conducted preoperatively and at 3, 6, and 12 months, and final follow-up postoperatively to assess graft continuity. RESULTS: The mean follow-up period was 27.6 months (range, 24-32 months). The shoulder ROM at final follow-up increased significantly in forward flexion (p = 0.023), external rotation (p = 0.018), internal rotation (p = 0.016), and abduction (p = 0.011). All patients showed improvement in VAS score (p = 0.005) and clinical scores (p < 0.001) compared with the preoperative state. Pseudoparalysis improved in all patients. The AHD was 3.88 mm (± 1.21 mm) preoperatively, 7.75 mm (± 1.52 mm, p = 0.014) at 6 months postoperatively, and 6.37 mm (± 1.72 mm, p = 0.031) at final follow-up. Graft removal and synovectomy were performed in 1 patient who developed postoperative infections. Radiological failure on follow-up MRI occurred in 2 patients at 6 and 12 months postoperatively, respectively. CONCLUSIONS: SCR using an Achilles tendon allograft in irreparable massive RCTs achieved functional and clinical improvement. The use of Achilles tendon allograft also has the advantages of short operation time without donor site morbidity, sufficient thickness, and robustness; therefore, this allograft can be a useful graft for SCR.

Do optional implants improve the femoral fit during total knee arthroplasty in Asians? Comparison of the femoral fit between single- and dual-option implants.

BACKGROUND: The purpose of this study was to evaluate the femoral fit in Asian patients during total knee arthroplasty by examining whether (1) the advanced single-option implant improves the fit of the femoral component in both the anterior flange and distal cutting surface and (2) the dual-option implants show better fit than the advanced single-option implant. METHODS: A total of 950 knees that underwent primary total knee arthroplasty were included. Two types of single-option implant systems (Optetrak Logic and Truliant) and three types of dual-option implant systems (Anthem, Attune, and Persona) were used. The difference between the resected surface of the femur and femoral component dimensions was analyzed in predefined six zones. Appropriateness of fit (good-fit, over-hang, under-hang) was also evaluated. RESULTS: The advanced single-option implant showed higher rates of good-fit and lower rates of over-hang and under-hang in almost all distinct zones than the old version single-option implant. The advanced single-option implant demonstrated similar good-fit, higher over-hang and lower under-hang rates than the dual-option implants. All single-option implants showed significantly higher over-hang rate (P < 0.05) and lower under-hang rate (P < 0.05) than all dual-option implant systems in the anterior flange transverse area. The narrow option was only selected in 12-20% of all cases. CONCLUSIONS: The advanced single-option implant system improved the fit of the femoral component when compared with the old version and showed similar good-fit rates when compared with the dual-option implant systems. The narrow option of the dual-option implant systems was used less frequently than expected in Asian knees.

Can Deep Learning Using Weight Bearing Knee Anterio-Posterior Radiograph Alone Replace a Whole-Leg Radiograph in the Interpretation of Weight Bearing Line Ratio?

Weight bearing whole-leg radiograph (WLR) is essential to assess lower limb alignment such as weight bearing line (WBL) ratio. The purpose of this study was to develop a deep learning (DL) model that predicts the WBL ratio using knee standing AP alone. Total of 3997 knee AP & WLRs were used. WBL ratio was used for labeling and analysis of prediction accuracy. The WBL ratio was divided into seven categories (0, 0.1, 0.2, 0.3, 0.4, 0.5, and 0.6). After training, performance of the DL model was evaluated. Final performance was evaluated using 386 subjects as a test set. Cumulative score (CS) within error range 0.1 was set with showing maximum CS in the validation set (95% CI, 0.924-0.970). In the test set, mean absolute error was 0.054 (95% CI, 0.048-0.061) and CS was 0.951 (95% CI, 0.924-0.970). Developed DL algorithm could predict the WBL ratio on knee standing AP alone with comparable accuracy as the degree primary physician can assess the alignment. It can be the basis for developing an automated lower limb alignment assessment tool that can be used easily and cost-effectively in primary clinics.

Can Additional Patient Information Improve the Diagnostic Performance of Deep Learning for the Interpretation of Knee Osteoarthritis Severity.

The study compares the diagnostic performance of deep learning (DL) with that of the former radiologist reading of the Kellgren-Lawrence (KL) grade and evaluates whether additional patient data can improve the diagnostic performance of DL. From March 2003 to February 2017, 3000 patients with 4366 knee AP radiographs were randomly selected. DL was trained using knee images and clinical information in two stages. In the first stage, DL was trained only with images and then in the second stage, it was trained with image data and clinical information. In the test set of image data, the areas under the receiver operating characteristic curve (AUC)s of the DL algorithm in diagnosing KL 0 to KL 4 were 0.91 (95% confidence interval (CI), 0.88-0.95), 0.80 (95% CI, 0.76-0.84), 0.69 (95% CI, 0.64-0.73), 0.86 (95% CI, 0.83-0.89), and 0.96 (95% CI, 0.94-0.98), respectively. In the test set with image data and additional patient information, the AUCs of the DL algorithm in diagnosing KL 0 to KL 4 were 0.97 (95% confidence interval (CI), 0.71-0.74), 0.85 (95% CI, 0.80-0.86), 0.75 (95% CI, 0.66-0.73), 0.86 (95% CI, 0.79-0.85), and 0.95 (95% CI, 0.91-0.97), respectively. The diagnostic performance of image data with additional patient information showed a statistically significantly higher AUC than image data alone in diagnosing KL 0, 1, and 2 (p-values were 0.008, 0.020, and 0.027, respectively).The diagnostic performance of DL was comparable to that of the former radiologist reading of the knee osteoarthritis KL grade. Additional patient information improved DL diagnosis in interpreting early knee osteoarthritis.

Spinal Flexibility Is an Important Factor for Improvement in Spinal and Knee Alignment after Total Knee Arthroplasty: Evaluation Using a Whole Body EOS System.

The purposes of this study were (1) to evaluate the relationship between lumbosacral flexibility and the effects of total knee arthroplasty (TKA) on whole-body alignment; and (2) to determine the prerequisites of the adjacent joints for successful TKA. A total of 116 patients (156 cases) who had whole-body X-ray and flexion-extension lumbar radiograph available were enrolled. For the sagittal alignment evaluation, hip-knee-ankle (HKA) angle, pelvic tilt (PT), sacral slope (SS), lumbar lordosis (LL), thoracic kyphosis (TK), and C7 plumb line-sacrum distance (SVA) were evaluated on the whole-body radiographs. Lumbar flexibility (LF) was evaluated using the flexion-extension lumbar radiographs, and pelvic flexibility (PF) was evaluated using the pelvic incidence (PI). The disparities in the knee joint between postoperative passive motion and weight-bearing posture were assessed. LF was significantly correlated with ΔLL and ΔSVA (LL: p = 0.039, SVA: p = 0.040; Pearson correlation coefficient (PCC): -0.206 and 0.205, respectively). There were correlations between PF and ΔSS (p < 0.001, PCC: -0.362), and between the disparity and LF (p = 0.005, PCC = -0.275). Linear regression analysis demonstrated that LF was significantly associated with the presence of disparity (p = 0.005, ß = -0.205). LF is an important factor for improved spinal and lower limb alignment after TKA. Additionally, reduced LF may result in knee joint disparity between passive extension and standing extension status. Therefore, surgeons should consider spinopelvic alignment, including lower limb alignment preoperatively, to be able to predict possible changes in whole-body alignment following TKA.

Screw-Home Movement of the Tibiofemoral Joint during Normal Gait: Three-Dimensional Analysis.

BACKGROUND: The purpose of this study was to evaluate the screw-home movement at the tibiofemoral joint during normal gait by utilizing the 3-dimensional motion capture technique. METHODS: Fifteen young males and fifteen young females (total 60 knee joints) who had no history of musculoskeletal disease or a particular gait problem were included in this study. Two more markers were attached to the subject in addition to the Helen-Hayes marker set. Thus, two virtual planes, femoral coronal plane (P f ) and tibial coronal plane (P t ), were created by Skeletal Builder software. This study measured the 3-dimensional knee joint movement in the sagittal, coronal, and transverse planes of these two virtual planes (P f and P t ) during normal gait. RESULTS: With respect to kinematics and kinetics, both males and females showed normal adult gait patterns, and the mean difference in the temporal gait parameters was not statistically significant (p > 0.05). In the transverse plane, the screw-home movement occurred as expected during the pre-swing phase and the late-swing phase at an angle of about 17°. However, the tibia rotated externally with respect to the femur, rather than internally, while the knee joint started to flex during the loading response (paradoxical screw-home movement), and the angle was 6°. CONCLUSIONS: Paradoxical screw-home movement may be an important mechanism that provides stability to the knee joint during the remaining stance phase. Obtaining the kinematic values of the knee joint during gait can be useful in diagnosing and treating the pathological knee joints.