Deep learning-based landmark recognition and angle measurement of full-leg plain radiographs can be adopted to assess lower extremity alignment.

PURPOSE: Evaluating lower extremity alignment using full-leg plain radiographs is an essential step in diagnosis and treatment of patients with knee osteoarthritis. The study objective was to present a deep learning-based anatomical landmark recognition and angle measurement model, using full-leg radiographs, and validate its performance. METHODS: A total of 11,212 full-leg plain radiographs were used to create the model. To train the data, 15 anatomical landmarks were marked by two orthopaedic surgeons. Mechanical lateral distal femoral angle (mLDFA), medial proximal tibial angle (MPTA), joint line convergence angle (JLCA), and hip-knee-ankle angle (HKAA) were then measured. For inter-observer reliability, the inter-observer intraclass correlation coefficient (ICC) was evaluated by comparing measurements from the model, surgeons, and students, to ground truth measurements annotated by an orthopaedic specialist with 14 years of experience. To evaluate test-retest reliability, all measurements were made twice by each measurer. Intra-observer ICCs were then derived. Performance evaluation metrics used in previous studies were also derived for direct comparison of the model's performance. RESULTS: Inter-observer ICCs for all angles of the model were 0.98 or higher (p < 0.001). Intra-observer ICCs for all angles were 1.00, which was higher than that of the orthopaedic specialist (0.97-1.00). Measurements made by the model showed no significant systemic variation. Except for JLCA, angles were precisely measured with absolute error averages under 0.52 degrees and proportion of outliers under 4.26%. CONCLUSIONS: The deep learning model is capable of evaluating lower extremity alignment with performance as accurate as an orthopaedic specialist with 14 years of experience. LEVEL OF EVIDENCE: III, retrospective cohort study.

Preoperative latent medial laxity and correction angle are crucial factors for overcorrection in medial open-wedge high tibial osteotomy.

PURPOSE: This study aimed to determine which preoperative factors affect the postoperative change in the joint line convergence angle (JLCA) by preoperatively quantifying soft tissue laxity. METHODS: Thirty-four patients who underwent medial open-wedge high tibial osteotomy (HTO) with a navigation were analysed. The JLCA change after HTO was calculated using standing long-bone anteroposterior radiographs taken preoperatively and 6 months postoperatively. Latent soft tissue laxity was defined as the amount of soft tissue that can be extended to valgus or varus from the weight-bearing position, and calculated by subtracting the JLCA on weight-bearing standing radiographs from that on stress radiographs. Multiple linear regression was performed to determine the preoperative factors that statistically correlated with the postoperative JLCA change. RESULTS: In multiple linear regression, JLCA change had a statistically significant correlation with latent medial laxity (R = 0.6) and a statistically borderline significant correlation with correction angle (R = 0.2). These imply that the postoperative JLCA change increased by 0.6° per 1° increase in latent medial laxity, and increased by 0.2° per 1° increase in correction angle. Latent medial laxity was the most crucial factor associated with postoperative JLCA changes. CONCLUSION: The JLCA change could be larger in patients with large latent medial laxity or severe varus deformity requiring a large correction, which could lead to unexpected overcorrection in HTO. Postoperative JLCA change should be considered in preoperative surgical planning. Target point shifting within the hypomochlion point could be a strategy to prevent overcorrection, especially in patients with large latent medial laxity. LEVEL OF EVIDENCE: Level IV.

The medial proximal tibial angle accurately corrects the limb alignment in open-wedge high tibial osteotomy.

PURPOSE: The purpose of this study was to detect the pre- and intra-operative influential factors for lower limb alignment correction error in open-wedge high tibial osteotomy (OWHTO). METHODS: This study involved 69 patients (71 knees) undergoing OWHTO for primary medial osteoarthritis. The weight-bearing line (WBL) ratio, medial proximal tibial angle (MPTA), and joint line convergence angle (JLCA) were measured on radiographs preoperatively and at 1 month after surgery, and the differences between the pre- and postoperative values were calculated. The correction angle during surgery was also investigated. The radiological correction angle was defined as the difference between the pre- and postoperative MPTA. The correction error was defined as the difference between the correction angle during surgery and the radiological correction angle. The ideal correction angle was defined as when the postoperative WBL passed through Fujisawa's point (WBL = 62.5%), and the alignment error was defined as the difference between the postoperative WBL ratio and 62.5. The correlations among the alignment error, the correction error, correction angle during surgery, pre- and postoperative WBL ratio, MPTA, and JLCA and the differences between the pre and postoperative WBL ratio, MPTA, and JLCA were investigated. In addition, the factor most influential on the alignment error was determined. RESULTS: The preoperative MPTA was the only predictor of the alignment error after OWHTO. The alignment error was positively correlated with the correction error and correction angle during surgery, and negatively correlated with pre- and postoperative WBL ratio, MPTA, and differences between the pre- and postoperative WBL ratio and JLCA. CONCLUSION: The preoperative MPTA was the only pre- and intra-operative predictor of the alignment error after OWHTO. The larger the correction angle, the greater the alignment error. The MPTA was recommended as an indicator for improving the correction accuracy. Accurate correction based on the MPTA provides good lower limb alignment and better clinical results. LEVEL OF EVIDENCE: III Case-control study/Retrospective comparative study.

Change of joint line convergence angle and other coronal alignments after total hip arthroplasty.

BACKGROUND: The purpose of this study was to determine the unloading effect of total hip arthroplasty (THA) on the ipsilateral knee joint through the joint line convergence angle (JLCA) change and determine the changes in other coronal radiographic parameters after THA, and evaluate the sensitivity of JLCA. PATIENTS AND METHODS: We retrospectively assessed 70 patients who underwent unilateral THA. Hip parameters such as neck shaft angle (NSA), neck length, and femoral offset and coronal alignment parameters of the lower extremity such as hip-knee-ankle angle (HKA), femur length, mechanical lateral distal femoral angle (mLDFA), JLCA, medial proximal tibial angle (MPTA), lateral distal tibial angle (LDTA), and plafond talus angle (PTA) were measured in the operative and non-operative sides. We compared all hip and coronal alignment parameters between before and 1 year after THA, and the amount of standardized changes (Δ) between all hip and coronal alignment parameters on the operative side, respectively. RESULTS: All mean hip and coronal alignment parameters on the operative side changed significantly 1-year after THA; however, those on the non-operative side did not. On the operative side, mean JLCA and PTA changed in the direction of closing the joint lateral space, from 1.81° and 0.47° to 1.22° and 0.11°, respectively (p<0.001 and 0.046, respectively). Mean NSA, HKA, and mLDFA changed in the valgus direction, from 129.39°, 2.62°, and 86.69° to 133.54°, 1.53°, and 85.91°, respectively (p<0.001). Mean femoral offset, neck length, and femur length increased from 36.45mm, 47.83mm, and 429.20mm to 39.85mm, 55.06mm, and 436.33mm, respectively (p<0.001). Mean MPTA and LDTA increased from 85.43° and 87.50° to 86.73° and 90.38, respectively (p<0.001). JLCA was more vulnerable to change than HKA, femur length, mLDFA, MPTA, and PTA (p<0.05). DISCUSSION: JLCA change on the operative side was more sensitive than changes of other coronal alignments after THA. According to the cohort, THA might have an unloading effect on the medial compartment of the knee joint. LEVEL OF EVIDENCE: IV; retrospective case-control and cohort studies.

Factors Associated with Postoperative Knee Joint Line Obliquity After Medial Open Wedge High Tibial Osteotomy.

BACKGROUND: There are still few reports on factors associated with postoperative knee joint line obliquity (KJLO). PURPOSE: The purpose was to determine preoperative radiographic factors that are associated with KJLO postoperatively after open wedge high tibial osteotomy (OWHTO) using multivariable linear regression analysis and multivariable logistic regression analysis. STUDY DESIGN: Case-control study; Level of evidence, 3. METHODS: A total of 60 patients with 65 varus knees who underwent OWHTO between December 2012 and June 2018 at a single institution were retrospectively enrolled in this study. The authors evaluated radiologic parameters including the weightbearing line ratio, femorotibial angle, medial proximal tibial angle, mechanical lateral distal femoral angle (LDFA), lateral distal tibial angle, joint line convergence angle (JLCA), KJLO, and ankle joint obliquity. They also categorized these radiographic parameters as preoperative and postoperative and calculated the difference (Δ) between preoperative and postoperative values. To determine which of the radiographic parameters were most associated with postoperative KJLO, multivariable linear regression analysis was performed using the stepwise method. Multivariable logistic regression analysis was used to examine the relative contribution of the preoperative radiographic parameters to an abnormal postoperative KJLO (>4°). RESULTS: In the multivariable linear regression analysis, the preoperative LDFA and JLCA showed a statistically significant correlation. Multivariable logistic regression analysis revealed that the mean preoperative LDFA was significantly larger in the group with abnormal KJLO than in the group with the control group (odds ratio, 1.84; 95% CI, 1.12-3.02; P = .02), while preoperative JLCA tended to be larger in the abnormal KJLO group than the control group but not statistically significantly different. CONCLUSION: KJLO after OWHTO was associated with preoperative LDFA and JLCA in multivariable linear regression analysis, and preoperative LDFA was the most important factor associated with abnormal KJLO after OWHTO in multivariable logistic regression analysis.