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.

Analysis of cervical spine alignment in currently asymptomatic individuals: prevalence of kyphotic posture and its relationship with other spinopelvic parameters.

BACKGROUND CONTEXT: Many studies tend to characterize cervical kyphosis as a significant clinical condition that needs to be treated. Moreover, opinions vary on whether cervical kyphosis should be considered a pathologic status or a natural occurrence in asymptomatic people. PURPOSE: This study aimed to determine the frequency of kyphotic posture of the cervical spine in currently asymptomatic individuals and to ascertain its relation with other spinopelvic parameters. STUDY DESIGN: A cross-sectional radiographic study was carried out. PATIENT SAMPLE: This study targeted 1,026 currently asymptomatic adult volunteers who agreed to participate in this study from January 2010 to March 2016. Only 958 were eligible for the study. OUTCOME MEASURES: Radiographic images, including the C-spine dynamic view and whole-spine lateral view, were measured. The sagittal parameters of the cervical spine and other parts of the spine and pelvis, such as the C2-C7 angle, C0-C2 range of motion (ROM), C2-C7 ROM, and C0-C7 ROM, thoracic kyphosis, lumbar lordosis, sacral slope, pelvic tilt, and pelvic incidence, were measured. METHODS: Based on the C-spine neutral lateral X-ray, a C2-C7 Cobb angle greater than 0 degree was defined as lordosis and an angle less than 0 degree was defined as kyphosis. Patients who showed kyphosis were further classified into the reducible or non-reducible group, depending on the ability of recovering neck motions (lordosis) in extension. The cervical and other global spine parameters between the two groups were analyzed, and the relation between the cervical alignment and other parts of the spine and pelvis were also examined. This study was not supported by any funding and had no conflicts of interest. RESULTS: Nearly one-fourth of the asymptomatic participants (26.3%) have kyphotic cervical posture, and almost one-sixth of the kyphotic individuals (16.7%) have non-reducible kyphosis. The prevalence increases with advanced age; non-reducible cases are mostly kyphotic, kyphosis stems from the C2-C7 region, and kyphosis is not correlated with any of the radiological parameters of the other parts of the spine except lumbar lordosis. CONCLUSIONS: Cervical kyphosis can be observed in normal healthy adults.

The natural course of prevertebral soft tissue swelling after anterior cervical spine surgery: how long will it last?

BACKGROUND CONTEXT: Prevertebral soft tissue swelling (PSTS) after anterior cervical spine surgery (ACSS) has been regarded as one of the critical complications that cause airway obstruction. Still, however, no research has dealt with how PSTS returns to presurgery status after ACSS; most recommendations are being performed without information about its natural course, focusing on acute-phase swelling after surgery. PURPOSE: The study aimed to examine how long postsurgery PSTS lasts and when it returns to its presurgery state, and to analyze the actual influence of a number of factors to observe the natural progress of postsurgery PSTS. STUDY DESIGN/SETTING: This is a prospective observational study. PATIENT SAMPLE: The sample included a total of 160 patients who underwent ACSS, including anterior cervical discectomy and fusion (ACDF) and cervical total disc replacement (TDR). OUTCOME MEASURES: The diameter of PSTS measured at each set time point after surgeries was compared with PSTS measurements before surgery, and analyzed with factors influencing PSTS. METHODS: Anterior and posterior diameters of the anterior soft tissue of C3 (pharyngeal airway) and C6 (laryngeal airway) were measured using simple lateral radiography before surgery, immediately after surgery, at 2 weeks, 1, 3, 6, and 12 months after surgery. The progress of postsurgery PSTS was analyzed according to patients' individual characteristics, such as age, gender, weight, body mass index (BMI), smoking status, use of antiplatelet therapy, hypertension and diabetes mellitus, complaints of dysphagia, along with surgical factors such as anesthesia time, operation time, numbers of involved operation segments, transfusion, estimated blood loss , and operation method. Multivariable analysis by generalized linear mixed model was used to perform additional univariable analysis on variables found to be related to PSTS. In addition, to find the postsurgery interval at which PSTS naturally stabilizes, repeated measures analysis of variance and Bonferroni method were used to perform post-hoc tests. There were no sources of funding and no conflicts of interest associated with this study. RESULTS: For ACDF, the mean values (95% confidence interval [CI]) of PSTS in C3 were 4.38 (4.04~4.71), 10.40 (9.64~11.17), 7.72 (7.10~8.35), 6.24 (5.74~6.69), 5.43 (5.03~5.82), 5.14 (4.77~5.50), and 4.96 (4.59~5.33) mm at each follow-up time, respectively. In C6, the average values (95% CI) of PSTS were 14.43 (13.96~14.91), 19.18 (18.59~19.77), 17.92 (17.37~18.47), 16.98 (16.45~17.51), 16.18 (15.67~16.69), 15.95 (15.50~16.40), and 15.49 (15.50~16.40) mm. For cervical TDR, the mean values (95% CI) of PSTS in C3 were 3.67 (3.45~3.89), 8.05 (7.17~8.93), 5.42 (4.92~5.91), 4.57 (4.21~4.92), 4.12 (3.99~4.36), 4.10 (3.87~4.34), and 3.90 (3.66~4.14) mm at each follow-up time, respectively. In C6, the average values (95% CI) of PSTS were 13.61 (12.96~14.25), 16.51 (15.80~17.21), 15.77 (15.13~16.42), 15.24 (14.61~15.87), 14.62 (14.01~15.22), 14.52 (13.88~15.17), and 13.94 (13.20~14.68) mm. It is discovered that PSTS after surgery returned to presurgery status within 1 to 3 months in the pharyngeal airway (C3) and within 3 to 6 months in the laryngeal airway (C6), and gender, BMI, and surgery method (ACDF) were determined to be the factors having influence on PSTS after surgery. CONCLUSIONS: It is necessary to pay attention to PSTS and patient conditions after ACSS for at least 1 to 6 months postsurgery, depending on surgical method and operation levels.

Outcome and Complications in Surgical Treatment of Lumbar Stenosis or Spondylolisthesis in Geriatric Patients.

Development of anesthesiology and improvement of surgical instruments enabled aggressive surgical treatment even in elderly patients, who require more active physical activities than they were in the past. However, there are controversies about the clinical outcome of spinal surgery in elderly patients with spinal stenosis or spondylolisthesis. The purpose of this study is to review the clinical outcome of spinal surgery in elderly patients with spinal stenosis or spondylolisthesis. MEDLINE search on English-language articles was performed. There were 39685 articles from 1967 to 2013 regarding spinal disease, among which 70 dealt with geriatric lumbar surgery. Eighteen out of 70 articles dealt with geriatric lumbar surgery under the diagnosis of spinal stenosis or spondylolisthesis. One was non-randomized prospective, and other seventeen reports were retrospective. One non-randomized prospective and twelve out of seventeen retrospective studies showed that old ages did not affect the clinical outcomes. One non-randomized prospective and ten of seventeen retrospective studies elucidated postoperative complications: some reports showed that postoperative complications increased in elderly patients, whereas the other reports showed that they did not increase. Nevertheless, most complications were minor. There were two retrospective studies regarding the mortality. Mortality which was unrelated to surgical procedure increased, but surgical procedure-related mortality did not increase. Surgery as a treatment option in the elderly patients with the spinal stenosis or spondylolisthesis may be reasonable. However, there is insufficient evidence to make strong recommendations regarding spinal surgery for geriatric patients with spinal stenosis and spondylolisthesis.