Machine Learning for Detecting Total Knee Arthroplasty Implant Loosening on Plain Radiographs.

(1) Background: The purpose of this study was to investigate whether the loosening of total knee arthroplasty (TKA) implants could be detected accurately on plain radiographs using a deep convolution neural network (CNN). (2) Methods: We analyzed data for 100 patients who underwent revision TKA due to prosthetic loosening at a single institution from 2012 to 2020. We extracted 100 patients who underwent primary TKA without loosening through a propensity score, matching for age, gender, body mass index, operation side, and American Society of Anesthesiologists class. Transfer learning was used to prepare a detection model using a pre-trained Visual Geometry Group (VGG) 19. For transfer learning, two methods were used. First, the fully connected layer was removed, and a new fully connected layer was added to construct a new model. The convolutional layer was frozen without training, and only the fully connected layer was trained (transfer learning model 1). Second, a new model was constructed by adding a fully connected layer and varying the range of freezing for the convolutional layer (transfer learning model 2). (3) Results: The transfer learning model 1 gradually increased in accuracy and ultimately reached 87.5%. After processing through the confusion matrix, the sensitivity was 90% and the specificity was 100%. Transfer learning model 2, which was trained on the convolutional layer, gradually increased in accuracy and ultimately reached 97.5%, which represented a better improvement than for model 1. Processing through the confusion matrix affirmed that the sensitivity was 100% and the specificity was 97.5%. (4) Conclusions: The CNN algorithm, through transfer learning, shows high accuracy for detecting the loosening of TKA implants on plain radiographs.

Comparison of Gene Expression of Inflammation- and Fibrosis-Related Factors Between the Anterior and Posterior Capsule in Patients With Rotator Cuff Tear and Shoulder Stiffness.

BACKGROUND: Arthroscopic capsular release is an effective treatment for refractory shoulder stiffness, yet there are no basic studies that can explain the extent of the release. PURPOSE: This study aimed to compare the genetic expression of inflammation- and fibrosis-related factors between the anterior and posterior capsules in patients with shoulder stiffness and rotator cuff tear. STUDY DESIGN: Descriptive laboratory study. METHODS: Enrolled in this study were 35 patients who underwent arthroscopic capsular release for shoulder stiffness along with the rotator cuff repair. Anterior and posterior glenohumeral joint capsular tissues were obtained during the capsular release. For the control tissue, anterior capsule was obtained from 40 patients without stiffness who underwent arthroscopic rotator cuff repair. The gene expression of collagen types I and III, fibronectin, extracellular matrix, basic fibroblast growth factor, transforming growth factor-ß, connective tissue growth factor, matrix metalloproteinase (MMP)-1, MMP-2, MMP-9, tissue inhibitor of metalloproteinase (TIMP)-1, TIMP-2, intercellular adhesion molecule 1, interleukin 1, and tumor necrotizing factor-α were analyzed using real-time reverse transcription polymerase chain reaction. Differences in gene expression between the anterior capsule, the posterior capsule, and the control tissue were compared with the Kruskal-Wallis test. RESULTS: The expression levels of collagen types I and III were significantly higher in the anterior capsule with stiffness com (pared with both the posterior capsule with stiffness (P = .010 for both) and the control (P = .038 and .010, respectively). The levels of fibronectin, MMP-2, and MMP-9 in the anterior capsule were significantly higher than in both the posterior capsule (P = .013, .003, and .006, respectively) and the control (P = .014, .003, and .005, respectively). CONCLUSION: Genetic analysis of the shoulder capsule revealed that more fibrogenic processes occur in the anterior capsule compared with the posterior capsule in patients with shoulder stiffness. CLINICAL RELEVANCE: Capsular release for shoulder stiffness should be more focused on the anterior capsule than on the posterior capsule.

Surgical outcomes for late neurological deficits after long segment instrumentation for degenerative adult spinal deformity.

OBJECTIVE: The most catastrophic symptom of proximal junctional failure (PJF) following long instrumented fusion surgery for adult spinal deformity (ASD) is neurological deficits. Although previous reports have shown that PJF usually developed during the early postoperative period, some patients showed late neurological deficits. The aim of this study was to report the incidence, characteristics, and surgical outcomes of PJF with late neurological deficits. METHODS: Patients surgically treated for ASD at a single institution were retrospectively reviewed. Among them, the patients requiring revision surgery for newly developed neurological deficits at least 6 months after the initial surgery were included. Patient demographic, radiographic, surgical, and clinical data were investigated. Neurological status was assessed using the Frankel grading system. RESULTS: PJF with late neurological deficits developed in 18 of 385 patients (4.7%). The mean age at the onset of neurological deficits was 72.0 ± 6.0 years, and the median time from the initial surgery was 4.5 years. The most common pathology of PJF was adjacent disc degeneration and subsequent canal stenosis (11 patients). Five patients showed disc degeneration with aseptic bone destruction. Fractures at the upper instrumented vertebra (UIV), UIV + 1, and UIV + 2 occurred in 2, 3, and 2 patients, respectively. Ossification of the yellow ligament, which had not been found at the first surgery, was identified in 6 patients. Eight patients showed improvement of their neurological deficits and 10 patients showed no improvement by the final follow-up. Perioperative major complications occurred in 8 of 18 patients. CONCLUSIONS: The incidence of PJF with late neurological deficits following ASD surgery was 4.7% in this cohort. The patients showed several morphological features. After revision surgery, perioperative complications were common and the prognosis for improved neurological status was not favorable.

Sacral insufficiency fracture after instrumented lumbosacral fusion: Focusing pelvic deformation -A retrospective case series.

The purpose of this study was to report the characteristics of SIFs after ILSF and discuss its management focusing on pelvic deformation. We retrospectively reviewed all consecutive patients who underwent ILSF for degenerative disc diseases during the period between 2000 and 2017 and were diagnosed as SIF at our institute. The clinical and radiographic data were reviewed on their medical charts. Treatment outcomes for SIF were also investigated. Eight patients (all females) were included in this study. Mean age at SIF diagnosis was 72 years, and the mean follow-up period was 3.8 years (range 1-7 years). SIF developed average 7.5 years (range 1 month-17 years) after the index ILSF. Fracture patterns were unilateral vertical in four, bilateral vertical in three, and horizontal in 1 patient. Unlike patients with unilateral vertical SIF, patients with bilateral vertical or horizontal SIF showed a marked increase of pelvic incidence (PI) by mean 17.0°±5.0° and sagittal vertical axis (SVA) by mean 4.5 ± 2.2 cm, compared to the respective values before the onset of abrupt pain. All patients with unilateral vertical SIF were treated favorably by conservative management, however sacropelvic fixation was inevitable in patients with bilateral vertical or horizontal SIF. Bilateral vertical or horizontal SIF showed marked changes on sagittal radiographic parameters including PI and SVA. Although unilateral vertical SIF has benign courses that responded well to conservative management, bilateral vertical or horizontal SIF is likely to need surgical treatment. Treatment plan should be determined depending on fracture pattern and pelvic deformation.

Effects of Restoration of Sagittal Alignment on Adjacent Segment Degeneration in Instrumented Lumbar Fusions.

STUDY DESIGN: Retrospective case-control study. OBJECTIVE: To investigate the effects of postoperative sagittal alignment on radiographic adjacent segment degeneration (ASD) after lumbar fusion surgery. SUMMARY OF BACKGROUND DATA: ASD is one of inherent problems with fusion surgery. Many confounding factors are related to the development of ASD. Recently, sagittal alignment has been emphasized for its significance on ASD. METHODS: Seventy-three patients who underwent four-level lumbar fusion surgery (L2-S1) were divided into two groups according to postoperative sagittal alignment (pelvic incidence-lumbar lordosis [PI-LL] ≥ or <9°): 44 patients (matched group, including 10 patients who underwent pedicle subtraction osteotomy [PSO] at L4) and 29 patients (mismatched group). The general demographics, radiographic parameters, and clinical outcomes were recorded. Preoperative disc degeneration at L1-2 was evaluated by Pfirrmann grade and Kellgren-Lawrence (K-L) grade. Disc degeneration at L1-2 was evaluated by the K-L grade on 2-year postoperative X-rays. RESULTS: The incidence of radiographic ASD (11 [25%] vs. 16 patients [55%], P = 0.02) and Oswestry Disability Index (ODI) scores (36.9 ±â€Š19.9 vs. 49.4 ±â€Š20.7, P = 0.015) at postoperative 2 years were significantly higher in the mismatched group. There were no significant differences in other demographic and radiographic parameters between the two groups. On subgroup analysis between 10 PSO patients and the mismatched group, the mismatched group showed a higher incidence of radiographic ASD (16 [55%] vs. 1 patient [10%], P = 0.041) and worse ODI scores (49.7 ±â€Š20.5 vs. 39.0 ±â€Š20.7, P = 0.040). Preoperative Pfirrmann grade at L1-2 (odds ratio [OR] = 4.191, 95% confidence interval [CI]: 1.754-10.013, P = 0.001) and postoperative PI-LL mismatch (OR = 4.890, 95% CI: 1.550-15.427, P = 0.007) showed significant relationships with the development of radiographic ASD at postoperative 2 years. CONCLUSION: The restoration of optimal sagittal alignment, even with PSO, may provide a protective effect on the development of radiographic ASD, although the preoperative disc degeneration grade was a risk factor for radiographic ASD. LEVEL OF EVIDENCE: 3.

The efficacy and safety of polydeoxyribonucleotide for the treatment of knee osteoarthritis: Systematic review and meta-analysis of randomized controlled trials.

INTRODUCTION: The purpose of this study was to use meta-analysis techniques to evaluate the efficacy and safety of polydeoxyribonucleotide (PDRN) injections for knee osteoarthritis (OA) treatment. METHODS: Multiple comprehensive databases, including MEDLINE, EMBASE, and the Cochrane Library, were searched in November 2018 for studies that compared the effectiveness and safety of intra-articular PDRN injection for the knee joint with hyaluronic acid (HA) injection. Two reviewers independently determined study inclusion and they extracted data using a standardized data extraction form. The predefined primary outcome was Visual Analogue Scale. Secondary outcomes included Knee Injury and Osteoarthritis Outcome Score (KOOS), Knee Society Score (KSS), and adverse events. RESULTS: Five randomized controlled trials were included in the meta-analysis. After 1 and 2 months, patients in the PDRN group showed significantly better improvement in pain than the HA group (P = .04 and P = .02, respectively). There was no significant difference in pain after 4 months. The pooled analysis showed that no significant differences were seen in function (KOOS and KSS) scores between the PDRN and HA groups (all P > .05) at all time points. There was no significant difference in adverse events between 2 groups (relative risks = 2.15, 95% confidential interval: 0.17-26.67, P = .55). CONCLUSION: The intra-articular use of PDRN was similar in function to HA, and the pain-relief effect was superior to HA for 2 months post-injection. Therefore, it could be a favorable alternative to HA to treat persistent pain associated with knee OA while avoiding side effects.Level of evidence I.

Serum cystatin C is associated with subclinical atherosclerosis in patients with type 2 diabetes: A retrospective study.

BACKGROUND: The aim of this study was to investigate the association between the serum cystatin C level and cardiovascular disease risk in patients with type 2 diabetes mellitus. METHODS: We studied 523 patients with type 2 diabetes mellitus and calculated estimated 10-year risk of atherosclerotic cardiovascular disease (%). Subclinical atherosclerosis was defined as brachial-ankle pulse wave velocity ⩾1700 ms, indicating the presence of arterial stiffness. RESULTS: Cystatin C level was significantly higher in the subclinical atherosclerosis group (brachial-ankle pulse wave velocity ⩾ 1700 ms) than in the non-subclinical atherosclerosis group (brachial-ankle pulse wave velocity < 1700 ms) (7.54 ± 3.15 mg/L vs 10.04 ± 5.12 mg/L, p < 0.001). Subclinical atherosclerosis was mainly determined by age, duration of diabetes and cystatin C level, but not by serum creatinine, 10-year risk of atherosclerotic cardiovascular disease score and estimated glomerular filtration rate in the multiple linear regression analysis. In addition, an increase in cystatin C level was independently associated with the risk of subclinical atherosclerosis after adjusting for age, sex, duration of diabetes, smoking, hypertension, 10-year risk of atherosclerotic cardiovascular disease risk score, serum creatinine level, total cholesterol, high-density lipoprotein cholesterol and haemoglobin A1c (odds ratio = 1.200, 95% confidence interval: 1.04-1.38, p = 0.011). CONCLUSION: Serum cystatin C level was significantly associated with subclinical atherosclerosis. This result suggests that an increase in cystatin C level could be a valuable surrogate marker for the risk of cardiovascular disease in patients with type 2 diabetes mellitus.