DESIGN: MEDLINE, EMBASE, CINAHL were systematically searched (1990-April 2020) for studies describing the prevalence of NP and PS in knee and hip osteoarthritis using self-report questionnaires. Random-effects meta-analysis was performed. Statistical heterogeneity between studies and sub-groups (affected joint and population source as a proxy for disease severity) was assessed (I2 statistic and the Chi-squared test).
RESULTS: From 2,706 non-duplicated references, 39 studies were included (2011-2020). Thirty-six studies reported on knee pain and six on hip pain. For knee osteoarthritis, the pooled prevalence of NP was: using PainDETECT, possible NP(score ≥13) 40% (95%CI 32-48%); probable NP(score >18) 20% (95%CI 15-24%); using Self-Report Leeds Assessment of Neuropathic Symptoms and Signs, 32% (95%CI 26-38%); using Douleur Neuropathique (DN4) 41% (95% CI 24-59%). The prevalence of PS using Central Sensitization Inventory (CSI) was 36% (95% CI 12-59%). For hip osteoarthritis, the pooled prevalence of NP was: using PainDETECT, possible NP 29% (95%CI 22-37%%); probable NP 9% (95%CI 6-13%); using DN4 22% (95%CI 12-31%) in one study. The prevalence of possible NP pain was higher at the knee (40%) than the hip (29%) (difference 11% (95% CI 0-22%), P = 0.05).
CONCLUSIONS: Using self-report questionnaire tools, NP was more prevalent in knee than hip osteoarthritis. The prevalence of NP in knee and hip osteoarthritis were similar for each joint regardless of study population source or tool used. Whether defining NP using self-report questionnaires enables more effective targeted therapy in osteoarthritis requires investigation.
MATERIALS AND METHODS: Results that are possible to be compared in more than two articles were presented as forest plots. A 95% confidence interval was calculated for each effect size, and we calculated the I 2 statistic, which presents the percentage of total variation attributable to the heterogeneity among studies. The random effects model was used to calculate the effect size.
RESULTS: Seven articles were included to the final analysis. Case groups were composed of HTO without concurrent procedures and control groups were composed of HTO with concurrent procedures such as marrow stimulation procedure, mesenchymal stem cell transplantation, and injection. The case group showed a higher hospital for special surgery score and mean difference was 4.10 [I 2 80.8%, 95% confidence interval (CI) - 9.02 to 4.82]. Mean difference of the mechanical femorotibial angle in five studies was 0.08° (I 2 0%, 95% CI - 0.26 to 0.43). However, improved arthroscopic, histologic, and MRI results were reported in the control group.
CONCLUSION: Our analysis support that concurrent procedures during HTO for medial compartment OA have little beneficial effect regarding clinical and radiological outcomes. However, they might have some beneficial effects in terms of arthroscopic, histologic, and MRI findings even though the quality of healed cartilage is not good as that of original cartilage. Therefore, until now, concurrent procedures for medial compartment OA have been considered optional. Nevertheless, no conclusions can be drawn for younger patients with focal cartilage defects and concomitant varus deformity. This question needs to be addressed separately.
METHODS: Knee image is first oversegmented to produce homogeneous superpixels. Then, a ranking model is developed to rank the superpixels according to their affinities to standard priors, wherein background superpixels would have lower ranking values. Finally, seed labels are generated on the background superpixel using Fuzzy C-Means method.
RESULTS: SAGE has achieved better interobserver DSCs of 0.94 ± 0.029 and 0.93 ± 0.035 in healthy and OA knee segmentation, respectively. Good segmentation performance has been reported in femoral (Healthy: 0.94 ± 0.036 and OA: 0.93 ± 0.034), tibial (Healthy: 0.91 ± 0.079 and OA: 0.88 ± 0.095) and patellar (Healthy: 0.88 ± 0.10 and OA: 0.84 ± 0.094) cartilage segmentation. Besides, SAGE has demonstrated greater mean readers' time of 80 ± 19 s and 80 ± 27 s in healthy and OA knee segmentation, respectively.
CONCLUSIONS: SAGE enhances the efficiency of segmentation process and attains satisfactory segmentation performance compared to manual and random walks segmentation. Future works should validate SAGE on progressive image data cohort using OA biomarkers.