Background context: The gold standard diagnostic test for Adolescent Idiopathic Scoliosis (AIS) involves manually measuring the spinal column deformity by determining the Cobb angle on a full-spine X-ray image. This measurement involves a subjective interpretation of vertebrae position and angle calculation, and inter-observer variability is widely accepted as one of the main causes of diagnosis uncertainty.
Purpose: Our objective was to develop an automated and reproducible system based on artificial intelligence (AI) to assist in Cobb angle estimation on full-spine radiographs without human intervention.
Study design: Retrospective, observational, multicenter study.
Methods: We performed a multicenter study involving four tertiary hospitals in which we collected full-spine anteroposterior/posteroanterior (AP/PA) X-ray images from AIS patients with Cobb angles ranging from mild to severe. Images were analyzed by three independent clinicians in each center (first dataset). Any discrepancies in clinician-reported measurements prompted reevaluation of images and data curation. We developed a deep learning pipeline featuring two specialized AI models designed to detect the spine's curvature from X-ray images, identify the individual vertebrae, and accurately estimate the Cobb angles of all curves detected in the spine.
Results: From a total of 484 X-ray images collected, spine surgeons reached consensus on 1,054 curves. Initial analysis identified 86.4% of these curves, with a mean absolute error (MAE) of 2.41° ± 3.24° relative to the consensus measurement after reevaluation and with error values ranging from -41.30° to 40.7°. In comparison, our SPinal Autonomous Radiological Cobb-assessment (SPARC) AI system detected 94.0% of the consensus curves, with a MAE of 3.01° ± 2.71°, which is within the clinical acceptance threshold (≤6°), and with a more constrained range of error showing values from -14.6° to 20.3°.
Conclusion: SPARC is an AI-based system developed for automatic, reproducible, and accurate calculation of Cobb angles in full AP/PA spine radiographs without human intervention. SPARC demonstrates superior performance by detecting a higher proportion of spinal curves (94.0% vs. 86.4%) and achieving a lower error range in Cobb angle estimation (±20.3° vs. ±41.3°) compared to the initial evaluation by three specialists with more than 10 years' experience.
Clinical significance: SPARC removes the intra-observer error and inter-observer variability inherent to manual measurements, and significantly decreases radiograph measurement and interpretation times, thus supporting clinicians in patient management and providing a reliable tool for less experienced practitioners involved in the care of patients at all stages of the AIS journey.
Background context: Outcomes of lumbar spine surgery in dialysis-dependent patients remain poorly characterized, particularly regarding patient-reported outcome measures (PROMs).
Purpose: To evaluate PROMs and complications in dialysis-dependent patients undergoing lumbar degenerative surgery.
Study design/setting: Multicenter retrospective cohort study.
Patient sample: We analyzed 19,911 patients, including 361 dialysis-dependent and 19,550 non-dialysis-dependent patients, who underwent elective lumbar degenerative surgery between 2017 and 2023.
Outcome measures: Outcomes included 1-year changes in Numeric Rating Scale (NRS) for pain, EuroQol 5-Dimension (EQ-5D), and Oswestry Disability Index (ODI), as well as patient satisfaction at 1 year, and perioperative complications and reoperations within 30 days.
Methods: Outcomes were compared between dialysis-dependent and non-dialysis-dependent patients. Propensity score matching was performed to adjust for baseline characteristics. Noninferiority of improvements in PROMs (NRS for pain, EQ-5D, ODI) was tested with predefined noninferiority margins (Δ = 1.5, -0.2, 15, respectively).
Results: In the overall cohort, the dialysis-dependent group was older, had a lower BMI, had higher American Society of Anesthesiologists Physical Status (ASA-PS) classes, and had a higher prevalence of comorbidities. They also underwent more extensive surgeries, with higher complication and reoperation rates and worse PROMs compared with the non-dialysis-dependent group. The non-dialysis-dependent group demonstrated a significantly greater improvement in EQ-5D (0.20 vs. 0.17; p=.03) and a higher satisfaction rate (83% vs. 74%; p=.004) than the dialysis-dependent group. However, both groups showed significant postoperative improvements in PROMs. Notably, 74% of the dialysis-dependent patients were satisfied with their surgery. In the matched cohort, despite baseline and postoperative PROMs were worse in dialysis-dependent group, improvements in PROMs were noninferior to those in the non-dialysis-dependent patients. These findings were consistent in multivariable and sensitivity analyses.
Conclusions: Despite poorer baseline characteristics and higher complication rates, dialysis-dependent patients achieved significant and noninferior improvements in PROMs and satisfaction after lumbar spine surgery. These findings support lumbar spine surgery as a feasible treatment option for appropriately selected dialysis-dependent patients.
Background context: Spinopelvic fixation is commonly used to reduce the risk of distal adjacent segment disease in adult spinal deformity surgery. However, it may impose increased stress on the hip joint, potentially leading to hip osteoarthritis (OA). The concept of adjacent joint disease has been proposed to describe hip joint degeneration due to compensatory overload following rigid spinopelvic fixation, but longitudinal data supporting this are lacking.
Purpose: To investigate the prevalence and risk factors for the progression of hip OA after spinal fusion surgery, and to evaluate adjacent joint disease as a distinct pathological entity.
Study design/setting: Retrospective single-center cohort study with 5-year follow-up.
Patient sample: A total of 290 patients (580 hips) who underwent spinal fusion surgery between 2011 and 2018, and met inclusion/exclusion criteria.
Outcome measures: Radiographic hip OA progression was defined by an increase in Kellgren-Lawrence (KL) grade at 5 years postoperatively. Spinopelvic parameters and hip morphology were also assessed.
Methods: Radiographs were evaluated preoperatively, and at 1 month and 5 years postoperatively. Patients were divided based on the presence or absence of OA progression. Logistic regression was used to identify independent risk factors. A post hoc exploratory subgroup analysis was performed in patients with long fusion (≥6 levels) and baseline KL grade 0-1.
Results: In the full cohort, OA progression was observed in 13.8%, and new-onset OA in 10.8%. Risk factors for progression in the overall cohort included female sex (OR 3.95, p=0.0048), pelvic incidence (PI) (OR 1.04, p=0.0012), sacral slope (SS) correction (OR 1.04, p=0.018), pelvic fixation (OR 5.04, p=0.0016), and baseline KL grade (OR 2.49, p=0.00001). In the long fusion subgroup (n=122; 244 hips), new-onset OA was observed in 21.7%. Within this subgroup, pelvic fixation (OR: 8.48, p=0.0069), larger SS correction (OR: 1.04, p=0.027), and higher PI (OR: 1.05, p=0.0017) remained significant predictors of hip OA.
Conclusions: Spinopelvic fixation was associated with an increased risk of hip OA progression, especially in female patients and those with high PI or large SS correction. These findings support adjacent joint disease as a clinically relevant entity and suggest the need for surgical caution and postoperative strategies to protect the hip joint in vulnerable individuals.
Background context: Lumbar spine fusion is frequently performed to eliminate motion between vertebrae and thereby relieve symptoms. However, there is currently no clinically validated, biomechanically rational standard for diagnosing failure to achieve this surgical goal. A strain-based method has recently shown promise in assessing fusion status after cervical spine surgery. Its applicability to the lumbar spine remains unknown.
Purpose: To evaluate the feasibility and performance of a strain-based approach for assessing lumbar spine fusion status.
Study design: Retrospective analysis of lumbar flexion-extension radiographs obtained following fusion surgery.
Methods: Using FDA-cleared automated software, intervertebral strain was calculated from anatomic landmarks on flexion-extension radiographs obtained at multiple time points (3-60 months) following posterior-lateral (PL) or PL plus interbody (PL+IB) fusion. Strain values were categorized as: Motion Compatible with Bridging (MCB), Uncertain, or Motion Incompatible with Bridging. The percentage of levels in each category was determined over time and compared between fusion types. Adjacent-level strain was also evaluated. A proof-of-concept convolutional neural network was trained on motion-stabilized image pairs to classify uncertain levels.
Results: Strain data were analyzed for 1,958 PL and 2,079 PL+IB fusion levels. PL+IB fusions demonstrated a faster reduction in intervertebral strain. By 60 months, average strain was <5% for both fusion types, with 86% of PL and 90% of PL+IB levels classified as MCB. Adjacent-level strain increased slightly after fusion surgery. The convolutional neural network correctly classified 96% of levels as MCB or Motion Incompatible with Bridging and reduced the proportion of uncertain cases from 21% to 5%.
Conclusions: A strain-based method provides an objective, biomechanically grounded, and automated approach for monitoring fusion progression after lumbar spine surgery. A neural network can enhance this method by reducing the need for subjective review of borderline cases.
Clinical significance: Strain-based fusion assessment enables standardized, reproducible, and scalable evaluation of postsurgical spinal motion. With further validation, it may improve clinical decision-making and facilitate more consistent outcomes reporting in spine surgery research.
Background context: Manual measurement of cervical sagittal parameters is time-consuming and exhibits significant interobserver variability. Existing artificial intelligence models fail when C7 is obscured by shoulder anatomy.
Purpose: To develop and externally validate a deep learning model for automated cervical alignment measurements under clinical conditions, including C7-obscured cases.
Design: Retrospective observational study.
Patient sample: Five thousand six hundred four lateral cervical radiographs were obtained from Chinese and Korean institutions.
Outcome measure: Intraclass correlation coefficient (ICC), Pearson correlation (r), and Bland-Altman agreement.
Methods: A Keypoint R-CNN with ResNet-50-FPN backbone was trained using multinational data, including C7-obscured cases. Model outputs were compared to consensus expert annotations using ICC, Pearson correlation, and Bland-Altman analysis. An independent dataset was used for external validation.
Results: In the external validation set (n=100), 62 patients (62.0%) had a partially obscured C7 and 20 patients (20.0%) had a fully obscured C7. The final model showed excellent reliability for the C2-C7 lordosis (ICC=0.95, r=0.95), C2 slope (ICC=0.99, r=0.99) and C7 slope (ICC=0.91, r=0.91). The mean errors for these parameters were clinically negligible at -0.44°, 0.06°, and -0.38°, respectively. The reliability for all disc height measurements were excellent in internal test set (ICC=0.97-0.99). Measurement errors slightly increased in few patients with complete C7 obscuration.
Conclusion: The Keypoint R-CNN model enables rapid, accurate, and clinically generalizable automated cervical alignment measurements; however, C7 obscuration remains a critical limitation that requires targeted improvement.
Background context: Spinal arthrodesis is widely used for degenerative, deformity, traumatic, and neoplastic conditions, yet nonunion remains a major source of pain, hardware failure, and revision surgery. Multimodal analgesia often includes ketorolac to reduce opioid exposure, but early single-center cohorts linked postoperative ketorolac to higher nonunion, while more recent randomized trials and large database studies using short, protocolized regimens have not shown increased nonfusion.
Purpose: To understand the effects of perioperative ketorolac on nonunion after spine fusion.
Study design: Systematic review with meta-analysis.
Methods: We conducted a PRISMA-compliant systematic review and meta-analysis (PROSPERO: CRD420251137564) of eligible studies that enrolled adults undergoing any spinal fusion, compared perioperative ketorolac with no ketorolac or no NSAID, and reported fusion outcomes. Risk of bias was assessed using ROBINS-I for observational studies and RoB-2 for randomized trials. Random-effects models pooled odds ratios for nonunion. Prespecified subgroups assessed study design, spine region, outcome definition, and exposure windows by time and dose.
Results: Across 41,365 patients (20,713 ketorolac vs 20,652 controls), perioperative ketorolac was not associated with higher nonunion overall (OR 1.10, 95% CI 0.82 to 1.49, p=.52, I²=57.0%). Statistically significant increases appeared only in specific contexts: older retrospective single-center cohorts (OR 2.59, 95% CI 0.68 to 9.91, p=.024), and exposures longer than 48 hours or exceeding 240 mg (each OR 2.17, 95% CI 1.21 to 3.90, p<.01), supported by significant subgroup contrasts for study type and exposure thresholds. Sub-analyses by study type and spine region did not show a significant difference.
Conclusions: Perioperative ketorolac, when limited to short, protocolized courses of less than 48 hours at moderate doses (less than 240 mg or 2.5 mg/hour), was not associated with a clinically meaningful increase in nonunion after spinal fusion. Elevated risk described in older single-center cohorts appears related to longer or less standardized exposure. These findings support ketorolac as a component of multimodal analgesia within defined time and dose limits and justify prospective dose-stratified trials to refine exposure thresholds for complex and multilevel constructs.
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：