This study aims to explore the application of the double-echo steady-state (DESS) sequence for the precise diagnosis of nerve root oedema in unilateral biportal endoscopic discectomy (UBED). This article explores the relationship between nerve root oedema and clinical symptoms, additionally investigating potential contributing risk factors. Postoperative clinical efficacy was evaluated using visual analogue scale (VAS) scores and the Oswestry Disability Index (ODI). DESS cross-sectional images were evaluated with ImageJ software. The oedema coefficient derived from DESS served as the principal quantitative metric. We adopted Lasso regression for feature selection. Subsequently, a multifactorial analysis was performed with the selected factors using the forward selection method. Forty-nine patients (27 males and 22 females) were included in the study. The average oedema coefficient was 1.25 ± 0.43. The postoperative lower limb VAS score was significantly correlated with the swelling coefficient (p = 0.000). However, there was no correlation between the postoperative ODI and the swelling coefficient (p = 0.106). The features selected for inclusion in the model were “Age,” “BMI,” “Type of protrusion,” “Disc height,” “Width of spinal canal,” “Lumbar curvature,” and “Course of disease. “Multiple stepwise regression analysis confirmed that disc height (r = 0.092, p = 0.008) was significantly positively correlated with the degree of oedema, and disease course (r=-0.010, p = 0.020) was significantly negatively correlated with the degree of oedema. The DESS sequence can indicate changes in the corresponding compressed nerve and correlates strongly with the postoperative leg VAS score. Disease progression and disc height may be significant factors influencing postoperative nerve root oedema.
Locating the posterior edge points of intervertebral discs is essential for evaluating the spatial relationship between the discs and adjacent neural structures, which is of great clinical significance for diagnosing disc protrusion. However, manual annotation of these points on MRI scans is time-consuming and labor-intensive. To overcome the limitations of low efficiency and accuracy in 3D posterior edge point localization, this study proposes a novel HRNet-ACAP-Offset framework. This framework adopts HRNet as the backbone and integrates a refined attention mechanism module (ACAP). Specifically, it employs a joint training paradigm with two task outputs and three loss functions. A 3D heatmap is utilized to regress the position of each landmark, while a landmark deviation estimation (Offset) method is adopted to correct the predicted positions. Furthermore, the regression of both positions and deviations is supervised by the original coordinate labels. Experimental results on a clinical dataset demonstrate that the proposed method achieves an average localization error of 3.84 mm (2.25 pixels), outperforming standard baseline architectures (e.g., vanilla HRNet and ResNet34). Furthermore, ablation studies validate that the multi-loss joint training strategy and the offset estimation module significantly enhance the localization accuracy of the posterior edge points for this task. The proposed HRNet-ACAP-Offset framework realizes efficient and high-precision automatic localization of intervertebral disc posterior edge points. It effectively solves the low efficiency and accuracy bottlenecks of traditional manual annotation and conventional model methods, possessing great potential for auxiliary clinical diagnosis of disc protrusion and related spinal diseases.
Multicenter retrospective cohort study. To compare 2-year sagittal alignment outcomes between two-row vertebral body tethering (2RVBT) and posterior spinal fusion (PSF) in adolescent idiopathic scoliosis (AIS). PSF remains the gold standard for AIS correction; however, concerns regarding motion loss and adjacent segment disease have encouraged development of fusionless techniques such as vertebral body tethering (VBT). While VBT has demonstrated promising coronal correction, evidence regarding its effects on sagittal alignment is limited. Ninety-nine AIS patients (49 2RVBT, 50 PSF) with ≥ 2-year follow-up were analyzed. Radiographic parameters included sagittal vertical axis (SVA), cervical sagittal vertical axis (cSVA), cervical lordosis (CL), pelvic tilt (PT), L4–S1 lordosis, T1 pelvic angle (TPA), and pelvic incidence–lumbar lordosis mismatch (PI–LL). The minimal clinically important difference (MCID) was prespecified as 5 mm for sagittal axis and 10° for lordosis, with pediatric thresholds of 3 mm and 5° based on AIS literature. Institutional Review Board approval was obtained at all participating centers. Inclusion criteria were AIS diagnosis, age ≤ 16 years at surgery, lumbar instrumentation, and ≥ 2-year follow-up. Exclusion criteria included neuromuscular/syndromic scoliosis and prior spinal surgery. 2RVBT patients were younger, more skeletally immature (Risser 1.6 ± 0.8 vs. 2.6 ± 1.8, p = 0.001), and more frequently female (97
PURPOSE:Whether research output is distributed across scoliosis subtypes in proportion to their burden of disease is unknown. We characterized the scoliosis literature published from 2010 to 2025 by etiologic subtype and compared scholarly output with disease prevalence. METHODS:We searched the Scopus database for articles and reviews on scoliosis published between January 1, 2010, and December 31, 2025. Records were classified by a predefined keyword algorithm into AIS, congenital, neuromuscular, degenerative/adult, syndromic, and non-adolescent idiopathic categories. The dataset was analyzed in R with the bibliometrix package. RESULTS:The search returned 27,190 documents (24,335 articles [89.5%] and 2,855 reviews [10.5%]) from 79,475 authors, with annual output rising from 858 in 2010 to 2,811 in 2025. Adolescent Idiopathic Scoliosis (AIS) was the largest named subtype, accounting for 6,818 documents (25.1% of the corpus and 43.8% of all output); 42.8% of documents could not be classified by etiology. Degenerative/adult scoliosis was second (4,176; 15.4%), grew the fastest, and had the highest mean citations per document (23.4), whereas AIS had a comparatively low mean citation rate (18.1). The AIS share of annual output remained between 20.2% and 27.2%, with no sustained directional change. Neuromuscular (11.4%), syndromic (6.0%), and congenital (3.6%) scoliosis remained comparatively small. CONCLUSIONS:The distribution of scoliosis research across etiologic subtypes doesn't mirror disease prevalence. Despite the substantially greater prevalence of degenerative scoliosis in adults, publication volume remained lower than that of AIS throughout the study period. These findings suggest congenital, neuromuscular, and degenerative scoliosis as priorities for future investigation.
To quantify the intraoperative adverse events (AEs) during surgical correction of Adolescent Idiopathic Scoliosis (AIS) and adolescent non-idiopathic scoliosis and identify predictors for such events using the British Spine Registry (BSR) dataset. Clinician-reported intraoperative AEs during AIS and non-idiopathic scoliosis procedures were collected from the BSR database. Data included deformity pathways for 10-18-year-olds diagnosed between 2012 and 2023. Intraoperative AEs included excessive bleeding, dural tears, implant malposition, neuromonitoring (IONM) changes, spinal cord injury, and cardiac complications. Hospital Episode Statistics (HES) were monitored to determine the reporting rate. Univariable analysis was utilised to calculate unadjusted Odds ratios (ORs) and 95
To investigate the clinical characteristics, neurological manifestations, and survival outcomes of patients with pathologically confirmed spinal lymphoma and to identify factors associated with prognosis. We retrospectively reviewed consecutive patients with pathologically confirmed spinal lymphoma treated at the First Affiliated Hospital, Zhejiang University School of Medicine, between Jan 2012 and Dec 2025. Clinical characteristics, treatment patterns, and survival outcomes were analyzed. Overall survival (OS) was estimated using the Kaplan–Meier method, and prognostic factors were evaluated using Cox proportional hazards models. Sixty-three patients were included. The median age was 63 years, and 90.5
To identify independent predictors of postoperative axial symptoms (AS) after cervical laminoplasty (CLP) and evaluate the predictive value of preoperative cervical range of motion. We retrospectively reviewed 191 patients with cervical spondylotic myelopathy who underwent unilateral open-door CLP between January 2019 and January 2024. Patients were classified into AS (n=73) and no AS (n=118) groups. Demographic, surgical, radiographic, and clinical variables were compared. Binary logistic regression identified independent predictors, and receiver operating characteristic analysis assessed predictive performance. Compared with the no-AS group, the AS group had higher BMI (26.5±3.7 vs. 25.4±3.3 kg/m2, P=0.027), more frequent T2-weighted spinal cord increased signal intensity (64.4
Cryoneurolysis and endoscopic facet denervation are established minimally invasive treatment options for lumbar facet joint–mediated pain; however, comparative evidence regarding their long-term effectiveness remains limited. This study compared the 36-month clinical outcomes of both procedures in a prospective multicenter randomized clinical trial. Eighty patients with lumbar facet joint–mediated pain confirmed by dual controlled medial branch blocks (≥ 70
Persistent cervical kyphosis after surgery for adolescent idiopathic scoliosis (AIS) has been associated with inadequate restoration of thoracic kyphosis and low T1 slope. However, the influence of intrinsic cervical factors on postoperative cervical alignment remains unclear. This study aimed to determine whether preoperative cervical disc degeneration is associated with residual cervical kyphosis after AIS surgery. A retrospective cohort study was conducted in 72 patients with AIS Lenke type 1 or 2 who were followed for at least two years after surgery (mean follow-up, 5.1 years). Patients were categorized into a cervical kyphotic group (C2–7 angle < 0° at final follow-up; n = 42) and a non-kyphotic group (C2–7 angle ≥ 0°; n = 30). Radiographic parameters included C2–7 angle, T1 slope, thoracic kyphosis (TK), pelvic incidence, lumbar lordosis, PI-LL mismatch, sagittal vertical axis, and major Cobb angle. Preoperative cervical MRI was used to assess disc degeneration using the Pfirrmann classification and to calculate the signal intensity ratio (SIR) of cervical discs. Multivariable analyses were performed to identify factors independently associated with cervical kyphosis at final follow-up. The cervical kyphotic group demonstrated significantly smaller preoperative C2–7 angles and lower postoperative T1 slope and TK(5–12) compared with the non-kyphotic group. Mid-cervical disc degeneration (Pfirrmann grade ≥ 3) was significantly more frequent in the cervical kyphotic group. The SIR of mid-cervical discs also correlated with the C2–7 angle. In multivariable logistic regression, preoperative C2–7 angle and Pfirrmann grade ≥ 3 at the mid-cervical discs were independently associated with residual cervical kyphosis at final follow-up. In contrast, postoperative T1 slope and TK(5–12) were not significantly associated with cervical kyphosis. Preoperative mid-cervical disc degeneration and preoperative C2–7 angle are independently associated with residual cervical kyphosis after AIS surgery, highlighting the importance of intrinsic cervical structural factors in postoperative cervical sagittal alignment.
Retrospective comparative cohort study. Accurate pedicle screw placement is critical for spinal stability and for preventing neurological or vascular injury. This study compared screw-placement accuracy and safety between robot-assisted and navigation-guided pedicle screw implantation and explored differences between intraoperative 3D-fluoroscopy and preoperative CT-based surface-matching registration. We assessed all consecutive patients who underwent robot-assisted or navigation-guided pedicle screw placement at our department between 2018 and 2022. For robot-assisted placement, screw trajectories were planned on preoperative CT, and intraoperative patient registration was performed using two radiographs fused with a preoperative CT. Navigation-guided placement used either intraoperative 3D-fluoroscopy registration in the final prone position or CT-based surface matching using preoperative supine CT images. Screw positions were evaluated on postoperative CT using a standardised ordinal grading scale from 0 to 4, with Grades 3–4 defined as clinically relevant malpositions. We analysed accuracy using generalised estimating equations with patient-level clustering and adjustment for covariates. We also assessed operative time, radiation exposure, complications, and revision rates. We analysed 1,718 pedicle screws from 268 patients: 728 were robot-assisted and 990 navigation-guided, including 579 using 3D-fluoroscopy and 411 using CT-based surface matching. Mean screw grades were 0.956 for robot-assisted placement, 0.864 for navigation overall, 0.807 for 3D-fluoroscopy, and 0.944 for surface matching. After accounting for patient clustering and adjustment, screw grade did not differ significantly between navigation-guided and robot-assisted placement (aOR 0.87, 95
Bilateral posterior C1 lateral mass–C2 pedicle screw fixation is the standard surgical treatment for unstable traumatic odontoid fractures. However, unilateral fixation may be required when bilateral instrumentation cannot be safely completed because of anatomical constraints or intraoperative complications. This study compared the clinical and radiological outcomes of unilateral and bilateral posterior C1–C2 fixation in patients with traumatic odontoid fractures. This retrospective comparative cohort study included 43 adult patients who underwent posterior C1 lateral mass–C2 pedicle screw fixation for traumatic odontoid fractures between 2016 and 2024. Fourteen patients underwent unilateral fixation and 29 underwent bilateral fixation. The primary outcome was CT-confirmed fusion. Secondary outcomes included pain, neurological function, radiological parameters, operative variables, complications, and reoperation. Longitudinal radiological outcomes were evaluated using baseline-adjusted linear mixed-effects models where appropriate. CT-confirmed fusion was achieved in 92.9
To synthesise patient-reported and reoperation outcomes after endoscopic decompression alone for lumbar spinal stenosis with degenerative spondylolisthesis, while accounting for paired measurements, linked reports, clinical importance and certainty. PubMed/MEDLINE, Embase, Scopus, the Cochrane Library, trial registries and supplementary sources were searched to 9 August 2026. Adult cohorts with lumbar stenosis, degenerative spondylolisthesis, endoscopic decompression alone and at least 12 months’ follow-up were eligible. Linked reports were clustered to avoid duplicate patients. Leg-pain visual analogue scale (VAS; 0–10) and Oswestry Disability Index (ODI; 0–100) were pooled as paired mean improvements using random-effects restricted maximum likelihood with Hartung-Knapp-Sidik-Jonkman inference. When change-score SDs were unavailable, a pre/post correlation of 0.60 was imputed and varied from 0.30 to 0.80. Reoperations were pooled with a binomial-normal generalized linear mixed model. ROBINS-I and GRADE were applied. Nine independent cohorts (335 patients) informed at least one synthesis. Leg-pain VAS improved by 5.25 points (95
Retrospective cohort study. To evaluate whether elevated 6-week postoperative pelvic tilt (PT) is associated with reduced 1-year age-adjusted sagittal alignment target achievement after TLIF, independent of baseline alignment and patient/surgical factors. Elevated postoperative PT has been associated with poor sagittal alignment, but prior analyses in degenerative TLIF cohorts have been criticized for reliance on simplified thresholds and limited adjustment for baseline alignment phenotype. The clinical importance of persistent or newly developed postoperative compensation remains incompletely defined. A retrospective analysis of 817 primary TLIF patients was performed. Patients were stratified by 6-week PT into high PT (≥ 25°) and low PT (< 25°) groups. The primary endpoint was 1-year age-adjusted PI–LL target achievement (Lafage age-specific thresholds). The primary multivariable logistic model adjusted for preoperative PT, preoperative PI–LL mismatch, age, sex, BMI, CCI, and number of fused levels; 6-week age-adjusted target status was not included, as it lies on the causal pathway between exposure and outcome, and the corresponding over-adjusted model is reported as a supplementary analysis. A prespecified sensitivity model used continuous 6-week PT. Supportive transition-phenotype analyses (Low→Low, Low→High, High→Low, High→High) were performed in patients with both preoperative and postoperative PT data. High PT was present in 306/817 patients (37.5
To compare anterior cervical discectomy and fusion (ACDF) pseudoarthrosis and reoperation rates derived from randomized controlled trials (RCTs) comparing ACDF to cervical disc arthroplasty (CDA) versus other non-CDA studies. This meta-analysis followed PRISMA guidelines. PubMed, Embase, Web of Science, Scopus, and Cochrane databases were reviewed for studies reporting reoperation and/or symptomatic pseudoarthrosis following ACDF. 61 studies involving 5,798 ACDFs were included. Analyses were stratified by Early (< 3 years), Mid (3–7 years), and Long (> 7 years) follow-up. Combined rates of reoperation and pseudarthrosis with 95
Treatment decisions for osteoporotic thoracolumbar vertebral fractures (OVF) in elderly patients are challenging. The AO Spine–DGOU Osteoporotic Fracture (OF) classification and OF-score support treatment allocation but do not explicitly account for chronological age, despite its known prognostic relevance. In addition, determinants of length of hospital stay (LOS) in older and very old patients with OVF remain incompletely defined. In this prospective single-centre cohort study, patients aged ≥ 65 years with low-energy OVF were systematically classified using the OF system and treated conservatively or surgically. Multivariable logistic regression was used to identify determinants of surgical treatment, including age-dependent effects. Determinants of LOS were analysed using linear regression of log-transformed LOS, including interaction analyses between age group and treatment. A total of 171 patients were included. Higher OF grades and neurological deficits strongly favoured surgical treatment (OF 3: OR 11.53, 95
Although lumbar spinal stenosis (LSS) surgery improves pain and function, some patients experience worsening low back pain (LBP). This exploratory study examined whether the preoperative lumbar multifidus (LMF) to lumbar erector spinae (LES, longissimus) activity ratio on surface electromyography (sEMG) is associated with postoperative LBP worsening after LSS surgery. This single-center prospective observational study included 70 patients with LSS who underwent surgery and completed 12-month follow-up. Preoperative sEMG signals from the LMF and LES were recorded during a trunk flexion–return task. The main exposure was the LMF/LES activity ratio during movement. The primary and secondary outcomes were change in LBP visual analog scale (VAS) from 6 to 12 months (ΔLBP 6–12 M) and 12-month LBP VAS, respectively. Recurrent LBP, defined as a ≥ 20-mm increase in LBP VAS from 6 to 12 months, was evaluated as an exploratory binary outcome. Multivariable linear regression was adjusted for prespecified covariates. Recurrent LBP occurred in 10 of 70 patients (14.3
To evaluate current evidence regarding the clinical reliability and reasoning capabilities of Large Language Models (LLMs) and Multimodal Large Language Models (MLLMs) within the spine surgery workflow. This scoping review utilizes a novel Vertical Integration Maturity Scale (VIMS) to map model maturity across five stages of the vertical workflow, identifying persistent research gaps and technical prerequisites for clinical implementation. A systematic search was conducted across PubMed, Embase, and Scopus for peer-reviewed studies published between January 2023 and January 2026. Utilizing the Population, Concept, and Context (PCC) framework, studies were selected based on their application of generative AI to the clinical evaluation and surgical management of spinal pathologies. Data were analyzed via independent dual-coding using a two-dimensional framework mapping five VIMS levels across five functional workflow stages. Synthesis included categorization by model architecture, input modality, and performance benchmarks, with inter-rater reliability calculated to validate the novel framework. Of 351 identified records, 40 studies met the inclusion criteria. Research density peaked in Stage III (Evaluative Decision Logic) at VIMS Level 1 (n = 15), indicating a methodological focus on evidence-based guideline retrieval over patient-specific synthesis. While ChatGPT-4 showed high concordance (61.1
Spinal endoscopy is associated with a steep learning curve, in part owing to its limited depth perception, which increases the need for training models. This study aimed to investigate construct and predictive validity of the Endobox as a low-fidelity training model for biportal endoscopy. To assess construct validity, we compared the performances of expertsand novices across four Endobox exercises. Performances were evaluated based on completion time and number of attempts, using a 0º endoscope and rongeur. To assess predictive validity of the Endobox, 16 students (n = 8 per group) were randomly assigned into EB+ (twelve training sessions) or EB- (no training) before performing six simulated discectomies on a non-anatomical simulation model. Completion time, technical skill (modified version of the ASSET score), and cognitive demand (NASA-TLX score) were compared. Statistical analysis was carried out using parametric and non-parametric tests using Stata 18 for Mac. Construct validity: experts were significantly faster in 3/4 exercises (2. p = 0.02, 3. p = 0.01, 4. p = 0.03), except in exercise 1 (p = 0.43). Experts needed fewer attempts for exercises 3 (p = 0.04) and 4 (p = 0.02), rated most difficult (9/10, 8/10 respectively). Predictive validity: the EB+ group was faster (p < 0.001) and improved across all observations, achieving higher scores for safety (p = 0.007), camera/bimanual dexterity, instrument handling, depth perception, and procedure quality (all p < 0.001). No difference in cognitive load was observed (p = 0.621). The present study provides preliminary evidence that the Endobox model may develop skills that are relevant to, and potentially transferable to, a simulated procedure representing endoscopic spine surgery.
Adolescent idiopathic scoliosis (AIS) requires follow-up during growth to detect progression, resulting in a burden of care. The Scolioscoop®V3 was developed as an alternative to the Bunnell Scoliometer, to facilitate home monitoring by parents/caregivers. This study evaluated the accuracy and precision of the Scolioscoop® V3 compared with the gold standard, both in the outpatient clinic and in the home setting. A single-center cross-sectional validation study and a multicenter prospective longitudinal study were combined. Scolioscoop® measurements performed by healthcare professionals and parents/caregivers were compared with Bunnell Scoliometer measurements. Accuracy and precision were assessed using intraclass correlation coefficients (ICC) and Bland–Altman analyses. 92 patients with AIS were included. Measurements performed with the Scolioscoop® V3 by healthcare professionals demonstrated an excellent correlation with the Bunnell Scoliometer (ICC 0.92, 95
Perioperative hypothermia is a recognized yet often overlooked complication of endoscopic spine surgery, where continuous fluid irrigation is required for adequate visualization and hemostasis. Despite the substantial volumes of saline used during unilateral biportal endoscopic (UBE) procedures, the systemic effects of irrigation fluid temperature remain poorly characterized. We evaluated its impact on perioperative body temperature, venous blood gas parameters, and pain outcomes after UBE surgery. In this retrospective cohort study sixty consecutive patients with lumbar disc herniation (LDH) or lumbar spinal stenosis (LSS) undergoing UBE surgery were analyzed in two groups: a warm group (n = 30) in which saline was warmed to 36–37 °C, and a cold group (n = 30) in which room-temperature saline (20–22 °C) was used. Tympanic body temperature, venous blood gas parameters (pH, lactate, bicarbonate, base excess), and back and leg Visual Analog Scale (VAS) scores were compared at standardized time points. No significant between-group differences were observed preoperatively (p > 0.05), but marked differences emerged from the 30th intraoperative minute onward and progressed throughout surgery (all p < 0.001). Postoperatively, the cold group showed lower body temperature (35.45 ± 0.37 vs. 36.25 ± 0.13 °C), lower pH (7.28 ± 0.03 vs. 7.38 ± 0.01), higher lactate (2.88 ± 0.35 vs. 1.40 ± 0.15 mmol/L), and lower bicarbonate (19.65 ± 1.56 vs. 23.26 ± 0.83 mmol/L). No significant differences were found in VAS scores or hemoglobin changes. Warming irrigation fluid to near-body temperature may prevent intraoperative hypothermia and associated metabolic acidosis during UBE surgery, representing a simple, low-cost strategy to support perioperative physiologic stability.
