Introduction: Intervertebral disc degeneration (IDD) is a primary cause of chronic back pain and disability, highlighting the need for precise detection and grading for effective treatment. This study focuses on developing and validating a convolutional neural network (CNN) with a You Only Look Once (YOLO) architecture model using the Pfirrmann grading system to classify and grade lumbar intervertebral disc degeneration based on magnetic resonance imaging (MRI) scans.

Methods: We developed a deep learning model trained on a dataset of anonymized MRI studies of patients with symptomatic back pain. MRI images were segmented and annotated by radiologists according to the Pfirrmann grading for the datasets. The segmentation MRI-disc image dataset was prepared for three groups: a training set (1,000), a testing set (500), and an external validation set (500) to assess model generalizability without overlapping images. The model's performance was evaluated using accuracy, sensitivity, specificity, F1 score, prediction error, and ROC-AUC.

Results: The AI model showed high performance across all metrics. For Grade I IDD, the model achieved an accuracy of 97%, 95%, and 92% in the training, testing, and external validation sets, respectively. For Grade II, the sensitivity was 100% in both training and testing sets and 98% in the validation set. For Grade III, the specificity was 95.4% in the training set and 94% in both testing and validation sets. For Grade IV, the F1 score was 97.77% in the training set and 95% in both testing and validation sets. For Grade V, the prediction error was 2.3%, 2%, and 2.5% in the training, testing, and validation sets, respectively. The overall ROC-AUC was 97%, 92%, and 95% in the training, testing, and validation sets, respectively.

Conclusions: The AI-based classification model exhibits high accuracy, sensitivity, and specificity in detecting and grading lumbar IDD using the Pfirrmann grading. AI has significantly enhanced diagnostic precision and reliability, providing a powerful tool for clinicians in managing IDD. The potential impact is substantial, although further clinical validation is necessary before integrating this model into routine practice.

Introduction: Elderly patients have a higher frequency of upper cervical fractures caused by minor trauma; nevertheless, the clinical differences between mid- and lower-cervical (C6-C7) injuries are unclear. The aim of this study was to compare the epidemiology of lower- and mid-cervical injuries in the elderly.

Methods: This multicenter, retrospective study included 451 patients aged 65 years or older who had mid- or lower-cervical fractures/dislocations. Patients' demographic and treatment data were examined and compared based on mid- and lower-cervical injuries.

Results: There were 139 patients (31%) with lower-cervical injuries and 312 (69%) with mid-cervical injuries. High-energy trauma (60% vs. 47%, p=0.025) and dislocation (55% vs. 45%, p=0.054) were significantly experienced more often by elderly patients with lower-cervical injuries than by patients with mid-cervical injuries. Although the incidence of key muscle weakness at the C5 to T1 levels were all significantly lower in patients with lower-cervical injuries than those with mid-cervical injuries, impairments at C5 occurred in 49% of them, and at C6, in 65%. No significant differences were found in the rates of death, pneumonia, or tracheostomy requirements, and no significant differences existed in ambulation or ASIA impairment scale grade for patients after 6 months of treatment.

Conclusions: Elderly patients with lower-cervical fractures/dislocations were injured by high-energy trauma significantly more often than patients with mid-cervical injuries. Furthermore, half of the patients with lower-cervical injuries had mid-cervical level neurological deficits with a relatively high rate of respiratory complications.

Introduction: This study aimed to standardize perioperative interruption of antiplatelet agents in patients undergoing cervical spinal surgery and investigate the incidence of epidural hematoma and thrombotic complications.

Methods: A total of 153 patients, consisting of 85 men and 68 women, were included in this study. Their mean age was 65.5 years. They were divided into two groups: Groups A and B. Group A (139 patients) did not receive preoperative antiplatelet agents, and Group B (14 patients) resumed antiplatelet agents from 7 or 14 days presurgery to 3 days postsurgery. Our analysis encompassed demographic data before surgery, postoperative magnetic resonance image-based assessment of radiological epidural hematoma (EH), and complications such as symptomatic hematoma, blood transfusion, stroke, and venous thromboembolism after surgery.

Results: The frequency of medical conditions, such as hypertension, diabetes, and hyperlipidemia, was significantly higher in Group B than in Group A. The CHADS2 scores, which serve as a clinical prediction rule for estimating stroke risk, were significantly higher in Group B than in Group A. In contrast, the intraoperative blood loss was significantly lower in Group B than in Group A. There was no significant difference in radiologically severe EH, hemorrhage, and thrombotic complications between the two groups. Interestingly, none of the patients in Group B had hemorrhagic and thrombotic complications.

Conclusions: Our standardized perioperative management of antiplatelet agents did not affect the incidence of radiological EH, hemorrhage, and thrombotic complications in patients undergoing cervical spinal surgery.

Introduction: Preoperative estimations of blood loss are important when planning surgery for cervical spine injuries in older adults. The association between ankylosis and blood loss in perioperative management is of particular interest. This multicenter database review aimed to evaluate the impact of ankylosis on surgical blood loss volume in elderly patients with cervical spine injury.

Methods: The case histories of 1512 patients with cervical spine injury at among 33 institutions were reviewed. After the exclusion of patients without surgery or whose blood loss or ankylosis status was unclear, 793 participants were available for analysis. Differences in blood loss volume were compared between the Ankylosis (+) group with ankylosis at the cervical level and the Ankylosis (−) group without by the inverse probability of treatment weighting (IPTW) method using a propensity score.

Results: Of the 779 patients (mean age: 75.0±6.3 years) eligible for IPTW calculation, 257 (32.4%) had ankylosis at the cervical level. The mean blood loss volume was higher in Ankylosis (+) patients than in Ankylosis (−) patients (P<0.001). This difference did not reach statistical significance when weighted by background factors, with mean blood loss of 244 mL and 188 mL, respectively, after adjustment.

Conclusions: This study revealed that ankylosis was significantly associated with increased blood loss volume when unadjusted by surgical time. Elderly patients with cervical spine injury accompanied by ankylosis appear predisposed to higher bleeding and severe hemorrhage, both as a result of the condition and their particular demographic characteristics.

Introduction: Ligamentum flavum (LF) hypertrophy is the main etiological factor in the development of lumbar spinal stenosis (LSS); however, its molecular pathology remains unclear. Histologically, LF hypertrophy is characterized by a reduction in elastic fibers and an increase in collagen fibers. We previously performed miRNA transcriptomic analysis on excised LF from elderly patients with LSS and identified the insulin receptor signaling along with TGFβ-mediated signaling as pathways involved in ligament hypertrophy. Therefore, this study aimed to investigate the involvement of endogenous insulin as a risk factor for LF hypertrophy in patients with LSS.

Methods: A total of 1,119 patients aged ≥65 years (average: 76.1±5.9 years) treated for LSS including surgery and conservative treatment were analyzed. The flavum canal ratio (FCR) was calculated in the MRI cross-sectional image, and an FCR of 0.4275 or greater was defined as ligamentous stenosis according to Sakai's criteria. Homeostatic model assessment for insulin resistance (HOMA-IR) was calculated and values ≥2.5 were indicative of insulin resistance in Japanese people.

Results: Fifty-one percent of patients with LSS exhibited LF hypertrophy, correlating with higher age, proportion of males and diabetic patients, BMI, HOMA-IR, and creatinine. Among LSS patients, 43.0% had insulin resistance, with 47.1% exhibiting LF hypertrophy and 38.6% without LF hypertrophy, with a significant difference (p<0.01). LSS patients with high insulin resistance also demonstrated significantly higher FCR (p<0.05) and a higher percentage of LF hypertrophy (p<0.01). Conditional logistic regression analysis, adjusting for age, identified HOMA-IR as a significant factor.

Conclusions: The study establishes an association between LF hypertrophy and insulin resistance. Considering LF hypertrophy as an inflammation-triggered degeneration of elastic fibers, age-related changes in LF may underlie the basis of inflammatory aging.

Introduction: Full endoscopic spine surgery continues to spread worldwide but has a long learning curve. Conventional endoscopy training uses live pigs or human cadavers, which has disadvantages such as high costs and limited availability. Therefore, this study aimed to develop and evaluate three-dimensional (3D)-printed models for endoscopy training.

Methods: Models for 3D printing were generated using raw imaging data from 1.0-mm slices of computed tomography scans, and each part was printed using a different colored material. The combined model was used for training as part of the full endoscopy training kit.

Results: This approach offers several advantages. First, it enables the creation of accurate disease models, such as lumbar disc herniation and other abnormalities, which are useful for both surgical training and preoperative simulations. Second, it is useful for learning surgical orientation. During surgical training, the surgical field can be viewed directly through an endoscope or with the naked eye. By using various colors, it becomes easier to recognize the orientation. Third, the amount of drilling resection can be easily confirmed, facilitating feedback. Finally, training for various surgical techniques is possible, including endoscopic holding techniques and using the endoscope's outer sheath to retract nerves. However, this approach also has some disadvantages, such as the lack of bleeding, inability to reproduce tissue hardness, and difficulty in faithfully recreating soft tissue, such as connective tissue, blood vessels, and fat. Therefore, it is difficult to reproduce the hardness of the calcified disc or disc herniation with apophyseal ring fracture. Moreover, 3D-printed models are not suitable for surgical training using the interlaminal approach because it is difficult to perform separation between the ligamentum flavum and dural matter or between the dural matter and intervertebral disc.

Conclusions: 3D-printed models are a useful complement to live pigs and human cadavers in surgical training and can reduce the time required to acquire endoscopic skills.

Introduction: Percutaneous vertebral augmentation techniques, such as balloon kyphoplasty (BKP) and vertebral body stenting (VBS), are commonly used for surgical intervention in osteoporotic vertebral fractures (OVFs). However, markedly unstable OVF cases require additional fixation procedures, prompting the exploration of combined percutaneous vertebral augmentation and posterior fixation. A novel surgical approach involving percutaneous vertebral augmentation with upward penetrating endplate screws (PES) and downward PES, complemented by a short fusion of one above one below, was developed. This study aimed to introduce and report the preliminary outcomes of this technique based on a retrospective analysis of 20 consecutive cases in the short and medium term.

Methods: Surgical indications are a vertebral wedge angle difference of 10° or more, vertebral pedicle fractures, posterior wall fractures, and diffuse low-signal changes exceeding 50% on T1-weighted magnetic resonance imaging. The procedure is reserved for highly unstable cases following a comprehensive health assessment. The surgical technique involves prone positioning, fluoroscopy-guided percutaneous vertebral augmentation, and the use of downward PES in the cranial vertebral body and upward PES for the caudal vertebral body by percutaneous technique. The fixation range is one above and one below.

Results: The case series of 20 patients, with an average follow-up period of 146.9 days, demonstrates a mean surgical time of 57 min and minimal complications. The advantages of the technique are as follows: ease of performance, minimal fixation range, and time efficiency. Risks, such as potential screw loosening and the need for prolonged follow-up, are acknowledged.

Discussion: The technique represents a promising surgical approach that balances the requirements of minimally invasive intervention and relatively robust initial fixation for elderly osteoporotic patients with unstable OVFs. While short- and medium-term results are favorable, long-term observations are needed to further assess its efficacy. This novel technique has a potential to be a valuable surgical option for unstable OVFs.

Objective: To evaluate curve correctability, complications, and rate of growth following treatment.

Background: Distraction-founded techniques such as traditionally growing rods or magnetically controlled growing rods are the almost globally accepted management patterns for early onset scoliosis. However, periodic lengthening operations are still needed. Moreover, an MCGR is difficult to contour, and implant-associated problems are common. We developed concave side apical control of the growing rod in which an additional anchor site is inserted at the apex to enhance stability and assist in the adjustment of axial deformity.

Methods: Entirely skeletally immature early onset scoliosis (EOS) cases with a progressive curve of >40° and without bone or soft tissue weakness were appropriate for this study. Coronal Cobb angle, sagittal parameters, complications, spinal length, and reoperations were documented with at least a 3-year follow-up.

Results: In this study, 15 patients were involved. The mean age was 7 years. The mean preoperative Cobb angle was 48°, which postoperatively became 12° with the percentage of coronal correction reaching 75.73%. The mean Cobb angle degrees of correction were 39°. T1-S1 height increased by 10 mm/year. Postoperative complications occurred in two cases with single rod technique and rod breakage.

Conclusions: The concave side apical control of the growing rod seems to be a hopeful surgical procedure for the management of EOS. Curve correctability in patients was 60% and can be sustained for a minimum of 2 years. Reoperations and complications might not be constricted, but the complication frequency looks more reasonable than in the current systems.

Introduction: Disc degeneration is a risk factor of pyogenic spondylitis. However, its degree in patients with pyogenic spondylitis is unknown. This study aimed to determine differences in disc degeneration between patients with pyogenic spondylitis and those with noninfectious lumbar spondylosis.

Methods: A total of 85 patients with lumbar pyogenic spondylitis (the infected group) and 156 with lumbar spondylosis who underwent posterior lumbar interbody fusion (the noninfected group) were retrospectively evaluated. Patients with a previous history of spinal fusion, tuberculous spondylitis, and multilevel infection and those receiving dialysis were excluded. Magnetic resonance imaging of the lumbar spine was conducted. Each disc at the L1/2-L5/S levels was graded. The total score of the four discs, excluding the affected disc, was used as the modified disc degenerative disease (DDD) score. Propensity score matching was performed using independent variables such as age, sex, diabetes mellitus, cancer, and steroid use. The modified DDD scores at all and each disc level were compared between the two matched groups.

Results: After matching, 48 patients in the infected group and 88 in the noninfected group were finally included in the study. The mean modified DDD scores of the infected and noninfected groups were 7.63 and 5.40, respectively. The modified DDD scores at all and each disc level were higher in the infected group than in the noninfected group.

Conclusions: The incidence of disc degeneration at all and each disc level was higher in patients with pyogenic spondylitis than in those with noninfectious lumbar spondylosis.

Introduction: This study aimed to investigate the clinical and radiological outcome of "indirect decompression" using lateral-posterior combined surgery for osteoporotic vertebral fracture (OVF) with neurological symptoms.

Methods: A total of 17 patients who underwent lateral and posterior combined indirect decompressive spinal reconstruction (LP-IDR) for single-level OVF with neurological symptoms were included in this study. The neurological symptoms (sensory disturbance and muscle weakness) and imaging findings (local angle and height of the fracture segment and bone fragment occupancy in the spinal canal) were investigated preoperatively, postoperatively, and at the 1-year follow-up.

Results: Muscle weakness was observed preoperatively in ten patients. Nine patients had complete recovery of muscle weakness (p<0.001), whereas one had residual muscle weakness at the 1-year follow-up. The presence of sensory disturbance was observed in 16 patients preoperatively, and it was significantly reduced to 8 patients at the 1-year follow-up (p=0.003). The bony fragment occupancy rate in the spinal canal was decreased from 44.0% to 40.2% postoperatively (p=0.04) and to 33.1% at 1 year (p=0.002). The local angle was corrected from 8.3° to −2.6° postoperatively (p=0.003) and to 1.2° at 1 year. The local height was corrected from 26.7 to 32.0 mm postoperatively (p<0.001) and to 29.8 mm at 1 year.

Conclusions: LP-IDR for OVF with neurological symptoms provided sufficient neurological improvement with expansion of the spinal canal over time.

Introduction: Patients with adult spinal deformity (ASD) lean forward with their trunks when walking, even if they can remain upright during static standing. However, it remains unclear which part of the spinal column is involved in forward trunk tilt and the details of the relationships between sagittal alignment during static standing and changes in dynamic parameters during walking. Therefore, this study aimed to clarify the above by analyzing the walking motion of ASD patients using inertial measurement units (IMUs).

Methods: Preoperative ASD patients were included in this study. Dynamic parameters during gait were measured by IMUs attached on the skin at the T1, T12, and S1 spinous processes, thigh, and lower leg. Walking data were divided into three phases of 10 s each (initial, middle, and final), and the average dynamic parameters at each phase were statistically compared. The relationships between the standing radiographic and dynamic parameters in the final phase were evaluated by linear regression analyses.

Results: A total of 34 patients were included in this study. Their mean age was 72 years. The inclination of IMUs on the T1, T12, and S1 and the flexion angle of T12-S1 IMUs significantly increased over time. Pelvic tilt (PT) of standing radiography was positively correlated with the inclination angles of T12 (r²=0.22, p=0.0048) and S1 (r²=0.16, p=0.0178) and the flexion angle of T12-S1 IMUs (r²=0.29, p=0.0011).

Conclusions: This study showed that anteversion of the trunk in patients with ASD is due to an increase in lumbar forward bending and anterior tilt of the pelvis. Lumbar forward bending was significantly correlated with PT on standing radiography. It is important to consider the presence of poorer posture during gait than during standing when we evaluate patients with high PT.

Introduction: Adult spinal Langerhans cell histiocytosis (LCH) presents a treatment challenge due to ongoing controversies. Traditional approaches such as curettage with bone grafting and internal fixation are preferred for severe cases involving mechanical instability, neurological deficits, or deformity. This study aimed to explore the efficacy of a customized approach involving simple posterior instrumentation without curettage or bone grafting in treating adult spinal LCH.

Methods: This retrospective study analyzed a prospectively maintained database of all spine surgeries conducted at our institute from April 2013 to December 2020. Adult patients (age≥20) diagnosed with LCH were included. We assessed surgical methods, adjuvant therapy, and clinical results, such as perioperative progression of disease, symptoms, and recurrence.

Results: Four male patients aged between 21 and 28, each with a single spinal LCH lesion (T6, T5, and C5) except one case (T5 and T7), were treated. Diagnoses were confirmed via biopsy (two open, two needle biopsies). Whole-body computed tomography or bone scintigraphy revealed no additional LCH lesions in any patient, except in one patient with a small lung nodule. All patients presented with severe back or neck pain and pathological fractures at the affected vertebra. Thoracic LCH cases received percutaneous pedicle screw fixation, while the cervical case was managed with conventional posterior instrumentation using lateral mass screws. After surgery, all patients experienced significant pain relief, halted bone lysis, and rapid new bone formation. One patient underwent chemotherapy postsurgery. Over 3 years of follow-up, imaging studies revealed no recurrences of the disease.

Conclusions: Posterior instrumentation, without the need for curettage or bone grafting, is a promising surgical treatment for adult spinal LCH. This method may effectively halt lesion progression, prevent spinal deformity, and avert neurological deficits in the patients with progressive spine lesion where conservative treatment may not adequately prevent vertebral fractures.

Introduction: Minimally invasive surgical treatment of myelopathy caused by central thoracic disc herniation (TDH) is challenging to carry out because reaching the herniation site is difficult and the thoracic spinal cord is fragile. In this study, using the posterior-lateral approach for central TDH with myelopathy, we present a novel procedure of transcostal microendoscopic discectomy (TCMED).

Technical Note: The patient was operated in a prone position under general anesthesia. At a preoperatively determined distance from the midline, an 18-mm-long longitudinal incision was conducted, and using a 25-degree microendoscope, the operation was carried out. The endoscope was placed at an inward angle of approximately 50 degrees in the vertical direction. The ribs adjacent to the disc were identified, and the disc was exposed by resecting the ribs using a surgical high-speed drill while preserving the cortical bone of the ribs on the pleural side. The herniation was identified by drilling the ventral side of the disc and was then successfully removed. After discectomy, the dura mater expanded ventrally.

Three male patients (mean age, 47.3 years) were treated, with 20 weeks of follow-up on average, 237.7-min mean operative time, and 26.7-mL mean blood loss. The average preoperative modified Japanese Orthopedic Association score was 5.2/11, which improved to 9.5/11 postoperatively, with a 75.6% average recovery rate. The 10-s step test score improved from an average of eight times preoperatively to 20 times postoperatively. No serious perioperative or postoperative complications or residual rib pain were observed.

Conclusions: The proposed TCMED approach for treating central TDH that causes myelopathy allows for safe access to the level of the posterior vertebral wall using the rib as a landmark for resecting the rib head without opening the chest. Using the angled microendoscope and curved surgical instruments, the central TDH, located ventral to the spinal canal, can then be safely and effectively resected without spinal cord retraction.