For middle-aged and elderly patients with conditions such as spinal fractures and degenerative spinal diseases, spinal internal fixation is a core surgical procedure for reconstructing spinal stability, heavily relying on the biomechanical stability provided by pedicle screw systems. Whereas, these patients are often complicated by osteoporosis that can significantly compromise the stability of the bone-pedicle screw interface, leading to a marked increase in pedicle screw loosening and surgical failure rates. The bone cement-augmented pedicle screw technique, which involves injecting bone cement into the vertebral body or screw trajectory to optimize the mechanical properties of the bone-pedicle screw composite, has been proven to significantly enhance fixation strength and effectively prevent screw-related failures, thereby reducing the incidence of internal fixation failure in high-risk populations undergoing spinal fusion. However, the widespread clinical application of this technique has faced challenges such as inaccurate clinical decision-making (indication and contraindication selection), non-standardized operative practices, and insufficient awareness of complication prevention, resulting in considerable variability in clinical outcomes and even severe complications. To address this, Prof. Luo Fei from First Affiliated Hospital of Army Medical University initiated the project and the Chinese Association Orthopaedic Surgeons organized relevant experts to develop the Evidence-based clinical practice guideline for bone cement-augmented pedicle screw technique ( version 2025), based on current evidence. The guidelines put forward 8 recommendations regarding the clinical value, scope of application, and operational standards of the technique, aiming to provide evidence-based medical support and technical standardization for clinical decision-making.

Image 1.

Objective:To evaluate the performance of three mainstream large language models (LLMs) in the review of drug reimbursement indications in hospitals, and to explore their potential in improving audit quality and efficiency, thereby safeguarding the medical insurance fund.Methods:A total of 3 247 outpatient prescription records were retrospectively collected from a specialized oncology hospital between January 2, 2022, and June 30, 2023. Manual assessment of the consistency between clinical diagnoses and drug reimbursement indications was used as the gold standard. Three LLMs, Baidu′s ERNIE Bot, Alibaba′s Tongyi Qianwen, and OpenAI′s ChatGPT-4o, were evaluated on the same task. Performance metrics included accuracy, precision, sensitivity, specificity, F1 score, and area under the curve (AUC).Results:The ERNIE Bot model returned 3 242 valid data, which took 314 min; The Tongyi Qianwen model returned a total of 3 162 valid data, taking 384 min; The ChatGPT-4o model returned a total of 3 218 valid data, taking 150 min. ChatGPT-4o demonstrated the best performance, with an accuracy of 88.41%, precision of 60.48%, sensitivity of 78.75%, specificity of 90.24%, F1 score of 0.68, and an AUC of 0.88.Conclusions:LLMs demonstrate stable performance in determining whether prescriptions align with reimbursement indications, with ChatGPT-4o approaching human-level accuracy and exhibiting more conservative specificity. These findings suggest that LLMs have practical value as auxiliary tools in drug indication reviews, contributing to improved audit efficiency and more refined management of medical insurance funds.

Objective To analyze the performances of electronic portal imaging device (EPID) in mechanical and dosimetry responses,and to investigate the feasibility of adopting EPID as a daily quality assurance (QA) equipment. Methods The EPID of Trilogy linear accelerator was tested for mechanical position repeatability and dose linear response. Then,a group of special fields were edited manually for simulations of field translation and symmetry deviations,and the sensitivities of EPID and MyQA Daily to field translation and symmetry deviations were evaluated,and moreover,the linear correlation between the set value and the measured value was analyzed. Finally,daily QA was conducted with EPID and MyQA Daily for 30 consecutive days,and the measurement results and time consumption were analyzed. Results The maximum deviation of EPID position repeatability was 0.4 mm,and the dose response was linearly correlated (R2>0.999). The symmetry deviation test results showed that both of EPID and MyQA Daily were capable to effectively detect the 0.2％ offset in symmetry,and a linear correlation was demonstrated between the set value and the measured results (R2>0.995). In translation test,the linear correlation of EPID was more explicit than that of MyQA Daily (R2=0.992 vs R2=0.976). In daily QA,all the measured results of EPID and MyQA Daily were within the clinic acceptance tolerance. Moreover,the mean value of Y-directional symmetry and flatness measured by EPID was approximately 1.00％ larger than MyQA Daily. The average time required by EPID was 50 s less than MyQA Daily[(129.97±4.16) s vs (184.53±4.23) s]. Conclusion EPID can be served as a reliable and efficient daily QA equipment for linear accelerator. However,it is not capable to be used as a criterion for evaluating linear accelerator performance.

Objective To analyze the performances of electronic portal imaging device (EPID) in mechanical and dosimetry responses,and to investigate the feasibility of adopting EPID as a daily quality assurance (QA) equipment. Methods The EPID of Trilogy linear accelerator was tested for mechanical position repeatability and dose linear response. Then,a group of special fields were edited manually for simulations of field translation and symmetry deviations,and the sensitivities of EPID and MyQA Daily to field translation and symmetry deviations were evaluated,and moreover,the linear correlation between the set value and the measured value was analyzed. Finally,daily QA was conducted with EPID and MyQA Daily for 30 consecutive days,and the measurement results and time consumption were analyzed. Results The maximum deviation of EPID position repeatability was 0.4 mm,and the dose response was linearly correlated (R2>0.999). The symmetry deviation test results showed that both of EPID and MyQA Daily were capable to effectively detect the 0.2％ offset in symmetry,and a linear correlation was demonstrated between the set value and the measured results (R2>0.995). In translation test,the linear correlation of EPID was more explicit than that of MyQA Daily (R2=0.992 vs R2=0.976). In daily QA,all the measured results of EPID and MyQA Daily were within the clinic acceptance tolerance. Moreover,the mean value of Y-directional symmetry and flatness measured by EPID was approximately 1.00％ larger than MyQA Daily. The average time required by EPID was 50 s less than MyQA Daily[(129.97±4.16) s vs (184.53±4.23) s]. Conclusion EPID can be served as a reliable and efficient daily QA equipment for linear accelerator. However,it is not capable to be used as a criterion for evaluating linear accelerator performance.

For middle-aged and elderly patients with conditions such as spinal fractures and degenerative spinal diseases, spinal internal fixation is a core surgical procedure for reconstructing spinal stability, heavily relying on the biomechanical stability provided by pedicle screw systems. Whereas, these patients are often complicated by osteoporosis that can significantly compromise the stability of the bone-pedicle screw interface, leading to a marked increase in pedicle screw loosening and surgical failure rates. The bone cement-augmented pedicle screw technique, which involves injecting bone cement into the vertebral body or screw trajectory to optimize the mechanical properties of the bone-pedicle screw composite, has been proven to significantly enhance fixation strength and effectively prevent screw-related failures, thereby reducing the incidence of internal fixation failure in high-risk populations undergoing spinal fusion. However, the widespread clinical application of this technique has faced challenges such as inaccurate clinical decision-making (indication and contraindication selection), non-standardized operative practices, and insufficient awareness of complication prevention, resulting in considerable variability in clinical outcomes and even severe complications. To address this, Prof. Luo Fei from First Affiliated Hospital of Army Medical University initiated the project and the Chinese Association Orthopaedic Surgeons organized relevant experts to develop the Evidence-based clinical practice guideline for bone cement-augmented pedicle screw technique ( version 2025), based on current evidence. The guidelines put forward 8 recommendations regarding the clinical value, scope of application, and operational standards of the technique, aiming to provide evidence-based medical support and technical standardization for clinical decision-making.

Objective:To evaluate the performance of three mainstream large language models (LLMs) in the review of drug reimbursement indications in hospitals, and to explore their potential in improving audit quality and efficiency, thereby safeguarding the medical insurance fund.Methods:A total of 3 247 outpatient prescription records were retrospectively collected from a specialized oncology hospital between January 2, 2022, and June 30, 2023. Manual assessment of the consistency between clinical diagnoses and drug reimbursement indications was used as the gold standard. Three LLMs, Baidu′s ERNIE Bot, Alibaba′s Tongyi Qianwen, and OpenAI′s ChatGPT-4o, were evaluated on the same task. Performance metrics included accuracy, precision, sensitivity, specificity, F1 score, and area under the curve (AUC).Results:The ERNIE Bot model returned 3 242 valid data, which took 314 min; The Tongyi Qianwen model returned a total of 3 162 valid data, taking 384 min; The ChatGPT-4o model returned a total of 3 218 valid data, taking 150 min. ChatGPT-4o demonstrated the best performance, with an accuracy of 88.41%, precision of 60.48%, sensitivity of 78.75%, specificity of 90.24%, F1 score of 0.68, and an AUC of 0.88.Conclusions:LLMs demonstrate stable performance in determining whether prescriptions align with reimbursement indications, with ChatGPT-4o approaching human-level accuracy and exhibiting more conservative specificity. These findings suggest that LLMs have practical value as auxiliary tools in drug indication reviews, contributing to improved audit efficiency and more refined management of medical insurance funds.

A case of secondary Burkitt lymphoma of prostate was reported. A 39-year-old patient was admitted to the hospital with no apparent cause for impaired urination with frequent/urgent urination. CT scan of the whole abdomen showed irregular soft tissue mass of the prostate, involving the bladder and rectum, and mild obstructive hydronephrosis of the renal pelvis and ureter. Biopsy of the prostatic mass was performed, and the pathological findings were consistent with Burkitt lymphoma. PET-CT examination showed multiple lesions and lymph node enlargement throughout the body, secondary prostate lymphoma was diagnosed. After diagnosis, the patient was treated with R-DA-EPOCH chemotherapy and followed up for 9 months. Prostate lymphoma is rare and needs to be combined with clinical manifestations and laboratory tests to determine whether it is primary or secondary, but also to distinguish from prostate diseases such as prostatitis and poorly differentiated prostate cancer.

Objective:To investigate the risk factors of pleural effusion after spinal separation surgery for patients with spinal metastatic tumors.Methods:A total of 427 patients with spinal metastatic tumors from January 2014 to January 2022 in the Second Affiliated Hospital of Zhejiang University School of Medicine were retrospectively analyzed. There were 252 males and 175 females, with an average age of 59±12 years (range, 15-87 years). All patients underwent separation surgery. Based on the chest CT within 1 month after surgery, the volume of pleural effusion was measured individually by reconstruction software. Pleural effusion was defined as small volume (0-500 ml), moderate volume (500-1 000 ml), and large volume (above 1 000 ml). Baseline data and perioperative clinical outcomes were compared between the groups, and indicators with statistically significant differences were included in a binary logistic regression analysis to determine the independent risk factors for the development of pleural effusion after isolation of spinal metastatic cancer. Receiver operating characteristic (ROC) curves were conducted to calculate the area under the curve (AUC) for each independent risk factor.Results:All patients successfully completed the operation. Among the 427 patients, there were 35 cases of large pleural effusion, 42 cases of moderate pleural effusion, and 350 cases of small pleural effusion. There were significant differences in tumor size (χ 2=9.485, P=0.013), intraoperative blood loss ( Z=-2.503, P=0.011), blood transfusion ( Z=-2.983, P=0.003), preoperative total protein ( Z=2.681, P=0.007), preoperative albumin ( Z=1.720, P= 0.085), postoperative hemoglobin ( t=2.950, P=0.008), postoperative total protein ( Z=4.192, P<0.001), and postoperative albumin ( t=2.268, P=0.032) in the large pleural effusion group versus the small and moderate pleural effusion group. Logistic regression analysis showed that decreased preoperative albumin ( OR=0.89, P=0.045) and metastases located in the thoracic spine ( OR=4.01, P=0.039) were independent risk factors for the occurrence of large pleural effusion after separation surgery. The ROC curve showed that the AUC and 95% CI for preoperative albumin, lesion location, and the combined model were 0.637 (0.54, 0.74), 0.421 (0.36, 0.48), and 0.883 (0.81, 0.92). The combined predictive model showed good predictive value. Conclusion:The volume of pleural effusion can be measured individually and quantitatively based on chest CT. Decreased preoperative albumin and metastases located in the thoracic spine are independent risk factors for the occurrence of large pleural effusion after separation surgery. The combined prediction of the two factors has better predictive efficacy.

Objective:To investigate the risk factors for 1-year survival rate in patients with spinal metastasis secondary to lung cancer.Methods:The data of 343 patients with spinal metastases secondary to lung cancer from January 2011 to December 2018 were retrospectively studied. There were 188 males (54.8%) and 155 females (45.2%) with an average age of 59.47±10.21 years old (range 23-91 years). The patients were divided into operation group (150 cases, 43.7%) and non operation group (193 cases, 56.3%). The demographics, types of primary tumor, non spinal metastasis, visceral metastasis, spinal metastasis and segments, pathological fractures of vertebra, Frankel classification, physical function status (Karnofsky performance scale, KPS), visual analogue score (VAS), the spinal instability neoplastic score (SINS) were recorded and analyzed. The impact of different treatments on the survival prognosis of patients with spinal metastasis was evaluated. The independent factors affecting survival in those patients were analyzed by Cox proportional hazards regression model.Results:The peak incidence of spinal metastases was found in the age group of 46-60 years (43.7%, 150/343). 38.5% (132/343) of the patients had pathological fractures of the involved vertebral body. 58.3% (200/343) of the patients had extraspinal bone metastasis. 36.2% (124/343) of the patients had visceral metastasis. Among the primary tumors, adenocarcinoma was the most common tpye (61.5%, 211/343), followed by large cell lung cancer (12.5%, 43/343), small cell lung cancer (6.4%, 22/343), squamous cell cancer (6.1%, 21/343) and mixed cell lung cancer (5.3%, 18/343). The type of lung cancer cells in about 8.2 (28/343) patients was unknown. Among the surgical patients, 21 patients underwent minimally invasive surgery (14.0%), 28 patients underwent simple decompression surgery (18.7%), 76 patients underwent separation surgery (50.7%), and 25 patients underwent radical surgery (16.6%). 59.3% (89/150) of the patients had a better neurological function than before surgery. The average survival time of all patients was 9.88 months with the median survival time of 8 (5,14) months. The survival rates were 62.1% (213/343), 30.0% (103/343), and 3.8% (13/343) at 6, 12, and 24 months, respectively. The average survival time of patients in the operation group was 10.24 months with the median survival time of 9 (5, 15) months, and the average survival time of patients in the non operation group was 9.41 months with the median survival time of 7 (5, 13) months with no significant difference between the groups (χ 2=0.300, P=0.584). Multivariate Cox proportional hazard regression model analysis showed that radiotherapy [ HR=1.913, 95% CI(1.471, 2.488), P<0.001], chemotherapy [ HR=1.313, 95% CI(1.040, 1.658), P=0.022], targeted drug therapy [ HR=1.683, 95% CI(1.221, 2.319), P=0.001], KPS [ HR=1.593, 95% CI(1.140, 2.225), P=0.006] and pathological type (non-small cell lung cancer) were independent factors affecting the 1-year survival rate of patients with spinal metastasis secondary to lung cancer [ HR=0.322, 95% CI(0.225, 0.460), P<0.001] with significant difference. Conclusion:Surgical treatment can improve both the neurological function and general status of patients with spinal metastasis. Treatments of radiotherapy, chemotherapy, and targeted drug therapy can significantly improve 1-year survival rate, while a KPS less than 50 points and a primary lung cancer other than adenocarcinoma were independent risk factors reducing 1-year survival rate.