ABSTRACT
Objective:To evaluate the value of preoperative prediction of vessel invasion (VI) of locally advanced gastric cancer by machine learning model based on the venous phase enhanced CT radiomics features.Methods:A retrospective analysis of 296 patients with locally advanced gastric cancer confirmed by pathology in the First Affiliated Hospital of Zhengzhou University from July 2011 to December 2020 was performed. The patients were divided into VI positive group ( n=213) and VI negative group ( n=83) based on pathological results. The data were divided into training set ( n=207) and test set ( n=89) according to the ratio of 7∶3 with stratification sampling. The clinical characteristics of patients were recorded, and the independent risk factors of gastric cancer VI were screened by multivariate logistic regression. Pyradiomics software was used to extract radiomic features from the venous phase enhanced CT images, and the minimum absolute shrinkage and selection algorithm (LASSO) was used to screen the features, obtain the optimal feature subset, and establish the radiomics signature. Four machine learning algorithms, including extreme gradient boosting (XGBoost), logistic, naive Bayes (GNB), and support vector machine (SVM) models, were used to build prediction models for the radiomics signature and the screened clinical independent risk factors. The efficacy of the model in predicting gastric cancer VI was evaluated by the receiver operating characteristic curve. Results:The degree of differentiation (OR=13.651, 95%CI 7.265-25.650, P=0.003), Lauren′s classification (OR=1.349, 95%CI 1.011-1.799, P=0.042) and CA199 (OR=1.796, 95%CI 1.406-2.186, P=0.044) were independent risk factors for predicting the VI of locally advanced gastric cancer. Based on the venous phase enhanced CT images, 864 quantitative features were extracted, and 18 best constructed radiomics signature were selected by LASSO. In the training set, the area under the curve (AUC) of XGBoost, logistic, GNB and SVM models for predicting gastric cancer VI were 0.914 (95%CI 0.875-0.953), 0.897 (95%CI 0.853-0.940), 0.880 (95%CI 0.832-0.928) and 0.814 (95%CI 0.755-0.873), respectively, and in the test set were 0.870 (95%CI 0.769-0.971), 0.877 (95%CI 0.788-0.964), 0.859 (95%CI 0.755-0.961) and 0.773 (95%CI 0.647-0.898). The logistic model had the largest AUC in the test set. Conclusions:The machine learning model based on the venous phase enhanced CT radiomics features has high efficacy in predicting the VI of locally advanced gastric cancer before the operation, and the logistic model demonstrates the best diagnostic efficacy.
ABSTRACT
Objective:To compare the effect of O-arm assisted and free-hand pedicle screw placement in the treatment of AO type C thoracolumbar fracture.Methods:A retrospective cohort study was used to analyze the clinical data of 34 patients with type C thoracolumbar fracture admitted to Henan Provincial People′s Hospital from January 2018 to June 2021, including 23 males and 11 females; aged 42-63 years [(50.4±7.4)years]. The fracture was located at T 11 in 4 patients, T 12 in 10, L 1 in 12, L 2 in 6, T 11~12/L 1 in 1 and T 12/L 1 in 1. Posterior reduction and internal fixation was carried out for all patients, of whom 18 were treated with O-arm assisted pedicle screw placement (navigation group) and 16 with free-hand pedicle screw placement (free-hand group). The operation time, single screw placement time, intraoperative bleeding volume, operation mode and screw placement accuracy were compared between the two groups. The kyphotic Cobb angle, visual analogue score (VAS) and American Spinal Injury Association (ASIA) score were compared between the two groups before operation, at 1 week after operation, at 3 months after operation and at the last follow-up. Postoperative complications were observed. Results:All patients were followed up for 12-29 months [(16.8±6.1)months]. There was no significant difference between the two groups in the operation time, intraoperative bleeding volume and operation mode (all P>0.05). The single screw placement time was (9.4±1.6)minutes in navigation group, but was (10.8±1.5)minutes in free-hand group ( P<0.05). The screw placement accuracy was 97.4% in navigation group, but was 81.5% in free-hand group ( P<0.01). The kyphotic Cobb angle and VAS had no significant differences between the two groups before operation (all P>0.05). The kyphotic Cobb angle in navigation group and free-hand group was (4.3±1.1)° and (5.9±1.1)° at 1 week after operation, (4.4±1.2)° and (5.7±1.3)° at 3 months after operation, and (4.4±1.2)° and (6.8±0.9)° at the last follow-up, decreased significantly from that before operation [(21.8±3.1)°, (22.2±3.2)°] (all P<0.01). The kyphotic Cobb angle in navigation group was significantly lower than that in free-hand group at 1 week, 3 months after operation and at the last follow-up (all P<0.01). The VAS in navigation group and free-hand group was (3.2±0.7)points and (4.1±0.7)points at 1 week after operation, (2.4±0.6)points and (3.0±0.8)points at 3 months after operation, and (1.8±0.9)points and (2.6±0.7)points at the last follow-up, decreased significantly from that before operation [(8.4±0.8)points, (8.3±0.9)points] (all P<0.01). The VAS in navigation group was significantly lower than that in free-hand group at 1 week, 3 months after operation and at the last follow-up (all P<0.01). The ASIA score showed no significant difference within and between the two groups before operation, at 1 week, 3 months after operation and at the last follow-up (all P>0.05). Postoperative incision infection occurred in 1 patient in both groups ( P>0.05). Implant failure such as loosening or displacement was not observed in navigation group, and only occurred in 2 patients in free-hand group ( P>0.05). Conclusion:Compared with free-hand pedicle screw placement, O-arm assisted pedicle screw placement in the treatment of AO type C thoracolumbar fracture has advantages of rapid and accurate screw placement, good reduction and notable pain relief.
ABSTRACT
Objective:To evaluate the efficacy of posterior subtotal vertebrectomy in the treatment of thoracolumbar vertebral refractures after vertebroplasty.Methods:A retrospective analysis was conducted in the 28 patients with refracture after percutaneous vertebroplasty(PVP) or percutaneous kyphoplasty(PKP) who had been treated at Department of Spinal Surgery, Henan Provincial People's Hospital from June 2017 to October 2019. They were 7 males and 21 females, with an average age of 62.4 years(from 61 to 76 years). A total of 28 vertebrae were involved, including 5 T11s, 9 T12s, 11 L1s and 3 L2s. Their previous operations were PKP in 17 cases and PVP in 11. After the spinous process, vertebral plate, articular process and transverse process were resected by posterior approach, the vertebral body, bone cement and upper and lower intervertebral discs were partially resected by trans-vertebral lateral approach. At the same time, nerve decompression was performed. Finally, the inter-vertebral support was fixated followed by the posterior screw-rod orthopedic fixation. The operation time and intraoperative bleeding volume were recorded. The cobb angles of kyphosis were compared on the X-ray films of the whole spine between preoperation and the last follow-up to evaluate correction. Functional improvement of the spine was evaluated by comparison of the visual analogue scale (VAS) and JOA(Japanese Orthopedics Association) scores between preoperation and the last follow-up.Results:The operation time averaged 182.1 min and intraoperative bleeding volume 996.2 mL. All the 28 patients were followed up for 8 to 29 months (mean, 19.8 months). No obvious neurological lesions or other serious complications were observed. The cobb angle was improved from preoperative 41.3°±10.3° to 6.4°±2.5° at the last follow-up, the VAS score from preoperative 7.3±1.8 to 2.5±1.0 at the last follow-up, and the JOA score from preoperative 8.4±2.3 to 21.3±2.5 at the last follow-up, showing a significant difference in all the comparisons ( P<0.05). Conclusion:The posterior subtotal vertebrectomy is effective for thoracolumbar vertebral refractures after vertebroplasty because it can remove bone cement, decompress the spinal canal, fuse the inter-vertebral graft and reconstruct the spinal stability in one stage.
ABSTRACT
Objective To investigate the clinical efficacy of non-fusion and fusion internal fixation with posterior pedicle screw for odontoid fractures. Methods A retrospective case control study was conducted to analyze the clinical data of 46 patients with odontoid fractures admitted to Henan Provincial People's Hospital from January 2013 to June 2015. There were 35 males and 11 females, aged 19-62 years [(36. 9 ± 1. 6)years]. The preoperative odontoid fractures were classified as type IIA in five patients, type IIB in eight, type IIC in 22, shallow type III in six according to Anderson and D'Alonzo typing. The fractures of five patients were not classified. According to the ASIA grading, the preoperative spinal cord function was graded as D or E in the 46 patients. Among the patients, 15 underwent posterior pedicle screw non-fusion internal fixation ( Group A) , with 60 pedicle screws removed 12-29 months after operation. A total of 31 patients were treated with fusion internal fixation with posterior pedicle screw ( Group B) , with 124 pedicle screws free from removal after operation. The follow-up time, operation time, intraoperative fluoroscopy time, intraoperative blood loss as well as visual analog scale (VAS), cervical rotation degree, and neck disability index ( NDI) before operation, 1 year after surgery ( when no internal fixation was removed ) and at the last follow-up were compared between the two groups. Results There were no significant differences between the two groups in the follow-up time, operation time, intraoperative fluoroscopy time, intraoperative blood loss, VAS score, neck rotation degree, NDI score before operation and at 1 year after operation (P >0. 05). There were significant differences between two groups in VAS scores, neck rotation degree, NDI scores indicating driving and walking and recreational activities at the last follow-up ( P <0. 05 ) . There were no significant differences between two groups in other NDI scores items, including neck discomfort, personal care, lifting heavy objects, reading & watching TV, headache situation, attention, working status, and sleep disorder at the last follow-up ( P> 0. 05 ) . Conclusions Posterior pedicle screw internal fixation for odontoid process fracture has good curative effect. Non-fusion internal fixation with posterior pedicle screw is more conducive to the recovery of cervical spine rotation function, the relief of neck and back pain, and the improvement of patients' driving, walking and entertainment activities.
ABSTRACT
Objective To develop a new type of lumbar prosthesis,movable artificial lumbar vertebral body (MALV) and evaluate the stability and activity of the prosthesis by in vitro biomechanical test.Methods Fifteen adult fresh lumbar spine specimens were randomly divided into three groups:intact group,fusion group and non-fusion group.The anatomical parameters of L2 vertebral body and adjacent intervertebral discs were collected by software (Mimics 16.0,Materialise Company,USA) and the new prosthesis was personalized by 3D printing and mechanical processing.All specimens in fusion group and in non-fusion group were implanted with titanium cage,titanium plate and MALV respectively after partial resection of L2 vertebra and adjacent intervertebral disc.No other intervention was conducted on physiological specimens.All specimens were examined by imaging to determine the position of the prosthesis and the condition of the spinal cord.Specimens in all groups were subjected to a 7.5 N · m load biomechanical test in flexion,extension,left and right lateral bending,left and right rotation aiming to evaluate the mobility of operative site and adjacent discs.Results The new prosthesis consists of three parts,vertebral body,intervertebral disc and composite material ball.All 15 cadavers were free of spinal deformity,fracture and osteoporosis.Postoperative radiographic examination indicated no prosthesis displacement and spinal compression without adverse phenomena.The results in vitro mechanical test were as followed.There was no statistical difference in ROM of the T12L1 intervertebral in non-fusion group (3.68°±0.86°,3.52°± 0.86°,2.64°±0.93°,2.58°±0.76° respectively) and in intact group (3.98°±0.90°,3.74°±0.91°,2.94°±1.10°,2.96°±0.86° respectively) in flexion,extension,left and right lateral bending (P＞0.05).The ROM of the T12L1 intervertebral in fusion group (5.28°±0.83°,5.32°±0.42°,4.72°±1.10°,4.94°±1.17° respectively) was significantly larger than that in intact group and non-fusion group (P＜0.05).There was no statistically significant difference between the ROM of L1.2 and L2.3 intervertebral in flexion,extension and left and rightlateral bending in the non-fusion group (ROM of L1.2 were 4.62°±0.51 °,4.34°±0.67°,3.16°±0.55°,3.28°±0.41 ° respectively;ROM of L2.3 were 4.54°±0.58°,4.36°±0.65°,3.26°±0.21°,3.42°±0.37° respectively) and the intact group (ROM of L1.2 were 4.10°±0.53°,3.72°±0.42°,2.74°±0.50°,3.04°±0.40° respectively;ROM of L2.3 were 4.26°±0.72°,4.08°±0.77°,3.00°±0.36°,3.20°±0.31° respectively) (P＞0.05).The ROM of L1.2 and L2.3 in intact group and in non-fusion group were significantly greater than those in the fusion group (ROM of L1.2 were 1.10°± 0.35°,1.28°±0.31°,0.84°±0.34°,0.80°± 0.43° respectively;ROM of L2.3 were 1.14°±0.30°,1.18°±0.28°,0.94°± 0.36°,1.06°±0.32°) (P＜0.05).There was no statistically significant difference in the ROM of the T12-L3 full segment in the left and right rotation of the three groups (Intact group:4.47°± 0.99°,4.40°±0.70°;Non-fusion group:4.60°±0.90°,4.50°±0.77°;Fusion group:3.85°±0.50°,3.72°±0.70°)(P＞0.05).Conclusion The MALV can retain the mobility of the operative site in flexion,extension,lateral bending and rotation.Furthermore,it can effectively avoid the increased mobility of adjacent segment intervertebral.After implantation,the prosthesis can obtain satisfied immediate stability.
ABSTRACT
Objective To compare the effects of movable artificial lumbar vertebra implantation and traditional vertebral excision and fusion on the stress of adjacent intervertebral discs,so as to provide reference for the biomechanical safety of new prosthesis implantation.Methods The total lumbar vertebra CT scan data of a healthy adult were imported into the compute r-aided software Mimics 16.0 for three dimensional reconstruction.The reconstructed three dimensional model was smoothed using computer-aided software Geomagic Studio 12.0.Subsequently,meshing,parameter setting and ligaments reconstructing were completed using computer-aided software Hypermesh 13.0 and Solidworks 2013,successfully constructing the finite element model of lumbar vertebral physiology group.On the basis of the finite element model of physiological group,three vertebral bodies and two intervertebral spaces were fused and fixed,and a simplified finite element model of fusion group was constructed.The partitioned prosthesis model was inserted into the established finite element model of physiological group,replacing the L 3 vertebral body and the adjacent two intervertebral discs in the model,constructing the finite element model of non fusion group.Three finite element models were imported into computer-aided software Abaqus/Explict and loaded under six motion modes including anteflexion,dorsiflexion,left and right rotation,and left and right bending.The stress changes of adjacent intervertebral discs were calculated.Results The maximum Mises stress of the adjacent intervertebral disc in the direction of forward flexure,dorsal extension,left and right rotation and left and right lateral bending of the three finite element models was located at the site of loading and increased with the increase of loading.Under the maximum loading,the Mises stress of adjacent intervertebral discs in the above directions in the fusion group (L1-2 respectively were 0.79,0.96,1.26,1.92,1.34,1.57 MPa while L4-5 respectively were 0.52,1.13,1.50,1.74,0.94,0.87 MPa) was significantly higher than that in the physiological group (L1-2 respectively were 0.42,0.53,0.57,0.66,0.64,0.72 MPa while L4-5 respectively were 0.23,0.29,0.68,0.63,0.37,0.34 MPa).The Mises stress of adjacent intervertebral disc in the non-fusion group (L1-2 respectively were 0.38,0.57,0.75,1.02,0.87,0.90 MPa while L4-5 respectively were 0.18,0.26,0.81,0.98,0.30,0.27 MPa) was similar to that in the physiological group,although there was some difference.Conclusion Movable artificial lumbar disc prosthesis implantation can better avoid the stress increase of adjacent intervertebral discs,and its long-term implantation in the human body is expected to reduce the incidence of degeneration of adjacent intervertebral discs.