BACKGROUND: Postoperative pulmonary complications (PPCs) are major adverse events in neurosurgical patients. This study aimed to develop and validate machine learning models predicting PPCs after neurosurgery. METHODS: PPCs were defined according to the European Perioperative Clinical Outcome standards as occurring within 7 postoperative days. Data of cases meeting inclusion/exclusion criteria were extracted from the anesthesia information management system to create three datasets: The development (data of Huashan Hospital, Fudan University from 2018 to 2020), temporal validation (data of Huashan Hospital, Fudan University in 2021) and external validation (data of other three hospitals in 2023) datasets. Machine learning models of six algorithms were trained using either 35 retrievable and plausible features or the 11 features selected by Lasso regression. Temporal validation was conducted for all models and the 11-feature models were also externally validated. Independent risk factors were identified and feature importance in top models was analyzed. RESULTS: PPCs occurred in 712 of 7533 (9.5%), 258 of 2824 (9.1%), and 207 of 2300 (9.0%) patients in the development, temporal validation and external validation datasets, respectively. During cross-validation training, all models except Bayes demonstrated good discrimination with an area under the receiver operating characteristic curve (AUC) of 0.840. In temporal validation of full-feature models, deep neural network (DNN) performed the best with an AUC of 0.835 (95% confidence interval [CI]: 0.805-0.858) and a Brier score of 0.069, followed by Logistic regression (LR), random forest and XGBoost. The 11-feature models performed comparable to full-feature models with very close but statistically significantly lower AUCs, with the top models of DNN and LR in temporal and external validations. An 11-feature nomogram was drawn based on the LR algorithm and it outperformed the minimally modified Assess respiratory RIsk in Surgical patients in CATalonia (ARISCAT) and Laparoscopic Surgery Video Educational Guidelines (LAS VEGAS) scores with a higher AUC (LR: 0.824, ARISCAT: 0.672, LAS: 0.663). Independent risk factors based on multivariate LR mostly overlapped with Lasso-selected features, but lacked consistency with the important features using the Shapley additive explanation (SHAP) method of the LR model. CONCLUSIONS: The developed models, especially the DNN model and the nomogram, had good discrimination and calibration, and could be used for predicting PPCs in neurosurgical patients. The establishment of machine learning models and the ascertainment of risk factors might assist clinical decision support for improving surgical outcomes. TRIAL REGISTRATION ChiCTR 2100047474; https://www.chictr.org.cn/showproj.html?proj=128279 .
Adult ; Aged ; Female ; Humans ; Male ; Middle Aged ; Algorithms ; Lung Diseases/etiology* ; Machine Learning ; Neurosurgical Procedures/adverse effects* ; Postoperative Complications/diagnosis* ; Risk Factors ; ROC Curve

Objective:To investigate the relationship between serum insulin-like growth factor-1 (IGF-1) levels and intracranial or extracranial atherosclerosis in patients with cerebral small vessel disease (CSVD).Methods:A total of 407 patients with CSVD admitted to Xiangya Hospital of Central South University between July 2021 and September 2023 were enrolled in the study. Carotid duplex ultrasound was used to measure the internal diameter, intima-media thickness (IMT), vascular wall thickness, plaque property score, stenosis index, and stenosis ratio of the bilateral common carotid arteries, internal carotid arteries, external carotid arteries, and vertebral arteries. Magnetic resonance angiography was used to assess the degree of stenosis in intracranial arteries. Patients were divided into 4 groups based on the serum IGF-1 levels (low level group:≤5.21 ng/ml, medium level group:>5.21 ng/ml and ≤10.73 ng/ml, high level group:>10.73 ng/ml and ≤24.26 ng/ml, extremely high level group:>24.26 ng/ml). The IMT of the common carotid artery, carotid plaques, diameters of various cervical vascular lumens, carotid artery diameter stenosis, and intracranial artery stenosis in 4 groups of the patients were compared. The relationship between IGF-1 and intracranial and extracranial atherosclerosis was analyzed by univariate Logistic regression analysis and multivariate Logistic regression analysis.Results:There were inter group differences among the 4 groups in internal carotid artery diameter [low level group 5.45 (0.50) mm vs medium level group 5.32 (0.55) mm vs high level group 5.30 (0.55) mm vs extremely high level group 5.30 (0.50) mm; H=8.210, P=0.042]. The carotid IMT [low level group 0.80 (0.05) mm vs medium level group 0.80 (0.05) mm vs high level group 0.83 (0.03) mm vs extremely high level group 0.83 (0.09) mm; H=8.107, P=0.044], the proportion of carotid artery vascular wall thickening [low level group 52.9%(54/102) vs medium level group 48.0%(49/102) vs high level group 68.3%(69/101) vs extremely high level group 60.8%(62/102); χ2=9.889, P=0.020], the carotid artery plaque property score [low level group 1 (2) vs medium level group 2 (2) vs high level group 2 (2) vs extremely high level group 2 (2); H=8.913, P=0.030] and the proportion of anterior cerebral artery stenosis [low level group 2.9%(3/102) vs medium level group 2.0%(2/102) vs high level group 4.0%(4/101) vs extremely high level group 10.8%(11/102); χ2=10.473, P=0.014] had inter group differences among the 4 groups, and the differences were statistically significant. Univariate Logistic regression analysis indicated that carotid artery vascular wall thickening ( OR=1.197, 95% CI 1.003-1.429, P=0.046), anterior cerebral artery stenosis ( OR=1.814, 95% CI 1.148-2.867, P=0.011), and basilar artery stenosis ( OR=1.530, 95% CI 1.084-2.159, P=0.015) were correlated with IGF-1 levels. Multivariate Logistic regression analysis revealed that after adjusting for age, gender, low-density lipoprotein cholesterol (LDL-C), and C-reactive protein, IGF-1 was positively correlated with the carotid artery vascular wall thickening ( OR=1.311, 95% CI 1.014-1.696, P=0.039); after adjusting for age, IGF-1 was positively correlated with the anterior cerebral artery stenosis ( OR=2.130, 95% CI 1.201-3.776, P=0.010); after adjusting for gender, low-density lipoprotein cholesterol, and cholesterol levels, IGF-1 was positively correlated with basilar artery stenosis ( OR=1.688, 95% CI 1.063-2.681, P=0.027). Conclusions:There is an association between IGF-1 levels and intracranial and extracranial atherosclerosis in patients with CSVD. IGF-1 may play a role in the development and progression of atherosclerosis in CSVD.

Objective:To investigate the clinical significance of serum galectin-9 (Gal-9) in patients with dermatomyositis (DM) or clinically amyopathic dermatomyositis (CADM) .Methods:This cross-sectional study included 105 newly diagnosed patients with DM or CADM who were admitted to the Department of Dermatology in Ruijin Hospital, Shanghai Jiao Tong University School of Medicine, from January 2015 to October 2024, among whom 53 had cancer-associated DM/CADM (CRDM). Additionally, an age-matched control group was included, consisting of 30 newly diagnosed cancer patients without autoimmune diseases, 27 systemic lupus erythematosus (SLE) patients, and 31 healthy controls. Serum levels of Gal-9 and transcriptional intermediary factor 1-gamma (TIF1-γ) were measured using enzyme-linked immunosorbent assay. The relationship between Gal-9 levels and laboratory indicators of DM disease activity was analyzed. Comparisons between different groups were performed using the t-test, Mann-Whitney U test, and chi-square test. Spearman correlation analysis was used to assess the association between Gal-9 levels and laboratory indicators. The diagnostic efficacy of Gal-9 and TIF1-γ for CRDM was evaluated using receiver operating characteristic (ROC) curve analysis. Results:Among the 105 DM/CADM patients, 35 were male (33.3%) and 70 were female (66.7%), with a mean age of 53.2 ± 15.1 years. In the 53 CRDM patients, the incidence rates of V-neck sign, dyschromia, and dysphagia were higher than those in non-CRDM patients (all P > 0.05). Serum Gal-9 levels in DM/CADM patients (21.2 [12.2, 32.3] ng/ml) were significantly higher than those in healthy controls (6.8 [5.4, 7.9] ng/ml, P < 0.001), SLE patients (12.3 [8.1, 15.5] ng/ml, P = 0.011), and cancer patients without autoimmune diseases (7.5 [4.9, 8.5] ng/ml, P < 0.001). Gal-9 levels were positively correlated with serum TIF1-γ antibody levels ( rs = 0.21, P = 0.029), serum ferritin ( rs = 0.29, P = 0.003), lactate dehydrogenase ( rs = 0.44, P < 0.001), creatine kinase ( rs = 0.28, P = 0.004), aspartate aminotransferase ( rs = 0.42, P < 0.001), C-reactive protein ( rs = 0.34, P < 0.001), and erythrocyte sedimentation rate ( rs = 0.46, P < 0.001). Among CRDM patients, those who had not received cancer treatment had higher Gal-9 levels (30.1 [23.3, 38.3] ng/ml) than those in stable condition after cancer treatment (13.5 [10.5, 27.9] ng/ml, P = 0.007). The area under the ROC curve (AUC) for serum TIF1-γ in diagnosing CRDM was 0.718, with an optimal cutoff value of 23.02 U/ml. The AUC for serum Gal-9 was 0.719, with an optimal cutoff value of 55.02 ng/ml. When combining both markers, the AUC increased to 0.783, with a sensitivity of 0.85 and specificity of 0.74. Conclusions:Gal-9 was highly expressed in serum of DM/CADM patients, particularly in CRDM patients. Dynamic monitoring of Gal-9 in CRDM patients may be helpful to monitor the therapeutic effect of malignancies.

Objective:To investigate the curative effects of the ulnar artery perforator flap carrying 2 and more perforators of same source artery on reconstruction of defects in hand.Methods:A retrospective observational study was conducted on 13 patients with hand injuries combined with bone or tendon exposure and met the inclusion criteria of the study were admitted to the Department of Hand Surgery, Wuxi NO.9 People’s Hospital Affiliated to Soochow University between April 2022 and September 2023. The patients were 8 males and 5 females with a mean age of 45.7 (25-67) years. Seven hand injuries were caused by mechanical crushing, 2 by hot crushing and 4 by machine strangulation. After debridement, the sizes of defect were found at 5.4 cm×5.1 cm to 8.2 cm×7.5 cm. Dopplor ultrasound was applied to locate the perforators before surgery. Perforator flaps carrying cutaneous branches of ulnar artery proximal to the wrist were designed for reconstruction of the defects in hand. The flaps were 6.0 cm×5.5 cm-8.5 cm×8.0 cm in size. Donor sites were covered by the ulnar artery perforator flaps sized 5.3 cm×2.7 cm-8.2 cm×4.0 cm and carryied 2 and more perforators of the same source artery. During harvest of the flaps, the number and calibre of perforators carried by flap, the calibre of the main trunk of the superior carpal branch of the ulnar artery and the length of its pedicle were recorded. A total of 35 perforators with 0.35-0.95 mm in calibre carried by 13 perforator flaps of ulnar artery were successfully retained. Five of the 13 flaps carried the perforators with a calibre smaller than 0.50 mm. Overall, there were 7 flaps with 2 perforators per flap, 4 with 3 perforators per flap, 1 with 4 perforators and 1 with 5 perforators. The cutaneous branch of ulnar artery in a caliber of 0.75-1.35 mm and proximal to the wrist was dissected as the vascular pedicle with 30.0-45.0 mm in length. All patients were included in the schedueled postoperative at outpatient clinic to observe the appearance and sensation of the flaps, and the occurrence of complication.Results:All flaps survived and the wounds healed at first intention without vascular compromises, wound dehiscence or obvious swelling. All patients received 8 - 20 months of follow-up, with 15 months in average. Flaps presented good appearance and colour, with soft texture and without bloating. TPD of the first flap was 8-18 mm, with an average of 10.5 mm±0.5 mm and that of the second or relay flap was 7-15 mm, with an average of 9.5 mm±0.5 mm. According to British Medical Research Council (BMRC) system, the sensory evaluation of the recipient sites achieved S 4 in 5 flaps, S 3 in 9 flaps and S 2 in 12 flaps. Conclusion:The ulnar artery perforator flap with 2 and more perforators of the same source artery has sufficient and reliable blood supply and is effective in reconstruction of hand defects.

Calcitonin-negative medullary thyroid carcinoma(CNMTC)is a rare subtype of medullary thyroid carcinoma,characterized by normal serum calcitonin levels,which often leads to misdiagnosis or missed diagnosis.The pathogenesis of CNMTC remains unclear and may involve impaired secretion mechanisms or assay-related false negatives.Diagnostic approaches include ultrasound-guided fine needle aspiration cytology,serum CEA and ProGRP measurements,and RET gene testing.Surgical resection remains the mainstay of treatment,while neoadjuvant therapy may be considered in selected cases.This review summarizes recent advances in the understanding,diagnosis,treatment,and prognosis of CNMTC,aiming to provide clinical guidance for better management of this challenging condition.

This study aims to investigate the effect of Lactiplantibacillus plantarum LP12 on obe-sity prevention.In our study,Lactiplantibacillus plantarum LP12 was added to the diet for feed-ing,and the blood biochemistry status of rabbit,as well as the antioxidant effect of serum and liver samples were analyzed by determining the body weight change and feed intake of Japanese White rabbits.The changes in colony structure and abundance were also analyzed by 16S rDNA sequen-cing.The results showed that supplementation of Lactiplantibacillus plantarum LP12 inhibits weight gain,decreases serum glucose and ALT levels,and increases SOD activity in the liver.16S RNA gene sequencing analysis showed that the addition of Lactiplantibacillus plantarum LP12 increases the abundance of Bacteroidetes and Desulfovibrioides at the phylum level,and the supple-mentation of Lactiplantibacillus plantarum LP12 increases the abundance of Muribaculaceae at the genus level.Predictive analysis of microbiota function revealed that the supplementation of Lactiplantibacillus plantarum LP12 positively regulated iron-sulfur clusters and Zn-dependent proteases.In conclusion,the addition of Lactiplantibacillus plantarum effectively inhibits weight gain in Japanese White rabbits,enhances the antioxidative activity of the liver,and induces altera-tions in the gut microbiota composition of these rabbits.These findings lay an experimental foun-dation for further exploring the mechanisms by which Lactobacillus plantarum LP12 exerts its preventive effects against obesity and promotes metabolic health.

Objective:To investigate the clinical predictive value of Ki67 proliferation index combined with preoperative serum Ctn for postoperative biochemical cure of medullary thyroid carcinoma (MTC) .Methods:Clinical data were collected from Dec. 2008 to Dec. 2024 from 90 patients with surgically confirmed MTC at China-Japan Union Hospital of Jilin University. The optimal cut-off value for preoperative Ctn prediction of biochemical cure (171.18pg/mL) was determined by the ROC curve; the Ki67 proliferation index cut-off value was adopted from the international MTC grading system standard (5%). Patients were divided into three groups based on the above cutoff values: double-low group (Ki67 <5% and Ctn <171.18pg/mL, n=23), single-high group (Ki67 ≥5% and Ctn <171.18pg/mL or Ki67 <5% and Ctn ≥171.18pg/mL, n=49), and double-high group (Ki67 ≥5% and Ctn ≥171.18pg/mL, n=18). The Kaplan-Meier method (Log-Rank and Trend test) was used to compare the differences in biochemical cure rates between groups, and the Cox proportional risk model was used to analyze the risk factors affecting biochemical cure. Results:The correlation between preoperative Ctn and Ki67 proliferation index was not significant. The three groups differed significantly in gender, tumor distribution, tumor size, vascular invasion, N stage, TNM stage, and biochemical cure ( P<0.05), with the double-high group being significantly associated with larger tumors, later N stage and TNM stage, and lower biochemical cure ( P<0.001). Kaplan-Meier analysis showed that the biochemical cure rate in the double-high, single-high, and double-low groups showed a stepwise improvement.Cox univariate analysis showed that tumor size, N stage, TNM stage, preoperative Ctn, and Ki67 combined with Ctn were risk factors for failure to biochemically cure; multivariate analysis confirmed that the double-high group was an independent risk factor ( P<0.05). In the single-high group, the biochemical cure rate of patients in the low Ki67-high Ctn group was lower than that of the high Ki67-low Ctn group and more malignant. Ki67 had less effect on biochemical cure and disease-free survival at the low Ctn level, and Ki67 was an independent risk factor for failure to biochemically cure at the high Ctn level ( P=0.023) and was significantly associated with disease-free survival ( P=0.004) . Conclusions:Serum Ctn is more sensitive than Ki67 index in predicting biochemical cure after MTC, and the correlation between the two was weak. Ki67 proliferation index alone has limited prognostic value, but combines with preoperative Ctn significantly optimize the prognostic assessment of patients.The role of Ki67 index varied at different Ctn levels.

The synthetic PDCoV N protein gene was optimized and cloned into the pET-30a vector to obtain the pET-30a-N plasmid.Thenthe recombinant plasmid was transformed into three strains of BL21 E.coli using heat-shock to explore protein expression conditions.The expressed proteins was purified using Ni Focurose 6FF(IMAC)and used as antigen to immunize the New Zealand White rabbit to prepare the polyclonal antibody against the PDCoV N protein.The antibody titer was measured by indirect ELISA method.The specificity for the antibody was identified by West-ern blot and indirect immunofluorescence(IFA).The results showed that the pET-30a-N plasmid showed high expression level in BL21 StarTM(DE 3).The optimal expression condition was 37 ℃ 4 h.The purity of the target protein could reach 90.3％after purification.Indirect ELISA showed that the antibody titers was up to 1∶204 800.Western blot and IFA showed that the produced rabbit polyclonal antibody exhibited good specificity.In conclusion,the polyclonal antibody was prepared which specifically recognized the PDCoV N proteins.The results provided some references for the subsequent exploration of PDCoV N protein function and laid a foundation for establishing a diag-nostic method for PDCoV.

Objective:To investigate the clinical predictive value of Ki67 proliferation index combined with preoperative serum Ctn for postoperative biochemical cure of medullary thyroid carcinoma (MTC) .Methods:Clinical data were collected from Dec. 2008 to Dec. 2024 from 90 patients with surgically confirmed MTC at China-Japan Union Hospital of Jilin University. The optimal cut-off value for preoperative Ctn prediction of biochemical cure (171.18pg/mL) was determined by the ROC curve; the Ki67 proliferation index cut-off value was adopted from the international MTC grading system standard (5%). Patients were divided into three groups based on the above cutoff values: double-low group (Ki67 <5% and Ctn <171.18pg/mL, n=23), single-high group (Ki67 ≥5% and Ctn <171.18pg/mL or Ki67 <5% and Ctn ≥171.18pg/mL, n=49), and double-high group (Ki67 ≥5% and Ctn ≥171.18pg/mL, n=18). The Kaplan-Meier method (Log-Rank and Trend test) was used to compare the differences in biochemical cure rates between groups, and the Cox proportional risk model was used to analyze the risk factors affecting biochemical cure. Results:The correlation between preoperative Ctn and Ki67 proliferation index was not significant. The three groups differed significantly in gender, tumor distribution, tumor size, vascular invasion, N stage, TNM stage, and biochemical cure ( P<0.05), with the double-high group being significantly associated with larger tumors, later N stage and TNM stage, and lower biochemical cure ( P<0.001). Kaplan-Meier analysis showed that the biochemical cure rate in the double-high, single-high, and double-low groups showed a stepwise improvement.Cox univariate analysis showed that tumor size, N stage, TNM stage, preoperative Ctn, and Ki67 combined with Ctn were risk factors for failure to biochemically cure; multivariate analysis confirmed that the double-high group was an independent risk factor ( P<0.05). In the single-high group, the biochemical cure rate of patients in the low Ki67-high Ctn group was lower than that of the high Ki67-low Ctn group and more malignant. Ki67 had less effect on biochemical cure and disease-free survival at the low Ctn level, and Ki67 was an independent risk factor for failure to biochemically cure at the high Ctn level ( P=0.023) and was significantly associated with disease-free survival ( P=0.004) . Conclusions:Serum Ctn is more sensitive than Ki67 index in predicting biochemical cure after MTC, and the correlation between the two was weak. Ki67 proliferation index alone has limited prognostic value, but combines with preoperative Ctn significantly optimize the prognostic assessment of patients.The role of Ki67 index varied at different Ctn levels.