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

Hypertensive intracerebral hemorrhage is in the critical condition. Surgical treatment can promptly remove cranial hematoma, reduce the compression to the intracranial nerve, and improve the patient's neurological function and prognosis. At present, there are many operating modes, from the traditional large bone flap craniotomy to remove hematoma, to minimally invasive surgery. Each has its own advantages. This paper reviews various minimally invasive hematoma removal procedures and clinical nursing care based on traditional surgical treatment, analyzes the advantages and disadvantages of surgical treatment for patients with hypertensive cerebral hemorrhage, selects appropriate surgical methods and formulates reasonable surgical strategies.

Objective: To know the detection rate of hereditary thrombocytopenia (HT) in children with chronic thrombocytopenia and its clinical and laboratory characteristics for an early clinical identification and diagnosis of HT in future. Methods: Data of the children with thrombocytopenia, who had been treated in Beijing Children′s Hospital from April 2016 to May 2018 and whose present history lasted for more than 1 year and had poor response to immunotherapy were retrospectively collected.HT was screened in these patients by adopting next generation sequencing (NGS). Finally, clinical and laboratory characteristics of these children with HT were summarized and analyzed. Results: A total of 161 children with chronic thrombocytopenia were included.Forty-three cases (26.7%) were found to have gene mutations.The genetic rules of the mutant gene, the family verification and the clinical manifestations of the proband and some related laboratory tests were analyzed and 24 cases (14.9%) can be diagnosed as HT.Among the HT patients, the proportion of males and females was 159, and the median onset of age was 0.58 years, which was significantly lower than that of non-HT cases (the median onset of age was 4.36 years), and the difference was statistically significant (P<0.001); the proportion of mucosal hemorrhage and visceral hemorrhage (31.8% and 13.7%) of HT was significantly higher than non-HT cases (15.3% and 0.6%), and the difference was statistically significant (P<0.001). Fifty percent of (12/24 cases) cases of HT had positive family history; according to the ave-rage platelet volume and platelet morphology in peripheral blood smear, HT could be divided into small platelet HT, po-sitive platelet HT and large platelet HT.Some cases had well response to immunotherapy but seemed easy to relapse du-ring the withdrawal period, while the others responded poorly to therapy.Different clinical manifestations of HT suggest different pathogenesis, which can be divided into the related types of megakaryocyte differentiation defect, megakaryocyte maturation defect, platelet release defect and platelet survival time shortening. Conclusions The pathogenesis and cli-nical phenotype of HT was different.Some of them were effective for immunotherapy, which were easily confused with immune thrombocy-topenla(ITP). It is clinically necessary to perform NGS in children with thrombocytopenia at early onset, abnormal platelet morphology, prolonged disease course or severe mucosal/visceral hemorrhage, in order to recognize HT timely to avoid delay in diagnosis and poor prognosis.

Objective To understand the characteristics of distribution on HIV-1 subtypes and the transmission clusters in Yiwu in Zhejiang province.Methods A cross-sectional study of molecular epidemiology was carried out on newly reported H1V/AIDS cases in Yiwu.RNA was extracted from 168 plasma samples,followed by RT-PCR and nest-PCR for pol gene amplification,sequencing,phylogenetic tree construction used for analyzing the subtypes and transmission clusters.Mutations on drug resistance was analyzed by CPR 6.0 online tool.Results Subjects were mainly males (86.3％,145/168),with average age as (39.1 ± 13.4) years old and most of them were migrants (66.7％,112/168).The major routes of transmission included homosexual (51.2％,86/168) and heterosexual (48.8％,82/168) contacts.The rate of success for sequence acquisition was 89.9％ (151/168).The dominant subtypes showed as CRF01_AE (74,49.0％) and CRF07_BC (64,42.4％),followed by CRF08_BC (5,3.3％),CRF55_01B (3,2.0％),each case of subtype B,CRF45_cpx,CRF59_01B,CRF85_BC and URF (B/C).CRF45_cpx and CRF85_BC were discovered the first time in Zhejiang province.Twenty-six transmission clusters involving 65 cases were found,with the total clustered rate as 43.0％ (65/151),in which the CRF01_AE clustered rate appeared as 54.1％ (40/74),higher than that of CRF07_BC (21/64,32.8％).The average size of cluster was 2.5 cases/cluster,with average size of cluster in CRF01_AE patients infected through heterosexual transmission as the largest (3.5 cases/cluster).The prevalence of transmitted drug resistance was 4.6％ (7/151).Seven cases with surveillance drug resistant mutations (SDRM) were found,including 5 cases of M46L (3.3％),and one case of F77L or Y181C.Conclusion HIV genetic diversity and a variety of transmission clusters had been noticed in this study area (Yiwu).Programs on monitoring the subtypes and transmission clusters should be continued and strengthened.

OBJECTIVE To evaluate the safety and clinical efficacy of cefoperazone/sulbactam in the treatment of biliary tract infections.METHODS In this prospective multicenter study,159 hospitalized patients with biliary tract infections received cefoperazone/sulbactam,and the clinical and bacteriological efficacy as well as the side effects were evaluated.RESULTS The clinical effective rate of cefoperazone/sulbactam in the treatment of biliary tract infections was 86.78%.After treatment,the body temperature reduced to normal rapidly,the average time of defervescence was 3.09?1.81 days.Pathogen eradication rate was 85.71%.No adverse reactions were reported during the study period.CONCLUSIONS Cefoperazone/sulbactam can be used as one of antibiotics of choice in the initial empirical therapy for biliary tract infections.