ObjectiveTo observe the effects of Guben Fangxiao decoction-containing serum on the expression of transcription factors and cytokines associated with calcitonin gene-related peptide （CGRP）， γ-aminobutyric acid （GABA）， and type Ⅱ innate lymphoid cells （ILC2） in human bronchial epithelial cells （BEAS-2B） stimulated with interleukin-4 （IL-4） combined with interleukin-13 （IL-13）， and explore the possible mechanism of Guben Fangxiao decoction-containing serum in alleviating type Ⅱ inflammation of bronchial epithelial cells. MethodsBEAS-2B cells stimulated with IL-4 combined with IL-13 were treated with the sera containing high （20%）， medium （15%）， and low （10%） doses of Guben Fangxiao decoction and the montelukast-containing serum （10%）. Real-time PCR was used to measure the mRNA levels of CGRP， GABA， mucin 5AC （MUC5AC）， thymic stromal lymphopoietin （TSLP）， interleukin （IL）-25， IL-33， IL-5， GATA-binding protein 3 （GATA3）， and B cell lymphoma/leukemia 11B （Bcl-11B） in each group. Western blot was employed to determine the relative protein levels of CGRP and MUC5AC. The immunofluorescence assay was employed to observe the relative expression of CGRP and GABA in the cells. Enzyme-linked immunosorbent assay was used to quantify the protein levels of IL-33， MUC5AC， and IL-5 in the culture and cell supernatants. Periodic acid-Schiff staining was performed to assess mucin secretion. ResultsCompared with the control group， the model group showed up-regulated mRNA levels of CGRP， GABA， MUC5AC， TSLP， IL-25， and IL-33 （P<0.05， P<0.01）. Compared with the normal group， the Guben Fangxiao decoction-containing serum down-regulated the mRNA levels of CGRP， GABA， MUC5AC， IL-5， GATA3， and Bcl-11B （P<0.05， P<0.01） in a dose-dependent manner. Compared with the normal group， the model group showed up-regulated protein levels of CGRP， GABA， MUC5AC， IL-33， and IL-5 （P<0.05， P<0.01）， which were down-regulated by the Guben Fangxiao decoction-containing serum （P<0.05， P<0.01）. Compared with the normal group， the model group showed enhanced average fluorescence intensity of CGRP and GABA （P<0.01）， which was weakened by the Guben Fangxiao decoction-containing serum （P<0.01）. Compared with the normal group， the model group showed increased secretion of mucin， which was reduced by the Guben Fangxiao decoction-containing serum. ConclusionGuben Fangxiao decoction can treat bronchial asthma by alleviating type Ⅱ immune airway inflammation and reducing airway mucus secretion， which is achieved by regulating the expression of CGRP， GABA， MUC5AC， and ILC2-related transcription factors.

Corn silk, a Traditional Chinese Medicine, has the effect of calming liver, cholagogue, detumescence and diuresis. Corn silk is also widely used as tea and functional food. Natural flavonoids have multiple biological activities, which are also the main bioactive components of corn silk. In the past decade, many new advances have been made in the chemistry, analysis, pharmacology, pharmacokinetics, and safety evaluation of corn silk flavonoids. The chemical composition research of flavonoids has enriched the variety of flavonoids in corn silk. Pharmacological studies have confirmed and expanded the efficacy of corn silk flavonoids. And safety evaluation has provided a theoretical basis for the safe application of corn silk flavonoids. Through literature search, the extraction, separation, compositional analysis, content determination, pharmacological effect, pharmacokinetics, and safety research progress of corn silk flavonoids in the past ten years were reviewed in this paper.

Objective: To explore the relationship between parenting styles and depressive symptoms in adolescents with nightmare disorder, so as to provide a scientific basis for formulating effective family intervention measures and psychological counseling. Methods: From January 2023 to August 2024, 90 adolescents diagnosed with nightmare disorder and admitted to Hangzhou Seventh Peoples Hospital, along with 176 healthy controls from the urban areas of Hangzhou, were recruited as participants in the study. All participants were assessed using the Nightmare Experience Questionnaire (NEQ), Family Relationship Questionnaire (FRQ), and Plutchik-van Praag Selfreport Depression Scale (PVP). The ttest and Chisquare test were conducted to compare two groups. Pearson correlation and stepwise multiple linear regression were employed to explore the correlations between PVP and NEQ or FRQ. The Process model was used to testing the mediating effects among NEQ/FRQ/PVP. Results: The nightmare disorder group had higher scores in nightmare frequency, the four factors of NEQ (physical effect, negative emotion, meaning interpretation, horrible stimulation), and PVP than the healthy control group (24.86±18.89, 10.12±3.67, 19.01±3.51, 17.02±3.31, 15.14±3.26, 14.02±4.38; 2.34±1.04, 6.49±2.18, 17.63±4.76, 13.91±4.24, 12.40±4.49, 9.39±3.28)(t=15.79, 10.11, 2.43, 6.09, 5.14, 27.46, P<0.05). The nightmare disorder group reported significantly lower scores in FRQ general attachment and maternal encouragement than the healthy control group (7.22±2.81, 16.39±3.28) (t=-5.53, -4.95). In contrast, they exhibited significantly higher scores in maternal abuse, maternal dominance, paternal freedom release, and paternal dominance than the healthy control group (8.23±1.80, 13.11±3.73, 18.36±3.37, 12.04±3.29; 6.07±1.85, 8.48±3.80, 15.15±2.51, 9.47±3.03) (t=6.70, 8.96, 5.90, 7.04, P<0.01). The results of Pearson correlation analysis showed that, in the nightmare disorder group, the PVP score was positively correlated with negative emotion, nightmare frequency, maternal abuse, and maternal dominance score (r=0.14, 0.63, 0.26, 0.51, P<0.05). The results of multiple linear regression analysis showed that when using FRQ score to predict NEQ score, the adjusted R2 in the nightmare disorder group was 0.01-0.59. Mother abuse could prediced physical effect (β=0.33); maternal dominance significantly predicted negative emotion, horrible stimulation, and nightmare frequency (β=0.29, 0.30, 0.79); paternal freedom release could predict negative emotion (β=0.26), paternal dominance predicted both negative emotion and nightmare frequency (β=0.22, 0.45) (P<0.05). Mediation analysis further revealed that, in the nightmare disorder group, PVP scores served as a mediating variable between FRQ and NEQ. Conclusion Abusive, controlling, and neglectful family upbringing styles as well as depression maybe are key factors that may contribute to the development of nightmare disorder among adolescents.

Objective:To explore the predictive value of admission serum homocysteine levels and quantitative electroencephalogram (qEEG) indicators for adverse outcomes in patients with cerebral hemorrhage.Methods:A retrospective study was conducted on 89 patients, who were collected as the study objects with hemorrhagic stroke treated in the neurology intensive care unit at Kailuan General Hospital from January 2017 to December 2022. Patients were categorized into two groups based on modified Rankin Scale (mRS) scores at discharge: a good prognosis group (mRS≤2) and a poor prognosis group (mRS 3-6). Clinical data and qEEG monitoring of various brain regions were collected. The impact factors of hemorrhagic prognosis were analyzed using multifactorial logistic regression. ROC curve analysis was performed to assess the predictive value of qEEG and admission homocysteine levels for adverse outcomes in hemorrhagic stroke patients.Results:(1) The age of the poor prognosis group was higher than that of the good prognosis group((66.51+13.64) to (60.53+11.69), t=2.15, P=0.034) and admission serum homocysteine levels were significantly higher in the poor prognosis group than in the good prognosis group (17.28(15.52,24.72)mmol/L to 14.50(10.28,16.00)mmol/L, Z=4.14, P<0.001). (2) In the poor prognosis group, power values of δ brain waves in leads Fp1-2, F4, C4, P4, F8, and T4 were higher than those in the good prognosis group (87.99(41.57,196.69) to 50.67(26.64,54.75), Z=2.76, P=0.006); (79.17(40.71,200.00) to 45.06(20.22,61.00), Z=2.10, P=0.036); (72.64(34.97,219.78) to 34.42(19.81,63.4), Z=2.03, P=0.043); (65.06(33.36,177.45) to 28.12(15.88,63.36), Z=2.08, P=0.038); (52.92(25.64,187.91) to 23.61(11.67,43.26), Z=2.21, P=0.027); (66.67(32.56,180.76) to 36.31(17.2,53.78), Z=2.46, P=0.014); (57.30(25.24,127.04) to 29.57(11.91,41.89), Z=2.26, P=0.024). Power values of θ brain waves in leads Fp1-2, F3, F4, C3, C4, P3-4, O1, F7-8, and T3-4 were higher in the poor prognosis group(77.45(47.63,138.72)比35.88(20.92,44.81), Z=3.50, P<0.001); (77.05(35.16,120.22) to 38.74(19.86,58.09), Z=2.27, P=0.023); (85.24(52.53,147.90) to 35.42(14.7,52.59), Z=2.61, P=0.009); (75.81(37.90,124.97) to 36.85(17.92,55.43), Z=2.30, P=0.021); (72.00(43.92,123.54) to 28.37(14.02,51.9), Z=2.22, P=0.027); (67.08(32.01,104.05) to 31.32(17.98,45.28), Z=2.10, P=0.035); (55.33(32.29,94.30) to 25.64(11.87,34.01), Z=2.24, P=0.025); (48.84(20.64,96.28) to 19.85(9.83,28.58), Z=2.30, P=0.022);(48.46(25.06,81.78) to 23.95(8.80,29.16), Z=2.51, P=0.012); (64.46(39.38,112.44) to 26.85(15.74,39.58), Z=2.80, P=0.005); (65.68(31.78,102.00) to 31.09(15.98,46.96), Z=2.38, P=0.017); (45.26(28.34,73.14) to 21.45(10.57,36.59), Z=2.04, P=0.042); (43.50(22.58,78.67) to 25.45(11.91,32.26), Z=2.22, P=0.027). Power values of slow-wave index in leads Fp1-2, F3-4, C3-4, P4, F7-8, and T4, as well as the overall brain average, were higher in the poor prognosis group (6.64(2.98,10.42) to 3.65(2.31,4.30), Z=2.65, P=0.01); (6.53(3.96,11.65) to 3.53(2.56,4.51), Z=2.30, P=0.022); (7.38(4.62,13.12) to 3.83(1.70,4.71), Z=2.38, P=0.017); (5.88(4.02,12.15) to 3.18(2.21,4.46), Z=2.29, P=0.022); (6.13(3.83,11.22) to 2.97(1.53,4.58), Z=2.01, P=0.044); (6.07(3.53,9.39) to 2.74(2.00,3.81), Z=2.40, P=0.016);(4.11(2.51,9.23) to 2.18(1.37,2.82), Z=2.25, P=0.024); (5.71(3.81,10.44) to 3.22(1.86,4.04), Z=2.28, P=0.023); (6.00(3.65,10.37) to 3.04(2.00,4.00), Z=2.39, P=0.017); (4.08(2.56,8.33) to 2.08(1.60,3.14), Z=2.50, P=0.013), with significant statistical differences noted (5.45(3.31,10.08) to 3.17(2.02,4.88), Z=3.62, P=0.005). (3) Logistic regression results showed that admission homocysteine levels ( OR 1.311,95% CI 1.008-1.705, P=0.044), admission NIHSS scores ( OR 1.588,95% CI 1.074-2.349, P=0.020), and overall brain average slow-wave index were influencing factors for poor prognosis in cerebral hemorrhage ( OR 8.596,95% CI 1.088-67.889, P=0.041). (4) ROC curve analysis revealed that the AUC for predicting adverse outcomes in cerebral hemorrhage was 0.768 (95% CI (0.665, 0.872)) for admission homocysteine levels, 0.743 (95% CI (0.634, 0.852)) for the overall brain average slow-wave index, and 0.896 (95% CI (0.827, 0.965)) for admission NIHSS. The cutoff values were 15.67, 3.62, and 8.5, respectively. Sensitivity was 77.8%, 71.1%, and 68.9%, and specificity was 59.4%, 68.7%, and 100%, respectively. The Youden indices were 0.372, 0.398, and 0.689. Conclusion:In the acute phase of cerebral hemorrhage, electroencephalographic physiological changes manifest shows an increase in the δ, θ, and slow-wave index throughout the entire brain. Higher admission homocysteine levels suggest a worse prognosis in patients with cerebral hemorrhage. Admission homocysteine levels and overall brain average slow-wave index have certain predictive value for adverse outcomes in acute cerebral hemorrhage.

Objective:This study aims to evaluate the quality and explore the preliminary clinical applications of a domestically developed hepatitis D virus nucleic acid quantification reagent (abbreviated as"domestic HDV RNA reagent").Methods:The sensitivity and accuracy of the reagent were evaluated in accordance with the WHO HDV RNA international standard, employing the Bio-Rad CFX Opus 96 real-time fluorescence quantitative PCR analysis system. Serial dilutions of pseudo-viruses or cell culture-derived virus were used to determine the linear range of the domestic HDV RNA reagent. Specificity was assessed using positive samples of HAV, HBV, HCV infection, and HEV national reference materials. Precision was evaluated with samples at both high and low concentrations. In a comparative analysis, 30 HDV IgG positive samples were tested using both the domestic HDV RNA reagent and the RoboGene HDV RNA kit based on the ABI 7500 FAST DX system. The Pearson correlation coefficient (r) was used to examine the correlation between the two reagents.Results:The domestic HDV RNA reagent demonstrated a high sensitivity of up to 6 IU/ml, consistent with that of the comparator reagent. The calibration curve for WHO HDV RNA standards had a slope of -3.286, with an amplification efficiency of 101.6%. The linear detection range spanned from 10 to 10 8 IU/ml for eight HDV genotypes. The domestic HDV RNA reagent exhibited exceptional specificity, without cross-reactivity observed with HAV, HBV, HCV, or HEV. Accuracy assessments at five concentration levels met the required standards, with intra-assay precision coefficient of variation ( CV) ranging from 1.20% to 4.20%, and inter-assay precision CV from 1.20% to 7.90%. The detection results for HDV IgG positive samples were highly correlated with the comparator reagent ( r=0.984, P<0.001), achieving a diagnostic accuracy of 100% compared to sequencing results. Conclusion:In this study, the domestic HDV RNA reagent possesses excellent specificity, accuracy, precision, and a broad linear range, attaining a sensitivity level on par with international reagents of the same type.

Objective:To explore the mediating chain effect between attachment and coping style of disease perception and hope in patients with advanced lung cancer, and to provide theoretical basis for improving coping style in patients with advanced lung cancer.Methods:From October 2021 to June 2022, 354 patients with advanced lung cancer in the First and Second Affiliated Hospitals of Anhui Medical University were selected by convenience sampling. The general information questionnaire, the Experiences in Close Relationships Inventory, the Brief Illness Perception Questionnaire, the Herth Hope Index, and the Medical Coping Modes Questionnaire were used to conduct cross-sectional questionnaire survey. SPSS 25.0 software and Bootstrap method were used to construct and verify the chain mediation model.Results:Finally, 336 patients with advanced lung cancer were included, including 214 males and 122 females, aged 27-79(59.43 ± 8.61) years old. Attachment avoidance score was (3.31 ± 1.01) points, attachment anxiety score was (3.86 ± 1.17) points, illness perception score was (40.07 ± 12.01) points, hope score was (34.05 ± 5.87) points, and face coping score was (18.75 ± 5.34) points in patients with advanced lung cancer. The avoidance coping score was (15.47 ± 1.97) points, and the yielding coping score was (9.62 ± 3.85) points. In patients with advanced lung cancer, attachment avoidance and attachment anxiety were positively correlated with yield coping ( r=0.448, 0.747, both P<0.01), positively correlated with illness perception ( r=0.356, 0.627, both P<0.01), and negatively correlated with hope ( r=-0.406, -0.670, both P<0.01). Illness perception was positively correlated with yield coping ( r=0.744, P<0.01), and negatively correlated with hope ( r=-0.628, P<0.01). Hope was negatively correlated with yield response ( r=-0.769, P<0.01). The mediation model showed that the chain mediating effect of attachment avoidance, illness perception, hope and yield coping was significant in patients with advanced lung cancer, with an effect value of 0.009 and an effect size of 13.95%. The chain mediating effect of attachment anxiety, illness perception, hope and yield coping were significant, with an effect value of 0.010 and an effect size of 8.27%. Conclusions:Attachment can not only directly predict submission coping in advanced lung cancer patients, but also indirectly predict submission coping through the chain mediation of illness perception and hope.

BACKGROUND:Astrocytes are the most abundant cells in the central nervous system,and various subsets of astrocytes are heterogeneous,performing a variety of special functions.Single-cell RNA sequencing(scRNA-seq)technology developed in recent years has extended our understanding of astrocyte heterogeneity from the perspective of transcriptome profiling. OBJECTIVE:To summarize the heterogeneity of scRNA-seq technology in different time and space,and pathological states and expand our knowledge of astrocyte heterogeneity on both molecular and functional levels. METHODS:The relevant articles on astrocyte heterogeneity and scRNA-seq were searched on PubMed,Elsevier,and CNKI databases.The search terms were"astrocytes,scRNA-seq,heterogeneity,Alzheimer disease,spinal cord injury,multiple sclerosis"in Chinese and English.Finally,74 articles were selected for viewing after screening according to inclusion criteria. RESULTS AND CONCLUSION:scRNA-seq studies related to the heterogeneity of astrocytes have shown that astrocyte is significantly heterogeneous across four aspects:species,developmental stage,central nervous system region,and pathological state.(1)Unique expression of certain genes occurs in astrocytes of different species,and the discovery of species-specific genes is beneficial for the translation of clinical studies.(2)During astrocyte development,differential gene expression emerged in the cellular subtypes identified at each stage,which further refined the cellular lineage of astrocytes and laid the foundation for the study of astrocyte developmental trajectories and mechanisms.(3)The discovery of differential gene expression allows regional localization of different astrocyte subpopulations and assists in the diagnosis and treatment of neurological diseases.(4)Astrocyte heterogeneity revealed by scRNA-seq can provide specific markers at the time of disease diagnosis and identify potential therapeutic targets.(5)The heterogeneity of astrocytes exists in many aspects,interacts with each other and is complex.The mechanisms of its generation,maintenance and transformation remain unclear.At present,molecular research on the single-cell level is still lacking.Linking transcriptionally defined astrocyte subpopulations to cellular activity,behavior and disease markers in real time remains one of the great challenges in the field.

Objective To explore the mechanism of somatostatin in improving acute lung injury associated with acute pancreatitis.Methods Wistar rats were randomly divided into sham operation group(injection of normal saline),model group(puncture of common bile duct and injection of 5％sodium taurocholate with wire ligation),somatostatin group(injection of somatostatin into tail vein of model group),somatostatin+miR-146a-5p inhibitor group(on the basis of somatostatin group,tail vein injection of miR-146a-5p inhibitor and somatostatin+oe-angiogenin-like protein 4(ANGPTL4)group(on the basis of somatostatin group,tail vein injection of oe-ANGPTL4 plasmid).Hematoxylin-eosin(HE)staining was used to observe the pathological changes of pancreatic and lung tissues;pathological score and tissue wet-dry weight ratio were determined,real-time fluorescence quantitative polymerase chain reaction(qRT-PCR)was used to detect miR-146a-5p and ANGPTL4 mRNA expression and Western blot was used to detect the expression of related proteins in lung tissues of rats.Tumor necrosis factor-α(TNF-α)was detected by enzyme-linked immunosorbent assay(ELISA).Results In sham operation group,model group and somatostatin group,the damage degree of pancreas tissue(based on modified computed tomography severity index)were 1.25±0.28,3.20±0.34,2.15±0.31,respectively;the damage degree of lung tissue(based on the Smith lung injury score system)were 1.40±0.13,5.10±0.58,3.10±0.38,respectively.The relative expression levels of ANGPTL4 mRNA in sham operation group,model group,somatostatin group and somatostatin+miR-146a-5p inhibitor group were 1.00±0.17,1.63±0.20,1.21±0.18 and 1.73±0.28.The levels of TNF-α in sham operation group,model group,somatostatin group,somatostatin+miR-146a-5p inhibitor group and somatostatin+oe-ANGPTL4 group were(76.33±7.25),(125.05±13.56),(80.11±10.68),(118.62±14.32)and(105.32±13.52)pg·mL-1,respectively;the relative expression levels of nuclear factor E2-related factor 2(Nrf2)protein were 1.00±0.27,0.51±0.07,0.88±0.14,0.68±0.12,0.51±0.09,respectively;the relative expression levels of heme oxygenase-1(HO-1)protein were 1.00±0.25,0.58±0.11,0.79±0.18,0.48±0.07 and 0.50±0.08,respectively.The above indexes of the model group were compared with those of the sham operation group,and the above indexes of the somatostatin group were compared with those of the model group,somatostatin+miR-146a-5p inhibitor group and somatostatin+oe-ANGPTL4 group,and the differences were statistically significant(all P＜0.05).Conclusion Somatostatin has antioxidant and anti-inflammatory effects and can ameliorate acute lung injury associated with acute pancreatitis.The mechanism may be related to Nrf2/HO-1 pathway mediated by miR-146a-5p/ANGPTL4.

Background::Risk assessment and treatment stratification for three-vessel coronary disease (TVD) remain challenging. This study aimed to investigate the prognostic value of left atrial volume index (LAVI) with the Synergy Between Percutaneous Coronary Intervention with Taxus and Cardiac Surgery (SYNTAX) score II, and its association with the long-term prognosis after three strategies (percutaneous coronary intervention [PCI], coronary artery bypass grafting [CABG], and medical therapy [MT]) in patients with TVD.Methods::This study was a post hoc analysis of a large, prospective cohort of patients with TVD in China, that aimed to determine the long-term outcomes after PCI, CABG, or optimal MT alone. A total of 8943 patients with TVD were consecutively enrolled between 2004 and 2011 at Fuwai Hospital. A total of 7818 patients with available baseline LAVI data were included in the study. Baseline, procedural, and follow-up data were collected. The primary endpoint was major adverse cardiac and cerebrovascular events (MACCE), which was a composite of all-cause death, myocardial infarction (MI), and stroke. Secondary endpoints included all-cause death, cardiac death, MI, revascularization, and stroke. Long-term outcomes were evaluated among LAVI quartile groups. Results::During a median follow-up of 6.6 years, a higher LAVI was strongly associated with increased risk of MACCE (Q3: hazard ratio [HR] 1.20, 95% confidence interval [CI] 1.06-1.37, P = 0.005; Q4: HR 1.85, 95%CI 1.64-2.09, P <0.001), all-cause death (Q3: HR 1.41, 95% CI 1.17-1.69, P <0.001; Q4: HR 2.54, 95%CI 2.16-3.00, P <0.001), and cardiac death (Q3: HR 1.81, 95% CI 1.39-2.37, P <0.001; Q4: HR 3.47, 95%CI 2.71-4.43, P <0.001). Moreover, LAVI significantly improved discrimination and reclassification of the SYNTAX score II. Notably, there was a significant interaction between LAVI quartiles and treatment strategies for MACCE. CABG was associated with lower risk of MACCE than MT alone, regardless of LAVI quartiles. Among patients in the fourth quartile, PCI was associated with significantly increased risk of cardiac death compared with CABG (HR: 5.25, 95% CI: 1.97-14.03, P = 0.001). Conclusions::LAVI is a potential index for risk stratification and therapeutic decision-making in patients with three-vessel coronary disease. CABG is associated with improved long-term outcomes compared with MT alone, regardless of LAVI quartiles. When LAVI is severely elevated, PCI is associated with higher risk of cardiac death than CABG.

Objective:To construct a risk prediction model for the inter-hospital transfer of critically ill children using machine learning methods, identify key medical features affecting transfer outcomes, and improve the success rate of transfers.Methods:A prospective study was conducted on critically ill children admitted to the pediatric transfer center of Hunan Children's Hospital from January 2020 to January 2021. Medical data on critical care features and relevant data from the Pediatric Risk of Mortality (PRISMⅢ) scoring system were collected and processed. Three machine learning models, including logistic regression, decision tree, and Relief algorithm, were used to construct the risk prediction model. A back propagation neural network was employed to build a referral outcome prediction model to verify and analyze the selected medical features from the risk prediction model, exploring the key medical features influencing inter-hospital transfer risk.Results:Among the 549 transferred children included in the study, 222 were neonates (40.44%) and 327 were non-neonates (59.56%). There were 50 children in-hospital deaths, resulting in a mortality rate of 9.11%. After processing 151 critical care medical feature data points, each model selected the top 15 important features influencing transfer outcomes, with a total of 34 selected features. The decision tree model had an overlap of 72.7% with PRISMⅢ indicators, higher than logistic regression (36.4%) and Relief algorithm (27.3%). The training prediction accuracy of the decision tree model was 0.94, higher than the accuracy of 0.90 when including all features, indicating its clinical utility. Among the top 15 important features selected by the decision tree model, the impact on transfer outcomes was ranked as follows based on quantitative feature violin plots: base excess, total bilirubin, ionized calcium, total time, arterial oxygen pressure, blood parameters (including white blood cells, platelets, prothrombin time/activated partial thromboplastin time), carbon dioxide pressure, blood glucose, systolic blood pressure, heart rate, organ failure, lactate, capillary refill time, temperature, and cyanosis. Eight of these important features overlapped with PRISMⅢ indicators, including systolic blood pressure, heart rate, temperature, pupillary reflex, consciousness, acidosis, arterial oxygen pressure, carbon dioxide pressure, blood parameters, and blood glucose. The decision tree was used to select the top 15 medical features with high impact on the neonatal and non-neonatal datasets, respectively. A total of 19 features were selected, among which there were 8 differences and 11 overlap terms between the important features of the neonatal and non-neonatal.Conclusions:Machine learning models could serve as reliable tools for predicting the risk of inter-hospital transfer of critically ill children. The decision tree model exhibits superior performance and helps identify key medical features affecting inter-hospital transfer risk, thereby improving the success rate of inter-hospital transfers for critically ill children.