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 effect and mechanism of isorhynchophylline (IRN) on angiotensin Ⅱ(Ang Ⅱ)-induced cardiac hypertrophy.Methods:H9c2 cells were co-cultured with Ang Ⅱ and different concentrations of IRN (0, 5, 10, 25, 50 μmol/L). The cell surface area and mRNA levels of cardiac hypertrophy markers atrial natriuretic peptide (ANP), brain natriuretic peptide (BNP) and β-myosin heavy chain (β-MHC) were detected to elucidate the effect of IRN on myocardial hypertrophy and the most effective concentration. H9c2 cells were co-cultured with Ang Ⅱ and IRN (25 μmol/L) at different times (0, 6, 12, 24 h) to elucidate the most effective time of inhibition. The phosphorylation levels of the signaling pathway were detected, and the effects of IRN and Akt inhibitor MK2206 on the phosphorylation levels of the signaling pathway were further explored to elucidate the underlying mechanisms.Results:Compared with the control group, the surface area of H9c2 cells, and the mRNA expression of myocardial hypertrophy markers ANP, BNP and β-MHC were significantly increased (all P<0.05). Pretreated with different concentrations of IRN (5, 10, 25, 50 μmol/L) could inhibit the increase in cell surface area induced by AngⅡ (all P<0.05), especially at the concentration of 25 μmol/ L ( P<0.01). IRN could time-dependently inhibit AngⅡ-induced activation of ANP, BNP, β-MHC mRNA (all P<0.05). AngⅡ caused increased phosphorylation levels of Akt, GSK3β, mTOR and FOXO3a. IRN could block AngⅡ-induced phosphorylation of the Akt signaling pathway. Conclusion:IRN attenuates AngⅡ-induced cardiomyocyte hypertrophy by inhibiting the Akt signaling pathway.

Objective:To analyze the change characteristics of creatinine level in the early stage of patients with diquat (DQ) poisoning, and to explore the early risk factors and the value of prognosis.Methods:A retrospective analysis was carried out on patients with DQ admitted to the the first affiliated hospital of Zhengzhou University from January 2020 to June 2022. The DQ patients were divided into death group and the survival group according to the 28 days survival status after posioning. The basic data and serum indexes and blood gas analysis of the patients on day 1 (D1), day 3 (D3) and day 5 (D5) were collected. The difference of clinical features between the two groups was analyzed, the variables were screened by multiple logistic regression analysis, and the predictive value of the variables was evaluated by drawing receiver operating characteristic curve (ROC curve).Results:A total of 88 patients were included, including 40 patients in the survival group and 48 patients in the death group. The toxic dose in death group was significantly higher than that in survival group [100(40.00, 120.00) mL vs. 50.00(20.00, 90.00) mL, P=0.003]. The higher the toxic dose, the higher the fatality rate. All 4 patients with oral doses greater than 200 mL died. Compared with the survival group, the levels of alanine aminotransferase (ALT) (D3, D5), creatinine (CR) (D3, D5), blood amylase (AMY) (D5) and oxygen partial pressure (PaO 2) (D5) in the death group were significantly higher than those in the survival group (all P<0.05). Multiple Logistic regression analysis showed that CR (D3) and AMY(D5) were independent risk factors for death after poisoning, and PaO 2(D5) was independent protective factor. ROC curve showed that the areas under ROC curve of CR (D3), AMY (D5) and PaO 2 (D5) were 0.814, 0.741 and 0.702, respectively. Conclusion:The higher the oral dose, the higher the death rate. After admission, CR(D3), AMY (D5) and PaO 2 (D5) were independent factors influencing the prognosis of DQ poisoning. In particular, CR (D3) is more effective in predicting death after poisoning.

Objective:To evaluate the effect of sleep fragmentation on postoperative cognitive dysfunction (POCD) and hippocampal glutaminergic metabolism in aged mice anesthetized with isoflurane.Methods:Forty healthy SPF-grade male C57BL/6J mice, aged 18 months, weighing 20-30 g, were divided into 4 groups ( n= 10 each) by the random number table method: normal control group (group C), sleep fragmentation group (group SF), isoflurane anesthesia/surgery group (group I/S), and sleep fragmentation plus isoflurane anesthesia/surgery group (group SF+ I/S). Group C did not received any treatment. Group SF received sleep fragmentation for 24 h. The right carotid artery exposure was performed under isoflurane anesthesia in group I/S. Group SF+ I/S received isoflurane anesthesia/right carotid artery exposure at 24 h after sleep fragmentation. The metabolic levels of glutamate (Glu), glutamine (Gln), Glu/Gln complex (Glx), and N-acetylaspartate (NAA) and their ratio to creatine (Cr) were measured by in vivo 9.4T hydrogen proton magnetic resonance spectroscopy at 2 h after anaesthesia. Y maze and Morris water maze tests were used to evaluate the cognitive function at 1-7 days after surgery. The mice were sacrificed after the behavioral testing, brain tissues were immediately obtained, and the number of Nissl bodies and density of dendritic spines in the hippocampal CA1 region were measured by Nissl staining and Golgi staining, respectively. Results:Compared with group C, the percentage of exploration time and shuttle times at the novel arm were significantly decreased, the number of crossing the original platform was decreased, the time of stay at the target quadrant was shortened, the ratios of Glu/Cr, Gln/Cr and Glx/Cr in the hippocampal CA1 region were increased, and the ratio of NAA/Cr was decreased, and the number of Nissl bodies and density of dendritic spines were decreased in SF, I/S and SF+ I/S groups ( P<0.05). Compared with group SF and group I/S, the percentage of exploration time and shuttle times at the novel arm were significantly decreased, the number of crossing the original platform was decreased, the time of stay at the target quadrant was shortened, the ratios of Glu/Cr and Glx/Cr in hippocampal CA1 region was increased, the ratio of NAA/Cr was decreased, and the number of Nissl bodies and density of dendritic spines were decreased in group SF+ I/S ( P<0.05). Conclusions:Sleep fragmentation exacerbates POCD in aged mice anesthetized with isoflurane, and the mechanism is related to nerve injury induced by abnormality in hippocampal glutaminergic metabolism excitability.

Objective:To explore the efficacy and safety of sivelestat, a neutrophil elastase (NE) inhibitor, in the treatment of acute lung injury (ALI) in the intensive care unit (ICU).Methods:A retrospective analysis was performed on 171 patients with ALI in the ICU of the First Affiliated Hospital of Zhengzhou University from June 2020 to June 2021, including 77 patients in the sivelestat group and 94 patients in the conventional treatment group. Acute physiology and chronic health evaluation (APACHE) Ⅱ score, Murray lung injury score, oxygenation index (PaO 2/FiO 2 ratio), inflammatory cytokines (IL-6, IL-10, TNF-α), ventilator-free days (VFD), the length of ICU stay, and the 28-day mortality were collected to assess the efficacy of sivelestat. At the same time, adverse reactions and laboratory test results within 30 days after the use of sivelestat were recorded to assess the safety. Results:Compared with conventional treatment, oxygenation index, Murray lung injury scores, IL-6, IL-10, and TNF-α were significantly improved after 7 days of sivelestat treatment. Compared with the conventional treatment group, the VFD was significantly longer ( P = 0.0119) and the length of ICU stay was significantly shorter ( P = 0.0269) in the sivelestat group. The mortality was 14.29% in the sivelestat group and 22.34% in the conventional treatment group and, with no statistically significant. In the meantime, sivelestat did not increase adverse reactions within 30 days after treatment. Conclusions:Sivelestat treatment is safe and more effective than conventional treatment for ALI patients in the ICU.

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 explore the serological feature and molecular mechanism for a case with A307 subgroup of the ABO blood group system. METHODS: Serological assay was carried out to determine the ABO blood group of the proband and his family members. Genotypes for exons 1 to 7 of the ABO gene were determined with sequence-specific primer polymerase chain reaction (SSP-PCR) and direct sequencing. The impact of the variant on the stability of alpha-1,3-N-acetylgalactosaminyltransferase (GTA) was predicted through construction of a 3D molecular model. RESULTS: The proband, his brother and daughter were diagnosed with Aend phenotype by serological analysis. Their ABO genotype was determined as A307/O02, with heterozygous c.467C>T (p.P156L) and c.745C>T (p.R249W) variants identified in exon 7 of the ABO gene. Molecular modeling suggested that the p.R249W variant may alter the number of hydrogen bonds between the amino acids. The protein was predicted to have a decreased Δ Δ G value of thermodynamic stability. CONCLUSION The p.R249W variant may give rise to the A307 subgroup by reducing the stability of the GTA enzyme, leading to serological features of Aend phenotype.

Objective:To investigate the effect of thymosin α1 on the differentiation of T lymphocyte and the secretion of inflammatory factors in septic mice, thus to explore the effect of thymosin α1 on the prognosis of sepsis.Methods:Adult female C57 mice were randomly (random number) divided into 3 groups: blank control group, sepsis group, and thymosin α1 treatment group. T cell counts and the corresponding inflammatory factors in the further differentiation of T lymphocytes as well as plasma and lung tissues were statistically analyzed, and the survival rate of the mice within 96 h was also analyzed. Graphpad 7.0 software was used for statistically analysis of the study results.Results:There was no significant difference in T cell counts among the three groups of mice, but in the further differentiation of T lymphocytes, the expression of Th17 in the thymosin α1 treatment group was significantly lower than that in the sepsis group, and the expression of Treg was significantly increased in the sepsis group. The expression of the inflammatory cytokine IL-10 was significantly increased in plasma and lung tissues of the thymosin α1 treatment group, while the expression of IL-17A in plasma and lung tissues of the thymosin α1 treatment group was significantly lower ( P <0.05). Survival analysis showed that the survival rate of the thymosin α1 treatment group increased significantly at 96 h, and the difference was significant statistically ( P <0.05). Conclusions:Thymosin α1 can enhance the cellular immunity in sepsis, ameliorate the systemic inflammation, and further protect against sepsis.

Objective To explore the effect of noninvasive ventilation (NIV) with helmet or facial mask on clinical efficacy, tolerability, and prognosis in patients with acute respiratory failure. Methods Fifty patients with acute respiratory failure according to the inclusion criteria were recruited from January 2018 to July 2018 in Emergency Intensive Care Unit of the First Affiliated Hospital of Zhengzhou University. Included patients were randomly allocated into the helmet group or facial mask group. Based on conventional drug therapy, pressure support mode was performed with the interface of the helmet or facial mask. Oxygenation index, arterial carbon dioxide partial pressure, and respiratory rates were measured before and after the treatment, and the data were compared and analyzed by the repeated measures ANOVA. Tolerance score, complication rate, tracheal intubation rate, and mortality rate were recorded at each observation time point of the two groups. Results The oxygenation index before NIV, at 4 h and at the end of NIV treatment of the helmet group were significantly increased from (160.29±50.32) mmHg to (249.29±83.47) mmHg and (259.24±87.09) mmHg; the oxygenation index of the facial mask group were increased from (168.63±38.63) mmHg to (225.00±74.96) mmHg and (217.69±77.80) mmHg, and there was no significant difference within the two groups (P <0.05). The respiratory rates before NIV, at 4 h and at the end of NIV treatment of the helmet group were obviously decreased from (27.60±7.64) breaths/min to (17.92±4.55) breaths/min and (16.88±3.90) breaths/min; the respiratory rates of the facial mask group were decreased from (24.68±6.14) breaths/min to (20.36±4.25) breaths/min and (19.68±3.34) breaths/min, and the differences within the two groups were statistically significant (P <0.05). However, there were no significant differences on oxygenation index and respiratory rates between the helmet group and facial mask group (P >0.05). Patients in the helmet was better tolerated than those in the facial mask group [ratio of good tolerance 96％ (24/25) vs 56％ (14/25) (P = 0.001) and fully tolerance 80％ (20/25) vs 36％ (9/25) (P =0.002)] and had less complications (1/25 vs 10/25, P = 0.002). 84％ patients in the helmet group and 76％ patients in the facial mask group were successfully weaned and discharged after NIV treatment (P =0.480). Conclusions Similar clinical efficacy in improving blood gas exchange and relieving dyspnea were observed in the helmet group and the facial mask group in patients with acute respiratory failure. However, the helmet is better tolerant, and had lower complication rate, which is especially suitable for patients with chest trauma combined with facial injuries.

OBJECTIVETo explore the molecular mechanism of a case of ABO discrepancies based on the results of blood group serology.

METHODSFive cases of the two-generation pedigrees were analyzed. ABO genotypes were determined using serological tests. DNA sequence analysis was performed on exon 6, exon 7 and intron 3 of the 5 cases to confirm the genotypes of a proband with B subgroup and 4 family members.

RESULTSThere were 3 cases of subgroup AB3 and 1 case of subgroup B3 among the 5 family members. The genotypes were identified as A102/B303 and O02/B303, respectively. B303 differed from B101 by intron 3 point mutation (intron3 + 5G>A).

CONCLUSIONThe point mutation of intron 3 (intron 3+5G>A) is specific in B303.