A machine learning stacking model accurately estimating gastric fluid volume in patients undergoing elective sedated gastrointestinal endoscopy.

BACKGROUND: The current point-of-care ultrasound (POCUS) assessment of gastric fluid volume primarily relies on the traditional linear approach, which often suffers from moderate accuracy. This study aimed to develop an advanced machine learning (ML) model to estimate gastric fluid volume more accurately. METHODS: We retrospectively analyzed the clinical data and POCUS data (D1: craniocaudal diameter, D2: anteroposterior diameter) of 1386 patients undergoing elective sedated gastrointestinal endoscopy (GIE) at Nanjing First Hospital to predict gastric fluid volume using ML techniques, including six different ML models and a stacking model. We evaluated the models using the adjusted Coefficient of Determination (R2), mean absolute error (MAE) and root mean square error (RMSE). The SHapley Additive exPlanations (SHAP) method was used to interpret the importance of the variables. Finally, a web calculator was constructed to facilitate its clinical application. RESULTS: The stacking model (Linear regression + Multilayer perceptron) performed best, with the highest adjusted R2 of 0.718 (0.632 to 0.804). The mean prediction bias was 4 ml (MAE: 4.008 (3.68 to 4.336)), which is better than that of the linear model. D1 and D2 ranked high in the SHAP plot and performed better in the right lateral decubitus (RLD) than in the supine position. The web calculator can be accessed at https://cheason.shinyapps.io/Stacking_regressor/. CONCLUSION: The stacking model and its web calculator can serve as practical tools for accurately estimating gastric fluid volume in patients undergoing elective sedated GIE. It is recommended that anesthesiologists measure D1 and D2 in the patient's RLD position.

Down-regulating nuclear factor of activated T cells 1 alleviates cognitive deficits in a mouse model of sepsis-associated encephalopathy, possibly by stimulating hippocampal neurogenesis.

Sepsis-associated encephalopathy (SAE) is a common complication of sepsis, and has been associated with increased morbidity and mortality. Nuclear factor of activated T cells (NFATs) 1, a transcriptional factor that regulates T cell development, activation and differentiation, has been implicated in neuronal plasticity. Here we examined the potential role of NFAT1 in sepsis-associated encephalopathy in mice. Adult male C57BL/6J mice received intracerebroventricular injections of short interfering RNA against NFAT1 or sex-determining region Y-box 2 (SOX2), or a scrambled control siRNA prior to cecal ligation and perforation (CLP). A group of mice receiving sham surgery were included as an additional control. CLP increased escape latency and decreased the number of crossings into, and total time spent within, the target quadrant in the Morris water maze test. CLP also decreased the freezing time in context-dependent, but not context-independent, fear conditioning test. Knockdown of either NFAT1 or SOX2 attenuated these behavioral deficits. NFAT1 knockdown also attenuated CLP-induced upregulation of SOX2, increased the numbers of nestin-positive cells and newborn astrocytes, reduced the number of immature newborn neurons, and promoted the G1 to S transition of neural stem cells in hippocampus. These findings suggest that NFAT1 may contribute to sepsis-induced behavioral deficits, possibly by promoting SOX2 signaling and neurogenesis.

A convenient machine learning model to predict full stomach and evaluate the safety and comfort improvements of preoperative oral carbohydrate in patients undergoing elective painless gastrointestinal endoscopy.

BACKGROUND AND AIMS: Assessment of the patient's gastric contents is the key to avoiding aspiration incidents, however, there is no effective method to determine whether elective painless gastrointestinal endoscopy (GIE) patients have a full stomach or an empty stomach. And previous studies have shown that preoperative oral carbohydrates (POCs) can improve the discomfort induced by fasting, but there are different perspectives on their safety. This study aimed to develop a convenient, accurate machine learning (ML) model to predict full stomach. And based on the model outcomes, evaluate the safety and comfort improvements of POCs in empty- and full stomach groups. METHODS: We enrolled 1386 painless GIE patients between October 2022 and January 2023 in Nanjing First Hospital, and 1090 patients without POCs were used to construct five different ML models to identify full stomach. The metrics of discrimination and calibration validated the robustness of the models. For the best-performance model, we further interpreted it through SHapley Additive exPlanations (SHAP) and constructed a web calculator to facilitate clinical use. We evaluated the safety and comfort improvements of POCs by propensity score matching (PSM) in the two groups, respectively. RESULTS: Random Forest (RF) model showed the greatest discrimination with the area under the receiver operating characteristic curve (AUROC) 0.837 [95% confidence interval (CI): 79.1-88.2], F1 71.5%, and best calibration with a Brier score of 15.2%. The web calculator can be visited at https://medication.shinyapps.io/RF_model/. PSM results demonstrated that POCs significantly reduced the full stomach incident in empty stomach group (p < 0.05), but no differences in full stomach group (p > 0.05). Comfort improved in both groups and was more significant in empty stomach group. CONCLUSIONS: The developed convenient RF model predicted full stomach with high accuracy and interpretability. POCs were safe and comfortably improved in both groups, with more benefit in empty stomach group. These findings may guide the patients' gastrointestinal preparation.

This study is the first model utilizing advanced ML techniques based on multiple clinical variables to identify full stomach. The model is suitable for patient-rich outpatient clinics, primary hospitals, remote regions, and specific clinical settings where POCUS is not available.The developed convenient RF model predicted full stomach with high accuracy and interpretability. The test cohort AUROC was 0.837. We further established an online accessible individualized risk calculator and provided waterfall plots to increase the interpretability of each prediction.The propensity score matching (PSM) showed that preoperative oral carbohydrate (POCs) were safe and comfortably improved in both groups, with more benefit in empty stomach group. These findings may provide information for anesthesiologists to guide patients on POCs.

Incorporating intraoperative blood pressure time-series variables to assist in prediction of acute kidney injury after type a acute aortic dissection repair: an interpretable machine learning model.

BACKGROUND: Acute kidney injury (AKI) is a common and serious complication after the repair of Type A acute aortic dissection (TA-AAD). However, previous models have failed to account for the impact of blood pressure fluctuations on predictive performance. This study aims to develop machine learning (ML) models combined with intraoperative medicine and blood pressure time-series data to improve the accuracy of early prediction for postoperative AKI risk. METHODS: Indicators reflecting the duration and depth of hypotension were obtained by analyzing continuous mean arterial pressure (MAP) monitored intraoperatively with multiple thresholds (<65, 60, 55, 50) set in the study. The predictive features were selected by logistic regression and the least absolute shrinkage and selection operator (LASSO), and 4 ML models were built based on the above features. The performance of the models was evaluated by area under receiver operating characteristic curve (AUROC), calibration curve and decision curve analysis (DCA). Shapley additive interpretation (SHAP) was used to explain the prediction models. RESULTS: Among the indicators reflecting intraoperative hypotension, 65 mmHg showed a statistically superior difference to other thresholds in patients with or without AKI (p < .001). Among 4 models, the extreme gradient boosting (XGBoost) model demonstrated the highest AUROC: 0.800 (95% 0.683-0.917) and sensitivity: 0.717 in the testing set and was verified the best-performing model. The SHAP summary plot indicated that intraoperative urine output, cumulative time of mean arterial pressure lower than 65 mmHg outside cardiopulmonary bypass (OUT_CPB_MAP_65 time), autologous blood transfusion, and smoking were the top 4 features that contributed to the prediction model. CONCLUSION: With the introduction of intraoperative blood pressure time-series variables, we have developed an interpretable XGBoost model that successfully achieve high accuracy in predicting the risk of AKI after TA-AAD repair, which might aid in the perioperative management of high-risk patients, particularly for intraoperative hemodynamic regulation.

In this study, we combined intraoperative blood pressure time-series data for the first time to build 4 machine learning (ML) models that successfully improve the accuracy of early prediction of postoperative AKI risk, with the XGBoost model displaying the best predictive performance.We explored the impact of multiple intraoperative hypotension thresholds (MAP <65, <60, <55 < 50 mmHg) on the occurrence of postoperative AKI in patients and attempted to provide clinicians with recommendations for hemodynamic management during surgery.Our study found that 65 mmHg showed a statistically superior difference to other thresholds in patients with or without AKI after undergoing TA-AAD repair (p < .001).

[Phosphate and tension homology-induced kinase 1/Parkin signaling mediates cognitive dysfunction in sepsis-associated encephalopathy through activation of hippocampal mitochondrial autophagy].

OBJECTIVE: To investigate the effects of gene of phosphate and tension homology (PTEN)-induced putative kinase 1 (PINK1)/Parkin pathway on hippocampal mitophagy and cognitive function in mice with sepsis-associated encephalopathy (SAE) and its possible mechanism. METHODS: A total of 80 male C57BL/6J mice were randomly divided into Sham group, cecal ligation puncture (CLP) group, PINK1 plasmid transfection pretreatment groups (p-PINK1+Sham group, p-PINK1+CLP group), empty vector plasmid transfection control group (p-vector+CLP group), with 16 mice in each group. The mice in CLP groups were treated with CLP to reproduce SAE models. The mice in the Sham groups were performed laparotomy only. Animals in the p-PINK1+Sham and p-PINK1+CLP groups were transfected with PINK1 plasmid through the lateral ventricle at 24 hours before surgery, while mice in the p-vector+CLP group were transfected with the empty plasmid. Morris water maze experiment was performed 7 days after CLP. The hippocampal tissues were collected, the pathological changes were observed under a light microscope after hematoxylin-eosin (HE) staining, and the mitochondrial autophagy was observed under a transmission electron microscopy after uranyl acetate and lead citrate staining. The expressions of PINK1, Parkin, Beclin1, interleukins (IL-6, IL-1ß) and microtubule-associated protein 1 light chain 3 (LC3) were detected by Western blotting. RESULTS: Compared with the Sham group, CLP group mice in Morris water maze experiment had longer escape latency, shorter target quadrant residence time, and fewer times of crossing the platform at 1-4 days. Under the light microscope, the hippocampal structure of the mouse was injured, the neuronal cells were arranged in disorder, and the nuclei were pyknotic. Under the electron microscope, the mitochondria appeared swollen, round, and wrapped by bilayer or multilayer membrane structures. Compared with the Sham group, CLP group had higher expressions of PINK1, Parkin, Beclin1, LC3II/LC3I ratio, IL-6 and IL-1ß in hippocampus, indicating that sepsis induced by CLP could activated inflammatory response and caused PINK1/Parkin-mediated mitophagy. Compared with the CLP group, p-PINK1+CLP group had shorter escape latencies, spent more time in the target quadrant and had more number of crossings in the target quadrant at 1-4 days. Under the light microscope, the hippocampal structures of mice was destroyed, the neurons were arranged disorderly, and the nuclei were pyknotic. Under transmission electron microscope, swollen and rounded mitochondria and mitochondrial structure wrapped by double membrane or multilayer membrane structure were observed. Compared with the CLP group, the levels of PINK1, Parkin, Beclin1 and LC3II/LC3 ratio in the p-PINK1+CLP group were significantly increased [PINK1 protein (PINK1/ß-actin): 1.95±0.17 vs. 1.74±0.15, Parkin protein (Parkin/ß-actin): 2.06±0.11 vs. 1.78±0.12, Beclin1 protein (Beclin1/ß-actin): 2.11±0.12 vs. 1.67±0.10, LC3II/LC3I ratio: 3.63±0.12 vs. 2.27±0.10, all P < 0.05], while the levels of IL-6 and IL-1ß were significantly decreased [IL-6 protein (IL-6/ß-actin): 1.69±0.09 vs. 2.00±0.11, IL-1ß protein (IL-1ß/ß-actin): 1.11±0.12 vs. 1.65±0.12, both P < 0.05], suggesting that overexpression of PINK1 protein could further activate mitophagy and reduce the inflammatory response caused by sepsis. There was no statistically significant difference in the above pathological changes and related indicators between Sham group and p-PINK1+Sham group, CLP group and p-vector+CLP group. CONCLUSIONS: PINK1 overexpression can further activate CLP-induced mitophagy by upregulating Parkin, thereby inhibiting inflammation response and alleviate cognitive function impairment in SAE mice.

Role of nuclear factor of activated T Cells-1 in Sepsis-induced behavioral deficits in mice.

INTRODUCTION: The nuclear factor of activated T cells-1 (NFAT1) is involved in both neuroinflammation and cognitive dysfunction. In this study, we examined the role of NFAT1 in sepsis-induced cognitive impairment in a mouse model. METHODS: Sepsis was established in adult mice by cecal ligation and puncture (CLP). Novel object recognition tests on days 14-21 and fear conditioning tests on days 22-23 post-surgery showed that CLP impaired both behaviors. BV2 microglia cells exposed to lipopolysaccharide (LPS) were used to examine the effects of short interfering RNA targeting NFAT1 on autophagy and inflammatory cytokines. RESULTS: CLP increased the expression of NFAT1 in hippocampal microglia and induced hippocampal autophagy by downregulating p62, upregulating beclin-1 and autophagy-related gene-5, and increasing the ratio of microtubule-associated protein 1 light chain 3-I (LC3-I) to LC3-II. In addition, CLP shifted microglial polarization from M2 to M1 and the production of inflammatory cytokines, similar to the effects of lipopolysaccharide on BV2 microglia cells. Conversely, NFAT1 knockdown or the autophagy inhibitor 3-methyladenine attenuated the effects of CLP on autophagy and inflammation in vitro and in vivo, while rapamycin partially reversed the protective effects of NFAT1 inhibition. CONCLUSION: This study suggests that NFAT1 downregulation attenuates sepsis-induced behavioral deficits by inhibiting autophagy, microglia polarization, and neuroinflammation..

Short Chain Fatty Acids Protect the Cognitive Function of Sepsis Associated Encephalopathy Mice via GPR43.

Objective: This study aims to analyze the changes of fecal short chain fatty acids (SCFAs) content and gut microbiota composition in sepsis associated encephalopathy (SAE) mice, further evaluating the effect of SCFAs on cognitive function and the underlying mechanism in SAE mice. Methods: A total of 55 male adult C57BL/6 mice (2-3 months of age, 20-25 g) were divided into four groups randomly: sham group (n = 10), cecal ligation and puncture group (CLP group, n = 15), CLP+SCFAs group (n = 15), and CLP+SCFAs+GLPG0974 group (n = 15). Seven days after surgery, fecal samples were collected for microbiota composition and SCFA analysis from 6 mice in each group randomly. Behavioral test was applied to assess cognitive impairment at the same time. After that, mice were sacrificed and brain tissue was harvested for inflammatory cytokines analysis. Results: The levels of acetic acid (.57 ± 0.09 vs 2.00 ± 0.24, p < 0.001) and propionic acid (.32 ± 0.06 vs .66 ± 0.12, p = 0.002) were significantly decreased in the CLP group compared with the sham group. The administration of SCFAs significantly increased the levels of acetic acid (1.51 ± 0.12 vs. 0.57 ± 0.09, p < 0.001) and propionic acid (0.54 ± 0.03 vs. 0.32 ± 0.06, p = 0.033) in CLP+SCFAs group compared with CLP group. Relative abundance of SCFAs-producing bacteria, including Allobaculum (0.16 ± 0.14 vs. 15.21 ± 8.12, p = 0.037), Bacteroides (1.82 ± 0.38 vs. 15.21 ± 5.95, p = 0.002) and Bifidobacterium (0.16 ± 0.06 vs. 2.24 ± 0.48, p = 0.002), significantly decreased in the CLP group compared with the sham group. The behavioral tests suggested that cognitive function was impaired in SAE mice, which could be alleviated by SCFAs pretreatment. ELISA tests indicated that the levels of IL-1ß, IL-6, and TNF-α were elevated in SAE mice and SCFAs could lower them. However, the GPR43 antagonist, GLPG0974, could reverse the cognitive protective effect and anti-neuroinflammation effect of SCFAs. Conclusion: Our study suggested that in SAE, the levels of acetate and propionate decreased significantly, accompanied by gut microbiota dysbiosis, particularly a decrease in SCFAs-producing bacteria. GPR43 was essential for the anti-neuroinflammation and cognitive protective effect of SCFAs in SAE.