Machine Learning for Predicting Risk and Prognosis of Acute Kidney Disease in Critically Ill Elderly Patients During Hospitalization: Internet-Based and Interpretable Model Study.

BACKGROUND: Acute kidney disease (AKD) affects more than half of critically ill elderly patients with acute kidney injury (AKI), which leads to worse short-term outcomes. OBJECTIVE: We aimed to establish 2 machine learning models to predict the risk and prognosis of AKD in the elderly and to deploy the models as online apps. METHODS: Data on elderly patients with AKI (n=3542) and AKD (n=2661) from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database were used to develop 2 models for predicting the AKD risk and in-hospital mortality, respectively. Data collected from Xiangya Hospital of Central South University were for external validation. A bootstrap method was used for internal validation to obtain relatively stable results. We extracted the indicators within 24 hours of the first diagnosis of AKI and the fluctuation range of some indicators, namely delta (day 3 after AKI minus day 1), as features. Six machine learning algorithms were used for modeling; the area under the receiver operating characteristic curve (AUROC), decision curve analysis, and calibration curve for evaluating; Shapley additive explanation (SHAP) analysis for visually interpreting; and the Heroku platform for deploying the best-performing models as web-based apps. RESULTS: For the model of predicting the risk of AKD in elderly patients with AKI during hospitalization, the Light Gradient Boosting Machine (LightGBM) showed the best overall performance in the training (AUROC=0.844, 95% CI 0.831-0.857), internal validation (AUROC=0.853, 95% CI 0.841-0.865), and external (AUROC=0.755, 95% CI 0.699-0.811) cohorts. In addition, LightGBM performed well for the AKD prognostic prediction in the training (AUROC=0.861, 95% CI 0.843-0.878), internal validation (AUROC=0.868, 95% CI 0.851-0.885), and external (AUROC=0.746, 95% CI 0.673-0.820) cohorts. The models deployed as online prediction apps allowed users to predict and provide feedback to submit new data for model iteration. In the importance ranking and correlation visualization of the model's top 10 influencing factors conducted based on the SHAP value, partial dependence plots revealed the optimal cutoff of some interventionable indicators. The top 5 factors predicting the risk of AKD were creatinine on day 3, sepsis, delta blood urea nitrogen (BUN), diastolic blood pressure (DBP), and heart rate, while the top 5 factors determining in-hospital mortality were age, BUN on day 1, vasopressor use, BUN on day 3, and partial pressure of carbon dioxide (PaCO2). CONCLUSIONS: We developed and validated 2 online apps for predicting the risk of AKD and its prognostic mortality in elderly patients, respectively. The top 10 factors that influenced the AKD risk and mortality during hospitalization were identified and explained visually, which might provide useful applications for intelligent management and suggestions for future prospective research.

[Retracted] Downregulation of ROS­FIG inhibits cell proliferation, colony­formation, cell cycle progression, migration and invasion, while inducing apoptosis in intrahepatic cholangiocarcinoma cells.

Following the publication of this paper, it was drawn to the Editors' attention by a concerned reader that the western blotting data shown in Fig. 9 were strikingly similar to data appearing in different form in other articles written by different authors at different research institutes that had either already been published elsewhere prior to the submission of this paper to International Journal of Molecular Medicine, or were under consideration for publication at around the same time. In view of the fact that certain of these data had already apparently been published previously, the Editor of International Journal of Molecular Medicine has decided that this paper should be retracted from the Journal. The authors were asked for an explanation to account for these concerns, but the Editorial Office did not receive a reply. The Editor apologizes to the readership for any inconvenience caused. [International Journal of Molecular Medicine 34: 661­668, 2014; 10.3892/ijmm.2014.1823].

METTL3-mediated N6-methyladenosine modification stimulates mitochondrial damage and ferroptosis of kidney tubular epithelial cells following acute kidney injury by modulating the stabilization of MDM2-p53-LMNB1 axis.

N6-methyladenosine (m6A) and MELLT3 assume a role in the development of acute kidney injury (AKI). However, their mechanism in AKI remains under-explored. On this basis, this study explored the mechanism of MELLT3 in mitochondrial damage and ferroptosis of kidney tubular epithelial cells after AKI. HK-2 cells were induced by lipopolysaccharide (LPS) to simulate AKI, followed by gain and loss of function of genes, detection of mitochondrial damage and ferroptosis indicators, and analysis of gene interactions. An AKI mouse model was developed using the cecal ligation and puncture (CLP) method to investigate the effect of METTL3 knockdown on kidney injury. MDM2 and LMNB1 were upregulated and p53 was downregulated in LPS-treated HK-2 cells. Mechanistically, the E3 ubiquitin ligase MDM2 increased p53 ubiquitination to activate LMNB1. METTL3 knockdown decreased m6A methylation of MDM2, thus diminishing YTHDF1-mediated MDM2 mRNA stability and translation in LPS-treated HK-2 cells. Knockdown of LMNB1, MDM2, or METTL3 reduced NO, MDA, iron ion, and ROS levels as well as mitochondrial damage and raised SOD, GSH, XCT, GPX4, FPN1, and TFR1 levels in LPS-treated HK-2 cells. The in vivo results showed that METTL3 knockdown reduced renal injury and ferroptosis in CLP mice. METTL3 knockdown prevents mitochondrial damage and ferroptosis of kidney tubular epithelial cells after AKI via the MDM2-p53-LMNB1 axis.

Exosomal microRNA-342-5p secreted from adipose-derived mesenchymal stem cells mitigates acute kidney injury in sepsis mice by inhibiting TLR9.

BACKGROUND: Sepsis-related acute kidney injury (AKI) is an inflammatory disease associated with extremely high mortality and health burden. This study explored the possibility of exosomes secreted by adipose-derived mesenchymal stem cells (AMSCs) serving as a carrier for microRNA (miR)-342-5p to alleviate sepsis-related AKI and investigated the possible mechanism. METHODS: Serum was obtained from 30 patients with sepsis-associated AKI and 30 healthy volunteers for the measurement of miR-342-5p, blood urea nitrogen (BUN), and serum creatinine (SCr) levels. For in vitro experiments, AMSCs were transfected with LV-miR-342-5p or LV-miR-67 to acquire miR-342-5p-modified AMSCs and miR-67-modified AMSCs, from which the exosomes (AMSC-Exo-342 and AMSC-Exo-67) were isolated. The human renal proximal tubular epithelial cell line HK-2 was induced by lipopolysaccharide (LPS) to construct a cellular model of sepsis. The expression of Toll-like receptor 9 (TLR9) was also detected in AKI cells and mouse models. The interaction between miR-342-5p and TLR9 was predicted by dual luciferase reporter gene assay. RESULTS: Detection on clinical serum samples showed that BUN, SCr, and TLR9 were elevated and miR-342-5p level was suppressed in the serum of patients with sepsis-associated AKI. Transfection with LV-miR-342-5p reinforced miR-342-5p expression in AMSCs and AMSC-secreted exosomes. miR-342-5p negatively targeted TLR9. LPS treatment enhanced TLR9 expression, reduced miR-342-5p levels, suppressed autophagy, and increased inflammation in HK-2 cells, while the opposite trends were observed in LPS-induced HK-2 cells exposed to AMSC-Exo-342, Rapa, miR-342-5p mimic, or si-TLR9. Additionally, the effects of AMSC-Exo-342 on autophagy and inflammation in LPS-induced cells could be weakened by 3-MA or pcDNA3.1-TLR9 treatment. Injection of AMSC-Exo-342 enhanced autophagy, mitigated kidney injury, suppressed inflammation, and reduced BUN and SCr levels in sepsis-related AKI mouse models. CONCLUSION: miR-342-5p transferred by exosomes from miR-342-5p-modified AMSCs ameliorated AKI by inhibiting TLR9 to accelerate autophagy.

Organizing Pneumonia in A Patient Double-Positive for ANCA and Anti-GBM Antibodies: A Case Report.

Both anti-glomerular basement membrane (GBM) disease and the anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) are common causes of pulmonary-renal syndrome. Organizing pneumonia (OP), a special pattern of interstitial lung disease, is extremely rare either in AAV or anti-GBM disease. We report an old woman presented with OP on a background of co-presentation with both ANCA and anti-GBM antibodies.

Development and deployment of interpretable machine-learning model for predicting in-hospital mortality in elderly patients with acute kidney disease.

BACKGROUND: Acute kidney injury (AKI) is more likely to develop in the elderly admitted to the intensive care unit (ICU). Acute kidney disease (AKD) affects ∼45% of patients with AKI and increases short-term mortality. However, there are no studies on the prognosis of AKD in the elderly. METHODS: Data from 2666 elderly patients with AKD in the Medical Information Mart for Intensive Care IV were used for model development and 535 in the eICU Collaborative Research Database for external validation. Based on 5 machine learning algorithms, 33 noninvasive parameters were extracted as features for modeling. RESULTS: In-hospital mortality of AKD in the elderly was 29.6% and 31.8% in development and validation cohorts, respectively. The comprehensive best-performing algorithm was the support vector machine (SVM), and a simplified online application included only 10 features employing SVM (AUC: 0.810 and 0.776 in the training and external validation cohorts, respectively) was deployed. Model interpretation by SHapley Additive exPlanation (SHAP) values revealed that the difference (AKD day - ICU day) in sequential organ failure assessment (delta SOFA), Glasgow coma scale (GCS), delta GCS, delta peripheral oxygen saturation (SpO2), and SOFA were the top five features associated with prognosis. The optimal target was determined by SHAP values from partial dependence plots. CONCLUSIONS: A web-based tool was externally validated and deployed to predict the early prognosis of AKD in the elderly based on readily available noninvasive parameters, assisting clinicians in intervening with precision and purpose to save lives to the greatest extent.

Establishment and external validation of an online dynamic nomogram for predicting in-hospital death risk in sepsis-associated acute kidney disease.

OBJECTIVES: Approximately one-third of patients with sepsis-associated acute kidney injury (AKI) progress to acute kidney disease (AKD) with higher short-term mortality. We aimed to identify the clinical characteristics that influence in-hospital death in sepsis-associated AKD and develop a nomogram to facilitate early warning. METHODS: Logical regression was applied to screen variables based on clinical data from the Medical Information Mart for Intensive Care IV (MIMIC-IV) database. A nomogram was established to predict in-hospital death risk in patients with sepsis-associated AKD. The eICU Collaborative Research Database (eICU-CRD) was used for external validation. The receiver operating characteristic and calibration curves were used to determine the model's performance. RESULTS: A total of 1,779 patients with sepsis-associated AKD were included from the MIMIC-IV and 344 from the eICU-CRD. Age, Glasgow coma scale score, systolic blood pressure, peripheral oxygen saturation, platelet count, white blood cell count, and bicarbonate levels were significantly correlated with death. The nomogram demonstrated high discrimination in the training (C-index, 0.829; 95% confidence interval [CI] [0.807-0.852]) and testing sets (C-index: 0.760; 95% CI [0.706-0.814]). At the optimal cut-off value of 0.270, the model's sensitivity in the training and validation datasets was 72.8% (95% CI [68.3-76.9%]) and 64.5% (95% CI [54.9-73.4%]), while the specificity was 79.2% (95% CI [76.9-81.4%]) and 74.8% (95% CI [68.7-80.2%]), respectively. CONCLUSION: We identified seven predictors of in-hospital death in patients with sepsis-associated AKD. In addition, we developed an online dynamic nomogram to accurately and conveniently predict short-term outcomes, which performed well in the external dataset.

Sarcopenia and Mortality Risk of Patients with Sepsis: A Meta-Analysis.

Background: The association between sarcopenia at admission and mortality in patients with sepsis has not been comprehensively evaluated. We performed a meta-analysis to systematically evaluate the above association. Methods: This meta-analysis included relevant observational studies from Medline, Embase, and Web of Science databases. A random-effect model after incorporation of the intrastudy heterogeneity was selected to pool the results. Subgroup analyses were applied to evaluate the influences of study characteristics on relationship. Results: Ten cohort studies including 2396 patients with sepsis were included, and 1496 (62.4%) of them had sarcopenia at presentation. Pooled results showed that compared to those without sarcopenia, septic patients with sarcopenia had a significantly increased early (in-hospital or 1-month) mortality risk (risk ration (RR): 2.14, 95% confidence interval (CI): 1.60-2.87, P < 0.001; I 2 = 46%). Subgroup analyses showed consistent association between sarcopenia and increased acute mortality risk in septic patients which were not affected by study characteristics such as study design, country of the study, clinical settings, diagnostic criteria for sepsis, age, gender of the patients, and methods for diagnosis of sarcopenia (P for all subgroup analyses >0.05). Further meta-analyses showed that sarcopenia was also associated with increased mortality risk in septic patients at 3-6 months (RR: 2.13, 95% CI: 1.58-2.89, P < 0.001; I 2 = 0%) and at 1 year (RR: 1.57, 95% CI: 1.09-2.24, P = 0.01; I 2 = 29%). Conclusions: Current evidence suggests that sarcopenia may be a predictor of mortality in patients with sepsis.

Depressive and Anxiety Symptoms of Healthcare Workers in Intensive Care Unit Under the COVID-19 Epidemic: An Online Cross-Sectional Study in China.

Background: Since the coronavirus disease-2019 (COVID-19) outbreak, intensive care unit (ICU) healthcare workers were responsible for the critical infected patients. However, few studies focused on the mental health of ICU healthcare workers. This study aimed to investigate the psychological impact of COVID-19 on ICU healthcare workers in China. Methods: We distributed the nine-item Patient Health Questionnaire (PHQ-9) and seven-item General Anxiety Disorder questionnaire (GAD-7) online to ICU healthcare workers in China. Respondents were divided into frontline and second-line according to whether they have contact with COVID-19 patients. Depressive and anxiety symptoms of all respondents were evaluated based on their questionnaire scores. Results: There were 731 ICU healthcare workers finally enrolled in our study, including 303 (41.5%) male, 383 (52.4%) doctors, and 617 (84.4%) aged 26-45 years. All in all, 482 (65.9%) ICU healthcare workers reported symptoms of depression, while 429 (58.7%) reported anxiety. There was no significant difference between frontline (n = 325) and second-line (n = 406) respondents in depression (P = 0.15) and anxiety severity (P = 0.56). Logistic regression analysis showed that being female, ICU work time >5 years, and night duty number ≥10 were risk factors of developing depressive and anxiety symptoms. Income reduction was separately identified as risk of anxiety. Additionally, ICU work time >5 years was also identified as risk of developing moderate-severe depressive and anxiety symptoms. Conclusions: Frontline ICU work was not associated with higher risk of depressive and anxiety symptoms during COVID-19 pandemic remission period in China. Actions like controlling night duty number, ensuring vacation, and increasing income should be taken to relieve mental health problem. Furthermore, we should pay close attention to those who had worked long years in ICU.

Common Carotid Artery Sonography Versus Transthoracic Echocardiography for Cardiac Output Measurements in Intensive Care Unit Patients.

OBJECTIVES: This study was designed to test the effectiveness of common carotid artery sonography in comparison with transthoracic echocardiography (TTE) for cardiac output measurements to provide an easier alternative for cardiac output monitoring in the intensive care unit. METHODS: This study included 148 patients who had common carotid artery Doppler examinations and TTE performed within 8 hours of each other, and the cardiac output measurement results were compared with each other. RESULTS: The mean age of the participants ± SD was 56.8 ± 16.2 years, with male patients composing 54.7% of the cohort. There was no significant difference in carotid and TTE cardiac output between different sexes, age groups, patients with and without mechanical ventilation, and primary indication groups. The overall intraclass correlation coefficient between the carotid and TTE cardiac output was 0.537. In patients with septic shock, multiple trauma, and respiratory failure, the intraclass correlation coefficients between TTE and carotid cardiac output were 0.241, 0.061, and 0.095, respectively. CONCLUSIONS: Carotid cardiac output shows moderate agreement with TTE cardiac output; thus, its use may be considered as an alternative for estimating cardiac output in emergencies and when TTE cardiac output is unobtainable. However, in patients with septic shock, multiple trauma, and respiratory failure, the use of carotid cardiac output is not recommended.

A study on new method of noninvasive esophageal venous pressure measurement based on the airflow and laser detection technology.

Using optics combined with automatic control and computer real-time image detection technology, a novel noninvasive method of noncontact pressure manometry was developed based on the airflow and laser detection technology in this study. The new esophageal venous pressure measurement system was tested in-vitro experiments. A stable and adjustable pulse stream was produced from a self-developed pump and a laser emitting apparatus could generate optical signals which can be captured by image acquisition and analysis system program. A synchronization system simultaneous measured the changes of air pressure and the deformation of the vein wall to capture the vascular deformation while simultaneously record the current pressure value. The results of this study indicated that the pressure values tested by the new method have good correlation with the actual pressure value in animal experiments. The new method of noninvasive pressure measurement based on the airflow and laser detection technology is accurate, feasible, repeatable and has a good application prospects.

[A new system for noninvasive esophageal varices pressure measurement based on airflow and laser technology].

OBJECTIVE: Combined the optical principle with automatic control technology and computer real-time image detection technology to develop a non-contact system for noninvasive esophageal varices pressure measurement. METHODS: The system included the adjustable air pump, laser device, image collection and analysis program. The feasibility and accuracy of the system were verified by in vitro experiments. RESULTS: The bionic vascular pressure measured by this system had good correlation and repeatability with the actual pressure. CONCLUSIONS: This system is accurate, feasible and has good application prospects.

Downregulation of ROS-FIG inhibits cell proliferation, colony­formation, cell cycle progression, migration and invasion, while inducing apoptosis in intrahepatic cholangiocarcinoma cells.

Intrahepatic cholangiocarcinoma (ICC) is the second most common primary liver cancer with poor responsiveness to existing drug therapies. Therefore, novel treatment strategies against ICC are required to improve survival. The aim of this study was to demonstrate the role of fused-in-glioblastoma-c-ros-oncogene1 (FIG-ROS) fusion gene in ICC. ROS was positively expressed in ICC tissues and HUCCT1 cells. Plasmids expressing ROS- and FIG-specific shRNAs were constructed and transfected into HUCCT1 cells. The results showed that single transfection of ROS- or FIG-specific shRNA inhibited HUCCT1 cell proliferation, colony formation, cell cycle progression, migration and invasion, while inducing apoptosis. Moreover, the co-inhibition of ROS- and FIG-specific shRNA exhibited stronger effects on HUCCT1 cell proliferation, apoptosis, colony formation, cell cycle progression, migration and invasion, when compared to single inhibition of ROS and FIG. Furthermore, findings of this study suggested that the AKT signaling pathway was involved in the ROS-FIG-mediated biological processes of HUCCT1 cells. In summary, the results suggest that FIG-ROS plays an oncogenic role in ICC. Additionally, ROS1-6290 and FIG-363 segments may become effective therapeutic targets for ICC harboring ROS-FIG fusion protein.

miR-31 promotes oncogenesis in intrahepatic cholangiocarcinoma cells via the direct suppression of RASA1.

MicroRNAs (miRNAs) are involved in the pathogenesis of intrahepatic cholangiocarcinoma (ICC). However, the role of microRNA-31 (miR-31) in ICC has yet to be elucidated. In this study, we demonstrated that the expression of miR-31 was significantly upregulated in ICC tissues and the human ICC cell line HCCC-9810, when compared with that in normal adjacent tissues. Bioinformatic analysis and a dual-luciferase reporter assay revealed RAS p21 GTPase activating protein 1 (RASA1) to be a direct target of miR-31 in HCCC-9810 cells. Further investigation showed that the protein expression level of RASA1 was significantly decreased in ICC tissues, suggesting an inverse correlation between miR-31 and RASA1 expression during the tumorigenesis of ICC. Moreover, the forced downregulation of miR-31 by its inhibitor in HCCC-9810 cells significantly inhibited cell proliferation and promoted cell apoptosis. However, when the cells were cotransfected with miR-31 inhibitor and RASA1-specific small interfering RNA (siRNA), these changes were attenuated. Further analysis of the molecular mechanism showed that the activity of the RAS-mitogen-activated protein kinase (MAPK) signaling pathway was significantly decreased in miR-31-downregulated HCCC-8910 cells, while cotransfection with miR-31 inhibitor and RASA1-specific siRNA attenuated this effect. These results indicate that the downregulation of RASA1 by miR-31 promoted cellular proliferation and inhibited cellular apoptosis, partially by upregulating the activity of the RAS-MAPK signaling pathway in ICC. In conclusion, the present study revealed important regulatory functions of miR-31 and RASA1 in ICC, indicating that miR-31 and RASA1 may become promising diagnostic and/or therapeutic targets for ICC.