Machine learning-based prediction of in-hospital mortality for critically ill patients with sepsis-associated acute kidney injury.

OBJECTIVES: This study aims to develop and validate a prediction model in-hospital mortality in critically ill patients with sepsis-associated acute kidney injury (SA-AKI) based on machine learning algorithms. METHODS: Patients who met the criteria for inclusion were identified in the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database and divided according to the validation (n = 2440) and development (n = 9756, 80%) queues. Ensemble stepwise feature selection method was used to screen for effective features. The prediction models of short-term mortality were developed by seven machine learning algorithms. Ten-fold cross-validation was used to verify the performance of the algorithm in the development queue. The area under the receiver operating characteristic curve (ROC-AUC) was used to evaluate the differentiation accuracy and performance of the prediction model in the validation queue. The best-performing model was interpreted by Shapley additive explanations (SHAP). RESULTS: A total of 12,196 patients were enrolled in this study. Eleven variables were finally chosen to develop the prediction model. The AUC of the random forest (RF) model was the highest value both in the Ten-fold cross-validation and evaluation (AUC: 0.798, 95% CI: 0.774-0.821). According to the SHAP plots, old age, low Glasgow Coma Scale (GCS) score, high AKI stage, reduced urine output, high Simplified Acute Physiology Score (SAPS II), high respiratory rate, low temperature, low absolute lymphocyte count, high creatinine level, dysnatremia, and low body mass index (BMI) increased the risk of poor prognosis. CONCLUSIONS: The RF model developed in this study is a good predictor of in-hospital mortality for patients with SA-AKI in the intensive care unit (ICU), which may have potential applications in mortality prediction.

Characteristics and function of the gut microbiota in patients with IgA nephropathy via metagenomic sequencing technology.

OBJECTIVE: The aim of this study was to investigate the characteristics and related functional pathways of the gut microbiota in patients with IgA nephropathy (IgAN) through metagenomic sequencing technology. METHODS: We enrolled individuals with primary IgAN, including patients with normal and abnormal renal function. Additionally, we recruited healthy volunteers as the healthy control group. Stool samples were collected, and species and functional annotation were performed through fecal metagenome sequencing. We employed linear discriminant analysis effect size (LEfSe) analysis to identify significantly different bacterial microbiota and functional pathways. Kyoto Encyclopedia of Genes and Genomes (KEGG) analysis was used to annotate microbiota functions, and redundancy analysis (RDA) was performed to analyze the factors affecting the composition and distribution of the gut microbiota. RESULTS: LEfSe analysis revealed differences in the gut microbiota between IgAN patients and healthy controls. The characteristic microorganisms in the IgAN group were classified as Escherichia coli, with a significantly greater abundance than that in the healthy control group (p < 0.05). The characteristic microorganisms in the IgAN group with abnormal renal function were identified as Enterococcaceae, Moraxella, Moraxella, and Acinetobacter. KEGG functional analysis demonstrated that the functional pathways of the microbiota that differed between IgAN patients and healthy controls were related primarily to bile acid metabolism. CONCLUSIONS: The status of the gut microbiota is closely associated not only with the onset of IgAN but also with the renal function of IgAN patients. The characteristic gut microbiota may serve as a promising diagnostic biomarker and therapeutic target for IgAN.

Electropositive Citric Acid-Polyethyleneimine Carbon Dots Carrying the PINK1 Gene Regulate ATP-Related Metabolic Dysfunction in APP/PS1-N2a Cells.

(1) Background: Alzheimer's disease (AD) is characterized by ß-amyloid (Aß) peptide accumulation and mitochondrial dysfunction during the early stage of disease. PINK1 regulates the balance between mitochondrial homeostasis and bioenergy supply and demand via the PINK1/Parkin pathway, Na+/Ca2+ exchange, and other pathways. (2) Methods: In this study, we synthesized positively charged carbon dots (CA-PEI CDs) using citric acid (CA) and polyethyleneimine (PEI) and used them as vectors to express PINK1 genes in the APP/PS1-N2a cell line to determine mitochondrial function, electron transport chain (ETC) activity, and ATP-related metabolomics. (3) Results: Our findings showed that the CA-PEI CDs exhibit the characteristics of photoluminescence, low toxicity, and concentrated DNA. They are ideal biological carriers for gene delivery. PINK1 overexpression significantly increased the mitochondrial membrane potential in APP/PS1-N2a cells and reduced reactive-oxygen-species generation and Aß1-40 and Aß1-42 levels. An increase in the activity of NADH ubiquinone oxidoreductase (complex I, CI) and cytochrome C oxidase (complex IV, CIV) induces the oxidative phosphorylation of mitochondria, increasing ATP generation. (4) Conclusions: These findings indicate that the PINK gene can alleviate AD by increasing bioenergetic metabolism, reducing Aß1-40 and Aß1-42, and increasing ATP production.

Under-utilization of statins among people with HIV who were aged 40 years or older.

From June 2022 to April 2023, 1629 HIV-positive participants were assessed for the risk of atherosclerotic cardiovascular disease (ASCVD). The 10-year ASCVD risk of <5 %, 5 % to <7.5 %, ≥7.5 % to <20 % and ≥20 % were 59.9 %, 14.4 %, 20.7 % and 5.0 %, respectively; 440 (27.0 %) participants met the criteria for statin therapy, but only 171 (38.8 %) were prescribed statins.

Hypogonadism among HIV-positive men who have sex with men in Taiwan: Prevalence and associated factors.

BACKGROUND: Male hypogonadism is not uncommon in people with HIV (PWH), with estimated prevalence ranging from 9% to 16%. Existing data are limited on the serum testosterone levels in PWH in Asian populations. METHODS: We enrolled HIV-positive men who have sex with men (MSM) and had been on stable antiretroviral therapy and MSM without HIV between February 2021 and November 2022. Serum free testosterone levels, sex hormone-binding globulins and other associated hormones were measured. Multiple linear regression analysis was performed to assess the association between serum free testosterone levels and clinical variables collected. RESULTS: A total of 447 MSM with HIV and 124 MSM without HIV were enrolled. Compared with MSM without HIV, MSM with HIV had a higher age (median, 41 versus 29.5 years) and prevalence of symptomatic hypogonadism (8.3% versus 1.6%). Among MSM who were aged <35 years, there were no significant differences in the serum free testosterone levels and prevalences of hypogonadism between the two groups. In multiple linear regression analysis, serum free testosterone level significantly decreased with advanced age (a decrease of 1.14 pg/mL per 1-year increase) and a higher body-mass index (BMI) (a decrease of 1.07 pg/mL per 1-kg/m2 increase), but was not associated with HIV serostatus. CONCLUSION: We found that MSM with HIV had a higher prevalence of symptomatic hypogonadism than MSM without HIV in Taiwan, which could be attributed to age difference. Serum free testosterone levels were negatively correlated with age and BMI, but did not show a significant correlation with HIV serostatus.

Effectiveness and evolution of anti-SARS-CoV-2 spike protein titers after three doses of COVID-19 vaccination in people with HIV.

BACKGROUND: Real-world vaccine effectiveness following the third dose of vaccination against SARS-CoV-2 remains less investigated among people with HIV (PWH). METHODS: PWH receiving the third dose of BNT162b2 and mRNA-1273 (either 50- or 100-µg) were enrolled. Participants were followed for 180 days until the fourth dose of COVID-19 vaccination, SARS-CoV-2 infection, seroconversion of anti-nucleocapsid IgG, death, or loss to follow-up. Anti-spike IgG was determined every 1-3 months. RESULTS: Of 1427 participants undergoing the third-dose COVID-19 vaccination, 632 (44.3%) received 100-µg mRNA-1273, 467 (32.8%) 50-µg mRNA-1273, and 328 (23.0%) BNT162b2 vaccine and the respective rate of SARS-CoV-2 infection or seroconversion of anti-nucleocapsid IgG was 246.1, 280.8 and 245.2 per 1000 person-months of follow-up (log-rank test, p = 0.28). Factors associated with achieving anti-S IgG titers >1047 BAU/mL included CD4 count <200 cells/mm3 (adjusted odds ratio [aOR], 0.11; 95% CI, 0.04-0.31), plasma HIV RNA >200 copies/mL (aOR, 0.27; 95% CI, 0.09-0.80), having achieved anti-spike IgG >141 BAU/mL within 3 months after primary vaccination (aOR, 3.69; 95% CI, 2.68-5.07), receiving BNT162b2 vaccine as the third dose (aOR, 0.20; 95% CI, 0.10-0.41; reference, 100-µg mRNA-1273), and having previously received two doses of mRNA vaccine in primary vaccination (aOR, 2.46; 95% CI, 1,75-3.45; reference, no exposure to mRNA vaccine). CONCLUSIONS: PWH receiving different types of the third dose of COVID-19 vaccine showed similar vaccine effectiveness against SARS-CoV-2 infection. An additional dose with 100-µg mRNA-1273 could generate a higher antibody response than with 50-µg mRNA-1273 and BNT162b2 vaccine.