Machine learning for the prediction of delirium in elderly intensive care unit patients.

PURPOSE: This study aims to develop and validate a prediction model for delirium in elderly ICU patients and help clinicians identify high-risk patients at the early stage. METHODS: Patients admitted to ICU for at least 24 h and using the Confusion Assessment Method for the Intensive Care Unit (CAM-ICU) in the Medical Information Mart for Intensive Care-IV (MIMIC-IV) database (76,943 ICU stays from 2008 to 2019) were considered. Patients with a positive delirium test in the first 24 h and under 65 years of age were excluded. Two prediction models, machine learning extreme gradient boosting (XGBoost) and logistic regression (LR) model, were developed and validated to predict the onset of delirium. RESULTS: Of the 18,760 patients included in the analysis, 3463(18.5%) were delirium positive. A total of 22 significant predictors were selected by LASSO regression. The XGBoost model demonstrated superior performance over the LR model, with the Area Under the Receiver Operating Characteristic (AUC) values of 0.853 (95% confidence interval [CI] 0.846-0.861) and 0.831 (95% CI 0.815-0.847) in the training and testing datasets, respectively. Moreover, the XGBoost model outperformed the LR model in both calibration and clinical utility. The top five predictors associated with the onset of delirium were sequential organ failure assessment (SOFA), infection, minimum platelets, maximum systolic blood pressure (SBP), and maximum temperature. CONCLUSION: The XGBoost model demonstrated good predictive performance for delirium among elderly ICU patients, thus assisting clinicians in identifying high-risk patients at the early stage and implementing targeted interventions to improve outcome.

The Role of Tryptophan Metabolism in the Occurrence and Progression of Acute and Chronic Kidney Diseases.

Acute kidney injury (AKI) and chronic kidney disease (CKD) are common kidney diseases in clinics with high morbidity and mortality, but their pathogenesis is intricate. Tryptophan (Trp) is a fundamental amino acid for humans, and its metabolism produces various bioactive substances involved in the pathophysiology of AKI and CKD. Metabolomic studies manifest that Trp metabolites like kynurenine (KYN), 5-hydroxyindoleacetic acid (5-HIAA), and indoxyl sulfate (IS) increase in AKI or CKD and act as biomarkers that facilitate the early identification of diseases. Meanwhile, KYN and IS act as ligands to exacerbate kidney damage by activating aryl hydrocarbon receptor (AhR) signal transduction. The reduction of renal function can cause the accumulation of Trp metabolites which in turn accelerate the progression of AKI or CKD. Besides, gut dysbiosis induces the expansion of Enterobacteriaceae family to produce excessive IS, which cannot be excreted due to the deterioration of renal function. The application of Trp metabolism as a target in AKI and CKD will also be elaborated. Thus, this study aims to elucidate Trp metabolism in the development of AKI and CKD, and explores the relative treatment strategies by targeting Trp from the perspective of metabolomics to provide a reference for their diagnosis and prevention.

Diagnostic accuracy of the 3-minute diagnostic interview for confusion assessment method-defined delirium in delirium detection: a systematic review and meta-analysis.

BACKGROUND: Delirium is a common complication clinically and is associated with the poor outcomes, yet it is frequently unrecognised and readily disregarded. Although the 3-minute diagnostic interview for confusion assessment method-defined delirium (3D-CAM) has been used in a variety of care settings, a comprehensive evaluation of its accuracy in all available care settings has not been performed. OBJECTIVE: This study aimed to evaluate the diagnostic test accuracy of the 3D-CAM in delirium detection through a systematic review and meta-analysis. METHODS: We systematically searched PubMed, EMBASE, the Cochrane Library, Web of Science, CINAHL (EBSCO) and ClinicalTrials.gov published from inception to 10 July 2022. The quality assessment of the diagnostic accuracy studies-2 tool was applied to evaluate methodological quality. A bivariate random effects model was used to pool sensitivity and specificity. RESULTS: Seven studies with 1,350 participants and 2,499 assessments were included, which were carried out in general medical wards, intensive care units, internal medical wards, surgical wards, recovery rooms and post-anaesthesia care units. The prevalence of delirium ranged from 9.1% to 25%. The pooled sensitivity and specificity were 0.92 (95% confidence interval [CI] 0.87-0.95) and 0.95 (95% CI 0.92-0.97), respectively. The pooled positive likelihood ratio was 18.6 (95% CI 12.2-28.2), the negative likelihood ratio was 0.09 (95% CI 0.06-0.14) and the diagnostic odds ratio was 211 (95% CI 128-349). Moreover, the area under the curve was 0.97 (95% CI 0.95-0.98). CONCLUSIONS: The 3D-CAM has good diagnostic accuracy for delirium detection in different care settings. Further analyses illustrated that it had comparable diagnostic accuracy in older adults and patients with dementia or known baseline cognitive impairment. In conclusion, the 3D-CAM is recommended for clinical delirium detection.

Advanced nitrogen elimination from domestic sewage through two stage partial nitrification and denitrification (PND) coupled with simultaneous anaerobic ammonia oxidation and denitrification (SAD).

The start-up, efficient, and secure operation of Anammox treating low ammonia sewage, is an important research focus. In this study, a partial nitrification-denitrification coupled with simultaneous Anammox and denitrification (PND-SAD) process was achieved in sequencing batch reactor/up-flow anaerobic sludge bed (SBR-UASB). The key measures to maintain high efficiency PND were: (i) controlling dissolved oxygen in the SBR below 0.5 mg/L, which is not only conducive to PN, but also promotes the contribution of simultaneous nitrification and denitrification to nitrogen removal; (ii) monitoring the nitrate (NO3--N) of SBR effluent and discharging sludge to wash out nitrate oxidation bacteria when the NO3--N exceeds 1.0 mg/L. The nitrite accumulation rate reached 97.6%. SBR effluent and domestic sewage entered the UASB. Although Candidatus Brocadia only accounted for 0.8%, its contribution to nitrogen removal reached 76.8%. In PND-SAD system, the aerobic HRT was only 3.8 h, nitrogen removal efficiency up to 97.3%.

Distribution characteristics, bioaccumulation and trophic transfer of heavy metals in the food web of grassland ecosystems.

In this study, according to the classification of biological "classes" and the different trophic levels of the food web, the distribution characteristics, bioaccumulation of heavy metals (HMs) and their trophic transfer in the food web of typical grassland ecosystems were studied and predicted. The results indicated that the accumulation of toxic As was the highest in small mammals and reptiles, Cu was the highest in insects, and the micronutrient Zn in large mammals was higher than that in plants. The metal transfer factor (MTF) by plants at the first trophic level showed that Leymus chinensis had the best ability to absorb HMs from soil. The trophic transfer factor (TTF) of HMs in the second-trophic level insects, birds and some mammals were Zn > As > Cu > Ni > Pb > Co = Cr > Mn > V, in which, biomagnified on Zn, As, and Cu. Organisms at the third trophic level including birds, reptiles and some mammals had the strongest accumulation ability for Pb, V and As, and all were biomagnified. The biomagnification on As and Co of the fourth trophic level Siberian weasel was obviously higher than that of Dione's rat-snake, which had significant biomagnification effect on As by preying on Steppe toad-headed agama. The study showed that the bioaccumulation levels of HMs in organisms at different trophic levels varied significantly with species, prey, and organ type, but they all showed strong bioaccumulation capacity to toxic As, which indicated that As could produce certain toxic effects on animals in the food web through trophic transfer. In addition, organisms at low-trophic levels were more likely to biomagnify Zn, while organisms at high-trophic levels were more likely to biodilute Pb.

[Study on effect of jiangya tongmai recipe on vascular activating substances in patients of hypertension with left ventricular hypertrophy].

OBJECTIVE: To study the effect of Jiangya Tongmai Recipe (JYTMR), a Chinese herbal medicine preparation for activating blood circulation to remove stasis, on vascular activating substances in treating patients of hypertension with left ventricular hypertrophy. METHODS: The 37 patients with hypertension were randomly divided into two groups, the treated group (n = 21) and the control group (n = 16). They were treated with JYTMR and captopril respectively for 8 weeks. Left ventricular mass weight (LVMI), peak flow velocity of early diastole (Emax), peak flow velocity of atrial contraction (Amax), Emax/Amax, ejection fraction (EF), as well as levels of plasma endothelin (ET), calcitonin gene-related peptide (CGRP) and angiotensin II (Ang II) were measured before and after treatment and compared. RESULTS: JYTMR could enhance the left ventricular dilation and contractile functions, lower the levels of plasma ET and Ang II and increase the level of plasma CGRP. CONCLUSION: JYTMR shows good effect in improving left ventricular function and regulating vascular activating substances, it could prevent and treat hypertension and its complications for prolonged treatment via multiple paths and links.