Purpose: This study aimed to establish an optimal model to predict vancomycin trough concentrations by using machine learning. Patients and Methods: We enrolled 407 pediatric patients (age < 18 years) who received vancomycin intravenously and underwent therapeutic drug monitoring from June 2013 to April 2020 at Xinhua Hospital affiliated to Shanghai Jiaotong University School of Medicine. The median (interquartile range) age and weight of the patients were 2 (0.63– 5) years and 12 (7.8– 19) kg. Vancomycin trough concentrations were considered as the target variable, and eight different algorithms were used for predictive performance comparison. The whole dataset (407 cases) was divided into training group and testing group at the ratio of 80%: 20%, which were 325 and 82 cases, respectively. Results: Ultimately, five algorithms (XGBoost, GBRT, Bagging, ExtraTree and decision tree) with high R 2 (0.657, 0.514, 0.468, 0.425 and 0.450, respectively) were selected and further ensembled to establish the final model and achieve an optimal result. For missing data, through filling the missing values and model ensemble, we obtained R 2=0.614, MAE=3.32, MSE=24.39, RMSE=4.94 and a prediction accuracy of 51.22% (predicted trough concentration within ± 30% of the actual trough concentration). In comparison with the pharmacokinetic models (R 2=0.3), the machine learning model works better in model fitting and has better prediction accuracy. Conclusion: Therefore, the ensemble model is useful for the vancomycin concentration prediction, especially in the population of children with great individual variation. As machine learning methods evolve, the clinical value of the ensemble model will be demonstrated in the clinical practice.
19 October 2021 in Pharmacology Research & Perspectives
To assess the pharmacokinetic parameters of vancomycin in Chinese critically ill pediatric patients, children treated with vancomycin, hospitalized in the intensive care unit were included. Samples to determine peak and trough serum concentrations were obtained on the third day of treatment. Half‐life was significantly longer in neonates and showed a decreasing trend in infants and children. In patients aged ≥1 month, AUC24/MIC ≥400 was achieved in 31.8% at the dose of 40 mg/kg/d, and in 48.7% at the dose of 60 mg/kg/d with an assumed MIC of 1 mg/L. Augmented renal clearance (ARC) was present in 27.3% of children, which was associated with higher vancomycin clearance and lower AUC values. A good correlation was observed between trough concentration and AUC24, and the trough concentration that correlated with AUC24 of 400 were varied according to the dosage regimens, 8.42 mg/L for 6‐hintervals, and 6.63 mg/L for 8‐h intervals. To conclude, vancomycin trough concentration that related to the AUC24 of 400 was much lower in critically ill children than that in adults. The dosage of 60 mg/kg/day did not enough for producing AUC24 in the range of 400–600 mg h/L in critically ill children, especially in those with ARC.
Original Article Factors influencing delayed high-dose methotrexate excretion and its correlation with adverse reactions after treatment in children with malignant hematological tumors Liting Yu, Jiayi Shen, Haonan Li, Min Zhang, Zhuo Wang, Yijin Gao, Jihui Chen, Junyu Li
Objectives: To explore the risk factors and clinical outcomes of carbapenem-resistant Klebsiella pneumoniae (CRKP) infection and establish nomograms to predict the probability of CRKP infection and mortality in adult patients.Methods: Patients infected with KP from August 2019 to April 2021 in a tertiary hospital in Shanghai were enrolled. Risk factors associated with CRKP and 30-day mortality were identified using multivariate logistic regression analysis and Cox regression analysis.Results: Overall, 467 patients with KP infection were enrolled, wherein 210 (45.0%) patients were infected with CRKP and 257 (55.0%) patients with carbapenem-susceptible K. pneumoniae (CSKP). Five factors, namely Charlson’s Comorbidity Index (CCI) ≥ 3, the use of central venous catheterization, prior hospitalization during the 3 months before infection, and previous exposure to carbapenems and broad-spectrum β-lactams, were found to be independently associated with CRKP infection. Based on these parameters, the nomogram showed a better performance as indicated by C-index of 0.94 (95% confidence interval [CI]: 0.92–0.96) and well-fitted calibration curves. CRKP was independently associated with 30-day mortality. Multivariate Cox regression analysis revealed that age ≥65 years, higher CCI scores, higher Sequential Organ Failure Assessment scores, the presence of respiratory failure, albumin levels ≤30 g/L, and non-appropriate treatments in 3 days, were associated with 30-day mortality.Conclusion: The predictive nomogram established in this study can facilitate the clinicians to make better clinical decisions when treating patients with KP infection.
Objective: This study was developed to assess the in vivo antimicrobial activity of specific drugs using a model system consisting of Caenorhabditis elegans (C. elegans) infected with Carbapenem-resistant Klebsiella pneumoniae (CRKP) in an effort to identify promising drugs for CRKP-infected patient treatment.Methods: A C. elegans-CRKP liquid assay platform was developed and used to conduct limited in vivo screening for antimicrobial agents with potential activity against CRKP. Time curves for 10 different concentrations of tested antimicrobial agents were tested in this model system at 0, 2, 6, 8, and 12 h after treatment. The protective effects of these different antimicrobial agents were compared at different time points. Furthermore, ten CRKP strains samples were isolated from clinical specimens to demonstrate the applicability of the nematode model method, and two typical clinical cases are presented.Results: CRKP bacteria were sufficient to induce C. elegans death in a dose- and time-dependent fashion, while effective antimicrobial agents improved the survival of these nematodes in a dose-dependent manner. Notably, PB and TGC exhibited robust antibacterial protection within 12 h even at low tested concentrations, and clear efficacy remained evident for high doses of CAZ at this same time point as mediators of improved nematode survival. The results of C. elegans model method were well consistent with that using the Kirby-Bauer method in 10 CRKP strains samples, and two typical clinical cases showed applicability, reliability and efficacy of C. elegans model method.Conclusion: Overall, nematode models in drug sensitivity testing have shown advantages in clinical settings. Our results highlight the value of C. elegans model systems as tools for the simultaneous screening of different agents for in vivo antibacterial efficacy and are deserved further study.