ABSTRACT
Early administration of proper antibiotics is considered to improve the clinical outcomes of Staphylococcus aureus bacteremia (SAB), but routine clinical antimicrobial susceptibility testing takes an additional 24 h after species identification. Recent studies elucidated matrix-assisted laser desorption/ionization time-of-flight mass spectra to discriminate methicillin-resistant strains (MRSA) or even incorporated with machine learning (ML) techniques. However, no universally applicable mass peaks were revealed, which means that the discrimination model might need to be established or calibrated by local strains' data. Here, a clinically feasible workflow was provided. We collected mass spectra from SAB patients over an 8-month duration and preprocessed by binning with reference peaks. Machine learning models were trained and tested by samples independently of the first six months and the following two months, respectively. The ML models were optimized by genetic algorithm (GA). The accuracy, sensitivity, specificity, and AUC of the independent testing of the best model, i.e., SVM, under the optimal parameters were 87%, 75%, 95%, and 87%, respectively. In summary, almost all resistant results were truly resistant, implying that physicians might escalate antibiotics for MRSA 24 h earlier. This report presents an attainable method for clinical laboratories to build an MRSA model and boost the performance using their local data.
ABSTRACT
The current processes used in clinical microbiology laboratories take ~24 h for incubation to identify the bacteria after the blood culture has been confirmed as positive and fa further ~24 h to report the results of antimicrobial susceptibility tests (ASTs). Patients with suspected bloodstream infection are treated with empiric broad-spectrum antibiotics but delayed targeted antimicrobial therapy. This study aimed to develop a method with a significantly shortened turnaround time for clinical application by identifying the optimal incubation period of a subculture. A total of 188 positive blood culture samples obtained from Nov. 2019 to Aug. 2020 were included. Compared to the conventional 24-h incubation for bacterial identification, our approach achieved 96.1% and 97.4% identification accuracy after shortening the incubation time to 4.5 and 3.5 h for gram-positive (GP) and gram-negative (GN) bacterial samples, respectively. Samples from short-term incubation without any intermediate step or process were directly subjected to analysis with the Phoenix M50 AST. Compared to the conventional disk diffusion AST, the category agreements for GP (excluding Streptococcus spp.), Streptococcus spp., and GN bacterial samples were 91.8%, 97.5%, and 92.7%, respectively. Our approach significantly reduced the average turnaround time from 48 h to 28 h for reporting bacterial identity and decreased average AST from 72 h to 50.3 h compared to the conventional methods. Accordingly, this approach allows a physician to prescribe the appropriate antibiotic(s) ~21.7 h earlier, thereby improving patient outcomes.
ABSTRACT
The production of bioethanol by the conversion of lignocellulosic waste has attracted much interest in recent years because of its low cost and great potential availability. However, the high cost of the enzyme required for this conversion is often considered to be the major bottleneck in the commercial lignocellulosic ethanol industry. In this work, the hydrolysis of rice straw by free and entrapped lignocellulolytic enzymes (cellulase, xylanase and laccase) was carried out at pH 5.5 and 37°C. The hydrolysis of rice straw by enzymes entrapped in a membrane produced a higher monosaccharide content: 601.05 mg/g rice straw for entrapped enzymes vs. 465.46 mg/g rice straw for free enzymes. This study has shown that enzyme entrapment is an important technique for the efficient use and reuse of enzymes in industrial applications and also for the rapid separation of saccharide products from the reaction medium, thus improving the remaining enzymatic activities.
