Domes Christian, Graul Lisa, Frosch Timea, Popp Juergen, Hagel Stefan, Pletz Mathias W, Frosch Torsten
Biophotonics and Biomedical Engineering Group, Technical University Darmstadt, Merckstraße 25, 64283, Darmstadt, Germany.
Leibniz Institute of Photonic Technology, 07745, Jena, Germany.
Commun Med (Lond). 2025 Apr 16;5(1):120. doi: 10.1038/s43856-025-00823-9.
Effective antibiotic therapy in critically ill patients requires precise dosing tailored to individual conditions. However, physiological changes in these patients can complicate drug exposure prediction, leading to treatment failure or toxicity. Therapeutic drug monitoring (TDM) is crucial in optimizing antibiotic therapy, with Raman spectroscopy emerging as a promising method due to its speed and sensitivity.
The utility of resonance Raman spectroscopy in analyzing clinical urine samples was investigated, specifically focusing on piperacillin concentrations. Samples subjected to various preparation techniques, including freezing, centrifugation, and filtration, were analyzed using deep UV resonance Raman spectroscopy. Data analysis involved preprocessing and chemometric modeling to assess concentration changes and the influence of sample matrix.
Sample preparation steps induce concentration changes in piperacillin, with freezing having the highest impact. Chemometric modeling reveals that freezing, filtration, and centrifugation, especially when combined, reduce drug concentration. Furthermore, the choice of urine reference for quantification impacts results, with sex-specific urine pools showing better accuracy compared to mixed pools.
Resonance Raman spectroscopy effectively quantifies piperacillin concentrations in urine. Freezing, centrifugation, and filtration during sample preparation influence drug concentration. Using sex-specific urine pools as references yields more accurate quantification results. These findings underscore the importance of considering sample processing effects and reference selection in TDM studies, offering insights for optimizing antibiotic dosing in critically ill patients. Further validation on a larger scale is warranted to confirm these observations.
重症患者的有效抗生素治疗需要根据个体情况进行精确给药。然而，这些患者的生理变化会使药物暴露预测变得复杂，从而导致治疗失败或毒性反应。治疗药物监测（TDM）对于优化抗生素治疗至关重要，拉曼光谱法因其速度和灵敏度而成为一种很有前景的方法。
研究了共振拉曼光谱法在分析临床尿液样本中的实用性，特别关注哌拉西林浓度。使用深紫外共振拉曼光谱法对经过各种制备技术（包括冷冻、离心和过滤）处理的样本进行分析。数据分析包括预处理和化学计量学建模，以评估浓度变化和样本基质的影响。
样本制备步骤会导致哌拉西林浓度发生变化，其中冷冻的影响最大。化学计量学建模表明，冷冻、过滤和离心，尤其是联合使用时，会降低药物浓度。此外，用于定量的尿液参考物的选择会影响结果，与混合尿液池相比，按性别分类的尿液池显示出更高的准确性。
共振拉曼光谱法能有效定量尿液中的哌拉西林浓度。样本制备过程中的冷冻、离心和过滤会影响药物浓度。使用按性别分类的尿液池作为参考物可产生更准确的定量结果。这些发现强调了在TDM研究中考虑样本处理效应和参考物选择的重要性，为优化重症患者的抗生素给药提供了见解。有必要进行更大规模的进一步验证以证实这些观察结果。
