基于"四联微透析-蒙特卡洛模拟"新方法的β-内酰胺类抗生素治疗脓毒症药动/药效学研究

Sepsis, with its high mortality, is a severe health menace to human being. Early, potent, and broad spectrum antibiotic therapy is an important factor that influences sepsis outcomes, but currently there is no recommended dose for clinical effective use and so empirical therapy is preferred. At the early stage of sepsis, insufficiency of drug concentration at infection target site is the main reason for treatment failure. And it is the crux of the matter that existing sampling techniques cannot accurately capture the suitable concentration. Microdialysis is a new in vivo sampling technique which can monitor the free drug concentration at the target site dynamically and continuously. In this study, a hypothesis that quadruple microdialysis, a new Monte Carlo simulation method, can be used to study the action law of antibiotic at sepsis target parts is put forward. A sepsis CLP classic rat model is to be established, using the first line drug of severe infection treatment piperacillin/tazobactam as model drug and applying "lung-celiac-skin- blood" four points microdialysis sync sampling technique and LC-MS/MS, so that the in vivo pharmacokinetics of drugs can be inspected accurately, and with the application of Monte Carlo simulation drug PK/PD as parameters, the optimal dosage regimen can be determined. The study aims at predicting the therapeutic effectiveness of β-lactam antibiotics and elucidating the distribution regulations of antibiotics in the target site of sepsis, thus providing reference and technology platform for PK/PD study of β-lactam antibiotic therapy for clinical sepsis.

脓毒症病死率高，严重威胁人类健康。早期、强效、广谱抗生素治疗是脓毒症转归重要因素，目前尚无有效临床推荐剂量，多凭经验治疗。脓毒症早期感染靶部位药物浓度不足是治疗失败的主要原因，而现有取样分析技术难以准确监测靶部位药物浓度是问题关键所在。微透析是一种新型活体取样技术，具有在体动态连续监测目标靶部位游离药物浓度的优点。本研究提出以四联微透析-蒙特卡洛模拟新方法研究脓毒症抗生素靶部位作用规律的设想。建立脓毒症CLP经典大鼠模型，以治疗重症感染一线药物哌拉西林/他唑巴坦为模型药，创建"肺-腹腔-皮肤-血液"四位点微透析同步取样技术，联合高效液相串联质谱，能够准确考察药物体内药动学过程，并应用蒙特卡洛模拟药物PK/PD参数，确定最佳给药方案。预测临床β-内酰胺类抗生素疗效，阐明抗生素在脓毒症靶部位分布规律，为临床脓毒症β-内酰胺类抗生素治疗PK/PD研究提供方法借鉴与技术平台。

脓毒症病死率高，严重威胁人类健康。早期、强效、广谱抗生素治疗是脓毒症转归重要因素，目前尚无有效临床推荐剂量，多凭经验治疗。脓毒症早期感染靶部位药物浓度不足是治疗失败的主要原因，而现有取样分析技术难以准确监测靶部位药物浓度是问题关键所在。本课题成功建立四联微透析取样技术，完成对脓毒症CLP经典大鼠模型血液、骨骼肌、肺部、皮肤4靶部位的游离美罗培南、头孢哌酮/舒巴坦钠药物浓度在体、连续、动态取样，监测美罗培南与头孢哌酮/舒巴坦钠在大鼠动物模型的4靶部位药代动力学全过程，时间分辨率高、动物个体间差异小、数据准确度高，运用WinNonlin模拟获得药动学参数、蒙特卡洛模拟获得药效学参数，结果证实模拟人1g美罗培南剂量治疗脓毒症大鼠时，脓毒症大鼠不同靶部位对特定MIC值达不到最优PTA，两步滴定优于延长输注，两者均优于传统输注。模拟人3 g头孢哌酮/舒巴坦钠剂量治疗脓毒症大鼠时，头孢哌酮不能达到有效杀菌作用，舒巴坦仅能于传统输注MIC≤2 mg/L，优化两步输注MIC≤1 mg/L时到达最佳杀菌效果。较正常状态下的大鼠，脓毒症大鼠无论是各靶部位浓度、药动/药效学参数均有较大变化，证实临床脓毒症患者的抗生素用药参考健康人体数据制定给药方案并不科学， 血液药物浓度变化不能反映各靶器官药物浓度，只检测血药的浓度很可能导致我们可能高估或低估了抗生素在靶部位的临床疗效。上述结果可为设计美罗培南、头孢哌酮/舒巴坦钠治疗脓毒症方案的临床医生提供重要的临床前数据参考，具有一定科学意义，同时本课题成功建立的新方法新技术为同类问题的研究提供参考性与借鉴性，具有较强的应用前景。