延長piperacillin/tazobactam靜脈輸注時間在重症病人之藥動學及藥效學研究

Abstract

背景： 感染症是常導致病人住院或死亡的重要原因，尤其是革蘭氏陰性菌造成的感染致死率高。隨著菌株抗藥性日益增加，若將現有之抗生素依照其藥動、藥效學特性最佳化，才能成功地治療感染症。Piperacillin/tazobactam臨床上常用於治療革蘭氏陰性菌的感染，其藥動學及藥效學性質為「時間依賴型」（time-dependent），自由型（free form）抗生素的血中濃度若能維持在病原菌的最小抑菌濃度（minimal inhibitory concentration，MIC）以上愈久，相對於每次給藥間隔（dosing interval）的比例愈高（%fT>MIC），治療感染症的效果愈好。國際上有許多以藥動學及藥效學模式模擬的研究，但完全以重症病人為研究對象的文獻相當少見。 研究目的： 分析在重症病人延長給予piperacillin/tazobactam輸注時間對藥動、藥效學參數的影響及與一般病人之差異。比較在重症病人延長輸注時間及傳統給藥方式給予piperacillin/ tazobactam後的piperacillin之% fT>MIC是否有顯著差異，並模擬不同給藥方式的藥效學指標達成率（probability of target attainment，PTA）。本研究同時評估重症病人延長輸注時間給予piperacillin/tazobactam時的臨床療效及安全性。 研究方法： 本研究為一前瞻性、開放性、隨機分配之藥品動態學及藥效學之臨床研究，自2012年2月20日至2012年7月31日在臺大醫院內科及呼吸加護病房進行收案。收納條件為大於18歲、住在加護病房、肌酐酸廓清率大於20 mL/min、且預期使用piperacillin/tazobactam藥品大於3天之病人。給藥劑量依照醫師根據病人感染部位及腎功能給予，病人簽署受試者同意書後，隨機分派至藥品靜脈輸注時間4小時（EI）或1小時（II）的組別，並皆於達到穩定濃度後的一個給藥區間抽取6個不同時間的血液檢體，利用高效能液相層析儀（high-performance liquid chromatography，HPLC）分析血中濃度，並利用WinNonlin之non-compartment model分析藥動學參數及藥效學結果，以蒙地卡羅模擬進行PTA的分析。 研究結果： 本研究共收案24人，兩組各12位病人，平均年齡為76.91 ± 10.53歲，男女比例14：10，平均體重53.78 ± 10.16公斤。在使用piperacillin/tazobactam時的APACHE II 分數中位數為26（15-32），多數病人以肺部感染為主，占83.3%。 共收集144筆血中濃度資料，使用3克的piperacillin最高及最低濃度平均分別為178.77 ± 63.64 mg/L及60.73 ± 51.06 mg/L；而4克則為205.43 ± 98.80 mg/L及56.83 ± 50.24 mg/L。EI組及II組的藥動學參數皆沒有差異，其piperacillin的排除速率常數分別為0.3 hr-1及0.43 hr-1、半衰期為2.34 hr及1.61 hr、廓清率為4.6 L/hr及6.16 L/hr，而擬似分佈體積為0.32 L/kg及0.30 L/kg。相較於健康人或一般住院病人，重症病人的Vd有偏大、CL偏低、t1/2偏長的趨勢。 在MIC較低時，兩組的% fT>MIC皆將近100%；隨MIC增高，EI組的% fT>MIC有較大的趨勢。當MIC為32 mg/L時，EI組平均有95.47%的時間達到% fT>MIC，而II組平均則有84.77%的時間，p值為0.0282。若MIC更高為64 mg/L或128 mg/L時，雖然EI組平均仍有較高的比例達到% fT>MIC，但都沒有統計顯著差異。蒙地卡羅模擬結果顯示，若給予相同劑量及給藥頻次，則EI組的PTA較高。在MIC為16 或32 mg/L時，4 g Q8H輸注4小時的PTA均高達90%以上，高於輸注1小時的86.96%與65.67%。 兩組病人在14天內死亡率或住院天數沒有顯著差異，也沒有病人發生癲癇或骨髓抑制等與劑量相關的副作用。 結論： 在重症病人使用piperacillin/tazobactam靜脈輸注4小時及1小時，除了血中濃度以外，其他藥動參數沒有差異。相較於健康人或一般住院病人，piperacillin在重症病人有Vd偏大、t1/2偏長的趨勢。EI組血中濃度變化不如II組大，因此病人的實際% fT>MIC在EI組略高，但兩組在MIC低時皆將近100%時間可達到目標，在MIC較高時差異較大，惟並不顯著。蒙地卡羅模擬看到相同趨勢，在MIC較高時，延長靜脈輸注推估可以最適化藥動學及藥效學結果。 在本針對重症病人的研究中，並未如文獻看到延長靜脈輸注比起30分鐘輸注在藥效學的顯著益處，但有觀察到輸注四小時之PTA比輸注一小時有較好的趨勢。儘管在MIC≤32 mg/L時，一個小時的輸注就有理想藥效，但由於在一開始經驗性給予piperacillin/tazobactam時並不知菌株的MIC，因此若病人疾病嚴重程度高或免疫較差，仍希望% fT>MIC越高越好，而使用較長的靜脈輸注時間可有效達到此目的。考量劑量、可行性等其他臨床因素，本研究建議重症病人經驗性可給予4 g Q8H且延長輸注時間至4小時；若有菌株培養結果，且菌株感受性為sensitive時可考慮使用3 g Q8H輸注4小時，以達到較佳的藥動/藥效學參數，而延長靜脈輸注時間給予piperacillin/tazobactam對於臨床療效的影響則需要未來更多的研究證明。

Background: With the increasing resistance of Gram-negative bacteria which is the major cause for morbidity and mortality, optimizing the existing antibiotics according to their pharmacokinetic (PK) and pharmacodynamic (PD) characteristics may optimize the treatment of infectious diseases. Piperacillin/tazobactam (pip/tazo) is one of the β-lactam antibiotics often used in the treatment of infections caused by Gram-negative bacteria. Pip/tazo exhibits“time-dependent”bactericidal activities. The killing effect is proportional to the time of the free form of pip/tazo serum concentration remaining above the minimal inhibitory concentration (MIC) relative to one dosing interval. When the goal of % fT>MIC of pip/tazo exceed 50% of the dosing interval, better treatment outcomes are reported. Many publications have demonstrated PK and PD models via simulation while studies in critically ill patients were few. Purpose: The purpose of this study was to evaluate PK and PD parameters in critically ill patients with extended infusion of pip/tazo. We also compared the % fT>MIC of pip/tazo in critically ill patients with extended infusion and traditional administration in ICU (1 hr-infusion). Furthermore, the probability of target (> 50% fT>MIC) attainment (PTA) in different dosing regimens was estimated via Monte Carlo simulation. The efficacy and safety of extended infusion in intensive care patients was also included in this study. Methods: This was a prospective, open-label, randomized clinical PK study in critically ill patients in medical and respiratory intensive care units (ICUs) at National Taiwan University Hospital from 20th February, 2012 to 31st July, 2012. Patients received pip/tazo were eligible if they were ≥18 years old, staying in an ICU, with CLcr ≥ 20 mL/min and expected on pip/tazo for ≥ 3 days. Dosing regimen was determined by physician according to patient’s infection site and renal function. Patients were randomized to either extended infusion (EI) group or intermittent infusion (II) control group after signing the informed consent. Six blood samples at different time points were collected in one dosing interval at steady state. Non-compartmental model in WinNonlin was used in analyzing PK and PD parameters, and Monte Carlo simulation was conducted for PTA. Results: Twelve patients were allocated to each group. Their average age was 76.91 ± 10.53 years old, female accounted for 41.7% (10/24), and an average weight was 53.78 ± 10.16 kilograms. Median of APACHE II score was 26 (15-32) when pip/tazo was prescribed, and 83.3% was lung infections. A total of 144 serum samples were collected. The average piperacillin Cmax and Cmin of 3 g (4g) was 178.77 ± 63.64 mg/L (205.43 ± 98.80 mg/L) and 60.73 ± 51.06 mg/L (56.83 ± 50.24 mg/L), respectively. The piperacillin elimination rate constant (k), half-life (t1/2), clearance (CL) and volume of distribution (Vd) of EI and II group were 0.3 hr-1 vs. 0.43 hr-1, 2.34 hr vs. 1.61 hr, 4.6 L/hr vs. 6.16 L/hr, 0.32 L/kg vs. 0.30 L/kg, respectively, and there was no statistically significant difference between two groups. These parameters had tendency to be larger in Vd and t1/2, and smaller in CL. The % fT>MIC were approaching to 100% in both groups, and became larger in EI group at higher MIC. When MIC was 32 mg/L, EI group had an average of 95.47% of the time above MIC compares to 84.77% in II group (p = 0.0282). There was no statistically significance between two groups though higher proportion of % fT>MIC in EI group when MIC ≥ 64 mg/L. Results from Monte Carlo simulation showed higher PTA in EI group under the same dosing regimen. The PTAs exceeded 90% for EI of 4 g Q8H compared with 86.96% and 65.67% for II of the same doing regimen at an MIC of 16 or 32 mg/L, respectively. There were no significant differences regarding the 14-day mortality, hospital length of stay, and dose-dependent side effects such as seizure or bone marrow suppression between two groups. Conclusion: Regarding to the PK parameters such as k, t1/2, CL and Vd, there were no statistically differences between EI and II group in critically ill patients treated with pip/tazo. Vd and t1/2 were larger, and CL was smaller. Serum concentrations fluctuated less during extending infusion, and this strategy was more likely to keep the target % fT>MIC. Although there were no significant differences in PD benefits in EI group vs. II group (1 hour) in this study compared to previous publications in which 30 min-infusion was administered in the II group.A 4-hr extended infusion regimen demonstrated a trend of better % fT>MIC and PTA at a higher MIC compared to 1-hr infusion. Therefore, extended infusion regimen may achieve a better PK-PD profile against pathogens with higher MIC when piperacillin/tazobactam was given as empirical treatment. Piperacillin/tazobactam 4 g Q8H with 4-hr IV infusion should be encouraged as empirical treatment for optimal PK-PD parameters considering daily dose, feasibility, and other clinical aspects. 3 g Q8H with 4-hr IV infusion can be considered when culture results with sensitive susceptibility are revealed. Further studies are needed for the impact of prolonged infusion on the treatment outcomes.