Vancomycin血中濃度對時間曲線下面積（AUC）與谷濃度療劑監測之比較

Comparison of Vancomycin Therapeutic Drug Monitoring Guided by Area under the Concentration-Time Curve (AUC) and Trough Concentration

徐曼晴(Man-Ching Hsu)

指導教授 ： 林淑文

國立臺灣大學/醫學院/臨床藥學研究所/碩士(2021年)

https://doi.org/10.6342/NTU202104235

摘要

背景
Vancomycin是一種糖肽類抗生素，為抗藥性金黃色葡萄球菌（methicillin-resistant Staphylococcus aureus, MRSA）等多重抗藥性革蘭氏陽性菌感染症的第一線治療藥物。過去在嚴重MRSA感染病人進行vancomycin的療劑監測（therapeutic drug monitoring，TDM）目標為trough濃度15-20 mg/L，但近年來研究發現trough濃度15-20 mg/L為急性腎損傷（acute kidney injury，AKI）之獨立風險因子。2020年vancomycin治療監測指引將建議的療劑監測指標由谷濃度改為24小時血中濃度對時間曲線下面積（area under the curve, AUC）除以最小抑菌濃度（minimum inhibitory concentration，MIC）之比值，期望提高安全性並確保療效足夠，並推薦以貝氏軟體估算AUC。
目前直接比較AUC和trough做為監測指標之研究大多樣本數較小，並且著重於安全性的比較，對於療效的評估較少。AUC的最佳範圍也尚未有定論。2020年8月臺灣大學醫學院附設醫院（簡稱臺大醫院）開始進行vancomycin之AUC監測，因此本研究分析AUC和trough濃度兩種TDM方式之臨床結果。
研究目的
主要研究目的為比較以AUC和trough濃度進行TDM之安全性與療效差異。次要目的包括尋找與安全性和療效相關的AUC及trough濃度切點，以及比較臺灣貝氏軟體tdm for R與梯形面積法所估算出的AUC差異。
研究方法
本研究為回溯性研究，納入18歲以上於臺大醫院使用vancomycin至少3天並在開始用藥後7天內有TDM之病人。用藥期間在2018年8月1日至2019年4月30日間的病人為trough組，2020年8月1日至2021年4月30日間為AUC組。排除接受腎臟替代療法、外院轉入並於當天開始使用vancomycin、AUC組抽血時間錯誤及血中肌肝酸濃度缺失者。
資料來源包括臺大醫療體系醫療整合資料庫、電子病歷系統與紙本TDM紀錄，蒐集病人基本資料、使用vancomycin前後之實驗室檢驗值、生命徵象和併用藥物等。
主要研究結果為AKI發生率，依照KDIGO criteria判斷。療效結果包括感染MRSA病人之臨床治療結果、微生物學治療結果、30日和90日全原因死亡率和90日再住院率。
統計方法以Student’ s t test或Wilcoxon rank-sum Test比較連續變項，以Chi-square test或Fisher’s exact test比較類別變項。採用傾向分數配對（propensity score matching，PS matching）來降低兩組病人基本特徵的差異。以多變項羅吉斯迴歸分析與結果相關之因子。利用ROC曲線分析尋找與結果相關之AUC和trough濃度切點。
研究結果
本研究共納入691位使用vancomycin並接受療劑監測之病人，AUC組別共177人，trough組別514人。經配對後AKI發生率在AUC組為21.0%，在trough組為21.7%。以羅吉斯迴歸分析校正配對後兩組間仍有差異之因子後發現使用AUC或trough進行療劑監測對AKI風險無顯著影響。在TDM範圍符合指引建議、SOFA score > 2和AKI前有接受TDM的次族群中同樣未觀察到療劑監測方式對於AKI風險之影響。與AKI相關之顯著風險因子包括女性、較高之SOFA score、Charlson comorbidity index > 2、baseline收治於ICU和較高的trough濃度。
在療效結果方面，排除提早停用vancomcyin者後本研究共有90位具有MRSA培養證據的病人納入療效分析。經過配對後發現使用AUC或trough進行療劑監測對於臨床治療失敗風險、微生物治療失敗風險、30日及90日全原因死亡率和90日再住院率都沒有顯著影響。與臨床治療失敗相關的顯著風險因子為較高之SOFA score和較大的vancomycin MIC。沒有找到與微生物學治療失敗相關的風險因子。30日死亡率和90日死亡率皆與較高之SOFA score和免疫不全有關。
由ROC曲線分析找到的trough濃度適當範圍約為13-17 mg/L。與AKI相關之AUC切點為446.9 mg•h/L，與療效相關的切點約為500 mg•h/L但模型預測不理想。
以兩點濃度利用梯形面積法計算出之AUC做為基準與貝氏軟體tdm for R利用單點trough濃度估算之AUC比較，發現兩者估算結果相近，accuracy中位數為0.9（IQR：0.9-1.0），bias中位數為8.5（IQR：4.3-14.0）。羅吉斯迴歸分析發現年齡較小者和女性較容易出現bias > 20% 的情形。
結論
本研究觀察到以vancomycin AUC和trough濃度做為TDM指標之安全性和療效結果相近。未來可進行前瞻性研究進一步減少潛在干擾因子，並收入更多樣本數，期望能確認適當的AUC及AUC/MIC範圍。

關鍵字

Vancomycin ； 療劑監測 ； AUC ； 谷濃度 ； 急性腎損傷 ； 貝氏

並列摘要

Background
Vancomycin, a glycopeptide antimicrobial agent, is the first-line therapy for infections of methicillin-resistant Staphylococcus aureus (MRSA) and other multi-resistant Gram-positive bacteria. The therapeutic drug monitoring (TDM) target of vancomycin was a trough level of 15-20 mg/L in patients with serious MRSA infection. Recent studies showed that a trough level of 15-20 mg/L was an independent risk factor of acute kidney injury (AKI). 2020 version of vancomycin TDM guideline recommended to monitor the ratio of 24-h area under the concentration-time curve (AUC) to minimum inhibitory concentration (MIC) instead of trough concentration to achieve efficacy while improving safety. The guideline also recommended to estimate AUC by Bayesian software.
Most studies comparing AUC-guided and trough-guided TDM had small sample sizes and focused on safety outcome. Few studies evaluated efficacy outcomes. There is no consensus on the optimal range of AUC.
National Taiwan University Hospital (NTUH) has implemented AUC monitoring of vancomycin since August, 2020. Therefore, we analyzed the clinical outcomes of patients receiving AUC-guided and trough-guided TDM.
Objective
The primary objective was to compare safety and effectiveness between AUC-guided and trough-guided TDM groups. Secondary objectives included identification of cut-off points of AUC and trough level associated with safety and effectiveness and to compare AUC estimated by Bayesian software “tdm for R” and trapezoidal rule.
Methods
This retrospective study enrolled patients more than 18 years old, receiving vancomycin for at least three days at NTUH and TDM within seven days after starting vancomycin therapy. Patients during 2018/8/1-2019/4/30 were assigned to the trough group and 2020/8/1-2021/4/30 to the AUC group. Patients who received renal replacement therapy, transferred from other hospitals and started vancomycin upon arrival, blood drawn at wrong time in the AUC group or without serum creatinine records were excluded.
Patients’ basic characteristics, lab data, vital signs and concomitant drugs of were collected from the Integrated Medical Database, National Taiwan University Hospital (NTUH-iMD), electronic medical records and paper records of TDM.
The primary outcome was AKI incidence based on KDIGO criteria. Effectiveness analyses included clinical outcome, microbiological outcome, 30-day and 90-day all-cause mortality and 90-day rehospitalization in patients with MRSA infections.
Student’s t test or Wilcoxon rank-sum test was used to compare continuous variables and categorical variables were compared by Chi-square test or Fisher’s exact test. Propensity score matching was used to reduce baseline difference between two groups. Factors associated to safety and effectiveness were analyzed with multivariable logistic regression.
Results
A total of 691 patients who received vancomycin and TDM were included with 177 in AUC group and 514 in trough group. AKI incidence in the matched cohort was 21.0% and 21.7%, respectively. The findings showed that using either AUC or trough level as TDM target did not affect risk of AKI significantly. TDM methods did not affect AKI risk in subgroup analysis in patients with normal TDM range, patients with SOFA score > 2 and patients who had TDM before AKI occurred. Factors significantly associated with AKI risk were female, higher SOFA score, Charlson comorbidity index > 2, baseline ICU stay and higher trough level.
After excluding patients with early discontinuation of vancomycin therapy, a total of 90 patients with positive MRSA culture results were eligible for effectiveness analysis. In the matched cohort, using either AUC or trough level as TDM target did not significantly affect clinical failure, microbiological failure, 30-day all-cause mortality, 90-day all-cause mortality or 90-day rehospitalization. Clinical failure was associated with higher SOFA score and elevated vancomycin MIC. We did not identify any significant risk factor associated with microbiological failure. 30-day and 90-day mortality were associated with higher SOFA score and immunocompromised state.
ROC curve analysis demonstrated the trough level of 13-17 mg/L was associated with favorable outcomes. The cut-off point of AUC associated with AKI was 446.9 mg•h/L. It was around 500 mg•h/L associated with treatment outcome, but the model prediction was not ideal.
Good concordance achieved when we compared AUC estimated by Bayesian software “tdm for R” using one trough level with AUC calculated by trapezoidal rule using 2 levels. The median of accuracy was 0.9 (IQR：0.9-1.0) and bias was 8.5 (IQR：4.3-14.0). Logistic regression analysis discovered that younger and female patients were associated with bias > 20%.
Conclusions
We observed that using AUC and trough level as TDM target for vancomycin resulted in similar outcomes. Future studies should be considered to conduct prospectively to reduce potential confounders. A larger population may also be necessary in order to explore the optimal AUC and AUC/MIC range of vancomycin.

並列關鍵字

Vancomycin ； Therapeutic drug monitoring ； AUC ； Trough ； Acute kidney injury ； Bayesian

參考文獻

1.Craig, W.A., Basic pharmacodynamics of antibacterials with clinical applications to the use of β-lactams, glycopeptides, and linezolid. Infectious Disease Clinics of North America, 2003. 17(3): p. 479-501.

2.Drusano, G.L., Antimicrobial pharmacodynamics: critical interactions of 'bug and drug'. Nature Reviews Microbiology, 2004. 2(4): p. 289-300.

3.Liu, C., et al., Clinical Practice Guidelines by the Infectious Diseases Society of America for the Treatment of Methicillin-Resistant Staphylococcus aureus Infections in Adults and Children. Clinical Infectious Diseases, 2011. 52(3): p. e18-e55.

4.Rybak, M.J., The Pharmacokinetic and Pharmacodynamic Properties of Vancomycin. Clinical Infectious Diseases, 2006. 42(Supplement_1): p. S35-S39.

5.Kullar, R., et al., Impact of Vancomycin Exposure on Outcomes in Patients With Methicillin-Resistant Staphylococcus aureus Bacteremia: Support for Consensus Guidelines Suggested Targets. Clinical Infectious Diseases, 2011. 52(8): p. 975-981.