Voriconazole 療劑監測於侵入性黴菌感染之成本效果分析

Abstract

背景： Voriconazole為新一代三唑類抗黴菌藥品，臨床上為治療侵入性麴菌感染的首選用藥。由於非線性的藥品動態學特性以及其他影響因子，造成個體間或個體內的血中濃度差異性大。研究皆顯示voriconazole血中濃度與臨床療效、藥品不良反應相關，且可改善臨床黴菌治療效果。在其他疾病的藥品療劑監測成本效果分析中，發現療劑監測可改善臨床治療效果等，但也可能造成臨床醫療花費負擔增加。目前尚無voriconazole 療劑監測的成本效果分析，本研究於國立臺灣大學醫學院附設醫院回溯性收案使用voriconazole的病人並執行voriconazole療劑監測的成本效果分析。 研究目的： 以醫院觀點分析執行voriconazole療劑監測前後，對於黴菌感染的治療反應影響、藥品不良反應、醫療花費的影響並且執行成本效果分析，提供醫療決策者將來是否納入voriconazole療劑監測作為醫院常規檢測項目以及實行方式之參考依據。 研究方法： 本研究回溯性收案於2010年1月1日至2010年12月31日在國立臺灣大學醫學院附設醫院新使用voriconazole治療侵入性黴菌感染並接受一次以上voriconazole療劑監測的病人作為療劑監測組 (Therapeutic drug monitoring ,TDM group)。以2008年1月1日至2008年12月31日新使用voriconazole但未接受療劑監測者為對照組 (Historical control , HC group)。自病人第一天用藥開始，觀察90天內病人voriconazole及其他抗黴菌藥品處方型態、藥物不良反應及處理方式、醫療直接成本。並請感染科醫師評估適應症與治療反應。以研究觀察期結束時的病人黴菌治療反應作為主要研究終點，以voriconazole治療天數、因發生不良反應導致病人停藥作為次要研究終點，並執行兩組之間的成本效果分析與敏感度分析。統計方法以卡方檢定、T檢定、Wilcoxon signed rank test、log rank test進行組間比較。 研究結果： 本研究HC組共有69位病人，TDM組為71位病人。兩組病人的年齡、性別、潛在疾病類似，皆以血液疾病為主，TDM組病人為neutropenia的比例與HC組相似 (45.1%與40.6%)，研究期間發生neutropenia則TDM組比HC組多22%。兩組病人接受骨髓移植的比例相近但TDM組接受周邊血液幹細胞移植比例較高 (4.4% 與17%)。在慢性肝臟疾病方面，TDM組有B型肝炎者比例也多於HC組。病人大多為因為確定或極有可能的肺部黴菌感染而開始用藥。 在持續使用voriconazole上，TDM組的持續使用率較HC組高出9.1 %且具有統計意義（p=0.04），voriconazole使用天數也顯著多於HC組。在黴菌感染治療反應方面，治療成功比例在TDM組與HC組分別為 45%與58%，治療無進展或疾病惡化為29.6%與33.3 %，但都不具有統計顯著意義。在死亡率則以TDM組較高 (22.5%與7.3%)。在藥物不良反應方面，TDM與HC兩組分別發生60筆與12筆與voriconazole使用相關性較高的不良反應，其中有8.3% (5/60)和50% (6/12)的不良反應導致病人停藥，具有統計顯著意義。若是因為不良反應而停止用藥的病人，則兩組分別為7% (5/71)與13% (9/69)，但不具有統計上意義。 醫療成本部分，TDM組總醫療花費中位數為734,014元，HC組為428,062元，抗黴菌藥品中位數分別為262,548與120,744元，兩組醫療成本差別主要發生在觀察的第一個月。 執行TDM後，若欲增加病人首選之抗黴菌藥物治療長度一天，依額外總醫療成本、抗黴菌藥品成本計算的ICER值分別為8,741元和4,052元。當要減少1% 病人因為藥物不良反應而導致停藥時，依額外總醫療成本與抗黴菌藥品成本計算的ICER值分別為50,992元和23,634元。敏感度分析顯示voriconazole成本可能影響ICER值的改變。 結論： Voriconazole 療劑監測可減少病人因發生藥物不良反應而導致停藥，並持續首選藥物的治療且增長病患接受藥物治療的長度，但需要額外的醫療成本支出。

Background： Voriconazole, a second generation triazole, is the drug of choice in the treatment of invasive aspergillosis (IA). Therapeutic drug monitoring (TDM) of voriconazole has been recognized as an effective way to improve clinical outcomes but may increase economic burden in the healthcare system. Currently, the impact of TDM on the pharmacoeconomic aspect has not been studied. Objective： The objective of this study is to evaluate the cost-effectiveness of voriconazole TDM from the hospital perspective in a 2500-bed medical center in Taiwan. Methods： New voriconazole users who underwent TDM in 2010 were enrolled in the TDM group. The historical control (HC) group consisted of new users in 2008 when pharmacists-managed TDM service was not established. Treatment of IA and clinical outcomes were collected from medical records. Treatment response of invasive fungal infections on 90th day or at the end of follow-up period was the primary endpoint. Adverse drug reactions which resulted in discontinuation of voriconazole and treatment duration of voriconazole were secondary endpoints. Direct medical costs included information from insurance reimbursement profiles, patient cash payment documents, and TDM costs. χ2 test and Fisher’s exact test were used for categorical variable analysis. T-test or Wilcoxon rank, wilcoxon signed rank test, log rank test were used for continuous variable analysis. Results The mean age, gender and underlying diseases were similar between 71 patients in the TDM group and 69 patients in the HC group. However, more patients in the TDM group had neutropenia and received peripheral blood stem cell transplantation. Most people used voriconazole for possible fungal infection in lungs. Treatment success rate was 45% vs. 58% in the TDM group and HC group, respectively. The percentage of stable disease or progression were 29.6% and 33.3% in 2 groups without statistical significance. More people died with fungal infection in the TDM group (22.5 %) compared to HC group (7.3%) with statistical significance. On the aspect of the continuation rate of voriconazole, TDM group was 9.1% higher than HC group with statistical significance (p=0.04). Treatment duration of voriconazole was also 35 days longer in TDM group compared with HC group. There were 60 and 12 drug adverse reactions in TDM and HC group and caused 8.3% (5/60) and 50% (6/12) discontinuation of voriconazole, respectively (p<0.001). The median of total medical cost and antifungal expenses were higher in TDM group compared with HC group, with NTD 734,014 v.s.428,062 in total cost and 262,548 v.s.120,744 in antifungal expense. The major difference of total medical costs between 2 groups occurred in the first observation month. The daily antifungal costs, however, were similar between two groups. The incremental cost effectiveness ratios (ICERs) were NTD 8,741 (total medical cost) and NTD 4,052 (antifungal expense) per voriconazole treatment day gained. ICERs were NTD 50,992 (total medical cost) and NTD 23,634 (antifungal expense) per 1 % patients who were prevented from discontinuation of voriconazole treatment due to adverse drug reaction. Sensitivity test analysis showed that ICER was sensitive to variation of voriconazole expenses. Conclusion： Voriconazole TDM did not affect the treatment outcomes in this study. It may reduce the discontinuation rate of voriconazole due to adverse drug reaction with the tendency of increased medical cost.