Amiodarone相關肝損傷自主通報案例分析：2000-2014年

Abstract

研究背景 藥物引致肝損傷的監測之於台灣等肝炎盛行區域格外重要。文獻顯示amiodarone在藥物引致肝損傷致死的排行中居前十名內，或許與其獨特的藥效學及藥動學特性有關，值得醫療專業人員多加留意其潛在之不良反應。 研究目標 本研究擬以國內藥物不良反應自主通報資料探討：（一）amiodarone相關肝損傷之通報案件的趨勢及其表徵；（二）amiodarone相關肝損傷之危險因子。 研究方法 收集並整理2000年迄2014年藥物不良反應通報中，使用amiodarone引致肝損傷（依MedDRA分類）案例，內容包括病人基本資料、潛伏期、不良反應結果、生化檢驗值む例如alanine aminotransferase (ALT)め、共病症及併用藥物等，敘述性討論案例資料並以統計軟體進行迴歸分析。 研究結果及討論 在十五年期間之肝損傷相關通報案例共3887件，其中由amiodarone引致的有55件（1.4%）。平均年齡為73.8 ± 11.5歲（mean ±SD）；男女比為3：2；潛伏期小於7天的案例有35件（63.6%）；平均每日劑量為582.3 ± 565.7 mg；大部分案例的受損類型為肝細胞型（81.0%）。單變項分析中，以不良反應結果嚴重度分組並沒有發現任何差異，但以潛伏期分組則發現急性組病人的直接膽紅素值（3.27 vs. 0.47 mg/dL; p = 0.040）、總膽紅素值（3.40 vs. 0.83 mg/dL; p = 0.048）及平均每日劑量（425.6 vs. 115.7 mg/dL; p < 0.001）的中位數都比慢性組高；多變項迴歸分析雖然皆未達到顯著，但可發現一些警訊，可能惡化預後的危險因子包含BSA校正後總劑量（OR = 1.00001; p = 0.413）、ALT最高值（OR = 1.0006; p = 0.168）、具肝病史者（OR = 3.41; 95% CI: 0.35–33.11; p = 0.290）及併用具肝損傷潛性藥品（OR = 1.70; 95% CI: 0.49–5.86; p = 0.403）。 結論 本研究呈現amiodarone相關肝損傷之長期通報狀況，雖因通報數有限及相關資料缺漏而難以提供明確定論，然而所得之初步警訊如劑量、膽紅素值等值得未來再進行更深入探討，亦期望藉此研究討論通報案件量及通報品質提升之意義。

Background Vigilance of drug-induced liver injury (DILI) is especially important in hepatitis-prevalent areas such as Taiwan. The special pharmacodynamic and pharmacokinetic characteristics of amiodarone make it among the top ten culprit drugs for mortality due to DILI. Objectives The study aims to: (1) describe reporting tendency and pattern; and (2) delineate possible risk factors related to amiodarone-associated liver injury by using data from the Adverse Drug Reaction Reporting System in Taiwan. Methods Cases of amiodarone-associated liver injury (categorized by MedDRA) reported during 2000~2014 were included. Data on baseline characteristics, latency, adverse drug reaction outcomes, biochemistries [such as alanine aminotransferase (ALT)], underlying diseases, and concomitant drugs were all collected and evaluated by descriptive analysis and multiple logistic regression using statistical software. Results and Discussion Adverse drug reaction (ADR) reports related to liver injury were 3887 in total; 55 cases (1.4%) were associated with amiodarone. The latter cohort had a mean age of 73.8 years old; male to female ratio was 3:2; latency of 35 cases (63.6%) were less than 7 days; average daily dose was 582.3 ± 565.7 mg; hepatocellular type was the most encountered type (81.0%). There was no statistically significant difference between 'severe' (e.g. death, prolonged hospitalization, disability) and 'non-severe' groups by univariate analysis; however, the medians of direct bilirubin (3.27 vs. 0.47 mg/dL; p = 0.040), total bilirubin (3.40 vs. 0.83 mg/dL; p = 0.048), and average daily dose (425.6 vs. 115.7 mg/dL; p < 0.001) were higher in the 'acute group' (latency ≤ 90 days). Further multivariate regression analysis revealed interesting signals for outcome prediction, including cumulative dose/body surface area (OR = 1.00001; p = 0.413), peak ALT level (OR = 1.0006; p = 0.168), preexisting liver disease (OR = 3.41; p = 0.290), and co-medication with drugs of DILI potential (OR = 1.70; p = 0.403), even though statistically significance were not reached. Conclusions Case numbers in the study might be too small to draw conclusive remarks, especially by using the often under-reported and sometimes incomplete nature of the data source. Certain predictors, such as dose and bilirubin, identified in the study deserve to be explored further. Discussion for ways of enhancing quality and quantity of ADR reporting will also be provided.