Benzodiazepine Receptor Agonists於老年慢性阻塞性肺部疾病藥物流行病學研究

Abstract

研究背景 慢性阻塞性肺部疾病（chronic obstructive pulmonary disease，COPD）為常見於老年人之慢性疾病，且其普遍伴有失眠與焦慮問題， 往往需benzodiazepine receptor agonists（BZRAs）治療之，推估此類病人BZRAs之用量應不低。針對2002年，台灣地區老年人使用benzodiazepines（BZDs）之盛行率約43%。BZRAs除潛藏中樞神經副作用外，亦有報導指出使用過久可能會影響病人之呼吸功能，但目前國內外之藥物流行病學研究相關資料尚缺乏。 研究目的 本研究擬針對老年COPD病人使用BZRAs之處方型態及用藥安全做探討，並進一步瞭解國人BZRAs之使用與COPD控制良好與否間之關聯。 研究方法 利用全民健康保險研究資料庫2003至2007年之百萬歸人檔，以ICD-9-CM篩選2004年七月至2007年六月間首次因COPD診斷而住院之病人。以其出院後首次門診為納入時間點，追蹤至研究指標發生或最長追蹤6個月。BZRAs用藥安全監視之研究指標共分為COPD治療處方改變與因COPD惡化入院治療兩面向探討。於COPD治療處方改變之監測，再細分為COPD藥品治療種類增加、藥品等級升級或藥品劑量增加等三項研究指標；而因COPD惡化入院治療之監測，則再細分為因COPD發生急診或住院等兩項研究指標。針對納入之病人進行背景資料及處方型態之描述性分析，依據BZRAs使用之有無進行分組，其中暴露組又再次分為BZRAs有無連續使用超過28天兩組，並以time-dependent Cox’s proportional hazards model進行BZRAs用藥安全之研究指標分析。 研究結果 符合本研究條件之病人共1,356位，其中，在納入追蹤期間未曾使用BZRAs、使用BZRAs未連續28天及連續使用BZRAs至少28天者分別有902位（66.5%）、215位（15.9%）及239位（17.6%）。病人平均年齡為78.6歲（男78.1歲，女80.0歲），男佔73%。於住院期間，治療COPD藥品以全身性固醇類（25%）為主，於門診期間，則以甲基茶鹼類（35%）為主；然而不論在住院或門診期間，BZRAs皆以中效型BZDs（50%）為主要用藥。 在BZRAs用藥安全監視之研究指標探討中，發現COPD病人使用中效型BZDs日劑量愈高，COPD治療藥品種類增加之風險提高（風險比，1.78；95%信賴區間，1.18-2.68；P值=0.0056），且COPD治療藥品等級升級之風險稍高（風險比，1.56；95%信賴區間，0.97-2.52；P值=0.0690）。長效型BZDs高日劑量使用時，容易促使COPD治療藥品劑量增加之事件（風險比，2.52；95%信賴區間，1.17-5.42；P值=0.0185）；使用BZRAs對於發生急診治療事件未達顯著意義；服用長效型BZDs日劑量愈高時，提升住院治療事件之發生（風險比，3.42；95%信賴區間，1.67-7.01；P值=0.0008）；服用中效型BZDs日劑量愈高，有較高發生COPD處方改變之風險（風險比，1.52；95%信賴區間，1.14-2.02；P值=0.0039）；使用長效型BZDs日劑量愈高時，增加因COPD入院治療之風險（風險比，3.74；95%信賴區間，1.95-7.20；P值< 0.0001）。 結論 在老年COPD病人中，BZRAs連續暴露達28日以上者有17.6%。使用中效型或長效型BZDs時，高日劑量者所面臨之COPD處方調整或因COPD入院治療事件之風險較高。因此，針對老年COPD病人選用中效型及長效型BZDs日劑量時，須特別留意。 為更清楚釐清BZRAs與呼吸功能之關聯性，建議未來研究可依據病人使用之COPD藥品類別進行分組（例如：使用吸入性固醇類藥品與長效型β2致效劑的複方藥者vs使用單方藥者），或將BZRAs依其使用劑量次分群，進一步分析以得更周詳之用藥安全資訊。

Background Chronic obstructive pulmonary disease (COPD) is a prevalent chronic illness in the elderly and usually accompanied with insomnia and anxiety disorders. Benzodiazepine receptor agonists (BZRAs) are widely used as hypnotics and anxiolytics. Therefore, it is common for elderly patients with COPD to be exposed to the BZRAs. Previous studies in 2002 showed that the prevalence of benzodiazepines (BZDs) used in the elderly was approximately 43% in Taiwan. Prolonged BZRAs may lead to central nervous system damage as well as respiratory deterioration function. To our knowledge, however, pharmacoepidemiologic studies on this issue is quite limited so far. Objective The aim of the study is to investigate the prescription patterns of BZRAs in the elderly with COPD; and further, to determine the association between BZRAs usage and the control of COPD. Methods A retrospective study was performed using a cohort of 1,000,000 randomly selected subjects from the National Health Insurance Research Database between 2003 and 2007. Subjects, hospitalized under the diagnosis of COPD during July 2004 and June 2007, were enrolled by their initial COPD admission during this 3-year period. Following discharge from that admission, the first ambulatory visit was selected as the index date for the study. All patients were followed up to study endpoint detected or to a maximal of 6 months. The study endpoints of pharmacovigilance of BZRAs included alteration of COPD treatment and exacerbation of COPD required hospitalization. Among the alteration of COPD treatment, increasing number of COPD agents, upgrade of COPD agents, and increasing dose of COPD agents were include. Among the exacerbation of COPD required hospitalization, both admissions and emergency visits with a COPD diagnosis were included. Patients’ demographics and the prescription patterns were analyzed. They were categorized as ever use and never use of BZRAs. Among ever use, patients were grouped according to prescribed duration with 28 consecutive days as cut-point. Besides, the study endpoints were analyzed by time-dependent Cox’s proportional hazards model survival analysis. Results A total of 1,356 eligible patients were enrolled in this study. Among these patients, 902 (66.5%) patients never used BZRAs, 215 (15.9%) patients had ever received BZRAs for less than 28 consecutive days, and 239 (17.6%) patients had received BZRAs for more than 28 consecutive days. Patients had a mean age of 78.6 years (M 78.1, F 80.0), and males were 73%. At admission, systemic steroids (25%) were the most frequently prescribed COPD agents. At outpatient clinics, methylxanthines (35%) were the most prevalent. Among BZRAs, intermediate-acting benzodiazepines (IABZDs) (50%) were the most frequently prescribed at admission and outpatient clinics. Our results showed that the risk of increasing number of COPD agents and upgrade of COPD agents were higher in patients receiving higher daily dose of IABZDs (hazard ratio (HR), 1.78; 95% CI, 1.18-2.68; p value for trend = 0.0056, and HR, 1.56; 95% CI, 0.97-2.52; p value for trend = 0.0690). The risk of increasing dose of COPD agents was significantly higher in patients receiving higher daily dose of long-acting benzodiazepines (LABZDs) (HR, 2.52; 95% CI, 1.17-5.42; p value for trend = 0.0185). There was no significant effect of BZRAs on COPD emergency visits. The risk of admissions was significantly higher in patients receiving higher daily dose of LABZDs (HR, 3.42; 95% CI, 1.67-7.01; p value for trend = 0.0008). The risk of alteration COPD treatment was significant higher in patients with higher daily dose of IABZDs. (HR, 1.52; 95% CI, 1.14-2.02; p value for trend = 0.0039). The risk of exacerbation of COPD required hospitalization was significantly higher in patients receiving higher daily dose of LABZDs (HR, 3.74; 95% CI, 1.95-7.20; p value for trend < 0.0001). Conclusions Among elderly COPD patients who had received BZRAs for more than 28 consecutive days were about 17.6%. Higher daily dose of IABZDs or LABZDs were with significant higher risks for adjustment of COPD treatment and exacerbation of COPD required hospitalization Therefore, the daily dose of IABZDs and LABZDs in the elderly COPD patients are worth noting. In the future, in order to clarify more clearly the association between BZRAs and the breathing function, categorizing patients according to their COPD medications [e.g.: combination inhaled corticosteroids (ICS) and long-acting β2 agonists (LABA) vs non-combination] and dose of BZRAs are suggested. It is necessary to conduct more research to investigate for acquiring detailed medication safety information.