李子奇Lee, Tzu-Chi陳萱Chen, Hsuan2019-08-282020-01-302019-08-282019http://etds.lib.ntnu.edu.tw/cgi-bin/gs32/gsweb.cgi?o=dstdcdr&s=id=%22G060505015E%22.&%22.id.&amp;http://rportal.lib.ntnu.edu.tw:80/handle/20.500.12235/87484背景 全世界的人口正快速地老化，據估計失智症影響全球約5000萬人口；然而，至今卻沒有一種治療方式能夠治癒失智症或改變其病程。故本研究主要的目的是篩選出對降低失智症風險有潛在效果的藥物。 方法 我們使用臺灣全民健康保險資料庫進行一項以全人口為基礎的病例對照研究。病例組包含13,459位年齡介於50至80歲、領有重大傷病卡、失智症初診日介於2005年1月1日至2013年12月31日的病人。以病例組的性別、年齡及投保薪資配對得到13,459位非失智症的病人做為對照組。排除失智症初診日前一年期間內所使用的藥物後，將前兩年至前5年間使用的藥物ATC代碼(Anatomical Therapeutic Chemical code)做分析。利用條件式邏輯斯迴歸及錯誤發現率(false discovery rate，FDR)篩選藥物。 結果 從1,353種藥物中，本研究篩選出811種藥物在病例組及對照組中的使用者皆超過30人。單變項分析結果顯示有5種Odds-ratio小於0.8的藥物分別與失智症風險呈負相關，446種Odds-ratio大於1.2的藥物與失智症風險呈正相關。單變項分析後，Odds-ratio小於0.8的5種藥物及Odds-ratio大於4的11種藥物分別進行多變項分析。多變項分析控制人口學變項(年齡及工作)、失智症相關疾病(糖尿病、高血壓、心血管疾病、憂鬱症、焦慮症及聽力損失)、住院天數及門診次數後，結果顯示單變項分析中顯著且Odds Ratio小於0.8的5種藥物中仍有4種藥物與失智症風險呈負相關。這四種藥物分別為phenoxymethyl penicillin (Odds-ratio = 0.70; 95% CI: 0.58-0.85), fluorouracil (Odds-ratio = 0.56; 95% CI: 0.42-0.75), ethenzamide (Odds-ratio = 0.68; 95% CI: 0.52-0.90), and butamirate (Odds-ratio = 0.73; 95% CI: 0.64-0.85)。此外，單變項分析中11種與失智症風險有正相關的藥物，在多變項分析中仍然有9種有顯著正相關。這9種藥物分別為trihexyphenidyl (odds ratio, 1.62; 95% CI, 1.33–1.97), guetiapine (odds ratio, 16.33; 95% CI, 13.91–19.17), risperidone (odds ratio, 15.74; 95% CI, 12.57–19.71), citalopram (odds ratio, 1.90; 95% CI, 1.46–2.47), levodopa (odds ratio, 2.37; 95% CI, 2.00–2.81), haloperidol (odds ratio, 1.55; 95% CI, 1.33–1.81), sertraline (odds ratio, 3.47; 95% CI, 2.90–4.16), escitalopram (odds ratio, 2.53; 95% CI, 2.00–3.21), and venlafaxine (odds ratio, 1.94; 95% CI, 1.44–2.61)。 結論 本研究發現4種藥物(penicillin、ethenzamide、butamirate及pseudoephedrine)值得針對失智症的初級預防或治療做進一步的研究。此外，本研究篩選出的神經系統藥物與失智症風險有正相關，可能是藥物導致或藥物治療的疾病為失智症的前驅期疾病所致。Background: Populations worldwide are aging rapidly. Currently, Dementia is estimated to affect approximately 50 million people worldwide. However, no drugs cure dementia or alter its progressive course. The primary aim of this study was to screen drugs that might potentially contribute to reduce the risk of dementia. Methods: We used data from Taiwan’s National Health Insurance Research Database to conduct a total population-based case–control study. The case group comprised 13,459 dementia patients, aged 50–80 years, who had a catastrophic illness certificate and obtained their first diagnosis of dementia between January 1, 2005, and December 31, 2013. These cases were compared with 13,459 sex–, age–, and insurance premium–matched controls. Drug use 2–5 years before the first diagnosis date was analyzed according to the Anatomical Therapeutic Chemical code. Conditional logistic regression models and false discovery rate were used for statistical analysis. Results: Of 1,353 drugs, 811 were identified with ≥30 users among both case and control groups. Univariate analysis revealed 5 drugs with OR< 0.8 and 446 with OR > 1.2 that were negatively and positively associated with dementia risk, respectively. Five drugs with OR < 0.8 and eleven drugs with OR > 4 selected from univariate analysis were conducted multivariate analysis, respectively. The multivariate analysis was conducted after controlling for demographic variables (age and job), dementia-related diseases (diabetes, hypertension, cerebrovascular disease, depression, anxiety, and hearing loss), hospital days, and outpatient visits. The results showed that of the five negative drugs (with OR<0.8) in univariate analysis, four were also negatively associated with the risk of dementia: phenoxymethylpenicillin (Odds ratio, 0.70; 95% confidence interval [CI], 0.58–0.85), 5-fluorouracil (Odds ratio, 0.56; 95% CI, 0.42–0.75), ethenzamide (Odds ratio, 0.68; 95% CI, 0.52–0.90), and butamirate (Odds ratio, 0.73; 95% CI, 0.63–0.84). In addition, of the eleven positively drugs (with OR>4) in univariate analysis, nine were found to be positively associated with the risk of dementia: trihexyphenidyl (Odds ratio, 1.62; 95% CI, 1.33–1.97), guetiapine (Odds ratio, 16.33; 95% CI, 13.91–19.17), risperidone (Odds ratio, 15.74; 95% CI, 12.57–19.71), citalopram (Odds ratio, 1.90; 95% CI, 1.46–2.47), levodopa (Odds ratio, 2.37; 95% CI, 2.00–2.81), haloperidol (Odds ratio, 1.55; 95% CI, 1.33–1.81), sertraline (Odds ratio, 3.47; 95% CI, 2.90–4.16), escitalopram (Odds ratio, 2.53; 95% CI, 2.00–3.21), and venlafaxine (Odds ratio, 1.94; 95% CI, 1.44–2.61). Conclusion: Four drugs, penicillin, ethenzamide, butamirate, and pseudoephedrine, warrant further research for dementia prevention or treatment. The nervous system drugs screened in this study may be caused by drug-induced or -treated prodromal diseases of dementia.失智症藥物全人口病例對照研究dementiadrugspopulation-wide case–control study