臺灣大學: 臨床藥學研究所高純琇卓怡慈Cho, Yi-TzuYi-TzuCho2013-04-162018-07-092013-04-162018-07-092010http://ntur.lib.ntu.edu.tw//handle/246246/257757研究背景 第二代抗精神病藥物因錐體外系統的副作用較少，而且對負向症狀、認知與情緒也有較佳的治療效果，目前已並列為精神分裂症的第一線治療藥物。過去研究發現，精神分裂症患者罹患代謝症候群（例如：糖尿病）和低骨密度的比例較高，除了精神分裂症本身會增加風險之外，使用抗精神病藥物也可能影響代謝症候群的發生和骨質健康，特別是第二代抗精神病藥物對血糖的影響逐漸被關注。目前國內尚無抗精神病藥物影響糖尿病和骨折發生之全國性藥物流行病學研究，本研究的進行，期望能提供臨床醫療人員在用藥選擇上的參考。 研究目的 比較精神分裂症新診斷者與非精神科族群發生糖尿病事件與骨折事件的差異。針對精神分裂症新診斷者，探討其使用第一代和各別第二代抗精神病藥物對於發生糖尿病事件、骨折事件和各部位骨折事件的影響。 研究方法 本研究為一回溯性世代分析，以健保資料庫2005年100萬人承保抽樣歸人檔（LHID2005）為研究材料，針對2001年至2008年間新診斷為精神分裂症的患者為研究對象，再依研究事件篩選糖尿病事件和骨折事件的研究族群，並從LHID2005找出1997年至2008年間無精神科相關診斷者，分別與兩事件研究族群的年齡及性別以1：4比例與進行配對，得到非精神科族群對照組。第一部份比較精神分裂症患者與非精神科族群，在糖尿病事件和骨折事件發生率和發生年齡的差異。第二部分根據是否使用抗精神病藥物將精神分裂症新診斷者分為用藥組和未用藥組，進行背景資料的描述性分析。第三部分以糖尿病事件、骨折事件以及各部位骨折事件為研究終點，利用time-dependent Cox’s proportional hazard model來進行存活分析，探討使用各種抗精神病藥物與用量對於終點事件風險的影響。 研究結果 針對糖尿病事件，符合研究條件的糖尿病事件研究族群（DM-Schizo group）共2649人，非精神科族群之糖尿病事件對照組（DM-Refere group）共9925人，第一部分研究結果，DM-Schizo group糖尿病事件發生率為7.1%，發生時的平均年齡為49.1歲，DM-Refere group發生率為3.7%，平均年齡為55.5歲，兩組間有顯著差異（p<0.0001）。第二部分研究結果，曾使用抗精神病藥物之用藥組共2414人，未用藥組235人，用藥組的年齡主要集中在11-50歲，未用藥組的分布則較為分散；用藥組在進入研究前後鋰鹽的使用率都較高，進入研究後使用valproate和罹患情感性精神病的比率也比較高。第三部分研究結果，校正可能之干擾因子後，在是否使用抗精神病藥物的模式中，第一代抗精神病藥物和amisulpride的hazard ratio (HR)分別為1.44（95% CI=1.06-1.95, p=0.0192）和2.37（95% CI=1.37-4.10, p=0.0020），顯示使用第一代抗精神病藥物或amisulpride都會增加糖尿病發生風險；抗精神病藥物的累積用量模式中，amisulpride、risperidone和ziprasidone的HR分別為1.005、1.002和1.007，顯示使用amisulpride、risperidone和ziprasidone的劑量越高，可能會增加糖尿病發生的風險。針對骨折事件，符合研究條件的骨折事件研究族群（FRA-Schizo group）共2501人，非精神科族群之骨折事件對照組（FRA-Refere group）共9009人，第一部分研究結果，FRA-Schizo group骨折事件發生率為8.0%，FRA-Refere group發生率為4.7 %，兩組間有顯著差異（p<0.0001），但發生事件時的平均年齡則沒有差異。第二部分研究結果，曾使用抗精神病藥物之用藥組共2284人，未用藥組217人，前者的年齡分布主要集中在11-50歲，後者的年齡分布則較為分散；用藥組在進入研究前後鋰鹽和BZD的使用率都較高，進入研究後使用抗憂鬱症和抗癲癇藥物，罹患情感性精神病和帕金森氏症的情況均較高。第三部分研究結果，校正可能之干擾因子後，是否使用抗精神病藥物和累積用量的模式中，均顯示不論使用何種抗精神病藥物都不會影響精神分裂症患者發生全部部位骨折的風險。然而在頸部和軀幹骨折事件的分析，顯示使用ziprasidone會增加頸部和軀幹骨折發生的風險，HR為5.44（95% CI=1.67-17.76, p=0.0051）。下肢骨折事件中，risperidone的使用者發生下肢骨折事件的風險反而比未使用者低，HR為0.58（95% CI=0.34-0.99, p=0.0473）。脊椎骨折事件中，顯示aripiprazole和zotepine的使用量越高，發生脊椎骨折事件的風險越高，兩者HR皆為1.002（95% CI均為1.000-1.003, p=0.0264和0.007）。 結論 精神分裂症新診斷者發生糖尿病和骨折事件的比率都高於其非精神科族群對照組，而且發生糖尿病事件的年齡明顯較低。不論在糖尿病或骨折事件研究族群當中，曾使用和未使用抗精神病藥物的精神分裂症患者，在年齡分布、用藥和疾病等背景資料上都有不同之處，例如用藥組的族群集中在青壯年，未用藥組的族群年齡分布則較為分散。與大部分國外研究不同，本研究結果顯示使用第一代抗精神病藥物和amisulpride的精神分裂症患者有較高發生糖尿病事件的風險；amisulpride、risperidone和ziprasidone的用量越高，風險也越高。全部部位骨折事件的分析中，抗精神病藥物均未達到顯著的影響。只有在各部位骨折事件中顯示，使用ziprasidone者發生頸部和軀幹骨折的風險較高；而aripiprazole和zotepine的用量越高，其發生脊椎骨折的風險可能越高。然而本研究卻發現使用risperidone發生對下肢骨折事有保護的效果。Background: Second-generation antipsychotics (SGAs), have been demonstrated efficacy in the treatment of schizophrenia with fewer extrapyramidal system side effects than first-generation antipsychotics (FGAs), and accordingly have also been approved as the first-line drugs in treating schizophrenia. However, recent studies found that schizophrenic patients have higher risk for diabetes mellitus (DM) and low bone mineral density, it is speculated that in addition to the disease of schizophrenia itself, the use of antipsychotics may also affect the blood sugar control and bone health, especially the second-generation antipsychotics. There is no national study examing the effects of antipsychotics on DM and fracture risks so far in Asia; therefore, we conducted the current study to provide a reference for clinical medical staffs in selecting drugs. Objectives: To compare the DM and fracture events rate between schizophrenic patients and non-psychiatric population; and to assess the DM and fracture risks between different antipsychotics in patients with newly diagnosed schizophrenia. Methods: This was a retrospective cohort study. We used a subset database (LHID2005) retrieved from National Health Insurance Research Database (NHIRD) to establish a cohort of patients with new onset of schizophrenia from 2001 to 2008, and screening for non-psychiatric population according to DM and fracture study groups. Descriptive analysis. We compared the DM and fracture event rate and age of onset between schizophrenic patients and non-psychiatric population. The DM and fracture study groups were divided into two groups (Schizo-A group and Schizo-N group), according to the exposure of antipsychotics, and then compared the demographic and clinical characteristics between the two groups. Survival analysis. Our study endpoints were DM, fracture at all sites and each part. We used time-dependent Cox’s proportional hazard model to evaluate the hazard ratio (HR) of all FGAs and individual SGAs for each study endpoints. Results: A total of 2649 new onset schizophrenic patients were included in the DM cohort (DM-Schizo group), and 9925 matched non-psychiatric reference were randomly selected (DM-Refere group). The crude DM event rates were 7.1% and 3.7%, and average DM onset ages were 49.1 y/o and 55.5 y/o in DM-Schizo group and DM-Refere group, respectively (p<0.0001). There were 2414 patients in DM-Schizo-A group and 235 in DM-Schizo-N group; DM-Schizo-A group mainly concentrated in 11-50 y/o, but the age distribution of DM-Schizo-N group is more dispersed. Compared with DM-Schizo-N group, patients in DM-Schizo-A group treated with more lithium and valproate, and had higher prevalence of episodic mood disorders. After adjusted with confounding factors, patients exposed to FGAs and amisulpride were more likely to have DM (FGAs HR=1.44; amisulpride HR=2.37). The higher accumulative dose of amisulpride, risperidone and ziprasidone were associated with a higher risk of DM (amisulpride HR=1.005; risperidone HR=1.002; ziprasidone HR=1.007). A total of 2501 new onset schizophrenic patients were included in the fracture cohort (FRA-Schizo group), and 9009 matched non-psychiatric reference were randomly selected (FRA-Refere group). The crude fracture event rates were 8.0% and 4.7 % in FRA-Schizo group and FRA-Refere group, respectively (p<0.0001). There were 2284 patients in FRA-Schizo-A group and 217 in FRA-Schizo-N group; FRA-Schizo-A group mainly concentrated in 11-50 y/o, but the age distribution of FRA-Schizo-N group is more dispersed. Compared with FRA-Schizo-N group, patients in FRA-Schizo-A group treated with more lithium, BZD, anti-depressants and anti-convulsants, and had higher prevalence of episodic mood disorders and Parkinson''s disease. After adjusted with confounding factors, patients exposed to any kinds of antipsychotics were not associated with the occurrance of fracture at all sites. However, patients exposed to ziprasidone was associated with a higher risk of neck/trunk fracture (HR=5.44, CI=1.67-17.76, p=0.0051). Risperidone was associated with a lower risk of lower limb fracture (HR=0.58, 95% CI=0.34-0.99, p=0.0473). The higher accumulative dose of aripiprazole and zotepine were associated with a higher risk of vertebral fracture (HR are both 1.002). Conclusions: The DM and fracture events rate are higher in schizophrenic patients than in non-psychiatric population, and the DM onset age was lower in schizophrenic patients. In both DM and fracture study population, there were some demographic and clinical characteristics differences between Schizo-A group and Schizo-N group. Different from previous study, the use of FGAs and amisulpride were found to be associated with significant increased risk of DM in our study; and the higher accumulative dose of amisulpride, risperidone and ziprasidone were also associated with a higher risk of DM. All antipsychotics seem not to be associated with the risk of fracture at all sites, but ziprasidone was associated with a higher risk of neck/trunk fracture, and the accumulative dose of aripiprazole and zotepine were also associated with a higher risk of vertebral fracture. However, risperidone seems to have a protective effect in lower limb fracture in our study.1330920 bytesapplication/pdfen-US精神分裂症抗精神病藥物糖尿病骨折schizophreniaantipsychoticsdiabetes mellitusfracture[SDGs]SDG3thesishttp://ntur.lib.ntu.edu.tw/bitstream/246246/257757/1/ntu-99-R97451001-1.pdf